IN VIVO AND IN VITRO NEUROPROTECTIVE EFFECTS OF THE PENTANE SOLUBLE COMPOUNDS FROM LEPIDIUM MEYENII (MACA) By

ALEJANDRO PINO-FIGUEROA A DISSERTATION Presented to the Graduate Faculty of the Massachusetts College of Pharmacy and Health Sciences In fulfillment of the requirements for the degree of Doctor of Philosophy Under the supervision of Dr. Timothy J. Maher September 2009 Dissertation approved by the Advisory Committee:

Timothy J. Maher, Ph.D., Research Advisor

Mark Bfchlke, Ph.D., Committee Member

Seth Finklestein, M.D., Committee Member Charles Keltgy, Ph.D.,

arbara LeDuc, Ph.D., Assistant Dean of Graduate Studies

UMI Number: 3385458

All rights reserved INFORMATION TO ALL USERS The quality of this reproduction is dependent upon the quality of the copy submitted. In the unlikely event that the author did not send a complete manuscript and there are missing pages, these will be noted. Also, if material had to be removed, a note will indicate the deletion.

UMT
Dissertation Publishing

UMI 3385458 Copyright 2009 by ProQuest LLC. All rights reserved. This edition of the work is protected against unauthorized copying under Title 17, United States Code.

ProQuest LLC 789 East Eisenhower Parkway P.O. Box 1346 Ann Arbor, Ml 48106-1346

To President Charles Monahan Jr. with gratitude

ACKNOWLEDGEMENTS

I wish to express my sincere gratitude and appreciation to the following for their scientific and technical support in the achievement of this research:

Dr. T. J. Maher Dr. M. Bohlke Dr. C. Kelley Mr.H. Wu Dr. M. Ren Dr. S. Finklestein Dr. D. Williams

Contents
ABSTRACT INTRODUCTION PURPOSE OF THE STUDY PHARMACOLOGICAL ACTIVITIES OF PENTANE EXTRACT SYNTHESIS OF MACA-CONTAINING AND RELATED COMPOUNDS HYPOTHESES 7 8 11 12 13 14

LEPIDIUM MEYENII (Maca) Description and botany Origin, distribution and cultivation Phytochemistry Pharmacology and Biological activity

15 15 15 17 20

Toxicology
Products and mode of usage THE ENDOCANNABINOID SYSTEM Background Elements of the Endocannabinoid System Ligands or endogenous agonists: Receptors: Synthetic enzymes: Mechanism for uptake and degradation: Exogenous mimics: Antagonists: Pharmacology Roles in health and disease Future applications PATHOPHYSIOLOGY OF NEURODEGENERATION Glutamate-mediated neurotoxicity and excitotoxicity Ion movements and calcium overload Neurotransmitter and hormone homeostasis Activation of enzymes and transcription factors Oxidative stress.

29
31 32 32 33 33 34 34 35 35 36 37 40 42 43 45 48 51 53 56

Neurovascular changes and inflammation Glial cell changes Cell death Gene expression EXPERIMENTAL MODELS TO STUDY NEUROPROTECTANTS In vitro models Cell culture Slice preparations In vivo models of brain injury Preclinical models of spontaneous stroke Hypoxic-Ischemic (H-l) brain injury in vivo Focal cerebral ischemia models in rodents or whole animal studies by middle cerebral artery occlusion (MCAO) Intraluminal occlusion model (Suture model or filament model): Transient focal ischemia and permanent focal ischemia MCA permanent occlusion plus common carotid arteries transient occlusion MCAO permanent occlusion (Tamura) MCA occlusion by perivascular microinjection of endothelin-1 (ET-1) Photothrombic ischemia model Embolic models Focal cerebral damage by intracerebral injection of toxins Evaluation of brain damage Infarct assessment Hematoxylin-Eosin (H&E) staining 2,3,5-triphenyltetrazolium chloride (TTC) staining Measurement of apoptosis MATERIALS AND METHODS Plant material and extract preparations Infrared (IR) spectral analysis of pentane extract High performance liquid chromatography (HPLC) Animals Development of focal infarction

58 60 62 67 71 71 71 73 73 74 75

75

76 77 78 80 80 81 82 83 83 83 83 84 85 85 85 86 86 87

Treatment Infarct Volume Evaluation Cells and culture media Trypan-blue exclusion cell counting Cell viability assay with MTS reagent ABTS test for antioxidant capacity Compounds for screening In vitro neuroprotection assays Statistical analyses RESULTS Yield of extracts HPLC analysis In vivo neuroprotective assay Crayfish neuron neuroprotective assay Neuroprotective concentration-dependent effects in neuroblastoma cells Involvement of the endocannabinoid system Evaluation of antioxidant effects DISCUSSION Yield of extracts HPLC analysis In vivo neuroprotective studies In vitro neuroprotective studies Involvement of the endocannabinoid system Evaluation of antioxidant activity CONCLUSIONS (* New findings) FUTURE RESEARCH FIGURES REFERENCES

88 88 88 89 90 90 91 92 93 94 94 94 95 96 97 100 102 103 103 104 105 107 113 116 119 123 125 167

ABSTRACT
The neuroprotective activity of extract of the plant Lepidium meyenii Maca was studied using in vivo and in vitro experimental models. The pentane extract obtained by continuous liquid-liquid extraction, was chemically characterized and administered intravenously to rats prior to and following middle cerebral artery occlusion produced with two models of focal cerebral infarction. While infarct volumes were decreased for the lower dose, higher doses increased infarct volumes compared to controls, indicating two types of activity. In vitro experiments were performed on crayfish neurons and rat neuroblastoma cells. Cells in well plates were pretreated with vehicle, the pentane

extract, isolated macamides (natural alkamides and possible active principles) or synthesized alkamides (with structures based on the natural alkamides), and then were subjected to H2O2 as a neurotoxic agent. Cell viability was determined microscopically and chemically. A significant neuroprotective effect was demonstrated by the pentane extract and purified alkamides. In other in vitro experiments, cells were pretreated with O-2050 [a high affinity silent CBi (cannabinoid) receptor antagonist], and later exposed to the treatments mentioned above. Anandamide and tetrahydrocannabinol (THC) were used as cannabinoid controls. O-2050 significantly blocked the neuroprotective effects of the pentane extract and alkamides as well as the effects of anandamide and THC. The pentane extract and the alkamide designated MCP-33 which significantly reversed the deleterious effects of O-2050 on the cells. Finally, when the tested compounds were evaluated for antioxidant activity using the ABTS method with trolox as a positive control, none of the compounds evaluated (pentane extract, alkamide MCP-33, anandamide and THC) showed any antioxidant effects. These results suggest the potential application of alkamides, which represent part of the active principles of Maca, as neuroprotectants. The mechanism of action appears to involve the endocannabinoid system.

INTRODUCTION
Today neurodegenerative diseases and stroke are diseases that the elderly most fear. No one looks forward to spending their crowning years dependent on others and with impaired mental and physical faculties. While over one hundred years ago neurodegenerative diseases and stroke were disorders of little medical interest, in recent years the focus has changed and there are many more scientists and pharmaceutical companies interested in prevention and treatment of neurodegenerative diseases and stroke. Despite the failures of many clinical trials to result in effective neuroprotectants or treatments, the search is still intensive and ongoing.

Ischemic stroke, one example of acute neurodegenerative pathology, is a major cause of morbidity and mortality throughout the world. There are more than 700,000 strokes in the U.S. every year, and by the year 2020 stroke is expected to be the cause of the fourth largest disease burden worldwide. In the U.S. the cost associated with the care of stroke victims exceeded $50 billion in 2005, and stroke is responsible for half the patients hospitalized for acute neurological disease (Read, 2001; Wang, 2007). Cell death is the neuropathological substrate of cerebral stroke, and its extent is a key factor determining disease course and prognosis for the stroke patient. Despite several possibilities for stroke care and therapy, only a small number of acute stroke patients actually receive specific stroke therapies. Sustained attempts to increase the percentages of treated patients include the use of new agents and strategies. The introduction of neuroprotective pharmacological therapies leading to a modulation of stroke-related cell loss is the ultimate target for stroke therapy. Recently, numerous potential neuroprotective drugs have been evaluated in phase II and III clinical trials, however thus far no clearly successful candidates have been identified.

The search for new effective drugs has also been focused on complementary and alternative medicines (Hartman, 2006). Medicinal plants have the great advantage that they have been evaluated in unofficial clinical trials during centuries of traditional use
8

and usually many of the adverse effects and toxicities are known. In addition plants have been important sources of chemical compounds to be used as drugs or to guide the synthesis of new drugs. Naturally occurring compounds have the advantage of having developed in living organisms during the course of evolution to protect them and to improve their survival. In light of the complexity of stroke, natural based drugs might be good candidates to target oxidative stress, stabilize membranes, inactivate enzymes, inhibit apoptosis or inflammation, and inhibit or decrease the neurodegeneration and neuronal injury observed in focal ischemic stroke and other neurodegenerative diseases.

The Maca plant (Lepidium meyenii Walp. Syn. Lepidium peruvianum Chacon) is found exclusively in Peru, in the central highlands of the Andes. The underground part of the plant is a storage organ which forms a rounded hypocotyl, ending in a fleshy root. Knowledge of this plant and its medicinal properties has been transferred from generation to generation, and this transfer continues today. The harsh climate in which it grows stimulates Maca to synthesize interesting chemical compounds. Maca possesses a unique chemical composition that has value in the role of Maca as a functional food and versatile medicine. Maca is known to contain alkaloids, glucosinolates, isothiocyanates, polyunsaturated fatty acids (macaenes), alkamides (macamides), sterols, minerals and vitamins. Several studies have been performed with Maca, most of them examining its effects on sexual and reproductive functions in rodents and humans. Other studies have focused on menopausal disorders, female fertility, anti-stress effects and actions on the central nervous system which affect memory and neuroprotection. Moreover, few studies have evaluated the toxicity of Maca. There is still a need for additional pharmacological studies to further characterize the properties of this plant and there remain questions about its efficacy, appropriate dosing, mechanism(s) of action, active components, biopharmaceutical properties, safety, toxicity, adverse effects and possible clinical applications for the hypocotyls, extracts, various preparations and isolated compounds.

The goal of this study was to determine the neuroprotective properties of the hypocotyls from Lepidium meyenii (Maca), especially the neuroprotective effects of the

pentane extract in vivo in rats subjected to focal ischemic stroke, and the effects and possible mechanisms of action of isolated and/or synthesized compounds tested in vitro in cell culture. The lipophilic nature of some Maca constituents could facilitate their ability to cross the blood-brain barrier (BBB) and act within the brain. Standardized extracts, purified constituents or synthetic compounds could act along with endogenous compounds, mimic endogenous compounds with demonstrated neuroprotective efficacy, or block neurodegenerative processes. Maca could be a source of a neuroprotective agent that might act to prevent neurodegeneration and/or reduce cell death in brain trauma, stroke and other neurodegenerative diseases.

10

PURPOSE OF THE STUDY

Maca, taxonomically designated Lepidium meyenii Walp and also nominated Lepidium peruvianum Chacon, is a plant which has attracted increasing interest throughout the world due to its chemical composition and pharmacological properties. As with many medicinal plants and natural drugs, Maca is considered enigmatic for its extensive traditional background, diverse pharmacological information and the current necessity for more studies to demonstrate its applicability in medicine. The entire body of medicinal information regarding Maca is a mixture of undocumented traditional properties, information found in non-scientific media, and scientific information focused on the chemical composition and pharmacological activities primarily concerned with its effects on reproductive functions.

The potential pharmacological activities of the constituents of Maca are based on the medicinal uses of the plant for centuries and the findings about its chemical composition. The promising medical uses of maca products have grown on the basis of pharmacological studies performed with crude extracts. Nevertheless, there is a lack of systematic organization in the sequences of studies, with respect to pre-clinical and clinical demonstrations of efficacy and safety.

The statement that Maca extracts are pharmacologically active is well supported. While several interesting types of compounds are contained in the hypocotyls, some of the compounds of most interest are very lipophilic, increasing the likelihood of their bioavailability and CNS pharmacological activity. Current and future research on Maca has generally taken two approaches: first, the search for new therapeutic applications based on the pharmacological effects of fractionated extracts (e.g. pentane extract) on the CNS, cardiovascular, reproductive, or immune systems; and second, the synthesis of active constituents or derivatives leading to the discovery of drugs that could be pharmacologically characterized and could potentially be used in clinical studies.

11

Some of the reported pharmacological activities including improvement of "mental energy", antioxidant, and "hormone balancer" suggest that Maca hypocotyls might have neuroprotective activity. Some of the constituents of Maca have been reported to have pharmacological activity in the CNS (6-carboline alkaloids), and others, because of their lipophilicity, are likely to reach the CNS (macaenes and macamides). The demonstration of neuroprotective effects produced by Maca constitutes an interesting research direction. Our understanding of the neuroprotective effects of the Maca hypocotyls would benefit from the following studies to demonstrate potential preclinical efficacy:

PHARMACOLOGICAL ACTIVITIES OFPENTANE EXTRACT

1. Preparation of a pentane extract (primarily lipophilic constituents)

from a crude methanol extract of Maca hypocotyls. 2. Chemical characterization and standardization of the pentane

extract to be used in the study, by HPLC analysis interpreted with purified standards of macamides and IR spectroscopy. 3. In vivo pharmacological study of the neuroprotective effects of the

pentane extract from Maca using animal models of stroke (Chen and suture models in rats), with determination of dose-dependent effects. 4. In vitro pharmacological determination of neuroprotective effects

using neuron cell cultures (Crayfish neurons and rat neuroblastoma cells), and concentration-effect studies. 5. In vitro determination of the possible cannabinomimetic

mechanisms of action of the pentane extract, using an antagonist to the CBi receptor. 6. In vitro evaluation of antioxidant activity.

12

SYNTHESIS OF MACA-CONTAINING AND RELATED COMPOUNDS

The goal of the study was to identify pharmacologically the constituents responsible for the neuroprotective activity of the hypocotyls of Lepidium meyenii (Maca). Considering the structural similarity of macamides to the endocannabinoids, the study has involved in vitro screening using cell cultures, and isolated macamides and synthesized alkamides. Studies with purified Maca-containing compounds (provided by Dr. Ikhlas Khan of the National Center for Natural Products Research at the University of Mississippi) and synthesized compounds (provided by Mr. Hui Wu and Dr. Charles Kelley of the Laboratory of Medicinal Chemistry at MCPHS) were performed to select the compound with highest activity, determine any neuroprotective effects, and investigate the potential involvement of the endocannabinoid system in its mechanism of action. One compound could be sufficiently active to provide the basis to develop a new therapeutic agent to protect the brain and/or reduce cell death in stroke and neurodegenerative diseases.

The research included:

1.

Cell culture screening of isolated and synthesized compounds to

determine the neuroprotective activity in vitro, using crayfish neurons and rat neuroblastoma cells. 2. Cell culture-guided selection of one compound with maximal

neuroprotective activity using rat neuroblastoma cells. 3. Cell culture characterization of neuroprotective activity in rat

neuroblastoma cells and comparation with the neuroprotective activity of cannabinoids (anandamide and THC). 4. Cell culture characterization of the mechanism of action with

experiments performed to investigate activity involving the endocannabinoid system. 5. In vitro screening of their antioxidant capability.

13

HYPOTHESES
It is hypothesized that the pentane extract from the hypocotyls of Lepidium meyenii is neuroprotective, preventing cell death in cultured neurons and decreasing infarct volume in animals subjected to focal ischemic stroke. This hypothesis is based on the pharmacological background of the plant, pharmacological information about some of its constituents, and the chemical nature of the constituents extracted by pentane.

It is hypothesized that one or more macamides (alkamides present in Maca) could be responsible for the neuroprotective effects. These alkamides could be synthesized, chemically and pharmacologically characterized, and used in the development of a new therapeutic agent for the treatment of stroke and neurodegenerative diseases.

14

LEPIDIUM MEYENII (Maca)

Description and botany

The genus Lepidium L. is widely distributed across all continents, with the exception of Antarctica. The genus includes approximately 175 species, which makes it one of the largest genera in the family Brassicaceae (Toledo, 1998). The leaves are green, fragrant, and lie at the soil surface in a rosette which is continuously renewed from the center as outer leaves die. The rosette of 12 to 20 leaves forms a mat on the ground. The main stem is short and is connected to an underground storage organ. The flowers are off-white, self-fertile and are borne on a central raceme; they are followed by 4-5 mm siliculate fruits, each containing two small (2-2.5 mm) reddish-gray ovoid seeds. The seeds are the only method of reproduction; they germinate within five days of maturity under ideal conditions. The seeds have no period of dormancy, which might reflect the fact that Maca's native habitat remains harsh during the entire year. Lepidium meyenii is the only species in the genus which produces fleshy roots. The hypocotyl, the underground part which is fused with the root to form a rough inverted pear-shaped body, is the main product of Maca. Hypocotyls can vary greatly in size and shape, and can be conical, ovoid, spherical or cylindrical. Maca hypocotyls can be gold/cream, red, purple, black or green. Each is considered a genetically unique variety, as seeds of the parent plants grow to have roots of the same color (Toledo, 1998; Valentova, 2003)

Origin, distribution and cultivation

The genus probably originated in the Mediterranean basin, where most of the diploid species are found. Little is known about the time of origin and the mechanisms

15

responsible for its current worldwide distribution. However, most evidence indicates that long-distance dispersal took place in the tertiary or quaternary periods, after continental drift had taken place, and was responsible for its colonization of America and Australia. Common genetic features observed in the immigrant species of Lepidium are autogamy and polyploidy, both of which probably helped these species become established in new habitats. The Andean Lepidium species belong primarily to the sections Dileptium and Monoploca. They are interesting because they grow at high altitudes (Toledo, 1998).

Lepidium meyenii Walp. is an Andean crop with a limited distribution, restricted to altitudes above 3500 m and often reaching 4500 m. The taxonomic status of the cultivated species has been questioned based on morphological observations. A change in the name to Lepidium peruvianum Chacon has been proposed; therefore, both names are used as synonyms for all ecotypes of Maca. None of the wild species of Lepidium are identical to cultivated Maca (Toledo, 1998; Valentova, 2003), and the harsh climatic conditions in which it grows contribute to Maca's unique chemical and pharmacological profiles (Valentova, 2003).

Peru is the only producer of Maca, and the only geographical region with optimal conditions for its production is the department of Junin, 230 km from Lima at an altitude of 4300 m. In Junin resides the Regional Association of Maca Producers, formed by the families involved in the cultivation of this plant. The typical producer family leases land for cultivation of Maca (Brinckmann, 2004).

The interval from seeding to harvest is one full year. At the time of harvest, the plants are uprooted and the hypocotyls are separated from the aerial parts. After cleaning, the hypocotyls are spread on mats for sun drying, which takes from several weeks to 3 months, and depends on the weather. The dried hypocotyls can be stored for years until they are consumed (Brinckmann, 2004).

Five percent of the annual crop is separately used for seed production, and these plants are transplanted to other fields to grow until the seeds ripen in 6 to 7 months, at which 16

time the plants are harvested and dried in the whole form. The seeds are separated by shaking them from the dried plants. To start a new cultivation requires 3.5 kg of seeds per hectare (Brinckmann, 2004).

Phytochemistry
Maca hypocotyls represent the storage organs of the plants, and contain a high proportion of carbohydrates. Hydrolysable saccharides constitute 50-60% of dried Maca hypocotyls (Dini, 1994; Valentova, 2003). Other primary constituents are proteins (1012%), lipids (2-3%), fiber (8-9%), and ash (4-5%). The water content of fresh hypocotyls can be as high as 80% and in the dried powdered hypocotyls the residual water content is 10-12% (Dini, 1994).

The amino acid composition of the proteins shows an excellent profile, with a high content of essential amino acids such as threonine (33.1), tyrosine (30.6), phenylalanine (55.3), valine (79.3), methionine (28.0), isoleucine (47.4), leucine (91), lysine (54.5), and tryptophan (10.0, expressed as mg amino acid per g protein; Dini, 1994).

The fat concentration is comparable to other roots, has a saturated/unsarurated ratio of 0.76, and after alkaline hydrolysis shows the presence of all fatty acids from 12 to 24 carbons in length, with the highest concentrations of linoleic acid (32.6%), palmitic acid (23.8%o), and oleic acid (11.1%). The steroid fraction consists primarily of brassicasterol (9.1%), ergosterol (13.6%), campesterol (27.3%), ergostadienol (4.15%), and sitosterol (45.5%; Dini, 1994).

Alkylamides are lipid derivatives present in Maca hypocotyls, which are considered interesting constituents due to their structural similarity to endogenous compounds frjund in vertebrates which form part of the endocannabinoid system. Benzylated alkylamides (macamides) have been isolated from hexane extracts of Maca; 7V-benzyl-5-oxo-61s,8'-octadecadienamide (Figure 2) and JV-benzylhexadecanamide 17

(Figure 1) were the first two macamides reported, along with the acyclic keto acid 5-oxo6',8'-octadecadienoic acid (Muhammad, 2002; Ganzera, 2002). The fatty acids that make up part of the macamides have unique chemical characteristics, and in the free form are designated macaenes (Valentova, 2003). Additional studies have led to the isolation and characterization of five new alkylamides: 7V-benzyl-9-oxo-12Z-octadecenamide, Nbenzyl-9-oxo-12Z, 15Z-octadecadienamide, TV-benzyl-13 -oxo-9.E, 11 is-octadecadienamide, 7V-benzyl-15Z-tetracosenamide, and iV-(m-metlioxybenzyl)hexadecanamide (Zhao, 2005). Further investigations have led to identification of TV-benzyloctadecanamide, Nbenzylpentadecanamide, TV-benzylheptadecanamide, 7V-benzyl-9Z-octadecenamide, Nbenzyl-9Z,12Z-octadecadienamide, and A^-benzyl-9Z,12Z,15Z-octadecatrienamide as well as the methoxylated forms of the unsaturated compounds (McCollom, 2005). Alkylamides have restricted distribution in the plant kingdom, and are found primarily in four plant families: Piperaceae, Aristolochiaceae, Rutaceae, and Asteraceae. Alkylamides isolated from Maca are an exception to this systematic distribution, and these compounds have not been reported in other Lepidium species (Zhao, 2005). Therefore, these compounds could be used as markers for standardization and their pharmacological screening is an important step in establishing the importance of Lepidium meyenii.

Alkaloids were discovered in Maca hypocotyls in 1961 and they were originally designated macaines 1, 2, 3 and 4, but they were later isolated again, characterized and renamed (Valentova, 2003). The chemical natures of these alkaloids are diverse. Macaridine (Muhammad, 2002) is a benzylated 1,2-dihydro-7V-hydroxypyridine (Figure 3). Lepidilin A and lepidilin B are two imidazole alkaloids identified as l,3-dibenzyl-4,5dimethylimidazolium chloride and l,3-dibenzyl-2,4,5-trimethylimidazolium chloride

(Figures 4 and 5; Cui, 2003). The indole alkaloid (li?,35)-l-methyltetrahydro-6carboline-3-carboxylic acid (Figure 6) has also been identified in organic extracts of Maca hypocotyls (Piacente, 2002), a component which belongs to the tetrahydro-Bcarboline group of alkaloids (THfiCs). Other species and products contain both diasteroisomers (li?,35 and 15,35). This is the case with chocolate, cocoa, and alcoholic beverages such as wine, beer and liquor, which contain both isomers in concentrations as high as several mg/kg. The origin of these THBCs is a reaction involving L-tryptophan 18

and aldehydes that are present or otherwise released during the processing of products. The chemical formation depends on the amounts of precursors and the processing conditions (Herraiz, 2000). The imidazole alkaloid lepidine has been identified in other

species of the genus Lepidium (Valentova, 2003) but it has not been reported in Maca.

Glucosinolates represent another important group of compounds found in the hypocotyls of Maca and contribute significantly to its pharmacological and biological properties. The principal glucosinolates in Maca hypocotyls are benzyl glucosinolate (glucotrapaeolin, Figure 7) and m-methoxybenzyl glucosinolate

(methoxyglucotrapaeolin, Figure 8; Piacente, 2002; Li, 2001). Other glucosinolates were also identified as 5-methyl-sulfinylpentyl glucosinolate (glucoalyssin), >-hydroxybenzyl glucosinolate (glucosinalbin), m-hydroxybenzyl glucosinolate, pent-4-enyl glucosinolate (glucobrassicanapin), indolyl-3-methyl glucosinolate (glucobrassicin), and 4-methoxyindolyl-3-methyl glucosinolate (4-methoxy glucobrassicin; Li, 2001). In addition to these aromatic glucosinolates some aliphatic types were also detected, such as 5methylsulfmylpentyl glucosinolate and allyl glucosinolate, among others. The highest content of glucosinolates is found in seeds, fresh hypocotyls, and fresh sprouts. The total content of glucosinolates in Maca is higher than that reported for other cruciferous crops (Brassicaceae), and constitutes approximately 1% of the fresh weight of hypocotyls. Dried, powdered Maca hypocotyls have a much lower content of glucosinolates due to hydrolysis, which occurs during processing. Myrosinase is the enzyme responsible for hydrolysis, and it is released from hypocotyls by the breakdown of membranes and organelles (Leoni, 1997; Li, 2001).

In intact tissues of the Maca plant, glucosinolates are separated from the endogenous myrosinase enzyme, a B-thioglucoside glucohydrolase which catalyzes the hydrolysis of glucosinolates. When cells are disrupted, myrosinase stimulates the reaction of glucosinolates with water, resulting in the formation of o-glucose and compounds such as isothiocyanate, thiocyanate, and nitriles; which of these compounds form depends on the substrate and reaction conditions (Leoni, 1997). Benzylisothiocyanate (Figure 9) and 4-methoxybenzylisothiocyanate (Figure 10) are the most prevalent isothiocyanates 19

isolated from dried powdered hypocotyls and Maca products (Piacente, 2002). These compounds are also responsible for some of the biological activities of Maca. The aerial parts of the plant have also been examined. Steam distillates of the aerial parts were continuously extracted with pentane and the pentane extracts were analysed by GC/MS to identify 53 essential oil constituents. Phenylacetonitrile (85.9%), benzaldehyde (3.1%) and 3-methoxyphenylacetonitrile (2.1%) were the major

components of the steam-distilled oil (Tellez, 2002). Phenylacetonitrile is known to be a degradation product of benzyl glucosinolate. Some other constituents have been reported to be possible thermally-formed secondary products (Tellez, 2002).

Catechins (2.5 mg/g dried hypocotyls) have also been identified in Maca (Valentova, 2003).

Pharmacology and Biological activity

This plant is traditionally used: to improve endurance, potency and stamina; to enhance fertility and sexual behavior in men and women; to prevent chronic diseases; and to retard the detrimental effects of aging. Traditional or popular reports describe the Maca plant as "adaptogenic", indicating that its constituents can contribute to an increased adaptation to adverse conditions. Several studies have already been conducted to demonstrate the pharmacological properties of this plant and its products.

Aqueous extracts of Maca have been evaluated with several methods for in vitro antioxidant activity. The abilities to quench peroxynitrite, to scavenge l,l-diphenyl-2picrylhydrazyl radical (DPPH), to decrease peroxyl radical formation from 2,2'-azobis(2amidinopropane) hydrochloride (AAPH), and to protect deoxyribose against hydroxyl radicals were measured to demonstrate the antioxidant activity of Maca (Sandoval, 2002). Macrophages (RAW 264.7 cells) exposed to oxidative stress (H2O2 at 1 mM) have also been used to evaluate the antioxidant or cytoprotective activity of aqueous extracts of
20

Maca. ATP production was decreased in cells exposed to an oxidative stress and this effect was reversed by pretreatment with an aqueous extract of Maca (1 mg/ml). The detection of apoptosis by ELISA was also performed in RAW 264.7 cells exposed to peroxynitrite. Aqueous Maca extracts were shown to decrease cell death induced by peroxynitrite (Sandoval, 2002). Maca might contain water soluble compounds that contribute to the scavenging and/or decomposition of reactive oxygen species (ROS) in inflammatory and degenerative states. Catechins, reported in Maca, were previously thought to be responsible for the antioxidant activity. However, a comparison of the catechin content of Maca with that of other natural products indicates that the antioxidant activity of Maca is most likely due to constituents other than catechins (Valentova, 2003; Sandoval, 2002).

More recently, antioxidant activities of methanol and aqueous extracts were compared using the DPPH radical scavenging test, demonstrating that the methanol extract was more active with an IC50 of 0.71 mg/ml compared to 3.46 mg/ml for the aqueous extract (Valentova, 2006). The cytoprotective effects of Maca extracts have been evaluated in rat hepatocytes exposed to ?-butyl hydroperoxide (tBH) as a model toxin. The tBH-induced damage was not reduced by Maca extracts as indicated by the MTT test of cell viability, LDH release from cells, and by the amount of lipoperoxidation products detected in the medium (Valentova, 2006). The estrogenic activity of Maca extracts has been evaluated in vitro using MCF-7 cells, an estrogen-positive cell line derived from a human breast cancer. In appropriate media, cells were exposed to Maca extracts in a range of concentrations to evaluate their proliferative activity as an indicator of estrogenicity. Estradiol was used as a positive control. Maximal stimulation of MCF-7 cells by estradiol occurred at concentrations of 10"10 to 10"12 M. The maximal proliferative effects of Maca extracts (suggested estrogenicity) were exhibited in the range of 100 to 200 ug/ml (Valentova, 2006). This possible estrogenic activity must be considered in the evaluation of in vitro or in vivo results, reported both in the literature and in traditional medicine. In contrast with this result, one study reported a lack of estrogenic and androgenic activity in vitro, using a

21

yeast-based hormone-dependent reporter assay in which yeast cells transfected with androgen-inducible and estrogen-inducible B-galactosidase reporter plasmids failed to hydrolyze lactose, when pretreated with a methanol Maca extract up to a concentration of 4 mg/ml (corresponding to a concentration of 200 mg/ml of hypocotyl powder). Estradiol and testosterone were used as positive controls (Brooks, 2008).

The androgenic activity of Maca extracts was also studied in androgenindependent prostate cancer cells (DU-145) transfected with two plasmids, one containing a gene for the glucocorticoid response element (GRE), controlling a luciferase gene, and another containing a gene for the human androgen receptor (hAR). This transfected cell was stimulated with testosterone to produce luminescence (expression of luciferase). However, none of the Maca extracts (hexane, chloroform, methanol or ethanol) at concentrations ranging from 1 to 50 p.g/ml were able to activate the expression of luciferase in this cell system (Bogani, 2006). These results exclude direct effects of Maca extracts on the androgen receptor in vitro; however, the relationship between Maca ingestion and its effects on the masculine reproductive system has yet to be ascertained.

The essential oil from the aerial parts of Maca has been tested for antitermite activity. Significant activity was observed with benzyl thiocyanate, 3methoxyphenylacetonitrile, and B-ionone, three minor components of the essential oil which showed dose-dependent antitermite activity (Tellez, 2002). None of these constituents had previously been reported to have antitermite activity.

The results of in vivo pharmacological investigations of Maca provide information about the effects of whole hypocotyls or crude extracts administered orally, especially to enhance fertility in male mice with malathion-induced spermatogenic damage, which showed improvements in spermatogenesis (Bustos-Obregon, 2005), and to increase litter size in normal female mice following a 15-day treatment with an aqueous extract of Maca (Ruiz-Luna, 2005). In contrast to this result, another study reported no change in the rate
22

of embryo implantation in normal mice when Maca was administered ad libitum in the drinking water, at a concentration of 5% for 30 days (Oshima, 2003).

Several studies in rats have focused on spermatogenesis, testicular function and sexual performance, demonstrating improvements in response to aqueous or lipophilic Maca extracts (Zheng, 2000; Gonzales, 2001; Cicero, 2001; Cicero, 2002; Gonzales, 2003; Gonzales, 2004; Chung, 2005; Gonzales, 2006; Rubio, 2006; Gasco, 2007; Yucra, 2008). These studies have led to consideration of Maca as a potential natural agent for treating erectile dysfunction, together with arginine, yohimbine, Panax ginseng, and Ginkgo biloba. However, specific data supporting its therapeutic value as a prosexual or erectile-enhancing agent in humans is minimal (MacKay, 2004; Drewes, 2003). Studies in animals indicated no effects of Maca on the levels of estradiol, progesterone, testosterone and other hormones. Nevertheless, one study reported significant increases in the plasma levels of progesterone and testosterone in normal mice exposed to Maca hypocotyls at 5 % (w/v) in the drinking water for 30 days (Oshima, 2003).

One study reported the effects of an ethanol Maca extract on osteoporosis in ovariectomized rats. Animals were were treated for 28 weeks with two daily orally administered doses of 0.096 and 0.24 g/kg dried ethanol extract, corresponding to 0.5 and 1.25 g/kg of dried powdered hypocotyls. At the end of the treatment period, samples of urine and blood were taken for analysis of total calcium, inorganic phosphorus and serum osteocalcin and alkaline phosphatase; and lumbar vertebrae and femurs were isolated to measure physical parameters, biomechanical properties, and mineral density, and for histopathological evaluation. The bone mineral density of the lumbar vertebrae and the middle femur diameter were increased by treatment with the higher dose of Maca. Macatreated groups had restored travecular networks with fewer intertravecular spaces. Other parameters evaluated were not significantly different from those of the ovarectomized control animals. However, these results suggest that Maca might be useful for the treatment of postmenopausal osteoporosis (Zhang, 2006).

23

Other studies have demonstrated anti-stress activity of a methanol extract, dried and solubilized in saline solution. Daily i.p. injections (doses of 125 and 250 mg/kg dried extract) for one week prior to tests of restraint stress showed the beneficial effects of abolishing stress-induced ulcers, decreasing plasma corticosterone levels, and

normalizing weights of adrenal glands and plasma levels of glucose and fatty acids (Lopez-Fando, 2004). There were no significant differences in the effects observed in response to the two doses administered and the methanol extract was not observed to alter white blood cell counts.

Glucosinolates and isothiocyanates have received increasing attention for their biological activities, particularly for a protective role against cancer (Leoni, 1997; Gonzales, 2006). The mechanism for this activity is not completely understood. It has been suggested that glucosinolates can react in the gut with bacterial myrocinase to produce isothiocyanates and other derivatives, which in turn could neutralize carcinogens by an antioxidant mechanism, or these compounds could activate hepatic enzymes to protect against carcinogens. Quinone reductase, glutathione S-transferase, and UDPglucuronosyl transferase are enzymes activated by glucosinolate derivatives (Piacente, 2002). Other studies have demonstrated that glucosinolate derivatives can activate cascades leading to apoptosis in early stages of neoplastic transformation. Myrocinase and native glucosinolates do not have effects on tumor cell growth in vitro (Leoni, 1997). Several isothiocyanates were compared for their abilities to inhibit erythroleukemic K562 cell growth. Benzylisothiocyanate (derivated from glucotrapeolin) was one of the most active of the compounds tested. Interestingly, the antiproliferative activity of these compounds is inversely proportional to their water/oil partition coefficients, which provide information about the abilities of compounds to pass through cell membranes (Leoni, 1997). In a previous study, benzylisothiocyanate was reported to inhibit 7,12dimethylbenz(a)anthracene-induced (Wattemberg, 1981). breast neoplasia in Sprague-Dawley rats

Another interesting activity attributed to glucosinolates is that of androgenic modulation. Indolyl-3-methyl glucosinolate (glucobrassicin) has been reported to be a
24

potential modulator of androgenic activity, as it can be hydrolyzed to indole-3-carbinol, which is then converted to one of several oligomeric products. One of these products is 2(indolyl-3-methyl)-3,3-diindolylmethane, which is known as a specific androgen receptor antagonist (Chang, 1999; Le, 2003). This compound, present in the gut after consumption of products containing glucosinolates, has in vitro antiproliferative properties in human prostate cancer cells (Le, 2003) and in estrogen-dependent (MCF-7) and -independent (MDA-MB-231) breast cancer cells (Chang, 1999). This compound might be involved in some of the reproductive system activities reported for Maca. For example, red Maca, but neither yellow nor black Maca, significantly reduced ventral prostate size in rats after 7 days of treatment (orally administered aqueous extract at 2 g/kg/day), and after 14 and 42 days of treatment also prevented testosterone-induced increases in prostate weight (Gonzales, 2005). Histological evaluation showed that prostate epithelial height increased after treatment with testosterone, and that red Maca was able to reduce this parameter, indicating interference with the androgenic action of testosterone (Gonzales, 2005). As mentioned before, glucosinolate content could be responsible for this activity and the composition of glucosinolates is different in the three varieties of Maca. In another study, water and ethanol extracts were compared for their abilities to reduce testosteroneinduced prostate hyperplasia in rats, but showed no significantly different effects (Gonzales, 2007). The dose-dependent effect observed in the preliminary study (Gonzales, 2005) was shown to be due to the amount of benzyl glucosinolates present in the Maca extracts. A dose-dependent reduction in prostate weight was observed with increasing doses of benzyl glucosinolates (0.02 to 0.08 mg/day, determined by HPLC). Seminal vesicle weights were not affected at any of the glucosinolate doses tested (Gonzales, 2007). The effect of red Maca on testosterone-induced prostate hyperplasia in rats was confirmed in mice, with a hydroalcoholic extract at 140 mg/kg/day. Red Maca reduced prostate weights after 21 days of treatment, as well as histological parameters such as prostatic stromal area after 7 days of treatment and prostatic acinar area after 14 days of treatment (Gonzales, 2008). The polyphenol content of the administered extracts was measured and it was suggested that these compounds could also be responsible for the effects of red Maca on prostate hyperplasia.

25

The influence of Maca on lipid and carbohydrate metabolism was evaluated in hereditary hypertriglyceridemic (HHTg) rats. Maca powder at 1 % (w/w) was added to a high sucrose diet. Maca significantly decreased the plasma levels of VLDL, LDL, total cholesterol and triglycerides compared to the control group. Maca also significantly lowered glucose levels and improved glucose tolerance. The HHTg rats treated with Maca showed increased superoxide dismutase (SOD) activity, increased liver glutathione levels and increased glutathione reductase activity in blood (Vecera, 2007). These results indicated that Maca improves lipid and glucose metabolism, as well as antioxidant capacity of the organism. Maca seems to show promise for chronic disease prevention.

Because they are lipophilic, many Maca constituents can cross the blood-brain barrier, and it has been hypothesized that some activities occur by the actions of Maca constituents on the nervous system. For example, the effects of Black Maca on learning and memory were evaluated in ovariectomized mice (Rubio, 2008) and in mice with scopolamine-induced memory impairments (Rubio, 2007). Mice were orally treated with doses of 0.5 and 2 g/kg/day of aqueous extract or with doses of 0.25 and 1 g/kg of hydroalcoholic black Maca extract. Memory and learning were evaluated with the Morris water maze and the step-down avoidance test. Treatment with black Maca extracts improved experimental memory which had been impaired by ovariectomy or scopolamine. At the end of the studies, mice were sacrificed to measure levels of malondialdehyde (MDA), acetylcholinesterase (AchE) and monoamine oxidase (MAO) to determine the possible mechanisms of action. Ovariectomized mice treated with Maca had lower brain levels of MDA, demonstrating the in vivo antioxidant activity of the aqueous extract. In the brains of ovariectomized and scopolamine-impaired mice, decreased Ache activity could be a possible mechanism for memory improvement. MAO activity was unchanged by the treatments with black Maca. There were no dosedependent effects observed during the evaluation. A previous study (Rubio, 2006) had shown differences between varieties of Maca in their effects on learning and depression, and had demonstrated that black Maca is the most active variety to induce learning improvements in ovariectomized mice.

26

Tetrahydro-B-carboline alkaloids (THBCs) are naturally occurring alkaloids produced by many different species of plants. These compounds have attracted the attention of neuroscientists, who have pointed out that they normally occur in biological tissues and fluids in physiological conditions (Buckholz, 1980; Gonzales, 2008). Because tetrahydro-B-carbolines have significant biological activities, their presence in Maca is of interest. Several THBCs are present in some foods and in alcoholic drinks, which indicates that their presence in tissues and fluids can be from dietary sources. There is speculation on their putative role in the central nervous system, where they can function as neuromodulators and where they might have a role in the etiology or addictive process of alcoholism, or in the cravings produced by chocolate and cocoa products. Their activities in the CNS have been thought to occur by the inhibition of monoamine oxidase (MAO), by modulating the uptake or release of serotonin or by affecting the affinity of GAB A for its receptor (Herraiz, 2000). It has not yet been demonstrated whether their presence in the body is wholly due to dietary origins or whether there is also endogenous production of these compounds.

Neuroprotective activities of Maca extracts have been demonstrated in vitro in crayfish neurons subjected to oxidative stress, and in vivo in rats subjected to focal ischemic stroke (Pino-Figueroa, 2009) in crayfish neurons subjected to oxidative stress with H2O2. Neuroprotective effects of the pentane extract of Maca were evaluated at concentrations ranging from 0.1 to 30 ng/ml, demonstrating a dose-dependent response. The pentane extract was also assessed in vivo in rats subjected to ischemic brain damage. Doses of 3, 10 and 30 mg/kg were administered by tail vein injection, 30 minutes prior to and one hour after the onset of stroke. Only the 3 mg/kg dose showed beneficial effects and higher doses actually led to increased infarct volumes. These results suggest that Maca might have two different kinds of activity on the brains of rats subjected to stroke. One activity at lower doses is beneficial, antioxidant, anti-apoptotic, and protects neurons from ischemic insult, while the activity at higher doses is excitatory, pro-apoptotic, and promotes cell death. These findings indicate new directions for research to develop agents from Maca that might be used to prevent or treat neurodegenerative disease and stroke (Pino-Figueroa, 2009).
27

Of interest is one recent report indicating that aqueous extracts of Maca hypocotyls could provide dermal protection from UV radiation when administered locally to rats (Gonzales-Castaneda, 2008). There have been no other reports concerning the topical administration of Maca as a dermatologic or cosmetic agent.

Clinical studies have demonstrated the ability of Maca to improve semen parameters such as volume, sperm count and sperm motility in healthy men, without affecting plasma hormone levels (Gonzales, 2001). Oral doses of 1.5 and 3.0 g/day of dried hypocotyl powder for 12 weeks had a beneficial effect on subjective sexual desire in healthy adult men (n=57). This effect on sexual desire was dose-dependent and did not affect plasma levels of FSH, LH, prolactin, progesterone, testosterone or estradiol (Gonzales, 2002; Gonzales, 2003).

A recent clinical study of healthy women, aged 50 to 60 years, examined the effects of Maca on menopausal symptoms. Each woman in the study received 3.5 g/day of powdered Maca hypocotyls or placebo for 6 weeks, in a random crossover design, in the 12 week study. At baseline and at weeks 6 and 12, plasma levels of estradiol, FSH, LH, and sex hormone binding globulin (SHBG) were measured and the Greene Climacteric Scale (GCS) was used to assess the severity of menopausal symptoms. The GCS is a well-validated, non-intrusive self report questionnaire which assesses physical and psychological symptoms associated with menopause. This study showed that Maca significantly reduced the scores of negative psychological symptoms associated with menopause, including significant effects on anxiety and depression. Additionally, Maca appeared to significantly reduce sexual dysfunction (Brooks, 2008). These effects do not appear to have an endocrine basis, since no change was observed in levels of sex hormones or SHBG. While traditional uses have recognized the effects of Maca on menopausal women, this is the first scientific study demonstrating these benefits of Maca. The failure to demonstrate hormonal effects of Maca indicates that the effects of Maca are not due to phytoestrogenic activity (Gonzales, 2002; Gonzales, 2003; Brooks, 2008).

28

Another recent clinical study has also demonstrated beneficial effects of dried powdered Maca hypocotyls on selective serotonin reuptake inhibitor (SSRI)-induced sexual dysfunction. Twenty patients (mean age 3613 years, 17 women and 3 men) who were currently taking a stable dose of an SSRI antidepressant for at least 4 weeks were recruited. Subjects were required to meet at least one of the following criteria: anorgasmia during sexual activity, significant orgasm delay compared to the usual time to achieve orgasm prior to antidepressant medication, inability to attain or maintain sexual arousal until completion, or decreased libido according to self-reports. None of the subjects had sexual dysfunction prior to antidepressant medication. Two doses of Maca powder (1.5 and 3.0 g/day) were randomly assigned to two groups of 10 patients each during a 12 week trial. Assessments of sexual function were made every two weeks with the Arizona Sexual Experience Scale (ASEX) and the Massachusetts General HospitalSexual Function Questionaire (MGH-SFQ). Ten patients completed the study and 16 patients were considered as "intent to treat" for the analysis of results. The results showed statistically significant improvements in sexual function with the high Maca dose, and improvement without statistic significance for the low Maca dose. The treatment was well tolerated overall, with reports of some transient adverse effects which were difficult to attribute solely to Maca administration (Dording, 2008).

When all clinical results are pooled, they indicate that Maca has activities on the sexual and reproductive systems, showing beneficial effects on libido, sexual dysfunction, and both physical and psychological symptoms of menopause. There might be dose-related effects and future research will improve our understanding of the mechanisms of action and constituents responsible for these activities of Maca.

Toxicology
The effects of Maca extracts have been evaluated in vitro in rat hepatocyte primary cultures exposed at concentrations of 0.0125 to 10 mg/ml. Lactate dehydrogenase (LDH) and aspartate aminotransferase (AST) leakages from cells were 29

measured after 24-72 hours of treatment as indicators of cytotoxicity. No morphological changes visible by microscopy or enzyme leakages were observed at Maca doses up to 10 mg/ml. In fact, the higher concentrations of Maca extract and longer exposure times to the extracts decreased leakage of the enzymes, demonstrating a protection from cytotoxicity (Valentova, 2006). These results provide evidence that exposures to Maca extracts, even at very high concentrations, do not cause in vitro hepatotoxicity.

Most of the studies in animals did not demonstrate toxicity (Valerio, 2005). Chronic treatment with three varieties of Maca at oral doses of 1 g/kg/day for 84 days had no effect on DNA integrity in the testis nor did it change the histo-pathological morphology of liver tissue (Gasco, 2007).

In humans, preliminary clinical toxicological assessments in a sample of 101 patients (58 female and 43 male) suffering from metabolic syndrome showed that consumption of 0.6 g/day of dried hypocotyl powder for 90 days led to significant increases in diastolic blood pressure and plasma aspartate aminotransferase levels (Valentova, 2008). No changes in lipids, glucose metabolism or other biochemical parameters were observed in these patients.

l-Methyl-l,2,3,4-tetrahydro-B-carboline-3-carboxylic acid (MCTA) is an alkaloid present as two stereoisomers: 15,35 and IR,3S. The li?,35 isomer has been identified as an alkaloid constituent of Maca hypocotyls (Piacente, 2002). The 15,35 isomer has been found to be neurotoxic in vitro, in one month old spinal cord cultures derived from fetal mice. The toxicity was associated with neuronal cell death in mature cultures, although the reverse occurred in immature cultures in which exposure to the compound increased cell growth. Treatment with the li?,35 isomer did not cause significant changes in either mature or immature cultures. The stereospecificity of this effect suggests that it might occur through a receptor-mediated mechanism (Brenneman, 1993). These observations require further investigation to determine the true activities associated with the MTCA present in Maca and whether this compound has beneficial or toxic effects.

30

High doses of the pentane extract of Maca (10 and 30 mg/kg), administered intravenously have shown detrimental effects in rats subjected to focal ischemic stroke, increasing the infarct volume (Pino-Figueroa, 2009), in contrast to the beneficial effect of decreased infarct volume observed in response to the low dose of pentane extract (3 mg/kg). Future research will hopefully identify the constituents responsible for this detrimental effect in ischemic conditions.

Products and mode of usage

Maca hypocotyls are a natural product and have been proven to have beneficial physiological effects in both animals and humans. Depending on the method of

preparation or the nature of derivatives, products can be classified as functional foods (such as cooked hypocotyls, powder, baked products or beverages) or dietary supplements (such as encapsulated powders, extracts or extracted fractions obtained from powder). Diverse products can be found in pharmacies (capsules, tablets) or in food markets (flour, cookies, candies, fermented beverages, and juices) produced from raw Maca alone or in mixtures with other plants and products.

There are no established scientific criteria for medicinal doses of Maca. The range of traditional consumption of Maca hypocotyls is 1 to 100 g fresh weight/day (Valerio, 2005) and this might guide dosing in future clinical studies.

31

THE ENDOCANNABINOID SYSTEM Background

Research on Cannabis sativa began about two centuries ago with the use of Cannabis in India and the study of its effects on humans. The identification of its active constituents was difficult due to the non-alkaloid structures and complexity of the mixture. The structure of the main active constituent was elucidated in the 1960s and was established as A9-tetrahydrocannabinol (THC). THC was proven to be the most active constituent and other less active compounds were also discovered, such as cannabidiol (CBD), cannabigerol, cannabichromene and cannabicyclol; together with other isomers of THC such as A6a'10a-THC and A8-THC (Mechoulam, 2002).

Since the 1980s, the pharmacodynamics of cannabinoids has been studied and the presence of a high affinity binding site (receptor) was demonstrated in the mammalian CNS. This receptor has been cloned and designated as Cannabinoid Receptor 1 (CBi). A peripheral receptor later identified in the spleen was designated CB2. The finding of receptors indicated the presence of endogenous ligands. In the 1990s, studies in porcine and canine brains demonstrated the presence of lipid derivatives acting as agonists on the CBi and CB2 receptors; these endogenous compounds were isolated and identified as Narachidonyl ethanolamide (Anandamide), 2-arachidonyl glycerol (2-AG) and 2arachidonyl glyceryl ether (noladin ether; Mechoulam, 2002). Since then, new endogenous active molecules have been discovered and related compounds have been synthesized. Other elements of the endocannabinoid system have since then discovered and they are currently under study to determine their involvement in physiological and pathological processes.

32

Elements of the Endocannabinoid System

The endocannabinoid system contains all the elements required for neurotransmission and signaling as shown in Figure 11, and decribed as follows:

Ligands or endogenous agonists: Endocannabinoids are a diverse group of lipid transmitters which can be chemically described as 7V-alkyl amides, formed by the conjugation of fatty acids and ethanolamine and designated as JV-acyl ethanolamides (NAEs). The most important example of an NAE is anandamide, a partial agonist at both CBi and CB2 receptors. Others are palmitoylethanolamide, estearoylethanolamide and oleoylethanolamide, all of which can also bind cannabinoid receptors. There is evidence indicating that palmitoylethanolamide can potentiate the activity of anandamide in the brain (Franklin, 2003). Other endocannabinoids are esters of arachidonic acid conjugated with glycerol such as 2-arachidonyl glycerol (2-AG), a full agonist at both CBi and CB2 receptors; with ethanolamine (virodhamine); or with dopamine to produce the compound 7V-arachidonyldopamine. An additional endocannabinoid is the ether produced by conjugation of arachidonic acid and glycerol (noladin ether; Rodriguez de Fonseca, 2005). Other compounds devoid of cannabinoid activity are single amides formed by conjugation of fatty acids with ammonia, such as oleamide (ODA) which has been demonstrated to bind to rat and human CB] receptors (Leggett, 2004; Maccarrone, 2005).

Anandamide

and

2-arachidonyl

glycerol

are

the

most

representative

endocannabinoids and also the first discovered. These compounds probably have different functions, as they are located in different tissues and regions of the brain. They are also produced by different biochemical pathways, and they have marked differences in their affinities for and activation of their receptors.

33

Receptors: The receptors of the endocannabinoid system belong to the superfamily of Gprotein coupled receptors (GPCR). The CBi receptor, the first described, is located primarily in neurons and glial cells in the nervous system in humans, mammals (Ho, 1996) and in invertebrates such as crayfish (Green, 2005). It has also been found in the reproductive system and in the microcirculation. The CB2 receptor was found initially in the immune system, expressed in white cells, lymphoid organs and microglial cells. Other cannabinoid receptors are nonCB]nonCB2 receptors (CB like receptors) which comprise new GPGR receptors with cannabinoid activity, such as GPR-55 and GPR-119 found in the human genes 55 and 119, respectively.

Synthetic enzymes: Different pathways are involved in the synthesis and release of anandamide and 2AG. The precursor for anandamide is the phospholipid iV-arachidonylphosphatidylethanolamine (NAPE) and the enzyme responsible for synthesis is phospholipase D (PLD; Yang, 1999; Hansen, 2002). This enzyme has no homology with other known PL enzymes and its concentration is highest in the brain, kidneys and reproductive system. The activity of PLD is regulated by the activation of glutamatergic NMDA receptors, nicotinic neuronal receptors and dopamine receptors, which indicates that major neurotransmitters such as glutamate, dopamine and acetylcholine are involved in the release of anandamide (Giuffrida, 1999; Rodriguez de Fonseca, 2005). The synthesis and release of 2AG is associated with the receptor-dependent activation of phosphatidylinositolspecific phospholipase C (PLC). It has been demonstrated that metabotropic receptors coupled to PLC increase the production of 2-AG. Once anandamide or 2-AG are formed and released via diffusion they target presynaptic or postsynaptic CB receptors.
34

Mechanism for uptake and degradation: Endocannabinoid signaling is terminated by a two step process which includes transport into the cells and hydrolysis. The reuptake transporter for

endocannabinoids works in a manner similar to other lipid carriers, is widely distributed in the brain and results in the uptake of both anandamide and 2-AG in an energy-independent manner. This transporter is saturable, substrate specific and can be blocked by specific reuptake inhibitors (Rodriguez de Fonseca, 2005). The degradation of endocannabinoids is performed by two enzymes: fatty acid amide hydrolase (FAAH), which degrades anandamide, 2-AG and other iV-alkyl amides; and monoacylglyceride lipase (MAGL), which also degrades 2-AG. FAAH is widely distributed in the body, with highest concentrations found in the brain and liver. MAGL has been found in nerve terminals and specific brain regions (Rodriguez de Fonseca, 2005; Seierstad, 2008)

Exogenous mimics: The term "endocannabinoid system" was derived from the name of the plant Cannabis sativa L. which contains the psychoactive A9-tetrahydrocannabinol (THC) as its most active constituent. THC is a tricyclic benzopyran derivative which is a partial agonist at both CBi and CB2 receptors. This compound was responsible for the discovery of the endocannabinoid system. Several exogenous compounds have been developed that act on CB receptors and might have potential medical applications. Some derivatives of THC are the compounds HU210 (ll-hydroxy-A8-THC-dimethylheptyl) which displays the highest affinity for the CBi receptor, and the compound HU-308, which is a selective CB2 receptor agonist (Rodriguez de Fonseca, 2005). Aminoalkylindoles were the first non35

cannabinoid molecules with cannabinomimetic

activity. Alkamides with

structures similar to anandamide but with different fatty acid or 7V-alkyl groups are found in nature in the plants Echinacea purpurea (Raduner, 2006) and Lepidium meyenii (Muhammad, 2002) and could lead to the development of new agonists for the CB receptors. There are compounds termed "indirect agonists", which act by potentiating or sustaining the action of endogenous ligands. These compounds are inhibitors of FAAH or are blockers of reuptake transporters, and they could be recommended for diseases in which increased cannabinoid activity is required, minimizing side effects produced by direct activation of CB receptors, such as the typical psychoactive effect (de Lago, 2007). Most indirect agonists are found in screenings for analgesic or anti-inflammatory agents. An example of an indirect agonist is the compound UCM707 (5Z,8Z,llZ,14Z-N-(3-

furylmethyl)eicosa-5,8,ll,14-tetraenamide),

a derivative of arachidonic acid

which has high potency and selectivity as an inhibitor of endocannabinoid reuptake (de Lago, 2007). Several inhibitors of FAAH have been developed and include trifluoromethylketones, ketoheterocycles, carbamates, arylureas, and thiohydantoin-based compounds (Seierstad, 2008). Some marketed drugs with analgesic or anesthetic properties have shown activity as FAAH inhibitors. This is the case for the anesthetic agent propofol and for some non-steroidal antiinflammatory agents (Seierstad, 2008).

Antagonists: Cannabinoid CBi receptor antagonists have been developed and investigated for potential therapeutic applications and for a role in the study of pharmacological properties of other drugs with potential cannabinoid activity. The two most relevant cannabinoid antagonists are 7V-piperidino-5-(4-chlorophenyl)-l-)2,4dichlorophenyl)-4-methylpyrazole-3-carboxamide (SR 141716) and (6a/?,10a/?)3-(l-methanesulphonylamino-4-hexyn-6-yl)-6a,7,10,10a-tetrahydro-6,6,936

trimethyl-6/f-dibenzo[&,cT|pyran (O-2050). SR 141716 is a competitive inverse agonist, which reverses cannabinoid agonist-induced hyperphagia and attenuates basal feeding in animals, suggesting that cannabinoid antagonists may have some utility in the treatment of obesity (Gardner, 2006; Canals, 2008). O-2050 is an
o

analogue of A -tetrahydrocannabinol, possessing high affinity for CBi receptors and showing no agonist or inverse agonist activity; because of this it has been designated as a "silent antagonist" (Gardner, 2006; Canals, 2008).

Pharmacology
Several neurotransmitters and signaling molecules in the CNS and periphery have been shown to interact with the endocannabinoid system. These interactions are designated as "cross-talk", which leads to a vast array of biological actions, impacting many aspects of mammalian pathophysiology (Maccarrone, 2005).

In vivo bioassays have demonstrated four effects that are considered as a signature of cannabimimetic activity in mice. These four effects are: analgesia, hypothermia, immobility and catalepsy. In vitro and in vivo experiments have demonstrated that CB receptors are responsible for this activity.

In the brain, CBi receptors are present primarily in neurons and are involved in ionic and metabolic processes. In the event of brain injury, CB] receptors might be involved in the initial stages of neuroprotection. CB2 receptors are also present in the brain, mostly in glial cells and on microvascular endothelial cells. Research suggests that CB2 receptors are involved in neuroprotective processes initiated at later stages, such as BBB restoration and vasodilation (Mechoulam, 2007). Another receptor activated by some cannabinoids is the vanilloid receptor 1 (TRPVi), which is a cation channel
9+ 4-

receptor and upon activation increases both intracellular Ca

and Na concentrations.

Anandamide is an endogenous ligand which binds to TRPVi and because of this it is also 37

considered an endovanilloid. This pharmacological property of anandamide enables it to have opposite actions on apoptosis, which is inhibited through CBi receptors and activated through the TRPVi receptor (Maccarrone, 2000). There are reports demonstrating the suppression of tumor growth by anandamide both in vitro and in vivo, indicating the potential use of anandamide as an anticancer agent (DeMorrow, 2008).

Activation of all cannabinoid receptors is coupled to similar transduction systems. CB receptors are coupled to G proteins. There is accumulated evidence of the coupling of a single CB receptor with multiple G proteins such as the Gi/0, Gs, or Gq/n subtypes. Depending on the type of G protein, CB receptor activation can inhibit or stimulate the enzyme cAMP synthetase (Bosier, 2007). G0if proteins are also associated with CB receptors and can be linked to ion channels. Inhibition (or stimulation) of cAMP formation leads to decreases (or increases) of protein kinase activities. Coupling to G0if proteins inhibits voltage-gated calcium channels and activates potassium channels (Figure 11; van der Stelt, 2001; Rodriguez de Fonseca 2005).

Several actions associated with CB receptors have been proposed, in addition to the activation of the above mentioned transduction systems such as: presynaptic inhibition of glutamatergic synaptic transmission; modulation of release or activity of other neurotransmitters such as GABA, adenosine, serotonin, dopamine, acetylcholine, and norepinephrine (Melis, 2004; Gardner, 2005; de Lago, 2007); inhibition of calcium channels; modulation of hormone release/activity; and inhibition of inflammatory processes. Endocannabinoids have been shown to have a crucial role in the control of excitability of dopaminergic neurons, in which modulation of glutamatergic transmission seems to be important in DA-related functions and disorders (Melis, 2004).

One interesting proposed mechanism by which endocannabinoids might regulate other neurotransmitters is the control of gene expression. There is evidence that endocannabinoids can control the expression of the tyrosine hydroxylase (TH) gene, affecting levels of TH mRNA and the intracellular activity of TH, which has a pivotal role in the synthesis of catecholamines (Bosier, 2007). The mechanisms proposed for the 38

control of gene expression would involve intracellular effectors of CB receptors such as PKA, MAP kinases and intracellular Ca2+.

Cross-talk between endocannabinoids and steroids has been shown or proposed to regulate stress, energy homeostasis, food intake, fertility, prolactin secretion, luteinizing hormone (LH) and testosterone levels, endothelial function and cytokine release from Tcells. Glucocorticoids secreted by the adrenal cortex can exert rapid inhibitory effects on several hypothalamic neuroendocrine systems, including negative feedback regulation of the HPA axis. These rapid actions are not mediated by intracellular receptors or by the regulation of gene transcription. Evidence suggests that there is interplay between endocannabinoids and glucocorticoids in the activation of G-protein-dependent signaling mechanisms (Maccarrone, 2005).

Cross-talk between different GPCRs in the same cell is another possible mode of regulation of activity by endocannabinoids. It has been demonstrated that GPCRs can dimerize and interact with one another, and by doing so they can alter their function and pharmacological effects. There are other mechanisms that allow cross-regulation between pairs of co-expressed GPCRs that do not require direct protein-protein interactions. Studies have demonstrated that opioid and cannabinoid receptors have the capacity to cross-regulate and that Mu opioid and CBi receptors are co-expressed in a variety of neurons in the mammalian brain (Canals, 2008).

Downregulation (desensitization and degradation) of CB receptors has also been demonstrated in vitro and in vivo by two different mechanisms. Homologous downregulation has been evidenced when the receptors are exposed to agonist at either higher concentrations or longer periods of time. For example, chronic exposures of animals to cannabinoid agonists decreased the density of CBi receptors in the brain. Heterologous mechanisms have also been confirmed in cell cultures, demonstrating that opioids can downregulate CB] receptors following prolonged exposure of cells to opioid agonists (Keren, 2003). It has frequently been observed that following agonist activation of GPCRs, the receptors undergo desensitization and internalization into endosomes, 39

following a B-arrestin-mediated mechanism. Receptors can later be resensitized and recycled to the plasma membrane or targeted for degradation (Canals, 2008). Lipid rafts are another mechanism for the endocannabinoid system to control the activity of CB receptors. Changes in membrane fluidity might affect ligand recognition by the CBi receptor and ligand binding properties. There is evidence that CBi receptors function in the context of lipid rafts, and the integrity of these compartments may be important to limit signaling. In vitro experiments have demonstrated that cholesterol depletion from the cell membrane by methylcyclodextrin disrupts lipid rafts and improves both ligand binding to CBi receptors and also subsequent interaction of the activated receptors with G protein (Bari, 2005).

Roles in health and disease

The cannabinoid system has an important role during cell differentiation and brain development, evidenced by the increased expression of receptors in the CNS during initiation of life. The CBi receptor is the most abundant GPCR in the mammalian brain, with 10-50 fold higher density than that of other classical neurotransmitter receptors, such as those binding dopamine or opioids, which indicates the important role of cannabinoids in brain function (Rodriguez de Fonseca, 2005). One relevant function of the endocannabinoid system is modulation of the release of neurotransmitters, primarily GABA, glutamate and dopamine, but also other neurotransmitters and endocrine molecules. Short-term synaptic plasticity is also facilitated by endocannabinoids in the CNS. Dysfunction in the role of the endocannabinoid system in dopaminergic synaptic transmission might be of considerable importance in the pathogenesis of many neurological and psychiatric disorders (such as Alzheimer's disease, Parkinson's disease and schizophrenia). Pharmacological manipulation of the endocannabinoid system could be useful to prevent or treat these pathologies (Melis, 2006)

40

Several studies have demonstrated that endocannabinoids have a role in protecting the brain from damage due to acute insults. Both in vitro and in vivo investigations suggest that endocannabinoids could be neuroprotective and research is ongoing to determine the medical importance of these observations (Mechoulam, 2002; Schabitz, 2002). In response to traumatic brain injury and focal cerebral ischemia in rats, there is local and transient accumulation of endocannabinoids and their precursors at the site of injury (Hansen, 2001). Alkamides such as anandamide, palmitoylethanolamide and

oleylethanolamide were detected in humans and were shown to be released after stroke, as measured by in vivo microdialysis. This release occurred in penumbral tissues and was associated with significant elevations in extracellular lactate, an early marker of hypoxic insult (Schabitz, 2002). It remains unclear by what mechanism endocannabinoids are accumulated in the brain in ischemic conditions and whether this accumulation has any beneficial or adverse effects (Berger, 2004). Neuroprotection provided by exogenous administration of endocannabinoids in animal studies has indicated that the formation of endocannabinoids is a molecular regulator of pathophysiological events, reducing brain damage, which is relevant in brain pathologies such as cerebral ischemia, post-traumatic brain injury and neurodegenerative disorders (van der Stelt, 2001; Franklin, 2003; Shouman, 2006; Mechoulam, 2007; Schomacher, 2008).

Focal cerebral ischemia induces neuronal necrosis in the core, and apoptosis that gradually expands to the penumbra, partly as the result of a neuroinflammatory response. Contradictory to its neuroprotective effect, anandamide has also been shown to contribute to the propagation of neuroinfiammation, and this effect is enhanced synergistically by the alkamide palmitoylethanolamide. It is well known that anandamide acts through CBi receptors to reduce the release of glutamate, preventing excitotoxicity; and acting through CB2 receptors anandamide also enhances cerebral blood flow. Therefore, the effect of anandamide to increase the motility of microglial cells, which in turn enhances neuroinfiammation, is an interesting topic for further study (Franklin, 2003).

Among the peripheral activities of anandamide, the regulation of fertility has attracted growing interest. Low FAAH levels in lymphocytes have been shown to be an early 41

predictor of spontaneous abortions in humans. In mice the uterus has the highest anandamide levels measured in any tissue so far. Anandamide acts through CBj receptors to bring about uterine epithelial changes, and to regulate blastocyst function and implantation. The mechanisms could involve MAP kinases and calcium channel activity coupled to CB receptors (Maccarrone, 2005b).

Future applications
Over the past few years, considerable progress has been made in our understanding of the role of endocannabinoids to prevent or to reduce the effects of progressive neurodegenerative diseases such as Parkinson's disease, Huntington's disease, multiple sclerosis, amyotropic lateral sclerosis, and Alzheimer's disease, as well as post-ischemic stroke and traumatic brain injury (Mechoulam, 2007). We still need answers to many questions about specific molecules that could provide neuroprotection and about molecular mechanisms of action, doses, pharmacokinetics and toxicities of new compounds, such as alkamide derivatives which have shown promising neuroprotective effects acting through the endocannabinoid system.

Another potential group of cannabimimetic drugs consists of FAAH inhibitors, which are being developed and evaluated for their analgesic and antiinflammatory effects. Future clinical research will hopefully confirm the applicability of FAAH inhibitors to the treatment of human diseases (Seierstad, 2008).

Effects on fertility could guide another interesting application of endocannabinoids and their derivatives. Selective pharmacological manipulations of the endocannabinoid system might restore fertility which has been impaired by age or disease (Maccarrone, 2005b).

42

PATHOPHYSIOLOGY OF NEURODEGENERATION
Neuronal cells are extremely sensitive to changes in their environments. Using ion pumps to regulate internal and external electrolyte concentrations, they maintain homeostasis and allow cell membranes to transmit electrical signals. Local oxygen and/or glucose deprivation, local pH changes or uncontrolled depolarizations can result in energy depletion, which rapidly leads to transient or permanent neuron injury.

Stroke and traumatic brain injury (TBI) are the most acute neurodegenerative conditions which involve rapid cell death by both necrosis and apoptosis. The current understanding of stroke and TBI pathogenesis is important to interpret chronic neurodegenerative processes and has been derived from studies with cell lines and animal models which simulate human stroke. There is evidence that sensitivity to neuron damage following stroke may be shaped by several factors, such as type of brain tissue, duration of ischemia, extent of reduced blood flow, extent of reperfusion, and gender. Cerebral ischemia triggers an array of mechanisms, starting with the depolarization of pre-synaptic neurons and glutamate release. Post-synaptic glutamate receptors open calcium channels to increase levels of intracellular calcium, activating pathological pathways. There are three zones of ischemic pathophysiology, as shown in Figure 12.

In the first zone, vessel occlusion causes release of cytokines from endothelial cells and promotes leukocyte adherence, initiating an inflammatory response. Astrocytes, oligodendrocytes and macrophages also produce chemokines which cause the breakdown of the blood-brain barrier. In the second zone the ischemic tissues (core and penumbra), cells die by necrosis and apoptosis. In the third zone, the recovery zone or peri-ischemic area, viable cells release recovery-enhancing factors inducing axo-, synapto- and neurogenesis in the surrounding viable tissue and sub-ventricular zone during the postishemic phase of stroke (Lois, 1993).

43

Declining blood flow initially affects functional neuronal activity, and as ischemia progresses, metabolic activity is suppressed. Oxygen and glucose reduction cause reduced production of ATP, which is needed to maintain ion transporter function. Membrane potential is maintained by cell metabolism, which if compromised, leads to depolarization, release of glutamate and the activation of destructive pathways. These events occur in sequence and overlap, depending on the intensity and duration of the ischemic event. After an intense ischemic insult, the detrimental cascade of events will include glutamate-mediated excitotoxicity, calcium overload, activation of enzymes, oxidative stress, neurovascular changes, astrocyte and tissue changes, altered gene expression and cell death.

The extent of reduced cerebral blood flow (CBF) determines the degree of deprivation of cellular oxygen and glucose, and therefore the extent of damage. In normal conditions CBF is 50 to 60 ml/100 g/min. During cerebral ischemia, damage spreads from the center to the periphery, forming a focal ischemic gradient in such a way that maximum damage is at the center or core, where CBF falls to less than 7 ml/lOOg/min and cell damage is irreversible. The ischemic core is surrounded by a region of moderate ischemia called the ischemic penumbra, in which blood flow varies from 7 to 17 ml/100 g/min. The penumbra remains metabolically active but electrically silent. Figure 13 illustrates this typical pattern of brain damage: cells undergo necrosis in the core and apoptosis in the penumbra. A transient zone between the core and penumbra is likely to merge with the core if blood flow is not restored quickly enough, or if this zone is not pharmacologically protected. The transient zone and penumbra are the targets of reperfusion and neuroprotective therapies, respectively (Astrup, 1981).

The different processes leading to neurodegeneration and cell death are described below. Understanding these processes is important to develop new targeted neuroprotective drugs.

44

Glutamate-mediated neurotoxicity and excitotoxicity

Excitotoxicity, in cultured neurons and neurological diseases such as epilepsy, schizophrenia, and also aging, results in a profound loss of dendritic spines, the sites of most NMDA receptors. Nevertheless, synaptic and extrasynaptic NMDA receptors have physiological and pathological roles (Arundine, 2003).

Glutamate is the major excitatory neurotransmitter in the mammalian brain and it is a key mediator of intracellular communication, plasticity, growth and differentiation (Aarts, 2002). In normal conditions, the extracellular concentration of glutamate is maintained in the micromolar range by glial uptake systems, and glutamate is responsible for postsynaptic signaling through ionotropic and metabotropic glutamate receptors. The ionotropic glutamate receptors include the 7V-methyl-.D-aspartate (NMDA), 2-amino-3-(3hydroxy-5-methylisoxazol-4-yl)-propionate (AMPA) and kainate (KA) receptor

subtypes, and activation of these receptors leads to increased sodium, potassium and calcium conductance (Arundine, 2003). Additionally, metabotropic receptors mediate glutamate actions through GTP-binding, protein-dependent mechanisms that result in the mobilization of calcium from internal storage sites (Arundine, 2003). Following cerebral ischemia, depolarization occurs due to a lack of ATP, which results in excessive release of glutamate (increasing 10- to 100-fold) and the breakdown of energy-dependent astrocytic functions, resulting in impaired glutamate uptake by astrocytes (Shashoua, 2003).

Severe focal ischemia, lasting more than 20 minutes, causes a 2.5- to 20-fold increase in the extracellular concentration of glutamate. The mechanism by which glutamate is released during ischemia is generally believed to be entirely calciumdependent, involving vesicular exocytosis from neurons. However, some authors have reported calcium-independent mechanisms, such as reverse operation of the glutamate uptake carrier due to reversed K+ and Na+ gradients. In normal conditions glutamate is co-transported with two Na ions into the cell, while one K and one OH" are transported

45

out of the cell. During anoxic depolarization, glutamate is pumped out of the cell until a new equilibrium is reached (Budd, 1998).

While cells in the ischemic core undergo anoxic depolarization and immediate necrosis, cells in the transient zone and penumbra can depolarize and repolarize successively. The numbers and frequencies of these depolarization-repolarization cycles determine the extent of cell death. Peri-infarct depolarizations (PID) are spontaneous waves of depolarization-repolarization that propagate to viable tissues from the transient and penumbral zones (Fabricius, 2006), triggering glutamate release and energy depletion. Sodium and calcium channels are involved in the production of PIDs (Ohta, 2001). PIDs can serve as important targets for neuroprotective therapy. It has been demonstrated that drugs which block sodium channels and NMDA receptors, thereby inhibiting PIDs, can decrease infarct size (Mies, 1993). In vitro blockade of sodium channels inhibits ischemia-induced glutamate and dopamine release (Callaway, 2003). The clinical impact of the pharmacological blockade of PID events might become the subject of future study, but it is currently necessary to establish the impact that these events have on human stroke outcomes (Fabricius, 2006).

In normal physiological conditions, the quiescent NMDA glutamate receptor channel is blocked in a voltage-dependent manner by Mg2+. Mild depolarization
9+

dislodges Mg from the channel. Overstimulation of NMDA receptors induces the

9+
_]_

intracellular accumulation of ions and second messenger molecules including Ca , Na , inositol triphosphate (IP3), and diacylglicerol (DAG), with the most important of these being the large influx of Ca2+ (Budd, 1998). Glutamate activation of the NMDA receptor causes a complete opening of the channel and is the predominant route of Ca influx.

Reports indicate that some AMPA receptors in selected brain regions can support calcium permeability (Budd, 1998; Arundine, 2003). Another possibility is the interaction of AMPA receptors with submembrane proteins to produce cell signaling. This is the case for GRIP1 and GRASP-1 proteins, which couple AMPA receptors to Ras and INK signaling (Ye, 2000; Sheng, 2001; Aurundine, 2003). GRASP-1 is a cellular substrate for 46

caspase-3, which in pathological conditions incapacitates the function of AMP A receptors, stimulating expression and increasing receptor numbers in the synapse, thereby rendering neurons more susceptible to excitotoxicity (Ye, 2000). The least studied of the glutamatergic receptors, the KA receptors, are heteromeric receptors formed by KA subunits which can be mixed with AMPA subunits, leading to diverse functions in the central nervous system. KA receptors, which are characterized by rapid desensitization, participate mostly in Na+-dependent depolarization, leading to Ca2+ influx through voltage-activated calcium channels (VACCs; Budd, 1998). Phosphorylation of KA receptors increases their permeability to Na+, leading to increased neuron firing. Calcium-calmodulin dependent kinases are responsible for the phosphorylation of subunit GluR6 of the KA receptor (Hao, 2005).

However, most neuron degeneration following stroke is produced by the glutamatergic activation of NMDA receptors and the consequent influx of Ca . The NMDA receptor has a key role in many functions of the CNS, including cognitive processes, memory acquisition and learning. As with AMPA and KA receptors, NMDA receptors are assembled from different subunits, resulting in specific ligand affinities and/or ion conductances with unique intracellular signaling characteristics. This diverse heteromeric conformation displays regional patterns in both developing and adult brains. These differences are directly related to physiologic properties such as the Ca2+/Na+ permeability ratio, voltage-dependent Mg +-blockade, affinity for glutamate, and requirement for glycine as a co-agonist (Budd, 1998).

NMDA agonists open channels with a linear current voltage relationship and a depolarizing potential near 0 mV. Mg blocks the channel at negative membrane

potentials and is not active during states of depolarization. The response of channels to glutamate is greatly enhanced by low concentrations of glycine acting on glycine-B receptors. Some have suggested that glycine binding is an absolute requirement for NMDA channel opening (Albers, 1999).

47

Protein-protein interactions govern signal transductions of many receptors. Ionotropic glutamate receptors are organized into multiprotein signaling complexes within the postsynaptic density (PSD) in a disk-like structure, which are involved in synaptic plasticity and anchoring of NMDA receptors to the post-synaptic cytoskeleton (Sheng, 2001). Other authors name these proteins membrane-associated guanylate kinases (MAGUK), members of a superfamily of sub-membrane proteins involved in receptor clustering on the plasma membrane (Aurundine, 2003). A prominent organizing protein belonging to the MAGUK superfamily is PSD-95 (Aarts, 2002), which couples the NMDA receptor to intracellular proteins and to signaling enzymes. PSD-95 allows the formation of a ternary complex between the NMDA receptor, PSD-95, and the nNOS enzyme (Sheng, 2001). This molecular scaffolding brings NMDA receptors into close proximity to nNOS and explains the preferential activation of nNOS by Ca entering the NMDA receptor as opposed to Ca2+ entry through other channels (Aarts, 2002; Arundine, 2003). This also helps to explain the relationship between glutamate, NMDA receptors, and Ca mediated excitotoxicity. It has been reported that perturbation of the interaction between PSD-95 and nNOS has a neuroprotective effect without affecting other actions of the NMDA receptor (Aarts, 2002). It has been demonstrated that the interaction of the NMDA receptor and PSD-95 activates nNOS, and that subsequently NO activates the stress-activated protein kinase p38, which leads to the activation of cell death mechanisms. Suppressing expression of PSD-95 in neuronal cell cultures does not alter expression of NMDA receptors, calcium loading through NMDA receptors, nor the expression of nNOS, but it does alter calcium-activated NO production (Aurundine, 2003).

Ion movements and calcium overload

The neuron membrane potential begins to change within 1 5 - 9 0 seconds after ischemic or traumatic insult. At this time, hyperpolarizing and depolarizing changes in membrane potential are reported, even in the damaged brain region. Some neurons are hyperpolarized and then depolarized, while other neurons are primarily depolarized
48

(Budd, 1998). In most neurons there is enhancement of K conductance, increasing the extracellular concentration of K+ from 3 mM to 50 - 80 mM, which leads to depolarization at -20 mV. This is secondary to reductions of intracellular concentrations of ATP and the consequent inhibition of Na+/K+ ATPase activity (Budd, 1998). In the first seconds after an ischemic/anoxic insult, there is also an interstitial and intracellular decrease in pH. This results from a lack of oxygen and stimulation of anerobic glycolysis, enhancing both lactate production and intracellular accumulation of H+ ions (Simon, 2006). While acidotic injury is one of the oldest theories proposed to explain brain damage following ischemic stroke, precisely how low pH could damage cells still remains unclear (Budd, 1998; Simon, 2006). It is generally accepted that Ca2+ ions are responsible for mediating neuron death due to glutamate excitotoxicity. However AMP A, KA and NMDA receptors also mediate the Na+ influx, which further contributes to the elevation of Ca + or directly produces neuron damage. Nevertheless, during a sustained excitotoxic insult, Na+ influx is influenced more by the lack of ATP-dependent Na+ extrusion than by receptor-mediated Na+ influx (Budd, 1998). Cellular sodium accumulation produces swelling of neuron bodies, presumably mediated by the osmotic potential of Na+ which causes the influx of CI" and water. Other ion exchanges altered by Na+ accumulation are Na+/H+ exchange involved in regulation of intracellular pH and mitochondrial Na+/K+ and Na+/Ca2+ exchangers, leading to altered mitochondrial function. Separating the toxic effects of Na+ and Ca2+ is difficult (Budd, 1998).

Calcium ions have important roles in the regulation of cellular physiological processes, as mediators of neurotransmitters and hormonal activity, as activators of enzymes, as modulators of gene expression and neurotransmitter release and as modulators of membrane plasticity in the central nervous system. There are specific processes to maintain both neuron homeostasis and Ca concentrations. The extracellular concentration of Ca2+ is approximately 1 mM, compared to 100 nM in the neuronal cytoplasm. This difference is achieved via mechanisms regulating Ca2+ influx, buffering, storage and extrusion. 49

Calcium ions can be derived from activity of the Na /Ca

exchanger. Membrane

depolarization, as occurs in PID, causes intracellular Na+ accumulation, which drives the Na+/Ca2+ exchanger in reverse. This causes neurotoxic calcium accumulation in the central nervous system in both gray and white matter (Budd, 1998; Arundine, 2003). It has been hypothesized that PID provides this mechanism of Ca influx into the

periinfarct cortical tissue, and that the gradual recruitment of damaged cells into the penumbra can be caused by Ca transients accompanying recurrent PIDs. If repeated

PIDs are generated, they result in a progressive breakdown of neuron function and ion homeostasis, contributing to the spread of infarction in cerebral focal ischemia. Recurrent Ca + influx is a mechanism that presumably contributes to this process, as has been demonstrated in cats (Ohta, 2001).

Neuronal nicotinic acetylcholine receptors possessing the alpha-7 subunit are highly permeable to Ca2+, to a greater extent than NMDA receptors. High levels of alpha7 subunit expression in the hippocampus and cerebral cortex make these brain regions more susceptible to ischemia (Arundine, 2003). The exact role of nicotinic receptors in brain damage following stroke is not fully understood.

Neurons also express ATP-gated ion channels (e.g. P2X receptors) distributed in the peripheral and central nervous systems, which mediate fast excitatory transmission at nerve-muscle and nerve-nerve synapses. P2X receptors are gated by ADP, a product of ATP hydrolysis which rises due to metabolic impairments during anoxia (Arundine, 2003). These receptor channels are also permeable to Ca2+ ions. Activation of the P2X receptor has been demonstrated to mediate rapid increases in intracellular Ca concentrations. Calcium-dependent neurotoxicity is triggered most efficiently when Ca2+ influx occurs through NMDA receptors and cannot be reproduced by loading neurons with equivalent amounts of Ca2+ through non-NMDA receptors or voltage-sensitive calcium channels (VSCC; Arundine, 2003). The localization of NMDA receptors is governed by 50

the interaction with a-actinin, an actin binding protein, which anchors the receptors to cytoskeletal elements at the synaptic border. Calcium/calmodulin has been found to antagonize this interaction (Shirasaki, 2006). Therefore, Ca entering through the

NMDA receptor interacts with calmodulin and displaces the NMDA receptor from the actin cytoskeleton, leading to redistribution of NMDA receptors to extrasynaptic sites (Arundine, 2003). NMDA receptors are equally capable of triggering excitotoxicity both within and outside the synaptic boundries. Glutamate-induced Ca2+ influx leads to the activation of a variety of potentially toxic Ca -dependent events such as protein kinase C (PKC) activation, phospholipase A2-induced release of arachidonic acid forming eicosanoids and ROS by metabolic processes, sustained protein phosphorylation, DNA fragmentation caused by

endonuclease activation, calpain-induced cytoskeletal breakdown, and depolymerization of microtubules (Budd, 1998).

Neurotransmitter and hormone homeostasis

The brain has various endogenous neuroprotectant factors which act to modulate neurotransmitter release or to activate defensive mechanisms preventing

neurodegeneration and cell death. This is the case for dopamine (DA; Callaway, 2003), (Gardner, 2005), adenosine, melatonin, estradiol and endocannabinoids (Gardner, 2005; Hansen, 2002).

Dopamine has been shown to contribute to subsequent neuron damage following ischemia primarily in the striatum, due to dopaminergic systems and to auto-oxidative degradation which produces ROS (Callaway, 2003). The striatum is generally the region which represents the core of the ischemic lesion following MCA occlusion, because the striatum receives its blood supply exclusively from the lenticulostriate end arteries. This limited blood supply makes the striatum particularly susceptible to oxidative stress and 51

also extremely resistant to neuroprotective treatments (Callaway, 2003). In pathological conditions DA can cause cell damage and neurotoxicity, as in age-related neurodegeneration, Parkinson's disease and stroke. Dopamine induces apoptosis through activation of the INK pathway and increases expression of the c-Jun protein, which is an activator of apoptosis. Another mechanism involves ROS produced during DA metabolism, which activate apoptosis by oxidative stress on the cell. This second mechanism is more likely involved in cases of age-related neurodegeneration (Luo 1998).

Levels of endogenous cannabinoid receptor ligands, anandamide and other Nacyl-ethanolamines (NAEs) increase in the brain during neuronal injury (Berger, 2004; Franklin, 2003; Shouman, 2006). This process could be linked to an endogenous neuroprotective response (Hansen, 2001; Hansen, 2002). This has been extensively discussed in the section on the endocannabinoid system.

Hormones are also involved in endogenous mechanisms of neuroprotection. Estradiol, progesterone (Cutler, 2007; Stein, 2008) and melatonin (Antolin, 1996, 2002) all have neuropotective effects on the central nervous system. These hormones protect and rebuilt the blood-brain barrier, down-regulate the inflammatory cascade, and limit cell necrosis and apoptosis in some cases (Stein, 2008).

Estrogens and estradiol are multifunctional molecules with multiple cellprotective mechanisms, and are present in the brain in both physiological and stress conditions. Estradiol acts through a nuclear hormonal receptor, affecting transcription of a variety of genes intended to maintain homeostasis in the body. The cells of the neurovascular unit are sensitive to changes in estrogen levels and may act as modulators of cerebral ischemia and cell death. There is a variety of mechanisms that might be involved in estrogen neuroprotection, such as increased expression of the neurotrophic factors NGF, BDNF, and bFGF and their receptors (Hum, 2003, 2005); increased expression of Bcl-2 (Alkayed, 2001); and increased expression of kinases (Hum, 2003). Nevertheless women continue to sustain low stroke rates beyond menopause, suggesting that hormonal factors are not solely responsible. New evidence suggests that cell death 52

mechanisms are not identical in genetically male (XY) and female (XX) cells. Activation of enzymes such as nNOS, iNOS and PARP-1 might result in different protective mechanisms in male and female cells (Hum, 2005).

The hormone melatonin has been considered to be exclusively involved in the control of circadian physiology and seasonal reproductive events. Nevertheless, in recent years melatonin has been identified as an antioxidant, capable of trapping free radicals in cells. Research has demonstrated that melatonin can be 5-fold more potent than glutathione as a hydroxyl radical scavenger, and that it can induce the activity of glutathione peroxidase in rat brains (Antolin, 1996, 2002). This antioxidant activity might account for some of the neuroprotective actions of melatonin in animal models of stroke and TBI.

Activation of enzymes and transcription factors

As mentioned above, glutamate receptor activation causes calcium influx and the activation of neuronal nitric oxide synthase (nNOS). Uncontrolled nNOS activity produces excess NO and other oxides which lead to production of the potentially harmful peroxynitrite, H 2 0 2 and other ROS (Iwashita, 2004). The nuclear enzyme poly-ADP ribose-polymerase-1 (PARP-1), an enzyme involved in DNA repair in normal physiological conditions, converts nicotinamideadenine-dinucleotide (NAD) into polymers of poly-ADP-ribose (PAR) which are intended to regulate nuclear homeostasis (Cipriani, 2005). Nevertheless, PARP-1 promotes cell death and neurodegeneration when overactivated by DNA damage caused by ROS or other genotoxic stressors. Activated PARP consumes NAD and consequently ATP, causing mitochondrial stress and translocation of the apoptosis inducing factor (AIF) from mitochondria to the nucleus (Iwashita, 2004; Cipriani, 2005). This observation led to the "suicide hypothesis", in which rapid catabolism of NAD affects 53

cellular energy metabolism and ultimately leads to cell death (Hong, 2004). Thus, PARP activation has been implicated in the pathogenesis of ischemic brain injury, traumatic brain injury and Parkinson's disease (Iwashita 2004). It has been reported that PARP-1 prompts a cascade of events leading to PAR-dependent mitochondrial dysfunction and rapid release of AIF, to activate cell death processes by caspase-independent apoptosis (Hong, 2004; Cipriani, 2005).

Kinases are another important family of signaling enzymes in neuronal cells. Some kinases support neuroprotection both during and after ischemia, while others are involved in the activation of apoptosis or autophagia in cell death. Calcium ions activate calmodulin and the resultant calcium-calmodulin (CaM) complex activates the CaMdependent kinases (CaMKs). CaMKs are involved in several calcium-mediated processes such as neurotransmitter synthesis and release, and modulation of receptors, ion channels, gene expression and neurite outgrowth (Mehta, 2007). Calcium/calmodulin-dependant protein kinase II (CaMK II) is present in the post-synaptic density (PSD) and is activated by Ca2+ influx through the NMDA receptor ion channel. Activated CaMK II phosphorylates and activates the GluR6 subunit of the kainite receptor, increasing its permeability and subsequent cell damage (Hao, 2005). There is also evidence that CaMKs phosphorylate transcription factors, including cAMP responsive element binding protein (CREB). This constitutive protein, found in cells, is activated by phosphorylation of serine-133 to promote transcription of genes with an up-stream cAMP responsive element known as CRE (Hu, 1999). Activation of CREB has been shown to regulate expression of Bcl-2, a survival protein which inhibits apoptosis. Additionally CREB activation is involved in changes in synaptic structure and function during long-term synaptic plasticity (Hu, 1999; Mehta, 2007). Therefore, CaMKs might be linked to neuroprotective processes and could constitute a possible target to treat both brain damage and neurodegeneration.

In contrast to the neuroprotective effect of CaM kinases, there are other proapoptotic kinases. Death-associated protein kinases (DAPKs) have been found to be regulated by CaM, and they can mediate both membrane blebbing during apoptosis and 54

formation of vesicles in autophagia. DAPKs, which have a key role in cell death during stroke, mediate apoptotic pathways, disrupt matrix survival signals and suppress integrinmediated cell adhesion (Shamloo, 2005). DAPKs have also been implicated as tumor suppressors, and their inactivation has been associated with malignancy. DAPK expression is frequently lost in tumors due to hypermethylation of the DAPK gene, and thus inhibition of DAPKs leads to neuroprotection, and DAPKs are a relevant target for therapeutical agents to treat brain injury (Shamloo, 2005). Phospholipase A2 is also activated by the influx of Ca +, which in turn releases arachidonic acid from cell membranes. Arachidonic acid is a substrate for the production of proinflammatory factors and oxidative species (stress factors) which activate mitogenactivated protein kinases (MAPKs). MAPKs are thought to be important contributors to the cell survival/damage cascade of events. MAPK cascades comprise the three kinase module consisting of MAPK, MAPK kinase (MAPKK or MEK) and MAPK kinase kinase (MAPKKK or MEKK). External signals such as increased glutamatergic transmission, calcium levels or stress factors can activate enzymes of the MEKK family. There are three well known enzymes in the MEKK family: extracellular signal-regulated kinase (ERK), p38, and c-jun-JV-terminal kinase (JNK). When activated, these enzymes phosphorylate members of the MEK family such as c-Fos, which activates an early gene inductor for cell survival, and/or Bcl-2, which activates Bax to activate apoptosis (Guan, 2005, 2006). In this way, ischemia/reperfusion injury activates JNK, which in turn activates DP5 (one member of the BH3-only proteins), and when activated DP5 interacts with Bcl-2, decreasing its activity as a survival factor or freeing Bax to activate apoptosis (Guan, 2006).

55

Oxidative stress
Oxidative injury and the resulting apoptotic and necrotic death of neurons is the major pathological factor involved in neurodegenerative diseases such as amyotrophic lateral sclerosis (ALS), Parkinson's disease (PD), Alzheimer's disease (AD), TBI and stroke. Protein oxidation and the aggregation of proteins resulting from oxidative stress have a role in cell death (Sarang, 2002).

During normal cell function, free radicals are generated by oxidation of nutrients for the generation of energy. The natural and most effective free-radical scavenger system is superoxide dismutase 1 (SODi). Early after the onset of ischemia, free radicals are generated in endothelial cells of microvessels at the site of occlusion. This initial generation of free radicals begins the damage to the blood-brain barrier and induces the recruitment of inflammatory cells, which in turn generate additional NO and free radicals (del Zoppo, 2006). Hypoxia results in the production of ROS in mitochondria in the forms of superoxide anion radicals (O2"), hydroxyl anions (OH") and hydroxyl free radicals (OH), which lead to the activation of apoptotic cascades within mitochondria. A critical target for superoxide anion could be the NO produced initially by endothelia, macrophages, and neutrophils. Superoxide anion and NO are known to react to form peroxynitrite anion, which decomposes to form stronger oxidative species such as hydroxyl free radicals and nitrogen dioxide free radicals (Beckman, 1990).

Brain tissue is highly susceptible to free radical insults, since its high lipid content results in extensive lipid peroxidation. Excessive free radical production has a detrimental effect on cell function and on processes required for neuroprotection. Secondarily, overactivation of NMDA receptors during cerebral ischemia increases the formation of NO to toxic levels, by degradation to peroxynitrite and other ROS. Moreover, oxidative processes activated by the deficiency of oxygen and glucose are accompanied by the generation of hydrogen peroxide (H2O2) and more superoxide anion radical (O2").

56

Although restoration of blood flow is the most important goal in stroke therapy, reperfusion is responsible for additional oxidative stress due to the generation of free radicals during the restoration of oxygen and glucose delivery to ischemic tissues. This leads to the overloading and inactivation of natural antioxidant defense mechanisms of cells (Salom, 2004). Obtaining evidence for the involvement of ROS in the pathophysiology of cerebral ischemia has been difficult, due to problems with direct measurements of these species which are extremely reactive and rapidly degrade. The measurement of salicylate or 4-hydroxybenzoate hydroxylation by ROS, which form stable and detectable adducts, has been used as an index of brain ROS formation (Callaway, 2003). Free radical scavenging molecules (ascorbic acid and a-tocopherol) and coupled antioxidant enzymes (superoxide dismutase, catalase, and glutathione peroxidase) have detoxifying functions during brain injury. In the pathogenesis of PD, a lower content of glutathione has been found in the substantia nigra pars compacta (Dringen, 2000). Injury to the endothelium is the major consequence of

ischemia/reperfusion, causing edema due to a loss of barrier function and an enhancement of platelet adhesion to the endothelium. SOD injection into the circulation is protective to the endothelium in experimental models of ischemia/reperfusion stroke (Beckman, 1990). Mice lacking the gene for SOD] show larger volumes of infarcted tissue when subjected to focal stroke, compared to wild-type mice. On the other hand, overexpression of SODi in animal models has a neuroprotective effect in stroke (Del Zoppo, 2006).

Excessive production of ROS results in peroxidation of lipids in cell membranes and generation of toxic aldehydes such as 4-hydroxynonenal (4-HNE), which damage a variety of cell structures such as ion channels, transporters, receptors, the cytoskeleton and DNA.

57

Neurovascular changes and inflammation

The study of the nervous system and its neuroprotection has recently led to the shift from a focus on neurons alone to a focus on the complex of neurons, microvessels, and supportive glial cells referred to as the "neurovascular unit". This nervous system unit can vary in structure and function from site to site in the brain, and it represents a better model to study neurodegeneration and brain injury and a more realistic target for neuroprotective agents. Damage to any part of the neurovascular unit could affect the other components. The vascular component of the unit consists of endothelial cells attached together by a proteic junctional complex that includes "adherens junctions" (AJ), "tight junctions" (TJ), and possibly gap junctions (Hawkins 2005), which form the bloodbrain barrier (BBB) and rest on a second anatomical barrier, the basal lamina. Both layers of cells constitute the microvessels. The AJ and TJ act to restrict permeability across the endothelium. The TJ are the most important proteic junctions (constituted by occludin, claudin and JAM proteins) in the BBB (Hawkins, 2005). Gap junctions, if present, mediate intercellular communication with astrocytes, which act as functional connector elements between microvessels and neurons. Neurons can control the function of microvessels through the astrocytes, which transfer supportive elements from the blood for the normal function of neurons (Hawkins, 2005; del Zoppo, 2006).

The extracellular matrix (ECM) of the basal lamina contains microglial cells, which are the resident immune cells of the brain. The ECM interacts with the endothelial cells which serve as an anchor, via laminins and other matrix proteins, to endothelial integrin receptors (Hawkins, 2005). Matrix proteins can influence the expression of TJ proteins, indicating that they are involved in maintenance of the BBB (Hawkins, 2005). Therefore the concept of the neurovascular unit creates a framework for describing brain function and mechanisms of brain injury, and facilitates the understanding of possible mechanisms of neuroprotection. The interaction between these components is complex, dynamic, and makes the CNS a closed system, isolated by two anatomical and functional barriers: the BBB and the basal lamina, both serving to control permeability of chemical compounds and the migration of circulating blood cells.
58

During brain injury or neurodegeneration, any vascular damage or occlusion activates microglia to induce the release of proteases, lipases and glutamate, and together with further activation of white blood cells leads to disruption of the TJ in the microvascular endothelium of the BBB, which initiates degradation of the basal lamina (Yenari, 2006).

Formerly, proteases were thought to be associated with a loss of vascular matrix and were thought to be limited to zinc-dependent matrix metallo-proteinases (MMPs or gelatinases) and plasminogen activators (Fukuda, 2004). However, cathepsins B and L are generated rapidly after ischemic injury and are involved in the degradation of perlecan, the most sensitive matrix component identified thus far in the basal lamina (Fukuda, 2004). In human stroke and other pathologies, there is an overactivation of MMP-9 into the infarcted area, and increased activation in the periinfarcted tissue (Rosell, 2006). Degradation of the basal lamina results in the loss of astrocytes and TJ, leading to hemorrhagic transformations. MMPs, in particular MMP-9, as well as plasmin, plasminogen, serine proteases and cysteine proteases or cathepsins, are all increased in the brain within hours after focal cerebral ischemia in experimental stroke and in humans (Rosell, 2006). Injury to the BBB or to the basal lamina permits the extravasation of inflammatory cells and factors, which damage neurons (Fukuda, 2004). The plasma activity of MMP-9 is strongly correlated with MRI measurements of lesion growth in humans, suggesting its early role as a predictor of eventual brain damage (Rosell, 2006).

Cytokines and NO formed immediately after ischemia stimulate the expression of adhesion molecules on endothelial cells and leukocytes, leading to leukocyte adherence and extravasation into the brain (Vemuganti, 2004). Transendothelial migration participates in the progression of cell injury during brain injury. Highly specific receptorligand interactions with endothelium and the extracellular matrix are involved in this process. The protein responsible for this specificity is the intercellular adhesion molecule1 (ICAM-1; Furuya, 2001). Blockade of these receptor-ligand interactions has succeeded in some preclinical studies (Furuya, 2001) to decrease entry of white cells into the brain. 59

Neutrophils are the major type of white cells entering the brain after ischemia; by releasing ROS and promoting lipid peroxidation products, they increase both BBB disruption and cellular insult. Polymorphonuclear leukocytes accumulate progressively for at least 24 hours after ischemic stroke, and these cells remain elevated for approximately one week. This accumulation correlates with the severity of brain damage and the likelihood of a poor neurological outcome (Lees, 2003; Baron, 2005).

The inflammatory cascade which follows cerebral ischemia involves several mediators such as cytokines, chemokines, adhesion molecules, NO, and eicosanoids, all of which interact to produce a long-lasting inflammatory reaction, observed both in animal models and in patients with ischemic stroke. The production of pro-inflammatory prostanoids is an injurious mechanism related to COX-2 enzymatic activity, and this process is associated with the release of arachidonic acid from cell membranes to produce PGE2, a potent mediator of inflammation (Candelario-Jalil, 2004). Cellular levels of COX-2 mRNA and iNOS mRNA are increased within neurons, glia, and vascular cells after ischemic onset, and persist for periods of hours or days. This increase in enzymatic activity is correlated with both levels of PGE-2 and extent of brain damage in animal models (Candelario-Jalil, 2004). Nevertheless, increased COX-2 activity by itself does not lead to cell death, suggesting that COX-2 would only mediate neuronal injury in the context of an overall inflammatory response (Candelario-Jalil, 2004).

Glial cell changes

Neurons, which constitute less than 5 % of the cells in cerebral gray matter, depend on glial cells for metabolic and trophic support. Astrocytes, oligodendrocytes, microglia and Schwann cells are the types of glial cells found in the brain. Astrocytes are the major cell type in the brain and are responsible for the control of fluid movement between the intracellular and extracellular spaces around neurons. Within the neurovascular unit, astrocytes assist neurons in controlling synthesis of neurotransmitters 60

by providing substrates, and synaptic signaling via specific reuptake transporters for neurotransmitters. Astrocytes also support the integrity of the BBB and constitute the bridge between blood vessels and neurons, transporting in both directions neurotrophic factors, cytokines, hormones, and basic substrates required for overall functioning of the nervous system (Nakase, 2003).

Moreover, astrocytes possess the highest antioxidant activity among systems intended for the protection of neurons (e.g., glutathione, SOD and others). During brain trauma or stress, astrocytes can release growth factors and other molecules to facilitate neurogenesis and regeneration. It has been observed that in focal ischemia, astrocytes within the ischemic core remain viable after neuron death. Astrocytes are responsible for the formation of glial scars, dividing viable tissue from death cells.

Microglia are considered to constitute the resident immune system of the brain, and are also associated with the BBB (Yenari, 2006). Stressors activate microglial cells to undergo proliferation, chemotaxis and the generation of immunomodulatory molecules. The role of microglia in maintaining or interfering with BBB integrity has not been well studied. Research has shown that blocking activation of microglial cells can reduce edema and hemorrhagic transformations in experimental stroke (Yenari, 2006). Stroke lesions increase progressively after ischemia. While cells in the penumbra may die by apoptosis causing expansion of cell death, the exact role of astrocyte involvement is still controversial. The wave of PIDs progresses through astrocytes and is suspected to be responsible for expansion of the infarcted volume (Nakase, 2003).

Gap junctions or nexi are connections between astrocytes and neurons that lead to increased communication and transfer of molecules. Some studies have demonstrated that gap junctions are involved in the spread of brain damage in rodents (Nakase, 2003). On the other hand, activation of gap junctions can reduce the progression of apoptosis in infarcted tissue. Mice which lack connexin 43, one of the proteins integral to gap junctions, display increased apoptosis and brain damage after stroke (Nakase, 2003).

61

Although astrocytes are more resistant than neurons to most stress conditions, certain types of astrocytes such as glial fibrillary acidic protein (GFAP)-negative protoplasmic astrocytes, which predominate in the gray matter, could be equally or more sensitive than neurons to ischemia in vivo (Giffard, 2005). Studies have demonstrated that astrocytes can undergo apoptosis in vitro and in vivo. Since GFAP has been used to identify astrocytes in apoptotic conditions, it was incorrectly assumed that astrocytes are not involved in apoptotic mechanisms of cell death. It has been demonstrated that GFAPnegative astrocytes develop apoptosis to the same extent as neurons (Giffard, 2005)

Cell death
Cell death is the final event in the cascade of neurodegeneration and brain damage. Traditionally, cell death after brain injury was considered to be necrotic. However recent research has shown that cell death within hours and days after TBI and stroke is essentially apoptotic, and is necrotic only initially after the brain insult. The period of time in which cell injury is reversible, before activation of the apoptotic cascade, is an important target and opportunity for therapeutic intervention.

After trauma and/or blood flow arrest, regions of metabolically unattended tissue evolve from an initial stage of reversible damage to a stage in which the core becomes necrotic. Necrosis occurs in response to acute anoxia, a sudden shortage of nutrients and rupture of osmotic homeostasis, which causes ATP to be unavailable to control internal ion concentrations. Cells swell as they take up water and the plasma membrane breaks, while the surrounding tissue undergoes an inflammatory response as cell contents leak. In this situation necrosis is a rapid and non-programmed cell death (Dietz, 2007).

If ischemia is incomplete, neuron survival is dependent on residual perfusion and oxygen availability. Residual perfusion in the ischemic area is dependent on collateral vessels and local perfusion pressures leading to the transient and penumbral zones of 62

ischemic tissue. In these zones the most important characteristic is delayed neuronal degeneration by apoptosis, an active process involving neuroactive factors and synthesis of new proteins, under very strict transcriptional control (Harrison, 2001). Morphological and biochemical features of apoptosis have been reproducibly detected in the ischemic brain, including cell membrane protrusion, chromatin condensation, formation of apoptotic bodies, and internucleosomal DNA degradation (Linnik, 1993; Abe, 2004).

Apoptosis, or programmed cell death, is an efficient physiological process responsible for the maintenance and homeostasis of living tissues and systems, involving normal cell turnover, immune system processes, embryonic development and hormonedependent atrophy (Cohen, 1997). The components of these processes are genetically encoded and remain inactive until they are stimulated by degeneration, stress or death signals. The function of apoptosis is to remove defective or harmful cells. Disordered apoptosis leads to pathological conditions such as cancer and autoimmune diseases; overactive apoptosis can result in neuron death in stroke and other neurodegenerative diseases. Apoptosis is essential and appropriate for modeling the developing nervous system. Half of the neurons produced in utero will be terminated via apoptosis, due to the insufficient reception of growth factors (Linnik, 1993). There are two classes of apoptosis: mitochondrial-mediated and death receptor-mediated. The mitochondrial mediated class can be further divided into two subclasses: caspase-dependent apoptosis and caspase-independent apoptosis (Hong, 2004).

Caspases are cysteine proteases characterized by almost absolute specificity for aspartic acid residues in the PI position. All caspases contain a conserved QACXG pentapeptide active site motif (where X is R, Q or G; Cohen, 1997). Caspases are synthesized as inactive proenzymes in polypeptidic chains, which will produce one large and one small subunit during activation (Cohen, 1997). The active enzyme is a heterotetramer, containing two small and two large subunits. The caspase family has been divided according to characteristics such as sequence homology, substrate specificity, structure and function (Harrison, 2001). According to function, the caspase family is divided into two groups: caspases activated during apoptosis: -2, -3, -6, -7, -8, -9, and 63

10, and caspases activated during inflammation: - 1 , -4, -5, -11, and -12. Apoptotic caspases can be divided into two sub-groups: initiator caspases: -2, -8, -9, -10 and -12, and effector or executor caspases: -3, -6, and -7, which are responsible for dismantling cell structure (Love, 2003).

Mitochondrial-mediated apoptosis is started by apoptogenic proteins such as cytochrome c, and second mitochondria-derived activator of caspases (Smac) or DIABLO, which are released from mitochondria into the cytosol. DNA damage and mitochondrial membrane perturbation are triggering factors for the release of both proteins. Calcium overload is the main factor of mitochondrial membrane perturbation which causes mitochondrial depolarization and the opening of permeability transition pores (PTP; Budd, 1998). Cytochrome c is involved in formation or multimerization of the apoptosome with participation of apoptotic protease-activating factor 1 (Apaf-1) and procaspase 9 in the presence of ATP. Such complexes lead to the activation of caspase 9 (Harrison, 2001). The apoptosome and caspase 9 cleave and activate downstream caspases such as caspase-3, caspase-6 and caspase-7, which have an executionery role in cell death (Cohen, 1997; Love, 2003).

Smac, a key apoptosis promoter, is released concurrently with cytochrome c, and interacts with inhibitory apoptotic proteins (IAPs), eliminating their inhibitory effects on apoptosis (Chai, 2000). Typical IAPs are XIAP, c-IAP-1, c-IAP-2, melanoma/livin-IAP, ILP-2 (IAP like protein 2), NAIP (neuronal apoptosis inhibitory protein), Bruce/Apollon, and survivin. IAPs are inhibited by physical interaction, so that apoptosis is promoted by at least two mechanisms: induction of the proteolytic activation of procaspase-3, and promotion of the enzymatic activity of mature caspase 3. Smac normally functions as a dimer, however monomelic Smac exhibits lower activation of procaspase-3 and is inactive in promoting the enzymatic activity of mature caspase 3. Thus, Smac appears to be a master regulator of apoptosis in mammals (Chai, 2000).

Activation of caspases during apoptosis results in the cleavage of critical cellular substrates, including PARP-1 (a DNA repair enzyme), Ul-70kD (an mRNA splicing 64

enzyme), DNA-PK (a DNA double strand break repair enzyme), Gas2 (a component of the microfilament system), protein kinase C-d (cleaved to its active form in apoptosis), Pro-IL-16 (cleaved to a mature active cytokine), lamins (which assist in nuclear shaping), huntingtin (Huntington disease gene product), fodrin (a membrane associated cytoskeletal protein), Rb (a cell cycle regulatory protein) and histones. Many of these precipitate the dramatic morphological changes seen during apoptosis (Cohen, 1997).

The Bcl-2 proteins, a family of proto-oncogenes which cause B-cell lymphocytic leukemia, constitute another molecular factor related to the activation of caspases and apoptosis. This family contains approximately 30 members, which can be divided into two groups: Bcl-2-like survival factors (Bcl-2 and Bcl-xL) or antiapoptotic Bcl-2 (Dietz, 2007), and Bcl-2-like death factors or proapoptotic Bax and Bad. Bcl-2-like survival factors contain hydrophobic domains, which permit the proteins to be anchored in the mitochondrial membrane, thus promoting membrane stability and preventing the release of cytochrome c and Smac (Abe, 2004). However, Bcl-2-like death factors do not contain the same number of hydrophobic domains, which prevents them from anchoring to the membrane, but allows them to interact with the Bcl-2-like survival factors forming clusters, removing them from the membrane, and contributing to the opening of pores in the mitochondrial membrane. The expression of Bax has the pro-apoptotic functions of perturbing membrane integrity and promoting release of cytochrome c and Smac. Bad and Bax are phosphorylated by the serine-threonine kinase Akt, which in the phosphorylated form is rendered inactive (Abe 2004). Interaction of Bad or Bax with Bcl2 or Bcl-xL is dependent on their dephosphorylation and translocation to the mitochondria (Abe, 2004). It has been shown that calcineurin, a calcium-dependent protein phosphatase, contributes to apoptosis by dephosphorylating Bad and Bax, which facilitates their translocation from mitochondria and the subsequent release of cytochrome c (Abe, 2004).

Another pathway to apoptosis activation is through the death receptors (DR), which are present in neuron cell membranes. Death receptors belong to the family of tumor necrosis factor receptors (TNFR). Activation of the death receptor by cytokines or 65

by agonistic antibodies activates caspase-8, which possesses an TV-terminus with FADD (Fas-associating protein with dead domain), similar to the death effector domains, thereby providing a direct link between the receptor and caspases. Caspase-8 recognizes adaptor molecules as receptor-interacting proteins (RIPs), which bind to the pro-domain of caspase-2 and recruit it to the signaling complex. The signaling complex activates caspase-1, which in turn activates the machinery of caspases such as caspase-3 and caspase-7 (Cohen, 1997; Love, 2003).

Poly-(ADP-ribose)-polymerase (PARP), also known as PARS (poly-ADP-ribose synthetase), is a chromatin-bound nuclear enzyme located in cell nuclei of various organs, including the brain. PARP is known to have an important role in cell damage after ischemic or oxidative injury. It is activated by single-strand DNA breaks, initiating an energy-consuming repair cycle by transferring ADP-ribose units to nuclear proteins intended for DNA restoration. An indirect result of this process is the rapid depletion of intracellular NAD and ATP pools, which increases the rate of glycolysis and mitochondrial respiration, leading to cell dysfunction and eventual cell death (Ding 2001). There are 17 types of PARP; type 1 or PARP-1 is present in the brain and is responsible for a different pathway of apoptotic activation. Other types have DNA repair functions in different tissues.

Caspase-independent apoptosis is another pathway of cellular self-destruction. The apoptosis-inducing factor AIF is a 67 kDa flavoprotein with significant homology to plant ascorbate reductases and bacterial NADH oxidases. Distributed in all tissues and cells of the mitochondrial inner membrane, its translocation from the mitochondria to the cytoplasm triggers catabolic reactions of apoptotic cell death. Overactivation of PARP-1 leads to AIF release, in turn recruiting proteases and nucleases and transporting them to the nucleus to cause cromatin condensation and DNA fragmentation, resulting in cell death (Hong, 2004; Cipriani, 2005). AIF has a dual function in cells. It is an important catalyst for redox reactions, as long as it is maintained in mitochondria. In conditions of stress, calcium overload, and perturbation of mitochondrial integrity, AIF can be released

66

as a death factor which cannot be blocked by caspase inhibitors or by the overexpression of Bcl-2 (Budd, 1998).

Ischemic neuronal cell death can be characterized by a mixture of apoptotic and necrotic cellular changes, as revealed by electron microscopy studies. Proteases which are not primarily involved in apoptosis, including cathepsins and calpains, are more relevant to necrosis. Nevertheless they have also been found in ischemic neurons undergoing apoptosis (Unal-Cevik, 2004). Evidence based on detection of caspases and cathepsins has led to the conclusion that necrosis predominates in the core ischemic area. In vitro findings suggest that routes of necrotic and apoptotic cell death are concomitantly activated in ischemic neurons in a gradient from the core to the mildly ischemic area where apoptosis predominates (Unal-Cevik, 2004). Cell death involving both mechanisms is referred to as necroptosis.

Gene expression
Most of the cellular changes in ischemia and neurodegeneration are mediated by transcriptional and translational activities. Cerebral ischemia is a powerful stimulus for the expression and up-regulation of numerous gene systems (Read, 2001). Preclinical models of spontaneous stroke can be applied to humans to identify genes associated with the risk of stroke, and to investigate how changes in gene expression during stroke are associated with post-stroke brain injury, resolution of brain injury, and brain recovery processes (Read 2001). Genes related to metabolism, cell communication and signal transduction are downregulated, and the expression genes involved in stress responses are increased for a number of hours following experimental brain damage. Gene expression after ischemia is observed as temporal episodes or waves of expression of different groups of genes. The gene families shown to be most activated are: immediate early genes (IEGs) such as NGFI, erg, NF-kB, activating transcription factor, c-fos, c-jun (Gillardon, 1996); cytokine and cytokine receptor genes such as interleukins IL-1, IL-2, 67

IL-6, IL-10, tumor-necrosis factor a (TNF-a), transforming growth factor (TGF), LIF and SOCS-3; inflammation genes such as COX-1, COX-2, monocyte chemoattractant protein-1 (MCP-1), z'-nos, MCP-3; apoptotic genes such as Bax (Gillardon, 1996), caspases, Fas, Fas-L, death receptor, and Arc (Read, 2001); and antiapoptotic genes such as Bcl-2 survival factors (Gillardon, 1996; Guegan, 1998). There is evidence that proteins encoded by the Bcl-2 gene family have a major role in the regulation of apoptosis, and that the ratio of cell death repressors Bcl-2 and Bcl-xL to cell death effectors Bax and Bak decreases at the onset of experimental ischemia (Gillardon, 1996; Love, 2003).

There are many effectors of apoptosis. Marked differences have been noted in the expression profiles of the caspase genes following induction of focal ischemia, with increased expression of caspases -1, -3, -6, -7, -8, -11, unchanged expression of caspase-2 and reduced expression of caspase-9 (Harrison, 2001). Growth factor genes such as vascular endothelial growth factor (VEGF), VEGF-receptor, neuroendocrine growth factor (VGF), brain-derived growth factor (BDGF) and nerve growth factor (NGF) are also involved (Guegan, 1998; Read, 2001). Additionally, other genes such as those coding for heat shock proteins (HSP), macrophage inflammatory protein (MIP), osteopontin, osteoactivin, TIMP-1, CD-14, and CD-44 appear altered (Read 2001). The detection of increased expression of genes after stroke has been accomplished in animal models using the standard techniques of Northern blotting, reverse transcriptionpolymerase chain reaction (RT-PCR), in-situ hybridization, and with more complex techniques such as subtractive library/subtractive hybridization, differential

hybridization, serial analysis of gene expression, representational difference analysis (RDA) and differential display (Read 2001). IEGs, transcription factors and HSPs are expressed as early as 30 min after cerebral ischemia in rats. Apoptosis and inflammation genes are expressed following cerebral ischemia in rats, with maximal expression from 4 to 24 hours after ischemia (Read, 2001).

Increased expression of pro-inflammatory genes, including cytokines (IL and TNF a), chemokines (MIP-1-a and MCP-1), and adhesion molecules (intercellular

68

adhesion molecule-1 [ICAM-1], E-selectin and P-selectin), contributes to the neuron damage following focal cerebral ischemia (Vemuganti, 2004).

Synaptic protein genes are downregulated during the seven days following ischemic stroke, but growth factor genes display increased expression, starting 24 hours after stroke and persisting for 7 days post-injury. Nerve growth factor (NGF; Guegan, 1998) and brain-derived neurotrophic factor (BDNF) are expressed and have a neuroprotective function (Shashoua, 2003), which involves the activation of transcription factor AP-1 that turns on neuronal growth genes and regulators of synaptic strength and synaptic number in the CNS (Shashoua, 2003).

The study of gene expression after cell damage or brain injury is important to identify potential targets for development of new drugs to treat or prevent brain damage and neurodegeneration. As the investigation of specific biological actions of drugs has been facilitated by the use of large-scale oligonucleotide microarray analyses in cell culture (Sarang, 2002), several genes have been identified as differentially expressed in response to neuroprotective drugs in cell cultures.

The gene for the tissue inhibitor of metallo-proteinase-1 (TIMP1), upregulated during periods of repair, is involved in cell proliferation and cell survival. TIMP1 can specifically inhibit apoptosis and confer resistance to oxidative stress. It has been proposed that matrix metalloproteinase-1 mRNA levels are increased in cells during oxidative stress (Gonzalez, 2000) and that neuroprotectants might control its activity through the TIMP genes (Han, 2001; Sarang, 2002). Levels of TIMP-1 mRNA are significantly elevated at 24 hours and 2 days after preconditioning, which corresponds well with the onset of ischemic tolerance (Read, 2001).

Ret-proto-oncogene (RET) encodes a cell membrane tyrosine kinase receptor protein, ligands of which belong to the glial cell line-derived neurotrophic factor family. RET was decreased more than two-fold in neuroblastoma cells subjected to oxidative

69

stress when treated with neuroprotective drugs. The role of the RET gene in oxidative stress and in neuroprotection needs to be elucidated (Sarang, 2002).

Clusterin (Apolipoprotein J) is a glycoprotein which has been implicated in cytoprotection, and which is induced by several cell stressors. Overexpression of clusterin can be associated with cell survival after oxidative injury, and clusterin has been thought to protect cells from apoptosis and neurodegeneration. In AD, lowered cellular expression of clusterin has been suggested to be associated with neuronal degeneration and death. In neuroblastoma cell cultures subjected to oxidative stress, clusterin was upregulated, confirming its likely involvement in neuroprotection (Sarang, 2002).

Galanin is a secreted 29-amino acid neuropeptide, which colocalizes with acetyltransferase and has been implicated in cell injury recovery processes in neurons (O'Meara, 2000). Cytokines and galanin have been suggested to function in a molecular cascade, mediating injury-induced regeneration. Neuronal growth factors induce the expression of the galanin gene, and mutations of this gene result in the loss of cholinergic neurons (O'Meara, 2000; Sarang, 2002). It was found that exogenous galanin could block cell damage induced by hydrogen peroxide in neuroblastoma cells (Sarang, 2002).

Growth associated protein (GAP43), a synaptic 43 kD protein involved in regeneration and neuronal plasticity, is an indicator of synaptic sprouting in the brain. In AD patients, GAP43 levels are decreased in the early progression of the disease (Masliah, 2001). Neuroblastoma cells subjected to oxidative stress display increased levels of GAP43 in response to neuroprotectants (Sarang, 2002).

70

EXPERIMENTAL MODELS TO STUDY NEUROPROTECTANTS

The discovery of new chemical entities with potential to be neuroprotectants requires the use of specific strategies mimicking neurodegeneration and neuron death. Several in vitro and in vivo models have been developed for drug screening in preclinical studies.

In vitro models
Cell culture and cell preparations have been used in neuroscience for a long time. The techniques of cell and molecular biology permit the study of neuron physiology, neuronal disorders and mechanisms of action of new drugs. In vivo models can fail to control complicating factors such as individual variability and cerebrovascular changes that might affect the assessment of pharmacological efficacy. In vitro techniques have the advantage of better controlling all tissue and cellular factors associated with neuron damage and possible mechanisms of action of pharmacological agents.

Cell culture
Recent studies have indicated the possibility to mimic neurodegenerative processes in cell culture, providing an alternative to study the inhibition of these processes by neuroprotective drugs. The use of PC-12 cells (Iwashita, 2004; Liu, 2007), human SH-SY5Y neuroblastoma cells (Sarang, 2002), rat neuroblastoma cells (Otey, 2003; Croslan 2008), crayfish neuron cells (Burov, 2000) and others have permitted the design of methods for drug screening and identification neuroprotective agents. of potentially new

71

Cells can be maintained in Dulbecco's modified eagle medium (DMEM) or minimum essential medium (MEM), supplemented with 10% fetal bovine serum (v/v) supplemented with penicillin-streptomycin, and incubated at 37C in a humidified atmosphere with 5% CO2. The cells can be seeded at approximately 1,000 to 1,000,000 cells per well in 24- or 96-well plates and grown until each well is 75-80% confluent. At this stage cells can be exposed to different injurious agents to mimic the various mechanisms of injury that are believed to occur in vivo. These injurious agents include hydrogen peroxide to mimic oxidative stress (Sarang, 2002; Liu, 2007), oxygen-glucose deprivation, glutamate (Croslan, 2008) or NMDA to mimic glutamate excitotoxicity and calcium overload.

The study drugs should be in contact with the cells before or after injury, according to the experimental design and drug screening strategy. Pilot studies are performed to optimize drug concentrations and the duration of exposure of cells to injurious agents as well as to study drugs. Cell viability is the initial indirect indicator of cell damage, such that prevention of cell damage with drugs can be interpreted as neuroprotection. The optimal dose of an injurious agent is that which causes an approximately 70 % loss of cell viability, which should be considered as the maximal injury or cell damage (negative control). This method makes it possible to measure the efficacy of treatments to rescue cells from eventual death.

Cell viability can be determined by various methods: direct microscopy to determine morphological changes such as shape (spherical or dendritic), dendrite length or spine density; dye exclusion, another microscopic procedure to evaluate the number of viable and non-viable cells by counting with a hemacytometer; and estimating cell viability by measuring changes in absorbance or fluorescence after treatment with reagents which can be modified by normal cellular metabolic processes. The use of wellplates and well-plate readers facilitates the application of these methods. More detailed information about these in vitro procedures is presented in the Materials and Methods section.

72

Slice preparations
Hippocampal slice preparations have been widely used in neuroscience for a variety of experiments in physiology, pharmacology and pathophysiology. In changing concentrations of oxygen and glucose, this versatile model can be used to mimic various ischemic injuries. While the CA1 region of the hippocampus is known to be one of the neuronal populations most sensitive to ischemia/hypoxia, other regions of the brain can also be used for similar studies (Dong, 1988). Brain slice-based screening (Wang, 2006) using brain slice explants, which maintain the three-dimensional architecture of the brain tissue, has increased the likelihood of reproducing mechanisms of cell damage and mechanisms of neuroprotection of surrounding viable cells, which are relevant and representative of the in vivo situation. Removing neurons from their native tissues and either immortalizing them or placing them in primary cultures can deemphasize mechanisms or targets which could have clinical importance (Wang, 2006).

The procedure includes the preparation of 400 urn brain slices from PND-7 rat pups, which are sacrificed, after which the brains are extracted and dissected. The slices are maintained in artificial cerebrospinal fluid until used in the experiments. Oxygen or glucose deprivation can be used to cause cell damage in the slices. After the experiments, slices can be treated with fluorescence reagents to study the effects of treatments on cell viability, or they can be homogenized to extract specific molecules to study the various mechanisms of neuroprotection and gene expression.

In vivo models of brain injury

The preclinical study of drugs is the most important pharmacological step for the demonstration of effects, relationships between doses and effects, characterization of possible mechanisms of action, determination of toxicity and side effects, and the screening and approval of drugs for clinical studies. Typically new neuroprotective drugs should have efficacy demonstrated in a variety of animal models, performed in different 73

species and in different laboratories. Outcome measures in acute models such as stroke or TBI should include both lesion size (histological) and behavior assessments (neurological), with attention to therapeutic time windows in relationship to time periods of penumbra survival in the models being used. Extended survival experiments should be performed to discard transitory effectiveness (Schabitz, 2006).

In an effort to stratify the various types of strokes that accurately represent human strokes, investigators have focused on the use of MRI "signatures" or perfusion/diffusion mismatches. In humans there are two principal groups of stroke patients, those with evolving infarcts in which perfusion imaging is superior to diffusion imaging, and those with stabilized infarcts in which perfusion imaging is inferior to or equal to diffusion imaging. Such perfusion and diffusion assessments have been proposed to indicate the extent of salvageable tissue. When the perfusion image is superior to the diffusion image, this indicates the presence of salvageable tissue and this situation occurs in 70% of patients at 6 hours post-stroke and 50% of patients at 24 hours post-stroke (Albers, 1999; Read, 2001). Comparisons between animal models of ischemia are difficult due to the uses of different rat strains, anesthesia methods, and models of ischemia induction. Development of the diffusion image is a marker of lesion volume with respect to time. Data from certain animal models demonstrates no mismatch between diffusion and perfusion images, which can likely be attributed in part to insufficient collateral blood flow and the peri-infarct depolarizations that eventually injure the poorly-perfused penumbra during evolution of the infarct (Hossmann, 1996; Read, 2001).

Preclinical models of spontaneous stroke

Genetic studies of stroke have focused on animal models of spontaneous stroke, in which environmental and genetic variables can be controlled. Homogeneous populations of stroke-prone rats have been isolated by inbreeding spontaneously hypertensive rats (SHR). Crossing the spontaneous hypertensive rat with the isolated stroke-prone rat allowed cosegregation of genes defining the stroke phenotypes STR1, 74

STR2 and STR3 (Read 2001). Another type of F-2 animal has been obtained by crossing SHR-stroke-prone rats with Wistar-Kyoto rats (WKY). SHR-stroke-prone and F-2 animals have been used to identify genes involved in the risk of human stroke (Read, 2001).

Hypoxic-Ischemic (H-I) brain injury in vivo

H-I injury to the prenatal and perinatal brain is a major contributor to morbidity and mortality in infants and children, and often leads to mental retardation, seizures and motor impairment. The PND7 rat brain is approximately equivalent to the human brain at 32 - 36 weeks of gestation (Han, 2000). In one study pups at PND7 were anesthetized with 2.5% halothane and the left common carotid artery was exposed and permanently ligated. The incision was then sutured and the pups were returned to their dams for a two hour recovery and feeding period. The pups were then placed in individual containers in a water bath at 37 C, with a humidified atmosphere containing 8% oxygen for 2.5 hours. The pups were then returned to their home cages. The pups were anesthetized with pentobarbital and perfused transcardically with PBS at pH 7.4 one week later, the brains were extracted, fixed overnight with 4% paraformaldehyde in 0.1 M phosphate buffer, frozen and cryoprotected with 30% sucrose in 0.1 M phosphate buffer, after which coronal sections were cut on a freezing sliding microtome and stained with cresyl violet for evaluation of either brain damage or neuroprotection (Han, 2000).

Focal cerebral ischemia models in rodents or whole animal studies by middle cerebral artery occlusion (MCAO)
The advantages of using the rat for stroke studies include the similarity of its intracranial circulation to that of humans, the abundant neurochemical data which have been derived from the rat brain, and the relatively low animal cost which is important for large studies and statistical analyses (Chen, 1986). 75

The several models of cerebral ischemia developed in the rat can be classified as global or focal by the extent of damage, and as reversible or irreversible by the chronology of ischemia-reperfusion. These methods use intravascular embolization and/or extravascular ligation or clotting.

Intraluminal occlusion model (Suture model or filament model): Transient focal ischemia and permanent focal ischemia. This model is a relatively non-invasive method, achieving reversible or permanent MCAO by use of an intraluminal suture. Some variations of this method can reliably produce regional infarcts. Nonetheless, brain injuries produced by this method can vary in size and distribution of infarct. This variability from animal to animal necessitates the use of large numbers of animals to discern statistical significance in drug testing. In order to prevent inconsistent infarct volumes, the use of coated sutures or heat-enlarged tip sutures is required, to increase adhesive forces on the vascular endothelium.

MCAO can be induced according to Longa (1989) in anesthetized animals using an operating microscope. The right common carotid artery (CCA) is exposed by a midline neck incision and is carefully dissected free from surrounding nerves and fascia, from its bifurcation to the base of the skull. The occipital artery branches of the external carotid artery (ECA) are then isolated, and these branches are dissected and coagulated. The ECA is dissected further distally and coagulated along with the terminal lingual and maxillary artery branches, which are then divided. The internal carotid artery (ICA) is isolated and carefully separated from the adjacent vagus nerve, and the pterygopalatine artery is ligated close to its origin with a 5-0 nylon suture. A 5-0 silk suture is tied loosely around the mobilized ECA stump, and a 4 cm length of 3-0 monofilament nylon suture is inserted through the proximal ECA into the ICA and thence into the circle of Willis, effectively occluding the MCA. The silk suture around the ECA stump is tightened around the intraluminal nylon suture to prevent bleeding. The suture is inserted 18 to 20 mm from the bifurcation of the CCA, according to the animal's body weight. After the

76

intraluminal suture is in place, the neck incision is closed with silk suture or clips. Blood flow can be restored by withdrawing the intraluminal suture.

MCA permanent occlusion plus common carotid arteries transient occlusion

This method has been developed to produce consistent and extensive cortical infarction without the disadvantage of the variability reported for the intraluminal suture method. The extensive intracranial collateral circulation in the rat resulting from the circle of Willis, leptomeningeal anastamoses and dorsal collateral junctions is the primary factor leading to the variability observed in other models of focal stroke. In this model, the unilateral or bilateral CCA ligation produces consistent ischemic lesions when performed following MCA occlusion (Chen, 1986). The disadvantage of this method is the invasive craniotomy necessary to occlude the MCA together with ligation of the CCAs. Another disadvantage could be production of a primarily cortical stroke, which only partially represents human stroke. The principal advantage of the method is a reproducible infarct volume, which enables drug efficacy to be measured in small animals.

Occlusion of the MCA is made through an incision at the midpoint between the right eye and ear. The temporalis muscle is separated in the plane of its fiber bundles and retracted to expose the zygoma and squamosal bone. A burr hole is made with a dental drill to the anterior junction of the zygoma and the squamosal bone. The bone is broken carefully to expose the cortex without damage to visualize the MCA. The dura mater is carefully pierced with a # 11 scalpel, and the exposed MCA is occluded with an electric coagulator. The craniotomy is then covered with a small piece of gelfoam and the muscles and skin are allowed to return to their normal positions and are sutured separately. Isolation of the CCAs is made via a ventral midline cervical incision, to expose both CCAs or the right CCA, which are then transiently ligated with sutures or surgical clips inmediately after occlusion of the MCA. A consistent time of reperfusion of the CCAs should be considered as an important factor leading to the reproducibility of
77

results. An optimal duration of occlusion and reperfusion of the CCAs appears to be 60 minutes (Chen, 1986). The production of a larger consistent cortical infarct is possible by interrupting blood supply to the right MCA territory and by permanent ligation of the right CCA together with transient ligation of the left CCA for one hour. This model provides an excellent quantitative measure of drug efficacy.

MCAO permanent occlusion (Tamura)

The Tamura model was developed to cause reproducible brain ischemia in the cortex and striatum by MCA occlusion at the origin of the MCA, or at a point just before it reaches the lateral edge of the olfactory tract (proximal MCA). The anatomy of arterial supply to the cerebral hemispheres in the rat is comprised of anterior, middle and posterior cerebral arteries, which then give rise to cortical and basal perforating branches, similar to the human anatomy. In the rat the three arteries are derived primary from the ICA, but they connect via an azygos anterior cerebral artery and via posterior communicating arteries to form a modified circle of Willis. The MCA in the rat runs laterally over the surface of the olfactory tract before it branches to the cerebral cortex. Between its origin and the lateral edge of the olfactory tract it provides some medial perforating, lenticulo-striate branches to the posterior part of the caudate-putamen complex and a variable branch to supply the olfactory tract and circumference of the hemisphere. The latter sometimes derives from the carotid artery. The anterior part of the caudate-putamen complex is supplied by lateral striate branches from the MCA and in addition, receives medial supply from Hubner's arteries which extend from the anterior cerebral artery. The cortical branches of the MCA are more variable distal to the olfactory tract (Tamura, 1981).

Direct MCA occlusion by electrocoagulation of the proximal part of the MCA produces an expanding diffusion lesion, with an initial marked expansion at 4 hours followed by an additional small increase from 4 to 24 hours (Gill, 1995; Read, 2001). A reproducible lesion occurs in the cortex and striatum when the artery is occluded at a 78

proximal site or subtemporal approach; this is technically feasible and its histological consequences are consistent.

Exposure of the MCA is made through a curved vertical 2 cm skin incision between the eye and ear. The parotid gland is exposed in the postero-inferior quadrant of the field. The gland is mobilized posteriorly, and an incision is made around the posterior and superior margins of the temporalis muscle. The muscle is reflected in a forward direction. A vertical incision is made down the anterior margin and across the attachment to the tip of the coronoid process of the mandible. This bone and the posterior half of the zygoma are then removed and the infero-temporal fossa exposed. The muscles are retracted downward, and the mandibular nerve is followed medially, across the anterior aspect of the temporo-mandibular joint to the foramen ovale. A craniotomy is then made using a dental drill, at the junction between the medial wall and the roof of the inferotemporal fossa. The position of the skull opening is critical, and it should be about 3 mm anterior and 1 mm lateral to the foramen ovale. The dura mater is opened with a fine scalpel to view the MCA as it runs forward initially and then over a white band of myelinated fibers that constitute the lateral edge of the olfactory tract. The artery is then occluded by electrocoagulation between the cortical branch to the rhinal cortex and the lateral striate arteries. The craniotomy and incisions are then closed as in the previous model (Tamura, 1981).

The consistency of focal ischemia produced with this model has been confirmed in procedures performed by investigators at different laboratories. Furthermore, the postoperative survival of animals for several days makes the model more attractive for screening neuroprotectants with extended therapeutic time windows after stroke or for time-course evaluation of brain damage and neuroregenaration.

79

MCA occlusion by perivascular microinjection of endothelin-1 (ET-1)

This model is the least invasive for production of focal ischemia by MCA occlusion. ET-1, the most potent vasoconstrictor known thus far, is used in this model which incorporates gradual reperfusion, avoids damage of the endothelium by the mechanical occlusion of arteries, and has the advantage that rats are conscious during the production of stroke. Other models have the disadvantage of anesthesia which can affect experimental outcomes such as ROS production or altered neurotransmitter release. Examples of this are the uses of barbiturates and isoflurane, both of which have been reported to be neuroprotectants. This model produces cortico-striatal damage, which can be measured histologically and by assessment of functional outcomes of the animals.

For surgical preparation of animals, a 23 gauge stainless steel guide cannula is stereotaxically implanted into the piriform cortex 2 mm dorsal to the proximal portion of the right MCA (0.2 mm anterior, -5.2 mm lateral to bregma and -6.1 mm ventral to the surface of the skull) of anesthetized rats. Following a three day recovery from surgery, stroke is induced in the conscious rat by administration of ET-1 (120 pmol in 6 \x\ of saline over 6 minutes) via a 30 gauge injector which protrudes 2 mm beyond the end of the previously implanted guide cannula (Callaway, 2003).

Photothrombic ischemia model

This model was developed to produce MCA occlusion at the distal portion, avoiding mechanical trauma and permitting the skull to remain intact. The process of vessel occlusion in this model involves the fluorescein derivative Rose Bengal, such that when it is injected into rats via the tail vein at the same time as the skull surface is focally illuminated, cerebral blood vessels in a confined area sustain photochemical injury. Singlet oxygen molecules generated by the dye-light reaction cause peroxidation of endothelial cell membranes and occlusive platelet aggregation. The subsequent thrombus

80

formation, vascular stasis, extravasation and cytotoxic edema lead to cerebral infarction and cell death (De Ryck, 1989).

This method of focal ischemia induction is non-invasive, allows for reproducible infarct size and location, and includes a cascade of stroke-like ischemic events. The neuron damage is essentially restricted to the cortex and results in selective and lasting deficits in tactile and propioceptive limb-placing. The advantage of this method is the survival of animals for several weeks allowing both histological and behavioral evaluations and the ability to start treatments at different times.

One modification of this method follows a procedure similar to that described by Chen (1989), consisting essentially of isolation and ligation of the CCAs and exposure of the MCA through a craniotomy. In this modification, the rupture of the dura mater is not necessary. Animals are prepared as previously described, with catheterization at the femoral vein for a more reliable i.v. administration of Rose-Bengal dye, followed by laser irradiation of the MCA with the laser source at three points to allow a better focalization of ischemia. Transient ligation of the CCAs is accomplished after the MCA is irradiated (Markgraf, 1993). When this modified procedure is followed, a highly consistent focal infarct is produced. A diffusion image results during the first 24 hours, with an expanding infarct volume continuing over the next 3 to 7 days. The extensive thrombosis produced by this technique, in association with profound blood-brain barrier breakdown, might limit applications of this model (Markgraf, 1993; Read, 2001).

Embolic m o d e l s

Available embolic models of focal stroke which utilize intra-arterial injections of thrombin, aged thrombin or fibrin-rich clots, have been reported to result in approximately similar expansions of diffusion lesions, first appearing 80 minutes postinjection with a gradually expanding volume over the next 24 hours (Read, 2001). The method was developed to induce intra-arterial formation of a thrombus by injection of

81

thrombin at the origin of the MCA. The injection of thrombin is performed via an intraluminal catheter placed in the intracranial segment of the ICA. The advantages of this model are: a thrombo-embolic ischemia is produced similar to that of human stroke; a reproducible and predictable infarct volume is produced; and the model can be useful for evaluation of thrombolytic therapies (Zhang, 1997).

Focal cerebral damage by intracerebral injection of toxins

Focal ischemia is not the only method to produce experimental brain damage for the evaluation of neuroprotectants. Other models of brain damage could include chemical or mechanical damage to the CNS. Measurable damage to the brain is an important goal for the validation of methods to study neuroprotectants.

One example of a reported method is neuron damage induced by intracerebral injection of ouabain in neonatal rats to test neuroprotective drugs. Neonatal rats are anesthetized and immobilized in a stereotaxic frame. A small burr hole is drilled in the cranium over the left hemisphere, 2.5 mm lateral to bregma. A 1 ul syringe is lowered into the left striatum to a depth of 4 mm. A 0.5 ul of 1 mM ouabain is injected at a rate of 0.125 ul/min using a microdrive. Control animals are injected with vehicle. After the injection, the needle is left in situ for 2 min to prevent the leakage of injection fluid. Animals can be pretreated or treated after brain damage has been produced with ouabain, according to the experimental design for the study. Evaluation of brain damage and drug efficacies are made in brain slices stained with hematoxylin-eosin. Measurements of infarct areas or damage volumes are used for statistical analysis (van der Stelt, 2001).

82

Evaluation of brain damage Infarct assessment Hematoxylin-Eosin (H&E) staining

After treatment infarcted animals are anesthesized and transcardially perfused with formalin for fixation and the brains are extracted and embedded in paraffin. Coronal plane sections are cut and stained with hematoxylin and eosin (H&E) for the measurement of infarct volume. The areas of infarction at nine coronal levels throughout the brain are traced and measured with the aid of a microscope camera attachment. Measurements of these infarcted zones are saved as digital images and computational procedures are performed to analyse the areas and to calculate infarct volumes as the products of cross-sectional areas for all sections and distances between sections (Bederson, 1986).

2,3,5-triphenyltetrazolium chloride (TTC) staining Another method of staining brain tissue to quantify infarct volume makes use of TTC. In this method an animal is anesthetized and sacrificed so the brain can be extracted without fixation. The brain is sliced into coronal sections of 2 mm thickness which are incubated in a 2% solution of TTC in the dark at room temperature or in the light at 37C. The slices are then fixed in 4% buffered formalin at pH 7.4 for 24 hours, after which the stained sections are photographed and images are processed as above (Candelario-Jalil, 2004). Unstained areas of the fixed brain sections are defined as infarcted.

TTC and H&E staining produce highly correlated results, indicating that in experimental conditions either method produces very similar results. TTC staining can be used to measure infarct volumes up to 4 days after MCAO, with the advantages of being inexpensive, easy to perform and fast. Staining with H&E is more expensive, but it can be employed at any time following the infarction (Bederson, 1986).

83

Measurement of apoptosis
Detection of apoptosis in stroke can be accomplished with radiolabeled annexin V, a protein that binds to phosphatidylserine on the surfaces of apoptotic cells. Its limitations include a high molecular weight, which prevents distribution and penetration into the cells (Reshef 2007). A significant improvement in detection is realized with jV,./V'-didansyl-/-cystine (DDC), a novel compound of low molecular weight and amphipathic properties, which only passes through membranes of cells undergoing the process of apoptosis. DDC can be detected by fluorescence, making possible the mapping of apoptosis in the rat brain (Reshef 2007). TUNEL (terminal deoxynucleotidyl

transferase-mediated dUTP nick end-labeling) is the method of choice for rapid identification and quantification of apoptotic cell fractions in cultured cell preparations (Negoescu, 1996).

84

MATERIALS AND METHODS Plant material and extract preparations

Dried hypocotyls of Lepidium meyenii, obtained from Arequipa, Peru, were botanically identified at the faculty of Pharmacy and Biochemistry of the Catholic University of Arequipa, and were then chemically characterized at the Massachusetts College of Pharmacy and Health Sciences (MCPHS) and the University of Mississippi. The dried hypocotyls (750 g) were ground into a powder and extracted with 3 L of methanol (HPLC grade, Pharmco-AAPER, Brookfield, CT) for 48 hours. The methanol extract was then filtered, concentrated in a rotary evaporator to 1 L and mixed with an equal volume of distilled water, followed by a continuous liquid-liquid re-extraction with 98% n-pentane (Sigma-Aldrich, St. Louis, MO), for 24 hours. After this the n-pentane fraction was dried in a rotary evaporator and the residue was refrigerated.

In all in vitro experiments, the pentane fraction was dissolved in phosphatebuffered saline (PBS, Mediatech, Manassas, VA) containing 10% DMSO (v/v, SigmaAldrich). For in vivo studies the pentane extract was re-dissolved in methanol together with 6 times its weight of polyvinylpyrrolidone (PVP 30 K, GAF Chemicals, Wayne, NJ), the solution was dried in a rotary evaporator until constant weight was attained, and the residue was dispersed in sterile water before administration.

Infrared (IR) spectral analysis of pentane extract

The pentane extract was analyzed by infrared spectroscopy to determine its transmittance spectrum. Approximately 0.2 g of dried pentane extract was spread on the surface of an NaCl plate which was then analyzed with an IR spectrophotometer (Nicolet Impact 410, GMI, Ramsey, MN), scanning from 500 to 4000 nm. The transmittance

85

spectrum was recorded and compared to spectra of chemical constituents reported from Lepidium meyenii (Maca).

High performance liquid chromatography (HPLC)

HPLC analysis of extracts and pure compounds was performed with a Hewlett Packard 1100 system (Agilent Technologies, Santa Clara, CA), equipped with a photodiode array detector. For all separations a Luna RP-Cig(2) column (250 x 4.6 mm, 5 |im particle size, Phenomenex, Torrance, CA) was used. The mobile phase used for standardization of Maca products (Ganzera 2002) consisted of water (A) and acetonitrile (B, Omnisolv, EMD Chemicals, Gibbstown, NJ), both containing 0.1% acetic acid (Sigma Aldrich). Separations were performed by linear gradient elution from 45% A/55% B to 5% A/95% B over a period of 45 minutes. The flow rate was adjusted to 1 mL/min, and UV detection at 210 and 280 nm was monitored. 10 [xi sample solutions were injected. All data were recorded and processed with Chemstation software (Agilent Technologies). Peaks were assigned by comparing the retention times with those of standard compounds.

Animals
Forty four adult male rats: 36 Sprague Dawley (CD) and 8 Wistar, all 300-350 g (Charles River Laboratories, Wilmington, MA) were housed in standard conditions of ambient temperature (743F), and humidity (5515%) with a 12 hour lightdark cycle, starting 5 days before each study. CD rats were divided into 4 groups of 9 animals each for the Chen model procedure. Wistar rats were divided into 2 groups of 4 animals each for the suture model procedure. Food and water were available ad libitum. All experimental animal protocols were reviewed and approved by the Institutional Animal Care and Use Committee (IACUC) at MCPHS.

86

Development of focal infarction

Chen model: Animals were anesthetized with chloral hydrate (400 mg/Kg i.p., Sigma-Aldrich). The common carotid arteries were isolated via a ventral midline cervical incision and both arteries were eventually ligated for one hour. Then, a 1.5 cm scalp incision was made at the midpoint between the right eye and the right ear. The temporalis muscle was separated and retracted to expose the zygoma and squamosal bone. A hole was made with a dental drill 1 mm rostal to the anterior junction of the zygoma and squamosal bone. Bone and dura mater were carefully pierced to expose the middle cerebral artery. The artery was occluded with a bipolar coagulator (Radionics 440E, LabX, Riverview, FL). The craniotomy was cleaned and the temporalis muscle and overlying skin were allowed to fall back and were sutured separately. After one hour the common carotid arteries were reperfused and the ventral incision was cleaned and sutured (Chen, 1986).

Suture model: Animals were anesthetized with chloral hydrate (400 mg/Kg i.p., Sigma-Aldrich). The right common carotid artery (CCA) was exposed through a midline neck incision. The branching of the external carotid artery (ECA) and internal carotid artery (ICA) was dissected. A 3-0 monofilament nylon suture with a rounded tip was inserted through one incision on the proximal ECA into the ICA and thence into the circle of Willis, effectively occluding the MCA. A silk suture around the ECA stump was tightened around the intraluminal nylon suture to prevent bleeding and disruption of the occlusion. The suture was inserted 18 to 20 mm from the bifurcation of the CCA, according to the animal's body weight. The neck incision was closed with clips (Longa, 1989).

87

Treatment
All groups of animals received double tail vein injections, the first when anesthetized before surgery (30 minutes prior to stroke), and the second one hour after the medial cerebral artery was occluded. In both models, one group of animals (Control) was treated with aqueous PVP solution and the other groups (Maca) were treated with the pentane extract at 3, 10 and 30 mg/Kg per injection in the Chen model procedure and at 3 mg/Kg in the suture model procedure.

Infarct Volume Evaluation

Twenty-four hours after treatment, animals were deeply anesthetized with i.p. chloral hydrate and then were decapitated. The brains were extracted and refrigerated in cold saline for 15 minutes. Brains were sliced and treated with 2% triphenyltetrazolium chloride (TTC, Sigma-Aldrich) for 30 minutes in darkness. The stained slices were transferred to a 10% formalin buffered solution (Sigma-Aldrich) for photographs after 24 hours. Infarct volumes measured using the Image J program were expressed as percent relative to total tissue of the brain hemisphere.

Cells and culture media

Orconectes limosus (crayfish) neuronal cells, OLGA-PH-J/92 (ATCC, Manassas, VA) were used to test neuroprotective activity. This cell line exhibits certain transformation features, such as anchorage independence and loss of contact inhibition. Its stable growth characteristics and morphology, from the primary culture through higher passages suggest that the cells are neuronal stem cells (ATCC product description).

The crayfish neurons were cultured in minimum essential medium (MEM, SigmaAldrich) supplemented with sodium bicarbonate (2.2 g/L, Sigma-Aldrich), HEPES (9.52 g/L, Sigma-Aldrich), sodium pyruvate (0.11 g/L, Sigma-Aldrich), and 10% fetal bovine 88

serum (v/v, FBS, ATCC) at 27 C in an incubator with 5% C0 2 . The pH of the supplemented basal media was adjusted to 6.8 with 0.1N hydrochloric acid (Burov, 2000).

B35 rat neuroblastoma cells (ATCC) are a cell line derived from tumors of the neonatal rat central nervous system in 1974. They have a number of advantages in the study of mammallian CNS neurons. They are simple to grow and differentiate, permitting the study of neuronal processes (Otey, 2003; Croslan, 2008).

B35 rat neuroblastoma cells were cultured in Dulbecco's modified eagle medium (DMEM, ATCC), 0.1 U/mL penicillin/0.1 mg/mL streptomycin (ATCC), and 10% fetal bovine serum (v/v, FBS, ATCC) at 37 C in an incubator with 5% C0 2 (Croslan, 2008).

Trypan-blue exclusion cell counting

Cell viability was measured using a trypan blue exclusion method to determine the number of living cells in each well. Previously, a 24-well plate had been prepared as mentioned above, with 1 x 105 cells seeded in 1 mL of DMEM per well.

Media was removed from the well and 200 uL of trypsin/EDTA buffer (ATCC) was added and incubated at 37C with shaking. Five minutes later 800 u.L of DMEM was added and the suspension was thoroughly mixed. Ten uL of cell suspension was mixed with 10 uL of trypan blue (Sigma-Aldrich) and transferred to a hemacytometer (Hausser Scientific, Horsham, PA), and cells were counted with an inverted microscope (Olympus CKX 31, Center Valley, PA) to determine the concentration and number of living cells per well. The numbers of cells in wells treated only with vehicle (no damage) were considered to represent 100% viability (Ma, 2009).

89

Cell viability assay with MTS reagent

Cell viability was also measured to determine proliferative and/or metabolic activity, using a colorimetric method with 3-(4,5-dimethylthiazol-2-yl)-5-(3salt (MTS, Promega,

carboxymethoxyphenyl)-2-(4-sulphophenyl)-2/f-tetrazolium

Madison, WI), which is reduced by viable cells to a formazan blue product that is soluble in the cell culture medium.

In the assay with crayfish neurons, 10 uL of MTS reagent was added to wells 3 hours after H2O2 treatment (Liu, 2007) and absorbance was measured after an incubation period of 24 hours on 96 well plates at 490 nm with a Synergy HT plate reader (Bio-Tek Instruments, Winooski, VT). The estimation of viable cells is possible as the absorbance of the blue color is directly proportional to the number of metabolically active cells. The percent viability is estimated by comparing the absorbance of treated wells to the average absorbance of wells treated only with vehicle (no damage to cells), which are considered to be 100% viable. A neuroprotective effect is estimated by the percent absorbance increase compared to the absorbance of wells treated only with H2O2 (maximal damage or 0% neuroprotection). Absorbance in wells treated only with vehicle (no damage) was considered to represent 100% neuroprotection.

In the assay with rat neuroblastoma cells, 10 \iL of MTS reagent was added to each well 24 hours after H2O2 treatment and absorbance measurements were performed after a four hour incubation on 96 well plates at 490 nm, with a Synergy HT plate reader (Bio-Tek Instruments). Estimates of viability and neuroprotection were made as above.

ABTS test for antioxidant capacity

This assay relies on the ability of antioxidants in a sample to inhibit the oxidation of 2,2'-azino-di(3-ethylbenzthiazoline sulphonate) (ABTS, Cayman Chemicals, Ann Arbor, MI) by oxidized metmyoglobin. H2O2 is used to oxidize metmyoglobin. The amount of oxidized ABTS is inversely proportional to the antioxidant capacity of the 90

compound tested and its absorbance is measured at 750 nm. The antioxidant capacity is compared to that of trolox, a tocopherol analog which is a recognized antioxidant that prevents oxidation of ABTS in a concentration-dependent manner. Test compounds and trolox standards were dissolved in DMSO at required concentrations. In each well of a 96-well plate, 10 \ii of test solution, standard or vehicle was mixed with 10 |iL of metmyoglobin and 150 |iL of ABTS reagent. The reaction was initiated by adding 40 uL of 0.441 mM H2O2. After 5 minutes absorbance was read at 750 nm with a Synergy HT plate reader (Bio-Tek Instruments).

Compounds for screening

Two isolated and purified macamides (natural alkamides) were provided by Dr. Ikhlas Khan of the National Center for Natural Products Research at the University of Mississippi. These compounds are: JV-benzylhexadecanamide (macamide-1) and 7Vbenzyl-5-oxo-6.E',8.E'-octadecadienamide (macamide-2), which were the first two

macamides isolated from the Maca plant (Muhammad, 2002; Ganzera, 2002)

Based on the chemical structures of the macamides, eight additional synthetic alkamides were provided by Mr. Hui Wu and Dr. Charles Kelley of the Laboratory of Medicinal Chemistry at MCPHS. These were designated MCP-11, MCP-12, MCP-13, MCP-14, MCP-15, MCP-31, MCP-32, and MCP-33.

In all in vitro experiments, alkamides and the endocannabinoid CBi antagonist O2050 (Tocris Bioscience, Ellisville, MO) were dissolved in phosphate-buffered saline (PBS, Mediatech) with 10% DMSO (v/v, Sigma-Aldrich). The final concentration of DMSO in the cell culture plate wells was 0.5%.

91

In vitro neuroprotection assays

The crayfish neurons were cultured in T-flasks at 27C until 80% confluent growth was achieved. The medium was then removed from the flask, a scraper was used to detach the cells, and cells were then suspended in supplemented MEM. Cells were counted with a hemacytometer after trypan blue staining to determine the number of viable cells per uL, to seed cells at 1 x 105 cells per well on a 96-well plate. MEM was subsequently added to yield a total volume of 180 uL per well. Twenty four hours after cell seeding each well was pre-treated with either 20 jxL of a test solution or with vehicle. Three hours later, 20 uL of hydrogen peroxide solution (H2O2, Sigma-Aldrich), which acts as a neurotoxic agent (Sarang, 2002; Liu, 2007), or vehicle was added to each well to produce a final concentration of 1 mM. Microscopic morphologies of cells were observed with an inverted microscope (Olympus CKX31) and photographs were taken with a digital camera (Sony MVC-CD400). An MTS viability assay was performed 24 hours later.

The rat neuroblastoma cells were cultured in T-flasks at 37C until 80% confluency. The medium was removed from the flask and a trypsin-EDTA solution (ATCC) was added and incubated for 12 minutes at 37C to detach the cells. The action of trypsin was then inhibited by the addition of DMEM, and the cells were pelleted by a 5 min centrifugation at 60 g (Dynac III centrifuge, BD, Franklin Lakes, NJ). The cell pellet was resuspended in supplemented DMEM. Cell density was then determined by cell counting on a hemacytometer after trypan blue staining to ensure a seeding density of 1 x 105 cells per well on a 96-well plate. Supplemented DMEM was subsequently added to yield a total volume of 180 uL per well. Forty eight hours after cell seeding the media was changed to DMEM without FBS. Twenty four hours later wells were pre-treated with 10 uL of test solutions or vehicle. In cases of additional treatments, solutions were added after 1 hour incubation periods. One hour after the final treatment, 10 |iL of hydrogen peroxide (H2O2, Sigma-Aldrich), or vehicle was added to each well to produce a final concentration of 0.3 mM. Microscopic morphologies of cells and photographic

92

evaluations were performed as above. Trypan blue exclusion cell counting or MTS viability assays were performed 24 hours later.

Statistical analyses
Data were analyzed for statistical significance and results are presented as mean SEM. Multiple comparisons were assessed using the one-way analysis of variance (ANOVA), or Kruskal-Wallis one way analysis of variance on ranks. A p-value less than 0.05 was considered significant. If the p-value in the ANOVA test was significant, post hoc Bonferroni, Dunn's or t-test methods were performed when indicated to isolate the group or groups that differed from control. All p-values were two-sided.

93

RESULTS Yield of extracts

From 750 g of dried powdered hypocotyls of Lepidium meyenii, 100 g of crude methanol extract and 5.10 g of pentane extract were produced. Thus the yield of dried crude extract was 13.3% and the yield of pentane extract was 0.68%.

The IR spectrum obtained from a sample of pentane extract (Figure 14) had the characteristic bands and peaks of transmittance of IR light which were reported in the literature for the alkamides isolated from Lepidium meyenii (Muhammad, 2002).

HPLC analysis
Figure 15 shows the HPLC chromatogram obtained for a sample of pentane extract using the method summarized above and described in the literature (McCollum, 2002; Ganzera, 2002). The chromatogram shows baseline separation of peaks within a 45 minute run, using a C\% stationary phase, an acidic mobile phase comprised of water and acetonitrile, with UV detection at 280 nm. The peak with a retention time of 27.4 minutes was confirmed to be macamide-2, as determined by injections of a standard provided by Dr. Ikhlas Khan at the University of Mississippi. The chromatogram of a macamide-2 standard of 0.16 ug/mL, detected with the same conditions, is shown in Figure 16. Correlations between peak areas and concentrations are presented in Figures 17 and 18 and represent the calibration curve for macamide-2 in extracts of Lepidium meyenii. A linear response was obtained in the concentration range of 0.16 to 160 |ig/mL (R2 = 0.9990). Reproducibility was verified with peak areas determinated on different days (n=2-5 per concentration).

94

This calibration curve was used for quantification of macamide-2 in samples of crude methanol extract and pentane extract from Lepidium meyenii (Maca). A sample of crude methanol extract was dissolved in methanol at 100 mg/mL and analyzed with the HPLC method described above. The peak with a retention time of 27.4 minutes (macamide-2) produced an area of 28.5 mAU.sec. The concentration of macamide-2 in the dried methanol extract was determined to be 0.027 ng/mg. A similar procedure was used for analysis of macamide-2 in the pentane extract, in which the average peak area of three determinations for a sample of 1 mg/mL in methanol was 42.8 mAU.sec, representing a concentration of 3.55 p.g/mg of macamide-2.

Estimates of the limits of detection and quantification were made with data for the lowest measured standard concentration of 0.16 ng/mL, which had a peak height of 18 mm and a noise level height of 2 mm. Accepted values for the signal/noise ratio are 3 for the limit of detection (LOD) and 10 for the limit of quantification (LOQ). The values calculated for macamide-2 with these criteria were: LOD = 0.05 |xg/mL and LOQ = 0.17 |ig/mL.

In vivo neuroprotective assay

The effects of the pentane extract were evaluated in rats subjected to focal ischemic stroke (n=9) using the Chen model. The results showed statistically significant differences for the three doses evaluated (3, 10, and 30 mg/Kg), as shown in Figure 19. The control group had a mean percent ischemic tissue of 231.5% and the pentane extract dose of 3 mg/Kg was the only dose resulting in a significant neuroprotective effect (13.52.2%; p=0.002). Doses of 10 and 30 mg/Kg actually resulted in significant increases in brain damage (31.72.4% and 29.42.4% respectively). The neuroprotective effect of the 3 mg/Kg dose on the brain of a representative animal is compared to the brain of a control animal in Figures 20a and 20b. The colorless area represents ischemic or dead tissue. The deleterious effect of the higher doses is shown in Figures 20c and 20d.

95

The effects of the pentane extract at 3 mg/Kg were evaluated in rats subjected to focal ischemic stroke (n=4) using the suture model, as a confirmation of the neuroprotective effect. The results also showed statistically significant differences, as shown in Figure 21. The control and treated groups had mean SEM ischemic tissues of 23.12.7% and 12.80.6% respectively.

Crayfish neuron neuroprotective assay

Following pretreatment of crayfish neurons with pentane extract solutions or vehicle, cells were treated with 1 mM H2O2. Differences in cell appearance could be detected microscopically as shown in Figures 23a to 23d. Compared to the dendritic shapes and dense concentrations of untreated cells (Figure 23a), and the spherical shapes and sparse concentrations of cells subjected to H2O2 (negative control, Figure 23b), cells treated with different concentrations of the pentane fraction exhibited varying cell shapes and densities. Cells treated with 30 ug/mL pentane extract displayed healthier dendritic cell shapes than cells treated with 3ug/mL pentane extract, as shown in Figures 23 c and 23d. A linear-regression analysis of six concentrations of pentane extract demonstrated a significant concentration-dependent effect (n=5, p=0.03 and R=+0.85). The percent

neuroprotection calculated from absorbances obtained in MTS-treated wells after experiments with various pentane extract solutions (0.1 to 30 ug/mL) and H2O2 in crayfish neurons are presented in Figure 24. The average absorbance of control wells treated with H2O2 is considered 0% neuroprotection. The average absorbance of control wells treated with vehicle is considered to be 100% neuroprotection. The neuroprotective effects on treated cells ranged from 71.6% with 0.1 ug/mL to 888.4% with 30 jig/mL pentane extract. The results indicated that the effective concentration 50% (EC50) was approximately 2.8 (xg/mL.

The percent viability of crayfish neurons, when pretreated with pentane extract and subjected to oxidative stress, indicated significant (p=0.05) increases at 1, 3, 10 and 30 |xg/mL (79.51.7%, 83.95.5%, 84.82.5%, and 95.73.2% respectively), as shown in 96

Figure 25. The percent viability was calculated assuming 100% viability for control cells treated only with vehicle, and statistical significance was demonstrated by comparison with viability of the H 2 0 2 treated group (65.42.1%). On the basis of the abovementioned in vitro results, a preliminary test was performed to investigate the neuroprotective effects of isolated macamides provided by Dr. Milas Khan of the University of Mississippi. Crayfish neurons pretreated with 10 uM macamide-2 and subjected to oxidative stress with H2O2 3 hours later, had significantly increased viability (p=0.05) as compared to cells treated only with H2O2. The percent neuroprotection for 10 uM treatments with macamide-1 and -2 were 11.15.7% and 23.46.3% respectively (Figure 26).

Neuroprotective concentration-dependent effects in neuroblastoma cells

The rat neuroblastoma B35 cell line is a type of mammallian neuron which has been used in several studies of neuroprotection and neurotoxicity. B-35 cells were included in this study to demonstrate the neuroprotective activity of lipophilic compounds from Maca in mammalian cells. The effects of oxidative stress (0.3 mM H2O2) in neuroblastoma cells were observed 24 hours after treatment (Figure 27b) and compared to control cells (Figure 27a). The neuroprotective effects of 3 [ig/ml pentane extract and lOuM anandamide are also shown in Figures 27c and 27d. Concentrationdependent effects of pentane extract in neuroblastoma cells are shown in Figure 28.

On the basis of the neuroprotective activities of pentane extract, preliminary results for isolated macamides, and the hypothesis that macamides are active in neuroprotective processes, natural macamides and derivatives were synthesized by Mr. Hui Wu and Dr. Charles Kelley at the Laboratory of Medicinal Chemistry at MCPHS. The synthetic compounds were designated MCP-11, -12, -13, -14, -15, -31, -32, and -33.

97

The effects of the MCP compounds on the viability of rat neuroblastoma cells subjected to oxidative stress with 0.3 mM H2O2, measured by the trypan blue exclusion cell counting method, are presented in Figure 29. Cell counts are expressed as percent relative to numbers of cells in control wells treated only with vehicle, which represent 100% viability. When cells were pretreated with 10 uM MCP compounds and subjected to H2O2 one hour later, significantly (p=0.05) increased cell counts were observed for MCP-15 (n=4), MCP-32 (n=8), and MCP-33 (n=4) treatments, with mean values of 71.910.1%, 73.33.9%, and 80.55.1% respectively, compared to cell counts in wells treated only with H 2 0 2 (34.83%). A lack of toxic or proliferative effects due to the pentane extract was observed in neuroblastoma cells, as shown in Figure 30. Cells were treated with pentane extract (n=8) at 0.5 to 15 ug/mL and cell viability was evaluated 24 hours later with MTS reagent. There were no statistically significant (p=0.05) differences in cell viability among any of the concentrations tested.

The MCP compound with maximal activity in the preliminary test of neuroprotection was MCP-33, so it was chosen as one of the purified Maca derivatives for further evaluation of neuroprotective potential. A lack of either toxic or proliferative effects of MCP-33 was demonstrated in neuroblastoma cells (n=8), as shown in Figure 31. Cells were treated with MCP-33 at 5 to 50 uM and cell viability was evaluated 24 hours later with MTS reagent. There were no statistically significant (p=0.05) differences in cell viability between any of the concentrations tested.

The pentane extract was evaluated for concentration-dependent neuroprotective effects in neuroblastoma cells treated at 0.5 to 5 |ig/mL and subjected to 0.3 mM H2O2 one hour later. Viability was measured after 24 hours with MTS reagent. Exposure to the pentane extract at concentrations of 1.5 and 5 )ig/mL resulted in significantly (p=0.05) increased viability, estimated to be 62.57% and 53.84.2% respectively, compared to

98

control cells treated only with H2O2, which had an estimated viability of 33.12.9% (n=8). This model did not show concentration-dependent effects.

Anandamide, an endocannabinoid with a chemical structure similar to those of the macamides and with demonstrated neuroprotective effects (Mechoulam, 2002; Schabitz, 2002; van der Stelt, 2001; Franklin, 2003; Mechoulam, 2007; Schomacher, 2008) was evaluated for concentration-dependent neuroprotective effects in neuroblastoma cells treated at 5 to 50 uM and subjected to 0.3 mM H2O2 one hour later (Figure 33). Viability was measured with MTS reagent. Anandamide at 5 uM resulted in significantly (p=0.05) increased viability (66.910%; n=4) compared to control cells treated only with H2O2 (30.94.3%; n=4). Higher concentrations of anandamide (15 and 50 u.M) did not show a neuroprotective effect. THC, a natural cannabinoid found in Cannabis sativa with a pharmacological activity related to endocannabinoids, was evaluated for concentration-dependent neuroprotective effects in neuroblastoma cells treated at 5 to 150 uM (Figure 34) and subjected to 0.3 mM H2O2 one hour later. Viability was measured with MTS reagent. THC treatment at 15 uM resulted in significantly (p=0.05) increased viability (54.44.8%; n=4) compared to control cells treated with only H 2 0 2 (30.94.3%; n=4). Other lower and higher concentrations of THC (5, 50 and 150 uM) did not demonstrate significant neuroprotective effects. MCP-33, a synthetic derivative based on the chemical structures of the natural macamides present in Lepidium meyenii and which showed neuroprotective activity in a preliminary test of the MCP compounds, was evaluated for concentration-dependent neuroprotective effects in neuroblastoma cells treated at 5 to 50 uM (Figure 35) and subjected to 0.3 mM H2O2 one hour later. Viability was measured with MTS reagent. Treatment at the concentrations of 15 and 50 uM resulted in significantly (p=0.05) increased viability (564.9% and 62.56.7%; n=8) compared to control cells treated only with H 2 0 2 (35.33.1%; n=8). The lowest concentration of MCP-33 tested (5 uM) did not result in a significant neuroprotective effect.

99

Involvement of the endocannabinoid system

The compound (6ai?,10ai?)-3-(l-methanesulphonylamino-4-hexyn-6-yl)(O-2050) is a synthetic

6a,7,10,10a-tetrahydro-6,6,9-trimethyl-6i/-dibenzo[5,<f]pyran

competitive antagonist at the CBi receptor (Figure 36). O-2050 was included in this study to demonstrate the participation of CB\ receptors in the pharmacological activities of Maca and alkamide derivatives such as compound MCP-33.

Figure 37 shows the effects of O-2050 at three different concentrations between 5 and 50 uM on the viability of neuroblastoma cells (n=8). There was no difference between control and treated cells at the lowest concentration of 5 uM. Concentrations of 15 and 50 uM significantly (p=0.05) reduced the viability of neuroblastoma cells (77.34.9% and 28.9 4.7% respectively) compared to control cells (1003.8%).

On the basis of the above results, the neuroprotective effects of the pentane extract, anandamide, THC and MCP-33 were evaluated in the presence of the CBi receptor antagonist O-2050 in one approach to demonstrate that macamides and synthetic alkamides might be considered cannabinergic compounds.

The neuroprotective effects of pentane extract were evaluated in neuroblastoma cells treated at 0.5 to 5 ug/ml, either with or without the prior addition of O-2050 at 50 uM, and subjected to 0.3 mM H2O2 one hour later (Figure 38). Viability was measured after 24 hours with MTS reagent. The significant (p=0.05) neuroprotective effect demonstrated by the pentane extract at 1.5 and 5 ug/mL was significantly blocked by the addition of O-2050 (n=4). Viability decreased from 55.94.9% to 24.18.6% and from 47.23.7% to 23.13.9%, respectively, when O-2050 was added. Control cells treated with O-2050 and H2O2 had a viability of 151.4%. The viability of cells without any treatment (only vehicle) was assumed to be 100%.

100

The neuroprotective effect of anandamide was evaluated in neuroblastoma cells treated at 5 to 50 uM, either with or without the prior addition of O-2050 at 50 |iM, and subjected to 0.3 mM H2O2 one hour later (Figure 39). Cell viability was measured after 24 hours with MTS reagent. The significant (p=0.05) neuroprotective effect provided by anandamide at 5 uM was significantly blocked by the addition of O-2050 (n=4). Viability decreased from 69.68.4% to 16.83.9% when O-2050 was added. Control cells treated with O-2050 and H2O2 had a mean viability of 151.4%. The viability of cells without any treatment (only vehicle) was assumed to be 100%.

THC was also evaluated in neuroblastoma cells treated at 5 to 50 uM, either with or without the prior addition of O-2050 at 50 uM, and subjected to 0.3 mM H2O2 one hour later (Figure 40). Cell viability was measured after 24 hours with MTS reagent. The significant (p=0.05) neuroprotective effect demonstrated by THC at 15 uM was significantly blocked by the addition of O-2050 (n=4). Cell viability decreased from 54.44.8% to 20.24.1% when O-2050 was added. Control cells treated with O-2050 and H2O2 had an estimated viability of 151.4%. The viability of cells without any treatment (only vehicle) was assumed to be 100%.

The evaluation of MCP-33 was also performed in neuroblastoma cells treated at 15 and 50 uM, either with or without the prior addition of O-2050 at 50 uM, and subjected to 0.3 mM H2O2 one hour later (Figure 41). Cell viability was measured after 24 hours with MTS reagent. The significant (p=0.05) neuroprotective effect demonstrated by MCP-33 at both concentrations evaluated was significantly blocked by the addition of O-2050 (n=4). Cell viability decreased from 60.17.7% to 15.52.7% and from 515.5% to 24.83.3% respectively when O-2050 was added. Control cells treated with O-2050 and H2O2 had an estimated viability of 151.4%. The viability of cells without any treatment (only vehicle) was assumed to be 100%.

101

Evaluation of antioxidant effects

Determination of the antioxidant potential of pentane extract from Lepidium meyenii, anandamide and MCP-33 were made with the ABTS method, using a trolox standard for comparison. Results of absorbance measurements are shown in Figure 42. Absorbance values are inversely proportional to the antioxidant potential of any compound tested. Trolox at concentrations between 0.05 and 1 mM demonstrated a concentration-dependent effect, inhibiting the oxidation of ABTS which was confirmed by absorbance readings at 750 nm. A control assay (vehicle only) produced an absorbance of 0.210.02 (maximal oxidation, n=4). Addition of trolox resulted in significant (p=0.05) reductions in absorbance at concentrations of 0.1 mM (0.0980.014) or higher.

The pentane extract did not inhibit oxidation of the ABTS reagent, when assayed at 0.02 to 5 ug/mL. This lack of an antioxidant effect was confirmed by absorbance measurements (n=4), which demonstrated no significant (p=0.05) differences from the control assay absorbance (Figure 42).

Anandamide and MCP-33 were also evaluated for antioxidant potential and compared to the trolox standard. None of the concentrations evaluated resulted in significantly (p=0.05) different absorbances compared to the control assay absorbance (n=4, Figure 42), demonstrating that neither compound inhibited the oxidation of ABTS reagent.

102

DISCUSSION
Traditional and popular information suggests that Maca hypocotyls might have diverse health-promoting properties including: improvement of male and female sexual performance, improvement of fertility and amelioration of menopausal symptoms, improvement of the ability to adapt to adverse conditions (adaptogenic) and prevention of chronic and degenerative diseases. The unique chemical composition of Maca, demonstrated by the isolation of diverse constituents such as glucosinolates, alkaloids, macamides and others, helps to explain the variety of biological activities attributed to Maca and suggests the merit of further research on the Maca plant.

Pharmacological studies in vitro and in vivo have substantiated some of the traditional claims for Maca hypocotyls. Nevertheless there still exists an overall need for additional pharmacological studies to demonstrate other properties of this plant and to answer questions remaining about the efficacy, doses, mechanisms of action, responsible compounds, biopharmaceutical properties, safety, toxicity, adverse effects and potential clinical applications of the plant, preparations or isolated compounds.

Yield of extracts
Based on the the weight of raw plant material used in the study and the extracts obtained the yield of dried crude methanol extract was 13.3% and the yield of pentane extract was 0.68%. These values were obtained from one extractive assay. It would be ideal to pool information from future extractions of the same plant material and of other Maca plant materials, to determine the variability of the species as well as the reproducibility of the extraction process. Other studies have reported methods of extraction and solvents with yields approximating the values found in this study (Muhammad, 2002).

103

The technique of IR spectroscopy was used as a primary approach to confirm the chemical identity of the plant material and second to verify the presence of alkamides in the pentane extract. The IR spectrum obtained for a sample of pentane extract prepared in this study (Figure 14) shows characteristic bands and peaks of transmittance reported in the literature for the alkamides isolated from Lepidium meyenii (Muhammad, 2002). This spectrum of the pentane extract provided evidence of the presence of alkamides.

HPLC analysis
In Figure 15 is presented the HPLC chromatogram of a sample of pentane extract analyzed with a previously validated method (McCollum, 2002; Ganzera, 2002), which shows a characteristic pattern of peaks reported for crude extracts of Lepidium meyenii. The peak with a retention time of 27.4 minutes was confirmed to be macamide-2, as demonstrated with the standard provided by the University of Mississippi. In Figure 16 is the chromatogram of a standard solution of macamide-2 at 0.0016 mg/mL detected at 280 nm. The calibration curve shown in Figures 17 and 18 was linear, with an R =0.9990, and was reproducible with peak areas determinated on different days (n=2-5 per concentration). The calibration curve was used for quantification of macamide-2 in samples of crude methanol extract and pentane extract from Lepidium meyenii in the present study. In a chromatogram of a sample of dried crude methanol extract, the peak with the retention time of the macamide-2 standard indicated a macamide-2 concentration of 0.027 ng/mg extract. Similarly analysis of macamide-2 in the pentane extract indicated a concentration of 3.55 |ig/mg extract. The significant difference between the two extracts is that enrichment by the process of continous liquid-liquid extraction preferentially concentrates the most lipophilic compounds such as the macamides in the pentane extract.

Estimates of the limits of detection (LOD) and quantification (LOQ) of 0.05 and 0.17 ug/mL, respectively, together with the calibration curve indicated that the analytical procedure was valid and supported the use of macamide-2 as a marker compound for the standardization of Maca extracts and preparations. These results also made possible the 104

calculation of actual concentrations or doses of macamide-2 during the in vitro or in vivo studies.

In vivo neuroprotective studies

Several studies have found that Maca extracts produce different pharmacological effects in animals and humans. Lipophilic extracts, which concentrate most of the active constituents of Lepidium meyenii, have been demonstrated to be the most

pharmacologically interesting materials (Cicero, 2002). Since traditional claims and preliminary studies have supported the suggestion that Maca extracts have activity in the CNS, the hypothesis that the pentane extract of the Maca root acts as a neuroprotective agent was then postulated. Focusing on this hypothesis, this study was designed to demonstrate the potential neuroprotective properties of the pentane extract from Maca in both in vitro and in vivo models, to demonstrate and characterize the activities of macamides (alkamides) and derivative compounds, in an attempt to demonstrate the relationship of the chemical structures of the alkamides with endogenous mechanisms of cellular and organic protection.

The neuroprotective effects of the pentane extract in vivo were first assessed in rats subjected to focal ischemic stroke. Occlusion of the middle cerebral artery (MCAO) in the rat produces histological consequences, which are reproducible in controlled experimental conditions (Hossmann, 2007). There are several models of rat focal ischemia. In the Chen model, the result is a consistently infarcted area, preferentially in the cortex. The suture model, a complementary method used in this study, produces a lesion similar to that observed in humans, with a final infarct volume reached at 7 hours (Read, 2001). To obtain consistent results this method necessitates the use of animals of homogeneous size, since the diameter and length of vessels vary with age and weight. In general, Wistar rats have been reported to result in less variability in this stroke model compared to other animal strains.

105

Drugs which result in a reduced infarct size in experimental stroke are considered potential neuroprotective candidates for treating stroke and neurodegenerative diseases. The effect of the pentane extract was evaluated in rats subjected to focal ischemic stroke using the Chen model. The results showed significant differences for the three doses evaluated (3, 10, 30 mg/Kg) as mentioned above and shown in Figure 19. Only the dose of 3 mg/Kg demonstrated a significant neuroprotective effect compared to the control group when administered intravenously 30 minutes prior and 1 hour after stroke onset, showing a reduction from 23% to 13.5% infarcted tissue resulting from MCAO. In contrast with this result, doses of 10 and 30 mg/Kg produced significant increases in brain damage (31.7% and 29.4%, respectively) demonstrating a neurotoxic effect in these stroke conditions. The neuroprotective effect of the 3 mg/Kg dose on the brain of a representative animal compared to a control brain is shown in the brain slices depicted in Figures 20a and 20b. The colorless area represents ischemic or dead tissue. The deleterious effects of the higher doses are shown in Figures 20c and 20d.

This is a preliminary study of the effects of a pentane extract of Lepidium meyenii administered intravenously in rats subjected to focal ischemic stroke. Previous pharmacological studies of the plant were performed with aqueous, hydroalcoholic or hexane extracts administered orally at equivalent doses without showing toxic effects. The pentane extract of Maca, obtained by the method described, contains mostly liposoluble constituents such as alkaloids, benzylated amides (macamides), polyunsaturated oxoacids (macaenes) and benzylisothiocyanates. When these compounds are

administered intravenously, they likely reach the brain at higher concentrations than by other routes of administration. The results suggest that there are two different kinds of activity in the brain during stroke in rats, and these activities might depend on the interactions of constituents with different cellular targets. One activity, at low doses, decreases infarct volume, indicating an inhibitory, antioxidative or antiapoptotic mechanism that prevents cell death or protects neurons from an ischemic insult. At higher doses another activity increases infarct volume, indicating that some excitatory, apoptotic or neurotoxic processes might be activated. For instance, glucosinolates and

106

isothiocyanates have reported pro-apoptotic and anti-proliferative effects (Johnson, 2002; Leoni, 1997).

Confirmation of the neuroprotective effects of the pentane extract at 3 mg/Kg was made in Wistar rats subjected to focal ischemic stroke (n=4) using the intraluminal suture model. The results revealed statistically significant differences as shown in Figure 21. The control and treated groups demonstrated overall percent ischemic tissues of 23.1% and 12.8% respectively. The results from two different models of stroke showing neuroprotective effects for the dose of 3 mg/Kg (containing 3.6 jxg of Macamide-2/mg extract), suggest that some of the constituents present in the pentane extract could be neuroprotectants, inhibiting processes of cell death. The results also lead to the hypothesis that one specific chemical compound might be isolated to produce this neuroprotective effect without causing the deleterious effects observed at higher doses of pentane extract.

In vitro neuroprotective studies

Discovery of the neuroprotective activity of a pentane extract has motivated the further study of macamides (alkamides), which represent the chemical feature which differentiates Maca from other species possessing some of the same constituents but lacking neuroprotective activity.

Alkamides are lipid derivatives, and those present in Maca hypocotyls are designated macamides. They are JV-benzylated alkanoyl amides, and are considered interesting constituents due to a chemical similarity with the endogenous lipid derivatives present in vertebrates which form part of the endocannabinoid system. Macamides have been isolated from hexane extracts of Maca. 7V-benzylhexadecanamide (Figure 1) and Nbenzyl-5-oxo-6,,8'-octadecadienamide (Figure 2) were the first two macamides reported, along with the acyclic keto acid 5-oxo-6is,8ii-octadecadienoic acid

(Muhammad, 2002; Ganzera, 2002). As previously mentioned, other studies have led to isolation and characterization of several additional alkamides, namely: jV-benzyl-9-oxo107

12Z-octadecenamide,

7V-benzyl-9-oxo-12Z, 15Z-octadecadienamide,

TV-benzyl-13-oxo-

9',ll.E,-octadecadienamide, jV-benzyl-15Z-tetracosenamide, and jV-(m-methoxybenzyl)hexadecanamide, 7V-benzyloctadecanamide, jV-benzylpentadecanamide, N-

benzylheptadecanamide,

./V-benzyl-9Z-octadecenamide,

7V-benzyl-9Z, 12Zas well as the

octadecadienamide, and iV-benzyl-9Z,12Z,15Z-octadecatrienamide

methoxylated forms of the unsaturated compounds (Zhao, 2005; McCollom, 2005). Therefore, these compounds have been proposed as markers for standardization (Zhao, 2005; McCollom, 2005) and their pharmacological screening has been considered as an important step in the study of Lepidium meyenii.

With the in vitro model of neurons subjected to oxidative stress with H2O2, the protective effects of the pentane extract were confirmed and the isolated and synthesized alkamides were evaluated for their neuroprotective activity (Sarang, 2002). Oxidative stress, defined as a disturbance in prooxidant/antioxidant balance, has been implicated in several neurodegenerative diseases which lead to cell death. Although reactive oxygen species (ROS) such as H2O2 are produced in normal metabolic processes, overproduction can result in oxidative stress injuries that damage the integrity of cell functions, causing subsequent death by either necrosis or apoptosis (Mehta, 2007). In optimal conditions of cell culture without oxidative stressors, neurons assume appropriate dendritic shapes, extending and forming synapses with surrounding neurons (Burov, 2000; Lau, 2006). When oxidative stressors are applied, these neurons lose their dendrites and transform into spherical shapes, suggesting a reversion to an undifferentiated state and eventually, depending on the intensity of stress, neurons can also undergo apoptosis or necrosis in vitro.

Development of cell culture assays of neuroprotection includes microscopic evaluations and cell viability assays to determine the extent of neuron damage. We considered two cell lines for screening and characterization of the neuroprotective effects of the pentane extract, macamides and synthetic alkamides. Crayfish neurons were used primarily because they are natural neuronal cells that do not transform, but with the disadvantage of their invertebrate origin. Subsequently, rat neuroblastoma cells were used 108

in further studies, since their mammalian origin makes them more similar to human neurons, but with the disadvantage of their cancer origin.

Studies with crayfish neurons revealed reductions in F^CVinduced stress in cells pretreated with different concentrations of pentane extract. Crayfish neurons treated with 30 ug/mL of pentane extract displayed more appropriate dendritic cell shapes and greater cell densities than those treated with 3 ug/mL. Moreover, the cell viability assay demonstrated a significant linear correlation between the concentration of pentane extract and percent neuroprotection, which ranged from 7% at 0.1 j^g/mL to 88% at 30 p.g/mL pentane extract, as shown in Figures 23 and 24. Significant effects were observed with concentrations of 1, 3, 10 and 30 jj,g/mL. Another way to interpret these results is to use percent viability, as shown in Figure 25. These results demonstrate that the pentane extract from Maca acts in a concentration-dependent manner and it is apparently not neurotoxic to crayfish neurons nor does it induce more damage in stressful conditions. These results also demonstrated that the neuroprotective activity observed in vivo could be reproduced in cell culture.

One advantage of cell culture as a model to evaluate neuroprotection is that it uses small amounts of study compounds. Macamide-1 and -2 were provided from the University of Mississippi in small amounts (1 mg), and were screened in cell culture for their possible neuroprotective effects. Crayfish neurons pretreated with 10 uM macamides and subjected to oxidative stress with H2O2 three hours later, had significantly increased viability with macamide-2 compared to cells treated with only H2O2. The percent neuroprotection for macamides 1 and 2 were 11.1% and 23.4% respectively, as shown in Figure 26. This was the first preliminary evidence that these compounds might be involved in the neuroprotective activity of the Maca plant. Another consideration is that macamide-2 has a longer fatty acid chain and more complex structure (carbonyl group in position 5 and two trans double bonds in positions 6 and 8) which could improve the neuroprotective activity.

109

The rat neuroblastoma B35 cell line was the other type of cells used in the study since these cells are mammalian neurons and they have been reported in several other studies of neuroprotection and neurotoxicity (Otey, 2003; Croslan, 2008). B-35 cells were included in our study to demonstrate the neuroprotective activity of lipophilic compounds found in Maca in mammalian cells. Figure 27b shows the effect of exposure to 0.3 mM H2O2 on neuroblastoma cells, observed 24 hours after treatment and compared to control cells treated with only vehicle (Figure 27a). The neuroprotective effects of 3 ug/mL pentane extract and lOuM anandamide were also confirmed with this model (Figures 27c and 27d). These results represent a validation of the method with neuroblastoma cells. The use of anandamide as a control compound supports the relationship between the neuroprotective activities of Maca and alkamides and the endocannabinoid system. The microscopic evaluations of the concentration-dependent effects of pentane extract on neuroblastoma cells are shown in Figure 28.

On the basis of the neuroprotective activity demonstrated by the pentane extract, preliminary results for isolated macamides and the hypothesis that macamides could be active in neuroprotective processes, some macamides and derivatives were synthesized by Mr. Hui Wu and Dr. Charles Kelley of the Laboratory of Medicinal Chemistry at MCPHS. (MCP-11, 12, 13, 14, 15, 31, 32, 33).

The neuroprotective effects of all MCP compounds were screened using a model of rat neuroblastoma cells subjected to oxidative stress with H2O2, followed by measurement of cell viability with the trypan blue exclusion cell counting method (Figure 29). Cells pretreated with MCP compounds at 10 uM and subjected to H2O2 one hour later, resulted in significantly increased cell counts for MCP-15, MCP-32, and MCP-33, with averages of 71.910.1%, 73.33.9%, and 80.55.1% of vehicle-treated cell viability respectively, compared to wells treated with only H2O2 (34.83%). Other MCP compounds did not demonstrate significant neuroprotective effects. There was a correlation between fatty acid complexity along with substitution of functional groups in the benzyl group and the neuroprotective effect. Further studies could involve synthesis of other derivatives to investigate improvement of the neuroprotective effects. 110

Since MCP-33 has demonstrated significant activity in preliminary tests of neuroprotection, it was chosen for evaluation of neuroprotective potential, together with the pentane extract as a source of a variety of alkamides.

In order to assess the possible toxic or proliferative effects of pentane extract and compound MCP-33, a study was completed to evaluate both materials in neuroblastoma cells (Figures 30 and 31). Cells were exposed to different concentrations of pentane extract or compound MCP-33, and cell viability was evaluated 24 hours later. There were no significant differences in cell viability at any of the concentrations tested, indicating that neither of these test materials have a direct effect on neuroblastoma cells within the range of concentrations evaluated.

Pentane extract at 1.5 and 5 p.g/mL demonstrated significant neuroprotective effects, estimated by the viability of the exposed neuroblastoma cells (62.57% and 53.84.2% respectively), compared to the control cells treated with only H2O2 (33.1=1=2.9%, Figure 32). The model did not show concentration-dependent effects as observed earlier in the crayfish neuron model, possibly due to differences in the constitution of mammalian cells, which are characterized by more complex mechanisms of cell damage and neuroregeneration.

Anandamide, an endocannabinoid with a chemical structure belonging to the group of alkamides and with demonstrated neuroprotective effects (Mechoulam, 2002; Schabitz, 2002; van der Stelt, 2001; Franklin, 2003; Mechoulam, 2007; Schomacher, 2008), was evaluated by the same procedure, at concentrations between 5 and 50 uM, and subjected to H2O2 one hour later (Figure 33). Five uM was the only concentration showing a significant neuroprotective effect, as indicated by increased viability (66.910%) compared to control cells treated only with H 2 0 2 (30.94.3%). Higher concentrations of anandamide (15 and 50 uM) did not demonstrate neuroprotective effects. We suggest that these results for anandamide can be explained by the inverted Ushaped dose-effect relationship called hormesis. This relationship could be explained by 111

the fact that anandamide as well as other endocannabinoids produce rapid desensitization by membrane rafts and endocytosis of receptors, and also by the fact that anandamide is an agonist at the vanilloid receptor and produces an effect opposite to CBi receptor activation. The presence of macamides in the pentane extract, the structural similarity between macamides and anandamide, and the lack of proportionality of concentrationdependent effects of both entities indicate that both might act by similar mechanisms in neuroblastoma cells. THC, a natural cannabinoid found in Cannabis sativa, with a pharmacological activity due primarily to the activation of the CBi receptor, was evaluated using the same model investigating concentration-dependent neuroprotective effects in neuroblastoma cells (Figure 34). THC at 15 uM demonstrated significant neuroprotective effects compared to control cells treated with only H2O2. Other lower and higher concentrations of THC (5, 50 and 150 uM) did not show significant neuroprotective effects, which are attributed to the inverted U-shaped concentration-dependent effects observed earlier with anandamide. THC, an agonist at CB receptors, has a structure very different from that of the alkamides and some of its actions on the CNS are still unclear. MCP-33, a derivative of the natural macamides found in Lepidium meyenii, was evaluated for concentration-dependent neuroprotective effects in neuroblastoma cells treated with concentrations of 5 to 50 uM (Figure 35). Treatment with concentrations of 15 and 50 uM showed significant neuroprotective effects compared to control cells treated with only H2O2. Exposure of cells to 5 (iM did not result in a significant neuroprotective effect. These results demonstrate a concentration-dependent effect with MCP-33, surprisingly without the inverted U-shaped concentration-dependent effects observed earlier with anandamide and THC. The potentially more efficient activity observed with MCP-33 could be confirmed in further studies designed to elucidate the mechanism of action of the alkamides as well as to clarify the structure-activity relationships of these compounds.

The reduction of H202-induced cell oxidative stress in vitro by pentane extract, macamides, anandamide, THC and MCP-33 demonstrated with crayfish neurons and 112

neuroblastoma cells in this study support the suggestion that alkamides have potential neuroprotective properties. Several mechanisms could be responsible and include direct scavenging of ROS, stabilizing cell membranes to prevent lipid peroxidation, alleviating damage by increasing intracellular defensive mechanisms, or inhibiting or blocking biochemical cascades which lead to cell death. The techniques and comprehensive methods developed to test neuroprotection in cell culture establish a foundation for future studies with these and other neural cells and with additional methods.

These results might indicate new directions for in vivo studies to determine the potential neuroprotective effects which should guide future research on MCP-33. Other derivatives and animal models of stroke and neurodegeneration need to be systematically applied to evaluate neuroprotective activity and toxic effects of these compounds. MCP33 and other derivatives could be potential agents to prevent or treat neurodegenerative diseases such as stroke, TBI, Alzheimer's disease and Parkinson's disease.

Involvement of the endocannabinoid system

The cannabinoid system has important roles in cell differentiation and brain development (Rodriguez de Fonseca, 2005). A relevant function of the endocannabinoid system is modulation of neurotransmitter release, primarily of GABA, glutamate and dopamine; however, other reports have indicated that this system modulates other neurotransmitters and endocrine molecules as well. Impairment of the endocannabinoid system in dopaminergic neurons might be involved in the pathogenesis of many neurological and psychiatric disorders (such as Alzheimer's disease, Parkinson's disease, depression and schizophrenia). Therefore pharmacological targeting of the

endocannabinoid system could be useful to prevent or treat these pathologies (Melis, 2006)

There have been several studies demonstrating that endocannabinoids are involved in protecting the brain from acute damage (Mechoulam, 2002; Schabitz, 2002; 113

Hansen,

2001).

Alkamides

such

as

anandamide,

palmitoylethanolamide

and

oleylethanolamide have been detected in humans after stroke (Schabitz, 2002; Berger, 2004). Exogenous administration of endocannabinoids in animal studies decreases brain damage, suggesting that they could interfere the pathophysiological events, reducing brain damage. This is relevant in the search for treatments of brain pathologies such as cerebral ischemia, post-traumatic brain injury and neurodegenerative disorders (van der Stelt, 2001; Franklin, 2003; Mechoulam, 2007; Schomacher, 2008). The structural similarities of endocannabinoids and macamides led to the hypothesis that they could be compounds capable of neuroprotection and that purified macamides or synthetic derivatives of macamides could potentialy act as neuroprotectants by way of the endocannabinoid system.

The

compound

(6a/?,10ai?)-3-(l-methanesulphonylamino-4-hexyn-6-yl)(O-2050) is a synthetic

6a,7,10,10a-tetrahydro-6,6,9-trimethyl-6i/-dibenzo[Z?,cT|pyran

competitive "silent" antagonist at the CBi receptor (Figure 36). O-2050 was included in this study as a tool to demonstrate the participation of CBi receptors in the neuroprotective activities of Maca, cannabinoids (anandamide and THC), and the compound MCP-33.

Figure 37 shows the concentration-dependent effects of O-2050 at three concentrations between 5 and 50 uM on viability of neuroblastoma cells. There was no difference between control and cells treated at 5 uM, the lowest concentration tested. Concentrations of 15 and 50 uM significantly reduced the viability of neuroblastoma cells (77.34.9% and 28.9 4.7% respectively) compared to control cells. This result indicates that CBi receptors are important to maintain viability of these cells. Saturating concentrations of O-2050 block the activity of CBi receptors, which could impair vital cellular processes affecting the overall viability of neuroblastoma cells.

The neuroprotective effects of pentane extract, anandamide, THC and MCP-33 were evaluated in the presence of O-2050 at the highest concentration investigated

114

earlier. The results obtained in this study demonstrated that macamides and synthetic alkamides might be considered cannabinergic compounds.

The neuroprotective effect of pentane extract was evaluated in neuroblastoma cells treated with 0.5 to 5 ug/mL extract with or without the prior addition of 50 uM O2050, and subjected to 0.3 mM H2O2 one hour later (Figure 38). Viability was measured with MTS reagent. The significant neuroprotective effect demonstrated by pentane extract at 1.5 and 5 ug/mL was significantly blocked by the addition of O-2050. Viability decreased from 55.9% to 24.1% and from 47.2% to 23.1% respectively when O-2050 was added, indicating that the CBi receptor participates in the neuroprotective effect. Control cells treated with O-2050 and H2O2 had 15% viability, a significantly lower value than when pentane extract was added to the wells at 5 |xg/ml (23.1%). This observation indicated that the pentane extract, by means of some of its constituents, could counteract the antagonism resulting from exposure to O-2050 or could competitively restore cell viability.

Anandamide, as a reference cannabinoid compound, was also evaluated in neuroblastoma cells treated with concentrations of 5 to 50 |j,M, with or without prior addition of 50 uM O-2050, and subjected to 0.3 mM H2O2 one hour later (Figure 39). Cell viability was measured with MTS reagent. The significant neuroprotective effect demonstrated by anandamide at 5 uM was significantly blocked when O-2050 was added. Viability decreased from 69.6% to 16.8%. Control cells treated with O-2050 and H2O2 had 15% viability which was not significantly different from the viability when anandamide was present. The failure of anandamide to counteract the antagonism of O2050 could be explained by the fact that anandamide is a partial agonist at CBi receptors and also because it has a lower affinity at CBi receptors than O-2050 (KjCBi 89 nM and 2.5 nM respectively).

THC was also evaluated in neuroblastoma cells treated with concentrations of 5 to 50 |iM, with or without the prior addition of 50 |xM O-2050, and subjected to 0.3 mM H2O2 one hour later (Figure 40). The significant neuroprotective effect demonstrated by 115

THC at 15 uM was significantly blocked by the addition of O-2050. Cell viability decreased from 54.4% to 20.2% when O-2050 was added. Control cells treated with O2050 and H2O2 had 15% viability, which was not significantly different from that when THC was present. THC is also a partial agonist at CBi receptors, and it has lower affinity for these receptors than O-2050 (KjCB!=5 nM).

The evaluation of MCP-33 was also performed in neuroblastoma cells treated with concentrations of 15 and 50 uM, with or without the prior addition of 50 uM O2050, and subjected to 0.3 mM H2O2 one hour later (Figure 41). Cell viability was measured with MTS reagent. The significant neuroprotective effects demonstrated by MCP-33 at both concentrations evaluated were significantly blocked by the addition of O-2050. Cell viability decreased from 60.1% to 15.5% and from 51.0% to 24.8%

respectively when O-2050 was added. Control cells treated with O-2050 and H2O2 had 15% viability. The significantly higher (24.8%) viability with exposure to 50 uM MCP33 in the presence of 50 uM O-2050 and H 2 0 2 indicated that MCP-33 could competitively counteract the antagonism of O-2050 at CBi receptors.

Previous studies have demonstrated that the compound O-2050 blocks the effects of cannabinergic compounds or produces effects opposite to those of endocannabinoids. The results presented in this study constitute a partial confirmation that alkamides such as macamides and derivatives (MCP-33) are neuroprotective by acting on the

endocannabinoid system. Further studies, such as binding studies, would be necessary to demonstrate if in fact these compounds act directly on CB receptors or indirectly by inhibiting the enzyme FAAH or by blocking reuptake transporters.

Evaluation of antioxidant activity

Since the method to demonstrate neuroprotection in cell culture employed H2O2 to produce oxidative stress and cell death, and since H2O2 and other peroxides are produced in pathophysiological conditions of brain injury, confirmation of the antioxidant potential of pentane extract, anandamide and MCP-33 was considered of 116

interest to demonstrate participation in the mechanism of action of neuroprotection. Previous studies had confirmed that antioxidant compounds could be neuroprotective in vitro and in vivo and their antioxidant activities could be complementary to other mechanisms of action.

In this study, evaluation of antioxidant activity was performed by the ABTS method, with a trolox standard for comparison. Trolox is a vitamin E derivative which possesses demonstrated antioxidant effects by its action as a reactive oxygen scavenger. Figure 42 shows that absorbance is inversely proportional to the antioxidant activity of a test compound. Trolox at concentrations of 0.05 to 1 mM demonstrated concentrationdependent effects inhibiting oxidation of ABTS, as confirmed by absorbances at 750 nm. Control assays produced maximal absorbances and trolox significantly reduced absorbance at concentrations of 0.1 mM or higher, confirming its antioxidant activity (Figure 42).

Pentane extract did not inhibit the oxidation of ABTS reagent at any of the assayed concentrations from 0.02 to 5 (ig/mL (Figure 42), demonstrating that the neuroprotectant activity both in vitro and in vivo is not produced by a direct antioxidant mechanism or by the scavenging of reactive oxygen species. Previous reports have indicated that aqueous Maca extracts are antioxidant. The lack of antioxidant activity of the pentane extract is an indicator that the antioxidant compounds that could be present in Maca are hydrosoluble and are not present in the pentane extract.

Anandamide and MCP-33 did not demonstrate antioxidant effects at any of the concentrations evaluated from 0.05 to 1 mM. Both compounds failed to inhibit the oxidation of ABTS, as absorbances did not change in response to any of the tested concentrations. These results demonstrated that the neuroprotective activities of both compounds are not related to direct antioxidant mechanisms or the scavenging of reactive oxygen species.

117

Based on the collective results, this study indicates that pentane extract from Lepidium meyenii and the alkamide MCP-33, an analogue of natural macamides, are potential neuroprotective products as demonstrated both in vitro and in vivo. These neuroprotective effects could involve the endocannabinoid system but are not produced by antioxidant activity. Further studies are needed to demonstrate other possible targets or other mechanisms involved in the activities of these products.

118

CONCLUSIONS (* New findings)

1. * The hypocotyls of Lepidium meyenii yielded 13.3% of crude methanol extract, which after re-extraction with pentane yielded 0.7% of dried pentane extract. 2. The IR analysis of the pentane extract yielded a characteristic transmitance spectrum, demonstrating a chemical composition as previously demonstrated by Muhammad et al, 2002. 3. HPLC analysis of the pentane extract by the method proposed by Ganzera (2002) and McCollom (2005), demonstrated a similar chromatographic pattern as shown previously for Lepidium meyenii extracts. 4. * HPLC quantification using a standard of macamide-2 provided by the University of Mississippi (Muhammad, 2002) demonstrated the presence of this macamide at concentrations of 3.55 and 0.027 ng/mg in pentane extract and crude methanol extract respectively. 5. Dried Maca hypocotyls contain aproximately 2.4 mg of macamide-2 in 100 g of plant material. 6. * Estimation of the limit of quantification (LOQ) and limit of detection (LOD) for macamide-2, resulted in values of 0.17 and 0.05 ng/ml, respectively. 7. The HPLC method of analysis for macamide-2 is reproducible and linear, as is shown by the calibration curve. 8. * In vivo studies with the Chen and filament models of focal permanent cerebral ischemia in rats were demonstrative for the evaluation of neuroprotective activity of pentane extract from Lepidium meyenii. 9. * The pentane extract, at a dose of 3 mg/Kg administered intravenously in rats 30 minutes prior to and 1 hour after the onset of ischemia, produced a reduction in infarct volume from 23% to 13% compared to controls in both models of focal cerebral ischemia. 10. * The pentane extract at doses of 10 and 30 mg/Kg, administered intravenously in rats 30 minutes prior to and 1 hour after the onset of ischemia, resulted in a

119

detrimental effect, increasing the infarct volume from 23% to 32% and 29%, respectively, in the Chen model of focal cerebral ischemia. 11. Crayfish neurons can be used as a reliable in vitro model for neuroprotection studies using H2O2 (1 mM) as a neurotoxic agent. 12. * Pretreatment of crayfish neurons with the pentane extract of Lepidium meyenii demonstrated a concentration-dependent neuroprotective effect when neurons were subjected to H2O2 (1 mM) as a neurotoxic agent. 13. * Concentrations of the pentane extract greater than 1 ug/ml demonstrated significant neuroprotective effects in crayfish neurons subjected to H2O2 (1 mM) as a neurotoxic agent. 14. * Macamide-2 at 10 |iM (provided by the University of Mississippi) demonstrated a significant neuroprotective effect when assayed with the crayfish neuron model described above. 15. * The laboratory of Medicinal Chemistry at MCPHS (Mr. Hui Wu and Dr. Charles Kelley) has been successful in synthesizing compounds chemically related to natural macamides (alkamides) present in Lepidium meyenii producing adequate amounts for study and optimizing the pharmacological activity by various derivatizations (MCP compounds). 16. The trypan blue exclusion method for counting the actual number of viable cells can be used to demonstrate neuroprotective activity in rat neuroblastoma cells subjected to neurotoxic agents such as H2O2 (0.3 mM). 17. * The MCP compounds showed neuroprotective activity in neuroblastoma cells evaluated with the trypan blue exclusion method. Chemical structure modification influenced the activity and might be applied in the future to improve the activity of new compounds. 18. * The compound MCP-33 showed the maximal neuroprotective effect with evaluation by the trypan blue exclusion method in rat neuroblastoma cells, using H2O2 (0.3 mM) as a neurotoxic agent. 19. * The pentane extract and MCP-33 compound by themselves do not show either proliferative or toxic effects when evaluated in rat neuroblastoma cells. 120

20. * The pentane extract had significant neuroprotective effects at 1.5 and 5 ug/ml in rat neuroblastoma cells subjected to 0.3 mM H2O2 and evaluated with MTS reagent. 21. * Anandamide at 5 uM showed significant neuroprotective effects in rat neuroblastoma cells subjected to 0.3 mM H2O2 and evaluated with MTS reagent. 22. * THC at 15 uM showed significant neuroprotective effects in neuroblastoma cells subjected to 0.3 mM H2O2 and evaluated with MTS reagent. 23. * MCP-33 compound at 15 and 50 uM showed significant neuroprotective effects in rat neuroblastoma cells subjected to H2O2 0.3 mM and evaluated with MTS reagent. 24. * O-2050 (CB1 receptor antagonist) by itself significantly reduced the viability of rat neuroblastoma cells at concentrations higher than 15 uM, evaluated with MTS reagent. 25. * O-2050 at 50 uM blocked the neuroprotective effect previously demonstrated with the pentane extract in rat neuroblastoma cells, when evaluated with MTS reagent. 26. * O-2050 at 50 uM blocked the neuroprotective effect previously demonstrated with anandamide in rat neuroblastoma cells, when evaluated with MTS reagent. 27. * O-2050 at 50 uM blocked the neuroprotective effect previously demonstrated with THC in rat neuroblastoma cells, when evaluated with MTS reagent. 28. * O-2050 at 50 uM blocked the neuroprotective effect previously demonstrated with MCP-33 in rat neuroblastoma cells, when evaluated with MTS reagent. 29. Trolox at 0.1 to 1 mM produced a significant concentration-dependent antioxidant effect as demonstrated by the ABTS method and it can be used as a control to evaluate antioxidant activity. 30. * The pentane extract at concentrations up to 5 ug/ml did not show any antioxidant effects, as determined by the ABTS method using trolox as a control antioxidant. 31. * Anandamide at concentrations up to 0.5 mM did not show any antioxidant effects, as demonstrated by the ABTS method using trolox as a control antioxidant. 121

32. * MCP-33 at concentrations up to 1 mM did not show any antioxidant effects, as determined by the ABTS method using trolox as a control antioxidant.

122

FUTURE RESEARCH
The present study has demonstrated that the pentane extract from Lepidium meyenii is a potential neuroprotective agent, as displayed both in vitro in crayfish neurons and rat neuroblastoma cells, and in vivo in rats subjected to focal ischemic stroke.

On the basis of the above results, and considering the presence of alkamides (macamides) in the pentane extract, this study has also demonstrated that the synthetic alkamide MCP33, an analogue of natural macamides, has neuroprotective effects in vitro in rat neuroblastoma cells. These neuroprotective effects of Lepidium meyenii and MCP-33 appear to involve the endocannabinoid system, and they are not the result of an antioxidant action.

Future research is necessary to obtain more information about the mechanisms involved in the neuroprotective activity of the alkamides, characterizing in more detail the involvement of the endocannabinoid system, investigating other possible targets and developing more active alkamides guided by the structure-activity relationships of these compounds.

Future research should include:

1. Biologically guided synthesis of new alkamides and neuroprotective screening using in vitro models described in the present study, emphasizing substitutions in the benzyl moiety and changes in the fatty acid moieties of new compounds. 2. Determination of the effects of MCP-33 and other promising new alkamides on the release of dopamine and glutamate in PC-12 cells. 3. Determination of the ability of MCP-33 and other promising new alkamides to bind and activate CBi receptors in rat neurons (whole brain membranes, neuroblastoma cells and other cell lines transfected with the CBi receptor gene).

123

Determinations should include CBi receptor antagonists, reuptake inhibitors and FAAH inhibitors when possible. 4. Determination of the effects of MCP-33 and other promising new alkamides in rats subjected to focal ischemic stroke, including a dose-escalation and a timewindow study. Determinations should employ CBi receptor antagonists, reuptake inhibitors and FAAH inhibitors, and should also include histological and behavioral measurements of brain damage and function. 5. Studies of the effects of MCP-33 and other promising new alkamides, in animal models of Alzheimer's disease. 6. Studies of the effects of MCP-33 and other promising new alkamides, in animal models of Parkinson's disease. 7. Studies of the effects of MCP-33 and other promising new alkamides, in animal models of traumatic brain injury (TBI).

124

FIGURES
Figure 1 Chemical structure of macamide-1

125

Figure 2 Chemical structure of macamide-2

126

Figure 3 Chemical structure of alkaloid macaridine

OH

127

Figure 4 Chemical structure of alkaloid lepidilin A: 1,3-dibenzyl-4,5-dimethylimidazolium chloride

128

Figure 5 Chemical structure of alkaloid lepidilin B: 1,3-dibenzyl-2,4,5-trimethylimidazolium chloride

cue
cr

129

Figure 6 Chemical structure of alkaloid (l/?,3-S)-l-methyltetrahydro-B-carboline-3-carboxylic acid.

COOH

130

Figure 7 Chemical structure of glucotropaeolin

H03S

131

Figure 8 Chemical structure of methoxyglucotropaeolin

OH

NV\-OCH 3
HO3S

Figure 9 Chemical structure of benzylisothiocyanate

133

Figure 10 Chemical structure of methoxybenzylisothiocyanate

H3CO

YX

134

Figure 11 Endocannabinoid system

PK cAMP PRESYNAPTIC TERMINAL ATP

1..2-

X
c

SiS

x.
Kt
OH <.a.lTAM,\Tt OABA. DOPAMINEN E | J V 1 I At.ETYlX'HOL >UNF I 1

NMDA

ANANDAMIDE s = ^ * = ^ ^ jj < J T ^

.V'
,t:--r-, ' AT
1A4H
j /

PE .OH
t 1*1 '

ortrp-^

r-1

^ ob
!
11

:'

Ck# vC + *-.

R
tAMP

ANANDAMIPE ETHANOLAMIDK

-"-'

__ ( \\m\
""""""" Ca2+

/ "

POSTSYNAPTIC: CELL

ACTIVn'Y OF PK <PKA / M A P K i

CELLULAR RESPONSES

Source: Rodriguez de Fonseca, 2005

135

Figure 12: Ischemic pathophysiology

Vessel Occlusion^/
Red used CfiF \ Hpaedysemia Hypwemla

Ischemic area

*jj3>;' Depolarization of

, 'GiuBmate release

* Giutemite receptors

Dying neuron

of pttsftflarewaton,' cytoUneg

Ca*+f>

we

tlnfiltraitionw. Free radfcai Act tva ting irtfiarrima*cFy" Q O G bload cells * o o

t
v

(
++

Protease

1
J

DMA + Protean, damage ,^_ Membwe damage ftenrjitic + IPSPtDtC Cell dssiBi

G
'

ff

Release: of recovery cnhanditg factors

Recovery zone .x-^

/'

V>^

IP Asogenesis Der. tfrifc 'branching

Neuroqeitesls

Source: Schabitz 2006

Figure 13: Zones of focal cerebral ischemia

VIABLE TISSUE

VIABLE TISSUE

VMMXTOSisE

Source: Mehta 2007

137

Figure 14 IR-SPECTRUM OF PENTANE EXTRACT FROM Lepidium meyenii Walp. (MACA)

z >-v*'"*'Nx 70-E \ 65 60^ 55-

^^_^
^"^ """w^

\ \
I
I
'i

\
\
\

/
I
1

50^ 4 5 -_

\ \
i

/ \ \ \ \
--~-s

/ /

I1A
i V

fu

| ; I ! !
I ' ' I

/ I

/ !

^ , | !
1 \
J
1

E E

8 g

40 E: 35 E: 30 -E 25 -E 20 -E
I

/ ;
j

j f

i
A

.'V J
j i

V ~^/
to <o

h
I
i ii j j

!
!;!

!Vi ^
! <D /

j'l ' i (
/
) i
1-

/A

/ < , i
1 A,
ir\
i
i

r /
/,

''i

j i

!i

\ 1
i

, /

15-E 10-E 5-= 0-E -5-E

O O

! 1
n
I ; /

I
2
f-

J
1

m
r.

!V Y
i
CM *"

I
s

/ j r^ K
S
r-

\
^

-A/ ty oo in
C M

v., *

*-

--

'

3500

3000

2500

2000

1500

1000

Wavenumbers (cm-1) Date: Wed Nov 10 14:40:18 2004 Scans: 8 Resolution: 4.000 Maea pentane-soluble fraction

138

Figure 15 HPLC chromatogram of pentane extract from Lepidium meyenii (Gradient: 55% to 95% acetonitrile in water + 0.1% acetic acid, 45 min, 1 ml/min, detection 280 nm, sample 10 ul)

^BWErSigBMWSee30.1TiWRKlpWASSSr
(nALI |

85

- J* ^ A - t l .\.LU

Figure 16 HPLC chromatogram of the standard of macamide-2 (JV-benzyl-5-oxo-6,8ii-

octadecadienamide) 0.16 |ig/ml

DAD1 E, Si8=280,l5.Ret= 380.100 (MARSTD5B.Dr mAU 1 r <e CSUN

W CFCti [NTS'

axe mm
tMTN V V

<dnb F l CSTM v w j A .wv

vv

-1.5 H

r*\ .f~

-1.6

-1.626 26 2S

Figure 17 Calibration curve for the HPLC analysis of macamide-2 standard at concentrations of 0.16, 1.6, 16 and 160 jig/ml (n=2-5, R2=0.9990)

3000

2500
2000
13

1500
a.

1000
500

./

50

100 Concentration ng/ml

150

200

141

Figure 18 Calibration curve for the HPLC analysis of macamide-2 at three concentrations from 0.16 to 16 ng/ml and averaged peak areas (R2=0.9995).

250

A
200

y
n

150

..Z-

tz

<u o.

1 0 0

_^r_

50

r_
0

10 Concentration iig/mi

15

20

142

Figure 19 Effect of pentane extract from Maca on % infarct in rats subjected to focal ischemic stroke, (n=9, Chen model). Values represent the mean S.E.M. Significance levels *P<0.05 compared to control.

40 35 30 25

*.

i- j ^
I
y" (13.5}
l

Infarct

20 15 10 5 0

'

{29,4}

control

3mg/Kg

lOrng/Kg 30rng/Kg

Doses of pentane extract

143

Figure 20 Representative subjected to focal is ections of rats treated i.v. with pentane ic stroke (Chen model). Animals

twice, 30 minutes prior and 1 hour after MCAO (n=9).

a) Control

b) 3 mg/Kg

c) 10 mg/Kg

30 mg/Kg

Figure 21 Effect of pentane extract from Maca on % infarct in rats subjected to focal ischemic stroke, (n=4, suture model). Values represent the mean S.E.M. Significance level *P<0.05 compared to control.

30 25 20 -i
Infarct

".

., _ _ .

_ _

:' -

10 ; 5 -

23.1:

I
12.8

n
Control Maca, 3 mg/Kg

Groups of study and treatment

145

Figure 22 Crayfish neurons (OLGA-PH-J/92) in culture.

146

Figure 23 In vitro model of neuroprotection with crayfish neurons.

5"

" f/ , y I }

v%T

a) Control crayfish neurons

b) Crayfish neurons treated with H 2 0 2 (1 mM)

:'...'tf.' ':"',"

-SK

sr

c) 3 (xg/ml of pentane extract of Lepidium meyenii + H2O2

d) 30 ng/ml of pentane extract of Lepidium meyenii + H2O2

147

Figure 24 Concentration-dependent response effect of pentane extract of Lepidium meyenii (Maca) Crayfish neurons were pretreated with the pentane extract and 3 hours later 1 mM H2O2 was added. The viability of the cells was measured with MTS reagent at 490 nm (n=5). Values represent the mean S.E.M.

120

100

80
to

w Q.

I o

60

40

iI

0.1

0.3

10

30

Concentration ng/ml

Figure 25 Viability of crayfish neurons treated with pentane extract and 3 hours later subjected to H2O2 at 1 mM (n=5). Values represent the mean S.E.M. Significance levels *P<0.05 compared to cells treated only with H2O2.

120 100 -

I
*

s
%Vi

80 60 100 40
65 fiwi

m. m- m

Pi

20 0 H202 ImM
Pent.

(-5

-<)

(4)

{+>

H
: 1 ,

(+}

<)

0.1

0.3

10

30

Treatments (Pent.=pentane extractng/ml)

149

Figure 26 Neuroprotective effect of pentane extract of Lepidium meyenii and isolated macamides, measured in crayfish neurons subjected to H2O2 1 mM. Values represent the mean S.E.M. Significance levels *P<0.05 compared to control cells (n=3).

90 80 70 60 50 40 30
2 0

f
!

*6l.7)
(0.0)

iiT.ir

10 i
0 4

J
H2021mM H202 1 m M

(23 4)

H202 1 mM Vehicle

H202 1 mM

iPentane 3.4 ng/mH Macamide-l, 10 j Macamide-2,10 nM j (iM

150

Figure 27 In vitro model of neuroprotection with rat neuroblastoma cells (B-35).

r
L
Control

r
i

^l
A
H 2 0 2 0.3 mM
^

A V

r
Pentane extract 3 ng/ml+H2C>2

A
Anandamide 10 uM + H2O2

151

Figure 28 Dose effect of pentane extract 1 hour prior to 0.3 mM H2O2 in rat neuroblastoma cells (B-35).

P
1

1 ug/ml

3 ug/ml

Ji.-kc

-&*

1R
A k
30 ug/ml

i**^ * *

10 ug/ml

Figure 29 Neuroprotective effect in neuroblastoma cells subjected to 0.3 mM H2O2 one hour after treatment with MCP compounds at 10 |uM. Cell counts were determined using the trypan blue exclusion method. Values represent the mean S.E.M. Significance level *P<0.05 compared to control cells treated only with H2O2

120 100
' *

80

I
"5 o

60 40 20 $4.1

fa
50.1 n=8

n.4
5/

57

n
m
n=4

r*i
rh
73.;
S3.7

SOi

i Control Comp 11 Comp 12 Comp 13 Comp 14 Comp 15 Comp 31 | Comp 32 Comp 33 : i=2C) n==8 r1 = 1:> n==4 n=4 | n=8 n=4

MCP compounds. 10 nM

153

Figure 30 Effect of pentane extract on the viability of neuroblastoma cells measured with MTS reagent (n=8). Values represent the mean S.E.M.

110 100 90 80 > 70 = 60

-I 100 I03.2

50 -

100

i n n fi lUU.C

> 40 s 30 20 10 n -

92.1

U 1

Vehicle

0.5|jg/ml

1.5ug/ml

5ug/ml

15(jg/ml

Pentane extract concentration

154

Figure 31 Effect of compound MCP-33 on the viability of neuroblastoma cells measured with MTS reagent (n=8). Values represent the mean S.E.M.

110 100 90 80 70 > 60 = 50 re 40 > 30 *- 20 10

-i -

100

94,7

102

U 1

>

Vehicle

5(JM

15UM

SOJJM

MCP-33 concentration

155

Figure 32 Neuroprotective effect of pentane extract in neuroblastoma cells subjected to 0.3 mM H2O2 one hour after treatment (n=8). Viability determinated with MTS reagent. Values represent the mean S.E.M. treated cells. Significance level *P<0.05 compared to vehicle-H202

120 100 > g > 3? 80 60 100 40 :

20 -

-i
I 33.1
:

I
5,3.8 '(+)

'

:A j3 ell:
H)
0.5 Treatments

62.5 *
(+;

H202

(+)

Pentane Hg/ml

1.5

156

Figure 33 Neuroprotective effect of anandamide in rat neuroblastoma cells subjected to 0.3 mM H2O2 one hour after treatment (n=4). Viability determinated with MTS reagent. Values represent the mean S.E.M. treated cells. Significance level *P<0.05 compared to vehicle-HaCh

120 100 ability 80 60

'I'
*

> 40 20 0 i H202

<i|Ki

I
30.9 ' (-} 0

66 9

I
34.S
l+)

.33.7

W
0

U)
5

(-0
50

iAnand. \iM

15

Treatments

157

Figure 34 Neuroprotective effect of THC in rat neuroblastoma cells subjected to 0.3 mM H2O2 one hour after treatment (n=4). Viability determinated with MTS reagent. Values represent the mean S.E.M. Significance level *P<0.05 compared to vehicle-H202 treated cells.

17Ci

100 Viability 80 60 -

_ J _
*
ir-

40 20 n

I
"' 30.9 H202 | THCuM

Wt
0} 5 1"reatments ,

I b4 j

'

I J8 2

_3Q g_

fS 0

(+)

(+) 15

(+)

1+) 150

50

158

Figure 35 Neuroprotective effect of compound MCP-33 in rat neuroblastoma cells subjected to 0.3 mM H2O2 one hour after treatment (n=8). Viability determinated with MTS reagent. Values represent the mean S.E.M. Significance level *P<0.05 compared to vehicleH2O2 treated cells.

120 100 bility 80 60 40 20 0

;

I *

* 100

>
0^

-II
35.3 47.4

I
S6

I
62 5

H202 MCP33 |iM :

<-) 0

(+)

(+)

w
15

(+)

50

Treatments

159

Figure 36 Chemical structure of (6a/?,10a/?)-3-(l-methanesulfonylamino-4-hexyn-6-yl)6a,7,10,10a-tetrahydro-6,6,9-trimethyl-6//-dibenzo[Z5, JJpyran (O-2050)

160

Figure 37 Effect of compound O-2050 on the viability of neuroblastoma cells measured with MTS reagent (n=8). Values represent the mean S.E.M. Significance level *P<0.05 compared to vehicle treated cells.

100 80 60 100 40 - i 20

>

ability

ifii

77 3

I
23 9 Vehicle 5uM 15 uM 50 uM

O-2050 concentration

161

Figure 38 Neuroprotective effect of pentane extract in presence of 50 \iM O-2050 (n=4) in rat neuroblastoma cells subjected to 0.3 mM H2O2. Viability measured with MTS reagent. Values represent the mean S.E.M. Significance level *P and aP <0.05 compared to O2050-vehicle-H2C>2 treated cells.

Bf)

7n

* 60 ~ % Viability 50 40 30 20 I 10 0 H202 j 15
2A

I
5F 1

a
17.2

I
23 1

M
0

(+) 1.5

{+} 1.5

(+) 5

{+)

Pentane \ Hg/ml i O-2050

(+)

(+) Treat tnents

(-)

t+)

162

Figure 39 Neuroprotective effect of anandamide in presence of 50 \xM O-2050 (n=4) in rat

neuroblastoma cells subjected to H2O2 0.3 mM. Viability measured with MTS reagent. Values represent the mean S.E.M. Significance level *P<0.05 compared to O-2050vehicle- H2O2 treated cells.

QO -

80
70 -

60 % Viability 50 40 30 20 10 0 H202 Anand. HM O-2050

S6 .5

* ~ w~ ~'
X
1

I 15
gp|ji

" "34 ""

J.

1.

1
H) 50

W
0

{+) 5

HI 5

(+) 15

Hi 15

t+)
50

(-)

H)

(-)

(-)

Treatimerits

163

Figure 40 Neuroprotective effect of THC in presence of 50 |iM O-2050 (n=4) in rat neuroblastoma cells subjected to 0.3 mM H2O2. Viability measured with MTS reagent. Values represent the mean S.E.M. Significance level *P<0.05 compared to O-2050-vehicle- H2O2 treated cells.

80 70 60

50
40

T
dfi

3? 30 20 10 0
H202 THC u M ; 0-2050 0 (+) (-) ! (+)

.I 15

6.5

n
W
(+) {+) (+) + 50 () (-.) 50 (+}

15
(-)

15
(+)

Treatments

164

Figure 41 Neuroprotective effect of compound MCP-33 in presence of 50 \xM O-2050 (n=4) in rat neuroblastoma cells subjected to 0.3 mM H2O2. Viability measured with MTS reagent. Values represent the mean S.E.M. Significance level *P and aP <0.05 compared to O2050-vehicle-H2O2 treated cells.

80 70 60 50
JO

> S 30 S

re

40

4
a

mm
-.I-.

20 10 0
H202 MCP33 \iM O-2050 (+) 0 (+) Treatments (+) 15

M-

I
24.8

15

m
(+) 15 50 (-)

50

165

Figure 42 Antioxidant effect of pentane extract, compound MCP-33 and anandamide determined with the ABTS method (n=4). Absorbance is inversely proportional to antioxidant effect. Values represent the mean S.E.M. Significance level *P<0.05 compared to oxidation in absence of trolox (0 mM).

0.3 0.25 .
u
C TO

0.2

f?

Irr

r r T
- >- in

e o.i5
o <r 0.1 0.05 0 ; OmM . O^g/rnl K Trolox B Pentane

ii- J I
0.02 : 0.05 tig/ml ; ng/ml
:

I *

;; MCP-33 mAnartd

:0.05mM. O.lmM . 0.2mM ; O.SmM ; I m M 0.5 \ 1 ng/ml ; 5 ng/ml ng/ml

Concentration (n=4)

166

REFERENCES

1. Aarts M, Liu Y, Liu L, Besshoh S, Arundine M, Gurd JW, Wang YT, Salter MW, Tymianski M. Treatment of ischemic brain damage by perturbing NMD A receptor- PSD95 protein interactions. Science. 2002 Oct 25;298(5594):846-50.

2. Abe T, Takagi N, Nakano M, Furuya M, Takeo S. Altered bad localization and interaction between bad and bcl-xL in the hippocampus after transient global ischemia. Brain Res. 2004 May 29;1009(l-2):159-68.

3. Albers GW. Expanding the window for thrombolytic therapy in acute stroke, the potential role of acute MRI for patient selection. Stroke. 1999 Oct;30(10):2230-7.

4. Albers GW, Clark WM, Atkinson RP, Madden K, Data JL, Whitehouse MJ. Dose escalation study of the NMDA glycine-site antagonist licostinel in acute ischemic stroke. Stroke. 1999 Mar;30(3):508-13.

5. Alkayed NJ, Goto S, Sugo N, Joh HD, Klaus J, Crain BJ, Bernard O, Traystman RJ, Hum PD. Estrogen and bcl-2: Gene induction and effect of transgene in experimental stroke. J Neurosci. 2001 Oct l;21(19):7543-50.

6. Antolin I, Mayo JC, Sainz RM, del Brio Mde L, Herrera F, Martin V, Rodriguez C. Protective effect of melatonin in a chronic experimental model of Parkinson's disease. Brain Res. 2002 Jul 12;943(2):163-73.

167

7. Antolin I, Rodriguez C, Sainz RM, Mayo JC, Una H, Kotler ML, Rodriguez-Colunga MJ, Tolivia D, Menendez-Pelaez A. Neurohormone melatonin prevents cell damage: Effect on gene expression for antioxidant enzymes. FASEB J. 1996 Jun; 10(8):882-90.

8. Arundine M, Tymianski M. Molecular mechanisms of

calcium-dependent

neurodegeneration in excitotoxicity. Cell Calcium. 2003 Oct-Nov;34(4-5):325-37.

9. Astrup J, Siesjo BK, Symon L. Thresholds in cerebral ischemia - the ischemic penumbra. Stroke. 1981 Nov-Dec;12(6):723-5.

10. Bari M, Battista N, Fezza F, Finazzi-Agro A, Maccarrone M. Lipid rafts control signaling of type-1 cannabinoid receptors in neuronal cells, implications for anandamideinduced apoptosis. J Biol Chem. 2005 Apr l;280(13):12212-20.

11. Baron JC, Warach S. Imaging. Stroke. 2005 Feb;36(2): 196-9.

12. Beckman JS, Beckman TW, Chen J, Marshall PA, Freeman BA. Apparent hydroxyl radical production by peroxynitrite: Implications for endothelial injury from nitric oxide and superoxide. Proc Natl Acad Sci U S A . 1990 Feb;87(4): 1620-4.

13. Bederson JB, Pitts LH, Germano SM, Nishimura MC, Davis RL, Bartkowski HM. Evaluation of 2,3,5-triphenyltetrazolium quantification chloride as a stain for detection and in rats. Stroke. 1986 Nov-

of experimental cerebral infarction

Dec;17(6):1304-8.

168

14. Belayev L, Alonso OF, Busto R, Zhao W, Ginsberg MD. Middle cerebral artery occlusion in the rat by intraluminal suture, neurological and pathological evaluation of an improved model. Stroke. 1996 Sep;27(9):1616,22; discussion 1623.

15. Berger C, Schmid PC, Schabitz WR, Wolf M, Schwab S, Schmid HH. Massive accumulation of N-acylethanolamines after stroke, cell signalling in acute cerebral ischemia? J Neurochem. 2004 Mar;88(5):l 159-67.

16. Bogani P, Simonini F, Iriti M, Rossoni M, Faoro F, Poletti A, Visioli F. Lepidium meyenii (Maca) does not exert direct androgenic activities. J Ethnopharmacol. 2006 Apr 6;104(3):415-7.

17. Bosier B, Tilleux S, Najimi M, Lambert DM, Hermans E. Agonist selective modulation of tyrosine hydroxylase expression by cannabinoid ligands in a murine neuroblastoma cell line. J Neurochem. 2007 Sep;102(6):1996-2007.

18. Braestrup C, Nielsen M, Olsen CE. Urinary and brain B-carboline-3-carboxylates as potent inhibitors of brain benzodiazepine receptors. Proc Natl Acad Sci U S A . 1980 Apr;77(4):2288-92.

19. Brenneman DE, Page SW, Schultzberg M, Thomas FS, Zelazowski P, Burnet P, Avidor R, Sternberg EM. A decomposition product of a contaminant implicated in Ltryptophan eosinophilia myalgia syndrome affects spinal cord neuronal cell death and survival through stereospecific, maturation and partly interleukin-1-dependent

mechanisms. J Pharmacol Exp Ther. 1993 Aug;266(2): 1029-3 5.

169

20. Brinckmann J, Smith E. Maca culture of the junin plateau. J Altern Complement Med. 2004 Jun;10(3):426-30.

21. Brooks NA, Wilcox G, Walker KZ, Ashton JF, Cox MB, Stojanovska L. Beneficial effects of Lepidium meyenii (Maca) on psychological symptoms and measures of sexual dysfunction in postmenopausal women are not related to estrogen or androgen content. Menopause. 2008 Nov-Dec;15(6):l 157-62.

22. Buckholtz NS. Neurobiology of tetrahydro-B-carbolines. Life Sci. 1980 Sep 15;27(ll):893-903.

23. Budd SL. Mechanisms of neuronal damage in brain hypoxia/ischemia: Focus on the role of mitochondrial calcium accumulation. Pharmacol Ther. 1998 Nov;80(2):203-29.

24. Burov YV, Uzdenskii AB, Robakidze TN. Comparative analysis of neuroprotective activity of new chemical agent vp and piracetam. Bull Exp Biol Med. 2000 Apr;129(4):362-4.

25. Bustos-Obregon E, Yucra S, Gonzales GF. Lepidium meyenii (Maca) reduces spermatogenic damage induced by a single dose of malathion in mice. Asian J Androl. 2005Mar;7(l):71-6.

26. Callaway JK, Lawrence AJ, Jarrott B. AM-36, a novel neuroprotective agent, profoundly reduces reactive oxygen species formation and dopamine release in the striatum of conscious rats after endothelin-1-induced middle cerebral artery occlusion. Neuropharmacology. 2003 May;44(6):787-800.

170

27. Canals M, Milligan G. Constitutive activity of the cannabinoid CBi receptor regulates the function of co-expressed mu opioid receptors. J Biol Chem. 2008 Apr

25;283(17): 11424-34.

28. Candelario-Jalil E, Gonzalez-Falcon A, Garcia-Cabrera M, Leon OS, Fiebich BL. Wide therapeutic time window for nimesulide neuroprotection in a model of transient focal cerebral ischemia in the rat. Brain Res. 2004 May 8;1007(l-2):98-108.

29. Chai J, Du C, Wu JW, Kyin S, Wang X, Shi Y. Structural and biochemical basis of apoptotic activation by Smac/DIABLO. Nature. 2000 Aug 24;406(6798):855-62.

30. Chang YC, Riby J, Chang GH, Peng BC, Firestone G, Bjeldanes LF. Cytostatic and antiestrogenic effects of 2-(indol-3-ylmethyl)-3,3'-diindolylmethane, a major in vivo product of dietary indole-3-carbinol. Biochem Pharmacol. 1999 Sep l;58(5):825-34.

31. Chen ST, Hsu CY, Hogan EL, Maricq H, Balentine JD. A model of focal ischemic stroke in the rat: Reproducible extensive cortical infarction. Stroke. 1986 JulAug;17(4):738-43.

32. Chung F, Rubio J, Gonzales C, Gasco M, Gonzales GF. Dose-response effects of Lepidium meyenii (Maca) aqueous extract on testicular function and weight of different organs in adult rats. J Ethnopharmacol. 2005 Apr 8;98(l-2):143-7.

33. Cicero AF, Bandieri E, Arletti R. Lepidium meyenii Walp. improves sexual behaviour in male rats independently from its action on spontaneous locomotor activity. J Ethnopharmacol. 2001 May;75(2-3):225-9.

171

34. Cicero AF, Piacente S, Plaza A, Sala E, Arletti R, Pizza C. Hexanic Maca extract improves rat sexual performance more effectively than methanolic and chloroformic Maca extracts. Andrologia. 2002 Jun;34(3): 177-9.

35. Cipriani G, Rapizzi E, Vannacci A, Rizzuto R, Moroni F, Chiarugi A. Nuclear poly(ADP-ribose) polymerase-1 rapidly triggers mitochondrial dysfunction. J Biol Chem. 2005 Apr 29;280(17): 17227-34.

36. Ciska E, Martyniak-Przybyszewska B, Kozlowska H. Content of glucosinolates in cruciferous vegetables grown at the same site for two years under different climatic conditions. J Agric Food Chem. 2000 Jul;48(7):2862-7.

37. Ciska E, Pathak DR. Glucosinolate derivatives in stored fermented cabbage. J Agric Food Chem. 2004 Dec 29;52(26):7938-43.

38. Cohen GM. Caspases: The executioners of apoptosis. Biochem J. 1997 Aug 15;326 ( Ptl)(Ptl):l-16.

39. Croslan DR, Schoell MC, Ford GD, Pulliam JV, Gates A, Clement CM, Harris AE, Ford BD. Neuroprotective effects of neuregulin-1 on B35 neuronal cells following ischemia. Brain Res. 2008 May 19;1210:39-47.

40. Cui B, Zheng BL, He K, Zheng QY. Imidazole alkaloids from Lepidium meyenii. J Nat Prod. 2003 Aug;66(8):l 101-3.

172

41. Cutler SM, Cekic M, Miller DM, Wali B, VanLandingham JW, Stein DG. Progesterone improves acute recovery after traumatic brain injury in the aged rat. J Neurotrauma. 2007 Sep;24(9): 1475-86.

42. de Lago E, Ortega-Gutierrez S, Ramos JA, Lopez Rodriguez ML, Fernandez-Ruiz J. Neurochemical effects of the endocannabinoid uptake inhibitor UCM707 in various rat brain regions. Life Sci. 2007 Feb 13;80(10):979-88.

43. De Ryck M, Van Reempts J, Borgers M, Wauquier A, Janssen PA. Photochemical stroke model: Flunarizine prevents sensorimotor deficits after neocortical infarcts in rats. Stroke. 1989 Oct;20(10): 1383-90.

44. del Zoppo GJ. Stroke and neurovascular protection. N Engl J Med. 2006 Feb 9;354(6):553-5.

45. DeMorrow S, Francis H, Gaudio E, Venter J, Franchitto A, Kopriva S, Onori P, Mancinelli R, Frampton G, Coufal M, Mitchell B, Vaculin B, Alpini G. The endocannabinoid anandamide inhibits cholangiocarcinoma growth via activation of the noncanonical wnt signaling pathway. Am J Physiol Gastrointest Liver Physiol. 2008 Dec;295(6):Gl 150-8.

46. Dietz GP, Dietz B, Bahr M. Bcl-xL protects cerebellar granule neurons against the late phase, but not against the early phase of glutamate-induced cell death. Brain Res. 2007 Aug 20;1164:136-41.

173

47. Ding Y, Zhou Y, Lai Q, Li J, Gordon V, Diaz FG. Long-term neuroprotective effect of inhibiting poly(ADP-ribose) polymerase in rats with middle cerebral artery occlusion using a behavioral assessment. Brain Res. 2001 Oct 12;915(2):210-7.

48. Dini A, et. al. Chemical composition of Lepidium meyenii. Food Chemistry. 1994;49:347-9.

49. Dong WQ, Schurr A, Reid KH, Shields CB, West CA. The rat hippocampal slice preparation as an in vitro model of ischemia. Stroke. 1988 Apr;19(4):498-502.

50. Dording CM, Fisher L, Papakostas G, Farabaugh A, Sonawalla S, Fava M, Mischoulon D. A double-blind, randomized, pilot dose-finding study of Maca root (L, meyenii) for the management of SSRI-induced sexual dysfunction. CNS Neurosci Ther. 2008 Fall;14(3):l82-91.

51. Drewes SE, George J, Khan F. Recent findings on natural products with erectiledysfunction activity. Phytochemistry. 2003 Apr;62(7): 1019-25.

52. Dringen R. Metabolism and functions of glutathione in brain. Prog Neurobiol. 2000 Dec;62(6):649-71.

53. Fabricius M, Fuhr S, Bhatia R, Boutelle M, Hashemi P, Strong AJ, Lauritzen M. Cortical spreading depression and peri-infarct depolarization in acutely injured human cerebral cortex. Brain. 2006 Mar;129(Pt 3):778-90.

174

54. Franklin A, Parmentier-Batteur

S, Walter L, Greenberg DA, Stella N.

Palmitoylethanolamide increases after focal cerebral ischemia and potentiates microglial cell motility. J Neurosci. 2003 Aug 27;23(21):7767-75.

55. Fukuda S, Fini CA, Mabuchi T, Koziol J A, Eggleston LL,Jr, del Zoppo GJ. Focal cerebral ischemia induces active proteases that degrade microvascular matrix. Stroke. 2004Apr;35(4):998-1004.

56. Furuya K, Takeda H, Azhar S, McCarron RM, Chen Y, Ruetzler CA, Wolcott KM, DeGraba TJ, Rothlein R, Hugh TE, del Zoppo GJ, Hallenbeck JM. Examination of several potential mechanisms for the negative outcome in a clinical stroke trial of enlimomab, a murine anti-human intercellular adhesion molecule-1 antibody: A bedsideto-bench study. Stroke. 2001 Nov;32(ll):2665-74.

57. Ganzera M, Zhao J, Muhammad I, Khan IA. Chemical profiling and standardization ofLepidium meyenii (Maca) by reversed phase high performance liquid chromatography. Chem Pharm Bull (Tokyo). 2002 Jul;50(7):988-91.

58. Gardner A, Mallet PE. Suppression of feeding, drinking, and locomotion by a putative cannabinoid receptor 'silent antagonist'. Eur J Pharmacol. 2006 Jan 13;530(12): 103-6.

59. Gardner EL. Endocannabinoid signaling system and brain reward: Emphasis on dopamine. Pharmacol Biochem Behav. 2005 Jun;81(2):263-84.

175

60. Gasco M, Aguilar J, Gonzales GF. Effect of chronic treatment with three varieties of Lepidium meyenii (Maca) on reproductive parameters and DNA quantification in adult male rats. Andrologia. 2007 Aug;39(4): 151-8.

61. Gasco M, Villegas L, Yucra S, Rubio J, Gonzales GF. Dose-response effect of red Maca (Lepidium meyenii) on benign prostatic hyperplasia induced by testosterone enanthate. Phytomedicine. 2007 Aug; 14(7-8):460-4.

62. Giffard RG, Swanson RA. Ischemia-induced programmed cell death in astrocytes. Glia. 2005 Jun;50(4):299-306.

63. Gill R, Sibson NR, Hatfield RH, Burdett NG, Carpenter TA, Hall LD, Pickard JD. A comparison of the early development of ischaemic damage following permanent middle cerebral artery occlusion in rats as assessed using magnetic resonance imaging and histology. J Cereb Blood Flow Metab. 1995 Jan;l5(1): 1-11.

64. Gillardon F, Lenz C, Waschke KF, Krajewski S, Reed JC, Zimmermann M, Kuschinsky W. Altered expression of bcl-2, bcl-X, bax, and c-fos colocalizes with DNA fragmentation and ischemic cell damage following middle cerebral artery occlusion in rats. Brain Res Mol Brain Res. 1996 Sep l;40(2):254-60.

65. Giuffrida A, Parsons LH, Kerr TM, Rodriguez de Fonseca F, Navarro M, Piomelli D. Dopamine activation of endogenous cannabinoid signaling in dorsal striatum. Nat Neurosci. 1999 Apr;2(4):358-63.

176

66. Gonzales GF, Cordova A, Gonzales C, Chung A, Vega K, Villena A. Lepidium meyenii (Maca) improved semen parameters in adult men. Asian J Androl. 2001 Dec;3(4):301-3.

67. Gonzales GF, Cordova A, Vega K, Chung A, Villena A, Gonez C. Effect of Lepidium meyenii (Maca), a root with aphrodisiac and fertility-enhancing properties, on serum reproductive hormone levels in adult healthy men. J Endocrinol. 2003 Jan;176(l):163-8.

68. Gonzales GF, Cordova A, Vega K, Chung A, Villena A, Gonez C, Castillo S. Effect of Lepidium meyenii (Maca) on sexual desire and its absent relationship with serum testosterone levels in adult healthy men. Andrologia. 2002 Dec;34(6):367-72.

69. Gonzales GF, Gasco M, Cordova A, Chung A, Rubio J, Villegas L. Effect of Lepidium meyenii (Maca) on spermatogenesis in male rats acutely exposed to high altitude (4340 m). J Endocrinol. 2004 Jan;180(l):87-95.

70. Gonzales GF, Gasco M, Malheiros-Pereira A, Gonzales-Castaneda C. Antagonistic effect of Lepidium meyenii (red Maca) on prostatic hyperplasia in adult mice. Andrologia. 2008 Jun;40(3): 179-85.

71. Gonzales GF, Gonzales-Castaneda C. The methyltetrahydro-{beta}-carbolines in Maca {Lepidium meyenii). Evid Based Complement Alternat Med. 2008 Jun 19

72. Gonzales GF, Miranda S, Nieto J, Fernandez G, Yucra S, Rubio J, Yi P, Gasco M. Red Maca (Lepidium meyenii) reduced prostate size in rats. Reprod Biol Endocrinol. 2005 Jan 20;3:5.

177

73. Gonzales GF, Nieto J, Rubio J, Gasco M. Effect of black Maca {Lepidium meyenii) on one spermatogenic cycle in rats. Andrologia. 2006 Oct;38(5):166-72.

74. Gonzales GF, Rubio J, Chung A, Gasco M, Villegas L. Effect of alcoholic extract of Lepidium meyenii (Maca) on testicular function in male rats. Asian J Androl. 2003 Dec;5(4):349-52.

75. Gonzales GF, Ruiz A, Gonzales C, Villegas L, Cordova A. Effect of Lepidium meyenii (Maca) roots on spermatogenesis of male rats. Asian J Androl. 2001 Sep;3(3):231-3.

76. Gonzales GF, Valerio LG,Jr. Medicinal plants from peru: A review of plants as potential agents against cancer. Anticancer Agents Med Chem. 2006 Sep;6(5):429-44.

77. Gonzales GF, Vasquez V, Rodriguez D, Maldonado C, Mormontoy J, Portella J, Pajuelo M, Villegas L, Gasco M. Effect of two different extracts of red Maca in male rats with testosterone-induced prostatic hyperplasia. Asian J Androl. 2007 Mar;9(2):245-51.

78. Gonzales-Castaneda C, Gonzales GF. Hypocotyls of Lepidium meyenii (Maca), a plant of the peruvian highlands, prevent ultraviolet A-, B-, and C-induced skin damage in rats. Photodermatol Photoimmunol Photomed. 2008 Feb;24(l):24-31.

79. Gonzalez W, Fontaine V, Pueyo ME, Laquay N, Messika-Zeitoun D, Philippe M, Arnal JF, Jacob MP, Michel JB. Molecular plasticity of vascular wall during N(G)-nitroL-arginine methyl ester-induced hypertension: Modulation of proinflammatory signals. Hypertension. 2000 Jul;36(l): 103-9.

178

80. Green S, Newman Z, Nordquist S. Pharmacological evidence for cannabinoid receptors in glutamatergic synapses at the crayfish neuromuscular junction. Pioneering Neuroscience. 2005;6:55-61. 81. Guan QH, Pei DS, Xu TL, Zhang GY. Brain ischemia/reperfusion-induced expression of DP5 and its interaction with bcl-2, thus freeing bax from bcl-2/Bax dimmers are mediated by c-jun N-terminal kinase (JNK) pathway. Neurosci Lett. 2006 Jan 30;393(23):226-30.

82. Guan QH, Pei DS, Zhang QG, Hao ZB, Xu TL, Zhang GY. The neuroprotective action of SP600125, a new inhibitor of JNK, on transient brain ischemia/reperfusioninduced neuronal death in rat hippocampal CA1 via nuclear and non-nuclear pathways. Brain Res. 2005 Feb 21;1035(l):51-9.

83. Guegan C, Ceballos-Picot I, Nicole A, Kato H, Onteniente B, Sola B. Recruitment of several neuroprotective pathways after permanent focal ischemia in mice. Exp Neurol. 1998Dec;154(2):371-80.

84. Han BH, Holtzman DM. BDNF protects the neonatal brain from hypoxic-ischemic injury in vivo via the ERK pathway. J Neurosci. 2000 Aug 1;20(15):5775-81.

85. Han X, Sun Y, Scott S, Bleich D. Tissue inhibitor of metalloproteinase-1 prevents cytokine-mediated dysfunction and cytotoxicity in pancreatic islets and beta-cells. Diabetes. 2001 May;50(5): 1047-55.

179

86. Hansen HH, Dconomidou C, Bittigau P, Hansen SH, Hansen HS. Accumulation of the anandamide precursor and other N-acylethanolamine phospholipids in infant rat models of in vivo necrotic and apoptotic neuronal death. J Neurochem. 2001 Jan;76(l):39-46.

87. Hansen HS, Moesgaard B, Petersen G, Hansen HH. Putative neuroprotective actions ofN-acyl-ethanolamines. Pharmacol Ther. 2002 Aug;95(2): 119-26.

88. Hao ZB, Pei DS, Guan QH, Zhang GY. Calcium/calmodulin-dependent protein kinase II (CaMKII), through NMDA receptors and L-voltage-gated channels, modulates the serine phosphorylation of GluR6 during cerebral ischemia and early reperfusion period in rat hippocampus. Brain Res Mol Brain Res. 2005 Oct 31;140(l-2):55-62.

89. Harrison DC, Davis RP, Bond BC, Campbell CA, James MF, Parsons AA, Philpott KL. Caspase mRNA expression in a rat model of focal cerebral ischemia. Brain Res Mol Brain Res. 2001 Apr 18;89(l-2):133-46.

90. Hartman RE, Shah A, Fagan AM, Schwetye KE, Parsadanian M, Schulman RN, Finn MB, Holtzman DM. Pomegranate juice decreases amyloid load and improves behavior in a mouse model of Alzheimer's disease. Neurobiol Dis. 2006 Dec;24(3):506-15.

91. Hawkins BT, Davis TP. The blood-brain barrier/neurovascular unit in health and disease. Pharmacol Rev. 2005 Jun;57(2): 173-85.

92. Herraiz T. Tetrahydro-6-carbolines, potential neuroactive alkaloids, in chocolate and cocoa. J Agric Food Chem. 2000 Oct;48(10):4900-4.

180

93. Ho BY, Zhao J. Determination of the cannabinoid receptors in mouse x rat hybridoma NG108-15 cells and rat GH4C1 cells. Neurosci Lett. 1996 Jul 12;212(2):123-6.

94. Hong SJ, Dawson TM, Dawson VL. Nuclear and mitochondrial conversations in cell death: PARP-1 and AIF signaling. Trends Pharmacol Sci. 2004 May;25(5):259-64.

95. Hossmann KA. Periinfarct depolarizations. Cerebrovasc Brain Metab Rev. 1996 Fall;8(3):195-208.

96. Hu BR, Fux CM, Martone ME, Zivin JA, Ellisman MH. Persistent phosphorylation of cyclic AMP responsive element-binding protein and activating transcription factor-2 transcription factors following transient cerebral ischemia in rat brain. Neuroscience. 1999 Mar;89(2):437-52.

97. Hum PD, Brass LM. Estrogen and stroke: A balanced analysis. Stroke. 2003 Feb;34(2):338-41.

98. Hum PD, Vannucci SJ, Hagberg H. Adult or perinatal brain injury: Does sex matter? Stroke. 2005 Feb;36(2): 193-5.

99. Iwashita A, Tojo N, Matsuura S, Yamazaki S, Kamijo K, Ishida J, Yamamoto H, Hattori K, Matsuoka N, Mutoh S. A novel and potent poly(ADP-ribose) polymerase-1 inhibitor, FR247304 (5-chloro-2-[3-(4-phenyl-3,6-dihydro-l (2H)-pyridinyl)propyl]-

4(3H)-quinazolinone) , attenuates neuronal damage in in vitro and in vivo models of cerebral ischemia. J Pharmacol Exp Ther. 2004 Aug;310(2):425-36.

181

100. Keren O, Same Y. Multiple mechanisms of CB1 cannabinoid receptors regulation. Brain Res. 2003 Aug 8;980(2): 197-205.

101. Le HT, Schaldach CM, Firestone GL, Bjeldanes LF. Plant-derived 3,3'diindolylmethane is a strong androgen antagonist in human prostate cancer cells. J Biol Chem. 2003 Jun 6;278(23):21136-45.

102. Lees KR, Dawson J. Advances in emerging therapies 2006. Stroke. 2007 Feb;38(2):219-21.

103. Leggett JD, Aspley S, Beckett SR, D'Antona AM, Kendall DA, Kendall DA. Oleamide is a selective endogenous agonist of rat and human CBj cannabinoid receptors. Br J Pharmacol. 2004 Jan;141(2):253-62.

104. Leoni O, Iori R, Palmieri S, Esposito E, Menegatti E, Cortesi R, Nastruzzi C. Myrosinase-generated isothiocyanate from glucosinolates: Isolation, characterization and in vitro antiproliferative studies. Bioorg Med Chem. 1997 Sep;5(9): 1799-806.

105. Li G, Ammermann U, Quiros CF. Glucosinolate contents in Maca (Lepidium peruvianum chacon) seeds, sprouts and several derived commercial products. Economic Botany. 2001 ;55(2):255-62.

106. Linnik MD, Zobrist RH, Hatfield MD. Evidence supporting a role for programmed cell death in focal cerebral ischemia in rats. Stroke. 1993;24:2002,2002-2009.

182

107. Liu CS, Chen NH, Zhang JT. Protection of PC 12 cells from hydrogen peroxideinduced cytotoxicity by salvianolic acid B, a new compound isolated from radix salviae miltiorrhizae. Phytomedicine. 2007 Aug; 14(7-8):492-7.

108. Lois C, Alvarez-Buylla A. Proliferating subventricular zone cells in the adult mammalian forebrain can differentiate into neurons and glia. Proc Natl Acad Sci U S A . 1993 Mar l;90(5):2074-7.

109. Longa EZ, Weinstein PR, Carlson S, Cummins R. Reversible middle cerebral artery occlusion without craniectomy in rats. Stroke. 1989 Jan;20(l):84-91.

110. Lopez-Fando A, Gomez-Serranillos MP, Iglesias I, Lock O, Upamayta UP, Carretero ME. Lepidium peruvianum Chacon restores homeostasis impaired by restraint stress. Phytother Res. 2004 Jun;18(6):471-4.

111. Love S. Apoptosis and brain ischaemia. Prog Neuropsychopharmacol Biol Psychiatry. 2003 Apr;27(2):267-82.

112. Luo Y, Umegaki H, Wang X, Abe R, Roth GS. Dopamine induces apoptosis through an oxidation-involved SAPK/JNK activation pathway. J Biol Chem. 1998 Feb 6;273(6):3756-64.

113. Ma R, Xiong N, Huang C, Tang Q, Hu B, Xiang J, Li G. Erythropoietin protects PC12 cells from beta-amyloid(25-35)-induced apoptosis via PI3K/Akt signaling pathway. Neuropharmacology. 2009 Mar 4

183

114. Maccarrone M. Central and peripheral interactions between endocannabinoids and steroids, and implications for drug dependence. Life Sci. 2005 Aug 19;77(14):1559-68.

115. Maccarrone M, Fride E, Bisogno T, Bari M, Cascio MG, Battista N, Finazzi Agro A, Suris R, Mechoulam R, Di Marzo V. Up-regulation of the endocannabinoid system in the uterus of leptin knockout (ob/ob) mice and implications for fertility. Mol Hum Reprod. 2005 Jan;ll(l):21-8.

116. Maccarrone M, Lorenzon T, Bari M, Melino G, Finazzi-Agro A. Anandamide induces apoptosis in human cells via vanilloid receptors, evidence for a protective role of cannabinoid receptors. J Biol Chem. 2000 Oct 13;275(41):31938-45.

117. MacKay D. Nutrients and botanicals for erectile dysfunction: Examining the evidence. Alternative Medicine Review. 2004 2004;9(1):4,5-16.

118. Markgraf CG, Kraydieh S, Prado R, Watson BD, Dietrich WD, Ginsberg MD. Comparative histopathologic consequences of photothrombotic occlusion of the distal middle cerebral artery in sprague-dawley and wistar rats. Stroke. 1993 Feb;24(2):286,92; discussion 292-3.

119. Masliah E, Mallory M, Alford M, DeTeresa R, Hansen LA, McKeel DW,Jr, Morris JC. Altered expression of synaptic proteins occurs early during progression of Alzheimer's disease. Neurology. 2001 Jan 9;56(1): 127-9.

184

120. McCollom MM, Villinski JR, McPhail KL, Craker LE, Gafiier S. Analysis of macamides in samples of Maca (Lepidium meyenii) by HPLC-UV-MS/MS. Phytochem Anal. 2005 Nov-Dec;16(6):463-9.

121. Mechoulam R. Discovery of endocannabinoids and some random thoughts on their possible roles in neuroprotection and aggression. Prostaglandins Leukot Essent Fatty Acids. 2002 Feb-Mar;66(2-3):93-9.

122. Mechoulam R, Shohami E. Endocannabinoids and traumatic brain injury. Mol Neurobiol. 2007 Aug;36(l):68-74.

123. Mehta SL, Manhas N, Raghubir R. Molecular targets in cerebral ischemia for developing novel therapeutics. Brain Res Rev. 2007 Apr;54(l):34-66.

124. Melis M, Pillolla G, Bisogno T, Minassi A, Petrosino S, Perra S, Muntoni AL, Lutz B, Gessa GL, Marsicano G, Di Marzo V, Pistis M. Protective activation of the endocannabinoid system during ischemia in dopamine neurons. Neurobiol Dis. 2006 Oct;24(l):15-27.

125. Melis M, Pistis M, Perra S, Muntoni AL, Pillolla G, Gessa GL. Endocannabinoids mediate presynaptic inhibition of glutamatergic transmission in rat ventral tegmental area dopamine neurons through activation of CBi receptors. J Neurosci. 2004 Jan 7;24(1):5362.

126. Mies G, Iijima T, Hossmann KA. Correlation between peri-infarct DC shifts and ischaemic neuronal damage in rat. Neuroreport. 1993 Jun;4(6):709-11.

185

127. Muhammad I, Zhao J, Dunbar DC, Khan IA. Constituents of Lepidium meyenii 'Maca'. Phytochemistry. 2002 Jan;59(l):105-10.

128. Nakase T, Fushiki S, Naus CC. Astrocytic gap junctions composed of connexin 43 reduce apoptotic neuronal damage in cerebral ischemia. Stroke. 2003 Aug;34(8):1987-93.

129. Negoescu A, Lorimier P, Labat-Moleur F, Drouet C, Robert C, Guillermet C, Brambilla C, Brambilla E. In situ apoptotic cell labeling by the TUNEL method: Improvement and evaluation on cell preparations. J Histochem Cytochem. 1996 Sep;44(9):959-68.

130. Ohta K, Graf R, Rosner G, Heiss WD. Calcium ion transients in peri-infarct depolarizations may deteriorate ion homeostasis and expand infarction in focal cerebral ischemia in cats. Stroke. 2001 Feb;32(2):535-43.

131. O'Meara G, Coumis U, Ma SY, Kehr J, Mahoney S, Bacon A, Allen SJ, Holmes F, Kahl U, Wang FH, Kearns IR, Ove-Ogren S, Dawbarn D, Mufson EJ, Davies C, Dawson G, Wynick D. Galanin regulates the postnatal survival of a subset of basal forebrain cholinergic neurons. Proc Natl Acad Sci U S A . 2000 Oct 10;97(21):11569-74.

132. Oshima M, Gu Y, Tsukada S. Effects of Lepidium meyenii Walp and Jatropha macrantha on blood levels of estradiol-17 beta, progesterone, testosterone and the rate of embryo implantation in mice. J Vet Med Sci. 2003 Oct;65(10):l 145-6.

186

133. Otey CA, Boukhelifa M, Maness P. B35 neuroblastoma cells: An easily transfected, cultured cell model of central nervous system neurons. Methods Cell Biol. 2003;71:287304.

134. Piacente S, Carbone V, Plaza A, Zampelli A, Pizza C. Investigation of the tuber constituents of Maca (Lepidium meyenii Walp.). J Agric Food Chem. 2002 Sep 25;50(20):5621-5.

135. Pino-Figueroa AJ, Nguyen D, Maher TJ. Neuroprotective effects of Lepidium meyenii (Maca). Ann N Y Acad Sci. 2009;In press

136. Raduner S, Majewska A, Chen JZ, Xie XQ, Hamon J, Faller B, Altmann KH, Gertsch J. Alkylamides from Echinacea are a new class of cannabinomimetics. cannabinoid type 2 receptor-dependent and -independent immunomodulatory effects. J Biol Chem. 2006 May 19;281(20): 14192-206.

137. Read SJ, Parsons AA, Harrison DC, Philpott K, Kabnick K, O'Brien S, Clark S, Brawner M, Bates S, Gloger I, Legos JJ, Barone FC. Stroke genomics: Approaches to identify, validate, and understand ischemic stroke gene expression. J Cereb Blood Flow Metab. 2001 Jul;21(7):755-78.

138. Reshef A, Shirvan A, Grimberg H, Levin G, Cohen A, Mayk A, Kidron D, Djaldetti R, Melamed E, Ziv I. Novel molecular imaging of cell death in experimental cerebral stroke. Brain Res. 2007 May 4;1144:156-64.

187

139. Rodriguez de Fonseca F, Del Arco I, Bermudez-Silva FJ, Bilbao A, Cippitelli A, Navarro M. The endocannabinoid system: Physiology and pharmacology. Alcohol Alcohol. 2005 Jan-Feb;40(l):2-14.

140. Rosell A, Ortega-Aznar A, Alvarez-Sabin J, Fernandez-Cadenas I, Ribo M, Molina CA, Lo EH, Montaner J. Increased brain expression of matrix metalloproteinase-9 after ischemic and hemorrhagic human stroke. Stroke. 2006 Jun;37(6): 1399-406.

141. Rubio J, Caldas M, Davila S, Gasco M, Gonzales GF. Effect of three different cultivars of Lepidium meyenii (Maca) on learning and depression in ovariectomized mice. BMC Complement Altern Med. 2006 Jun 23;6:23.

142. Rubio J, Dang H, Gong M, Liu X, Chen SL, Gonzales GF. Aqueous and hydroalcoholic extracts of black Maca {Lepidium meyenii) improve scopolamine-induced memory impairment in mice. Food Chem Toxicol. 2007 Oct;45(10):1882-90.

143. Rubio J, Qiong W, Liu X, Jiang Z, Dang H, Chen SL, Gonzales GF. Aqueous extract of black Maca {Lepidium meyenii) on memory impairment induced by ovariectomy in mice. Evid Based Complement Alternat Med. 2008 Oct 9

144. Rubio J, Riqueros MI, Gasco M, Yucra S, Miranda S, Gonzales GF. Lepidium meyenii (Maca) reversed the lead acetate induced damage on reproductive function in male rats. Food Chem Toxicol. 2006 Jul;44(7):l 114-22.

188

145. Ruiz-Luna AC, Salazar S, Aspajo NJ, Rubio J, Gasco M, Gonzales GF. Lepidium meyenii (Maca) increases litter size in normal adult female mice. Reprod Biol Endocrinol. 2005 May 3;3:16.

146. Salom JB, Perez-Asensio FJ, Burguete MC, Marin N, Pitarch C, Torregrosa G, Romero FJ, Alborch E. Single-dose ebselen does not afford sustained neuroprotection to rats subjected to severe focal cerebral ischemia. Eur J Pharmacol. 2004 Jul 8;495(1):5562.

147. Sandoval M, Okuhama NN, Angeles F. M., Melchor W , Condezo LA, Lao J, Miller M. J. S. Antioxidant activity of the cruciferous vegetable Maca {Lepidium meyenii). Food Chemistry. 2002;79:207-13.

148. Sarang SS, Yoshida T, Cadet R, Valeras AS, Jensen RV, Gullans SR. Discovery of molecular mechanisms of neuroprotection using cell-based bioassays and oligonucleotide arrays. Physiol Genomics. 2002 Oct 29;ll(2):45-52.

149. Schabitz WR, Fisher M. Perspectives on neuroprotective stroke therapy. Biochem Soc Trans. 2006 Dec;34(Pt 6): 1271-6.

150. Schabitz WR, Giuffrida A, Berger C, Aschoff A, Schwaninger M, Schwab S, Piomelli D. Release of fatty acid amides in a patient with hemispheric stroke: A microdialysis study. Stroke. 2002 Aug;33(8):2112-4.

189

151. Schomacher M, Muller HD, Sommer C, Schwab S, Schabitz WR. Endocannabinoids mediate neuroprotection after transient focal cerebral ischemia. Brain Res. 2008 Nov 13;1240:213-20.

152. Seierstad M, Breitenbucher JG. Discovery and development of fatty acid amide hydrolase (FAAH) inhibitors. J Med Chem. 2008 Dec ll;51(23):7327-43.

153. Shamloo M, Soriano L, Wieloch T, Nikolich K, Urfer R, Oksenberg D. Deathassociated protein kinase is activated by dephosphorylation in response to cerebral ischemia. J Biol Chem. 2005 Dec 23;280(51):42290-9.

154. Shashoua VE, Adams DS, Boyer-Boiteau A, Cornell-Bell A, Li F, Fisher M. Neuroprotective effects of a new synthetic peptide, CMX-9236, in in vitro and in vivo models of cerebral ischemia. Brain Res. 2003 Feb 14;963(l-2):214-23.

155. Sheng M. Molecular organization of the postsynaptic specialization. Proc Natl Acad Sci U S A . 2001 Jun 19;98(13):7058-61.

156. Shirasaki Y, Kanazawa Y, Morishima Y, Makino M. Involvement of calmodulin in neuronal cell death. Brain Res. 2006 Apr 14;1083(l):189-95.

157. Shouman B, Fontaine RH, Baud O, Schwendimann L, Keller M, Spedding M, Lelievre V, Gressens P. Endocannabinoids potently protect the newborn brain against AMPA-kainate receptor-mediated excitotoxic damage. Br J Pharmacol. 2006

Jun;148(4):442-51.

190

158. Simon R, Xiong Z. Acidotoxicity in brain ischaemia. Biochem Soc Trans. 2006 Dec;34(Pt6):1356-61.

159. Stein DG. Progesterone exerts neuroprotective effects after brain injury. Brain Res Rev. 2008 Mar;57(2):386-97.

160. Tamura A, Graham DI, McCulloch J, Teasdale GM. Focal cerebral ischaemia in the rat: 1. description of technique and early neuropathological consequences following middle cerebral artery occlusion. J Cereb Blood Flow Metab. 1981;l(l):53-60.

161. Tellez MR, Khan IA, Kobaisy M, Schrader KK, Dayan FE, Osbrink W. Composition of the essential oil of Lepidiwn meyenii (Walp). Phytochemistry. 2002 Sep;61(2):149-55.

162. Toledo J, et. al. Genetic variability of Lepidiwn meyenii and other andean lepidium species {brassicaceae) assessed by molecular markers. Annals of Botany. 1998;82:52330.

163. Unal-Cevik I, Kilinc M, Can A, Gursoy-Ozdemir Y, Dalkara T. Apoptotic and necrotic death mechanisms are concomitantly activated in the same cell after cerebral ischemia. Stroke. 2004 Sep;35(9):2189-94.

164. Valentova K, Buckiova D, Kren V, Peknicova J, Ulrichova J, Simanek V. The in vitro biological activity of Lepidium meyenii extracts. Cell Biol Toxicol. 2006 Mar;22(2):91-9.

191

165. Valentova K, Stejskal D, Bartek J, Dvorackova S, Kren V, Ulrichova J, Simanek V. Maca (Lepidium meyenii) and Yacon (Smallanthus sonchifolius) in combination with silymarin as food supplements: In vivo safety assessment. Food Chem Toxicol. 2008 Mar;46(3):1006-13.

166. Valentova K, Ulrichova J. Smallanthus sonchifolius and Lepidium meyenii prospective andean crops for the prevention of chronic diseases. Biomed Pap Med Fac Univ Palacky Olomouc Czech Repub. 2003 Dec;147(2):l 19-30.

167. Valerio LG,Jr, Gonzales GF. Toxicological aspects of the south american herbs cat's claw (Uncaria tomentosa) and Maca {Lepidium meyenii) : A critical synopsis. Toxicol Rev. 2005;24(1):11-35.

168. van der Stelt M, Veldhuis WB, van Haaften GW, Fezza F, Bisogno T, Bar PR, Veldink GA, Vliegenthart JF, Di Marzo V, Nicolay K. Exogenous anandamide protects rat brain against acute neuronal injury in vivo. J Neurosci. 2001 Nov 15;21(22):8765-71.

169. Vecera R, Orolin J, Skottova N, Kazdova L, Oliyarnik O, Ulrichova J, Simanek V. The influence of Maca (Lepidium meyenii) on antioxidant status, lipid and glucose metabolism in rat. Plant Foods Hum Nutr. 2007 Jun;62(2):59-63.

170. Vemuganti R, Dempsey RJ, Bowen KK. Inhibition of intercellular adhesion molecule-1 protein expression by antisense oligonucleotides is neuroprotective after transient middle cerebral artery occlusion in rat. Stroke. 2004 Jan;35(l): 179-84.

192

171. Wang JK, Portbury S, Thomas MB, Barney S, Ricca DJ, Morris DL, Warner DS, Lo DC. Cardiac glycosides provide neuroprotection against ischemic stroke: Discovery by a brain slice-based compound screening platform. Proc Natl Acad Sci U S A . 2006 Jul 5;103(27):10461-6.

172. Wattenberg LW. Inhibition of carcinogen-induced neoplasia by sodium cyanate, tert-butyl isocyanate, and benzyl isothiocyanate administered subsequent to carcinogen exposure. Cancer Res. 1981 Aug;41(8):2991-4.

173. Yang HY, Karoum F, Felder C, Badger H, Wang TC, Markey SP. GC/MS analysis of anandamide and quantification of N-arachidonoylphosphatidylethanolamides in various brain regions, spinal cord, testis, and spleen of the rat. J Neurochem. 1999 May;72(5): 1959-68.

174. Ye B, Liao D, Zhang X, Zhang P, Dong H, Huganir RL. GRASP-1: A neuronal RasGEF associated with the AMPA Receptor/GRIP complex. Neuron. 2000;26:603,603617.

175. Yenari MA, Xu L, Tang XN, Qiao Y, Giffard RG. Microglia potentiate damage to blood-brain barrier constituents: Improvement by minocycline in vivo and in vitro. Stroke. 2006 Apr;37(4): 1087-93.

176. Yucra S, Gasco M, Rubio J, Nieto J, Gonzales GF. Effect of different fractions from hydroalcoholic extract of black Maca (Lepidium meyenii) on testicular function in adult male rats. Fertil Steril. 2008 May;89(5 Suppl):1461-7.

193

177. Zhang Y, Yu L, Ao M, Jin W. Effect of ethanol extract of Lepidium meyenii Walp. on osteoporosis in ovariectomized rat. J Ethnopharmacol. 2006 Apr 21;105(l-2):274-9.

178. Zhang Z, Zhang RL, Jiang Q, Raman SB, Cantwell L, Chopp M. A new rat model of thrombotic focal cerebral ischemia. J Cereb Blood Flow Metab. 1997 Feb; 17(2): 123-35.

179. Zhao J, Muhammad I, Dunbar DC, Mustafa J, Khan LA. New alkamides from Maca (Lepidium meyenii). J Agric Food Chem. 2005 Feb 9;53(3):690-3.

180. Zheng BL, He K, Kim CH, Rogers L, Shao Y, Huang ZY, Lu Y, Yan SJ, Qien LC, Zheng QY. Effect of a lipidic extract from Lepidium meyenii on sexual behavior in mice and rats. Urology. 2000 Apr;55(4):598-602.

194