International Immunopharmacology 7 (2007) 383 391

www.elsevier.com/locate/intimp

Induction of inflammatory cytokines by cartilage extracts

Liza Merly a,b , Shabana Simjee b,1 , Sylvia L. Smith a,b,
a
Department of Biological Sciences, Florida International University, University Park, Miami, FL 33199, United States
b
Comparative Immunology Institute, Florida International University, University Park, Miami, FL 33199, United States
Received 14 June 2006; received in revised form 17 November 2006; accepted 28 November 2006

Abstract

Shark cartilage extracts were examined for induction of cytokines and chemokines in human peripheral blood leukocytes. Primary
leukocyte cultures were exposed to a variety of aqueous and organic extracts prepared from several commercial brands of shark
cartilage. From all commercial sources of shark cartilage tested the acid extracts induced higher levels of TNF than other extracts.
Different commercial brands of shark cartilage varied significantly in cytokine-inducing activity. TNF induction was seen as early as
4 h and IFN at detectable levels for up to four days. Shark cartilage extracts did not induce physiologically significant levels of IL-4.
Results suggest that shark cartilage, preferentially, induces Th1 type inflammatory cytokines. When compared to bovine cartilage
extract, collagen, and chondroitin sulfate, shark cartilage induced significantly higher levels of TNF. Treatment with digestive
proteases (trypsin and chymotrypsin) reduced the cytokine induction response by 80%, suggesting that the active component(s) in
cartilage extracts is proteinaceous. The induction of Th1 type cytokine response in leukocytes is a significant finding since shark
cartilage, taken as a dietary supplement for a variety of chronic degenerative diseases, would be contraindicated in cases where the
underlying pathology of the chronic condition is caused by inflammation.
2006 Elsevier B.V. All rights reserved.

Keywords: Cytokine; Shark cartilage; Immunomodulation; CAM; Natural product

1. Introduction
Abbreviations: AA, Alkaline extract; AE, Acid extract; BCA,
Bicinchoninic acid; BSA, Bovine serum albumin; Con A, Concanav-
alin A; FCS, Fetal calf serum; EDTA, Ethylenediaminetetraacetic acid; Intake of natural dietary supplements for the treat-
ELISA, Enzyme-linked immunosorbant assay; HPBL, Human periph- ment of various diseases or for general health improve-
eral blood leukocytes; IFN, Interferon gamma; IL, Interleukin; LPS, ment is a form of complementary and alternative medicine
Lipopolysaccharide; MWCO, Molecular weight cut off; OE, Organic that is gaining in popularity [13]. The biological activity,
extract; PBS, Phosphate buffered saline; PHA, Phytohemagglutin; SC,
Shark cartilage; SS, Salt soluble extract; TC, Tissue culture; TNF,
however, of many of the commercially available supple-
Tumor necrosis factor alpha. ments has not been thoroughly characterized, nor has the
Corresponding author. Department of Biological Sciences, Florida treatment efficacy of many natural products been deter-
International University, University Park, Miami, FL 33199, United mined [4,5]. This study was undertaken to assess the
States. Tel.: +1 305 348 3183; fax: +1 305 348 1083. biological activity of extracts of commercial shark carti-
E-mail address: smiths@fiu.edu (S.L. Smith).
1
Current address: International Center for Chemical Studies, H.E.J.
lage with respect to the effect on immuno-regulation.
Research Institute for Chemistry, University of Karachi, Karachi- While the initial interest in shark cartilage related to
75270, Pakistan. its alleged anticancer properties [611], recently, the
1567-5769/$ - see front matter 2006 Elsevier B.V. All rights reserved.
doi:10.1016/j.intimp.2006.11.011
384 L. Merly et al. / International Immunopharmacology 7 (2007) 383391
interest is on the effect of shark cartilage on oxidative
stress [12,13], the expression of adhesion molecules and
fibrinolytic enzymes [14,15], and on immune function
[1618]. For these studies, shark cartilage has been
purified preparations rather than commercial prepara-
tions of shark cartilage that are taken as oral supple-
ments by consumers for therapeutic and/or prophylactic
purposes. The purpose of the research conducted in our
laboratory was to evaluate the immuno-effectiveness
and bioactivity of commercial shark cartilage that is
available as a non-regulated dietary supplement.
Cytokines are extracellular protein/peptide messen-
gers that play a critical role in cell-to-cell communi-
cation. They are the primary mechanism by which
communication occurs between leukocytes such as
lymphocytes and macrophages, and between other
immune and non-immune cells and, thus, contribute
significantly to regulating and controlling immune
responses [19]. The profile of cytokines produced at
any one time during an immune response is largely
governed by two subsets of T-helper cells designated
Th1 and Th2. The differentiation of T-helper cells into
Th1 and Th2 cells is tightly controlled by the cytokines
present in the local environment and the type of infec-
tion and/or immune stimulus [20]. Polarized Th1 and
Th2 responses can contribute to the pathogenesis of
immune-mediated diseases. Consequently, natural pro-
ducts and other therapeutic agents that are able to cause
shifts in the Th1/Th2 balance could significantly in-
fluence the overall immune response [21]. We report
here on the cytokine profile induced in human peri-
pheral blood leukocytes following stimulation with ex-
tracts of commercial shark cartilage.
2. Materials and methods
Shark cartilage was obtained from several commercial
companies: (1) Cartilade (Solgar Laboratory, Leonia, NJ), (2)
Advanced Shark Cartilage Extract (Twin Laboratories Incor-
poration, Ronkonkoma, NY), and (3) Natural Shark Cartilage
(FDC Wholesale Corporation, Miami, FL). Bovine cartilage
was obtained from Sigma Chemical Company, St. Louis, MO.
All chemicals, enzymes, and enzyme inhibitors were purchased
from Sigma Aldrich (St. Louis, MO) unless otherwise stated.
Dialysis membranes (MWCO 68 kDa) were obtained from
Fisher (Fisher Scientific International, Inc. Liberty Lane,
Hampton, NH). Filtration was carried out using Whatman
(Whatman Inc., Florham Park, NJ) filter paper, and syringe
filters purchased from Fisher Scientific. RPMI medium was
purchased from Gibco/BRL (Invitrogen Corporation, Carlsbad,
CA). Protein determination was carried out using the BCA kit
purchased from Pierce (Pierce Biotechnology Inc., Rockford,
IL). ELISA and TC plates were purchased from Fisher Scientific
(Falcon, Becton Dickinson). The cytokine detection kits were
purchased from Pierce Endogen (Pierce Biotechnology Inc.,
Rockford, IL). The E-toxate kit from Sigma Aldrich was used to
detect endotoxin. HPBL were obtained from healthy donors
according to IRB approved protocol.
2.1. Cartilage extracts
The starting material for all extract preparations was 2 g of
commercial shark cartilage or bovine cartilage. Since com-
mercial shark products are heterogeneous in chemical compo-
sition, the level of bioactivity of extracts prepared from different
commercial sources was related to mg extract protein. Except for
studies where extracts from different commercial brands of shark
cartilage were compared, the acid extract used for all assays was
prepared from Solgar cartilage.
2.1.1. Acid extract (SCAE)
Acid extracts were prepared from both shark and bovine
cartilage. Two grams of cartilage were suspended in 50 ml of
0.5 M acetic acid with 0.03% Toluene, pH 4.2, and incubated
with stirring overnight at 4 C. The extract was centrifuged
(500 g, 45 min) and the supernatant carefully decanted and
filtered to remove undissolved material through a series of filters
that included Whatman filter paper followed by a 0.45 m and a
0.22 m syringe filter. The extract, in 15 ml aliquots was dia-
lyzed twice against 500 ml of RPMI-1640 medium supplemen-
ted with HEPES (25 mM) and glutamine (0.3 mg/ml) using a
molecular porous membrane with a MWCO of 68 kDa. The
dialyzed extract was then filtered through a 0.22 m filter and
stored at 20 C until further use. All cartilage extracts were
assayed for endotoxin before use in bioassays. Each extract was
also tested to determine whether it inhibited or interfered with
ELISA used to measure cytokine levels.
2.1.2. Salt soluble extract (SCSS)
Shark and bovine cartilage were dissolved in 50 ml of 0.1 M
TrisHCl, 0.15 M NaCl, pH 7.5, and incubated overnight at
4 C. Insoluble material from each extract was removed by
centrifugation and clarified by filtration. Fifteen (15) ml
aliquots of extract were dialyzed against culture medium,
filtered, and stored as described above.
2.1.3. Phosphate buffered saline extract (SCPBS)
Shark and bovine cartilage were suspended in 50 ml of
0.02 M PBS (0.5 mM NaH2PO4, 1.9 mM of NaHPO4, 17.9 mM
NaCl, pH 7.3) and incubated at 4 C for 2 h. The extract was
centrifuged and the supernatant containing soluble protein was
filtered as described above. Extract aliquots were dialyzed
against culture medium, filtered, and stored.
2.1.4. Alkaline extract (SCAA)
Shark cartilage was suspended in 50 ml of 5% sodium
hydroxide and incubated for 45 min at room temperature (22
25 C). The mixture was centrifuged and the supernatant filtered,
dialysed against culture medium, and stored in 15 ml aliquots as
described above.
 L. Merly et al. / International Immunopharmacology 7 (2007) 383391
2.1.5. Organic extract (SCOE)
Shark cartilage was dissolved in 10 ml of cold ethanol and
kept on ice during the drop wise addition of 40 ml of a 1:3
isopropanol-hexane solution with stirring. The mixture was
incubated on ice for no more than 30 min. The mixture was
then centrifuged and the supernatant cleared by filtration,
dialysed against culture medium, and stored in 15 ml aliquots.
2.2. Protein determination
The protein concentration of each extract preparation was
determined by the BCA protein assay using BSA as the standard
[22].
2.3. Assay for endotoxin
The E-toxate test (Limulus Amebocyte Lysate Assay) was
used for the detection and semi-quantitation of endotoxin in all
test samples according to the manufacturer's instructions [23].
Endotoxin-free water provided by the manufacturer was used
as a negative control and samples were also tested for the
presence of inhibitors for the E-toxate test. The E-toxate test
was chosen because it has a threshold sensitivity level between
0.05 and 0.1 endotoxin units per ml, a level considered extre-
mely low for physiological significance.
2.4. Isolation of peripheral blood leukocytes
For this study blood was provided by seven individuals. A
single donor bleed was used for each experiment. HPBL were
separated from freshly drawn heparinized peripheral blood of
healthy donors by separation on 3% dextran in physiological
saline. The harvested heterogeneous mixture of leukocytes was
washed with saline, centrifuged, and the cell pellet subjected to
hypotonic lysis to remove contaminating erythrocytes. The
leukocytes were resuspended in 7 ml of 0.2% NaCl solution
(1520 s), gently mixed by inverting the tube several times, then
an equal volume of 1.6% NaCl solution containing 0.2% dextrose
was added to the cell suspension to achieve a final NaCl
concentration equivalent to normal physiological saline (0.15 M).
The suspension was centrifuged at 250 g for 10 min and the
supernatant containing erythrocyte stroma was discarded. The cell
pellet was washed several times in PBS and suspended in culture
medium (RPMI-1640 medium supplemented with 0.3 mg/ml
glutamine and 25 mM HEPES) containing 100 g/ml of strep-
tomycin and 100 U/ml of penicillin to inhibit bacterial conta-
mination, and 10% fetal bovine serum. The cell suspension was
standardized to 2.02.5 105 cells/ml by hemacytometer. Cell
viability was checked by the Trypan Blue exclusion test, using
0.4% Trypan Blue in physiological saline. Only cell suspensions
with a viability of greater than 94% were used for cell cultures.
2.5. In vitro stimulation of human leukocytes
Primary cultures of HPBL were set up in triplicate in 24-
well flat bottom tissue culture plates. Each well contained 50 l
385
of test stimulant or control stimulant, 100 l of culture medium
(supplemented with 10% FBS), and 200 l of leukocyte
suspension (2.02.5 105 cells/ml). Depending on the indi-
vidual experiment, test stimulants consisted of shark or bovine
cartilage extract alone or in combination with mitogen, colla-
gen or chondroitin sulphate. LPS (E. coli, Difco laboratories),
PHA, or Con A served as stimulants in the positive control
cultures. Unstimulated or negative control cultures contained
culture medium in lieu of stimulant. Cultures were set up in
triplicate for varying periods and plates were incubated in a
humid chamber at 37 C in 5% CO2.
2.6. Assessment of biological activity in vitro
2.6.1. In vitro induction of cytokines and chemokine
HPBL culture supernatants were harvested at varying time
intervals by aspiration, centrifuged, and stored at 20 C.
Supernatants were assayed for TNF, IFN, IL-1, IL-4, and
IL-8, using commercially available ELISA kits for each indi-
vidual cytokine/chemokine. Assays were set up in triplicate
according to the manufacturer's instructions and cytokines were
quantified spectrophotometrically. A standard curve was con-
structed for each cytokine from standards provided in each kit.
2.6.1.1. TNF induction. Extracts (acid, alkaline, salt soluble,
PBS and organic) using different commercial brands of shark
cartilage were prepared. To make reliable comparisons between
acid extracts each was standardized to contain between 0.350 to
0.450 mg protein/ml. Culture supernatants were harvested at 0,
4, 8, 12, 20, 24, 30, and 48 h and assayed for released TNF. To
determine whether the cytokine-inducing property was unique to
shark cartilage, acid, salt soluble and PBS extracts were also
prepared from purified bovine cartilage, but the culture super-
natants were harvested at 24 h which represented peak level of
cytokine production noted for shark cartilage.
Since cartilage is a complex composite of proteins that in-
cludes collagen and chondroitin sulfate, TNF induction,
following stimulation of leukocytes by collagen and chondroitin
sulfate was compared to that induced by shark cartilage
treatment. HPBL cultures were set up as described previously
and stimulated with SCAE (0.360 mg protein/ml), collagen
(300 mg/ml) and chondroitin sulfate (100 mg/ml). A 1:2 dilution
of each stimulant was also tested. Cultures were incubated for
24 h at 37 C, 5% CO2. Culture supernatants were harvested and
assayed for TNF.
2.6.1.2. Miscellaneous cytokine induction. Cytokines IFN,
IL-1, IL-4 and IL-8 were assayed in HPBL cultures stimulated
with SCAE (0.365 mg/ml) at 37 C, 5% CO2. Culture super-
natants were harvested and assayed for cytokine at various time
intervals ranging from 0 h up to 96 h. In case of IFN induction
was followed up to six days. Positive control cultures were
stimulated with appropriate mitogens: LPS (5 g/ml) for IL-1
induction, PHA (5 g/ml) for IFN, Con A (5 g/ml), a T cell
mitogen, for IL-4 and all three mitogens individually for IL-
8 induction. Furthermore, to determine whether cartilage extract
386 L. Merly et al. / International Immunopharmacology 7 (2007) 383391
could affect the response of HPBL to PHA, cultures were also
stimulated with a combination of both PHA (5 g/ml) and SCAE
(0.350 mg/ml) and culture supernatants were harvested and
assayed for IFN daily for 6 days. Unstimulated cell cultures
consisted of medium alone. To rule out the possibility that the
extracts contained material that influenced and was reactive in
the cytokine assays, SCAE alone and in medium was assayed for
cytokine.
2.6.2. Protease susceptibility
Because shark cartilage is a dietary supplement, SC ex-
tracts were treated with digestive proteases. Acid extracts of
shark and bovine cartilage, dialyzed against 0.01 M TrisHCl
buffer, pH 7.5, were subjected to proteolysis using a com-
bination of trypsin and chymotrypsin. Enzymes (2 mg each of
trypsin and chymotrypsin) were added directly to 10 ml of
dialyzed extract and the mixture incubated for 3 h at 37 C.
Digestion was stopped by the addition of 4 mg of trypsin
chymotrypsin inhibitor directly to the mixture. The digested
extracts were dialyzed against RPMI-1640 culture medium
and filtered before use in bioassays. Each digest was tested for
endotoxin.
2.7. Statistical analyses
Each experiment was repeated at least once for each treat-
ment. Cultures were set up in duplicate and cytokine assays on
culture supernatants were performed in triplicate. Standard errors
above the mean were calculated and reported for each expe-
riment for each treatment. Statistical analysis was performed by
the Student's t-test, comparing induction level of cytokine in the
treated culture with that of cultured cells in medium alone, and
each treatment was compared to the positive control for each
experiment. A p-value of b0.05 was considered statistically
significant.
3. Results
3.1. Test for detectable endotoxin
All cartilage extracts were determined to be free of
detectable endotoxin by the E-toxate test. Extracts were also
tested and found to be free of inhibitors for the E-toxate test
(results not shown).
3.2. Cytokine-inducing activity of cartilage extracts
Results for all cytokine and chemokine assays are expres-
sed for each treatment as the mean of three replicate assays of
duplicate cultures for two experiments. The p-values for each
assay are provided in the figure captions and ranged from
p b 0.001 to p b 0.01. All experiments were carried out for
multiple time intervals and where significant cytokine induc-
tion occurred is reported for each cytokine. Extracts used as
stimulants were determined to be free of inhibitory/interfering
material for the cytokine ELISAs (results not shown).
3.2.1. Stimulation of TNF production
Fig. 1 shows that of all the various extracts prepared (AE,
SS and PBS) from each of three brands of shark cartilage, the
acid extract (SCAE) was the most effective inducer of TNF,
yielding high levels of TNF. This was true for acid extracts
prepared from all three commercial brands of shark cartilage.
When acid extract of three commercial brands of shark carti-
lage were compared to each other for cytokine-inducing acti-
vity, acid extract of shark cartilage from Solgar Laboratories
induced higher level of cytokine than that of Twin Labo-
ratories, but not significantly different from FDC acid extract.
The AE, SS, and PBS extracts prepared from Twin Labo-
ratories cartilage induced the least amount of TNF.
Similarly, when the cytokine-inducing properties of AE,
SS, and PBS extracts of shark (Solgar) and bovine cartilage
were compared (Fig. 2), the AE extracts from both sources
induced significant levels of TNF. When the level of TNF
produced was correlated to mg extract protein, the shark
extracts produced 35 fold higher levels of TNF than bovine
extracts. Comparing the three bovine extracts, the SS extract
induced the highest level of TNF. The alkaline (AA) extract
of shark cartilage induced TNF production albeit to a lesser
degree than SCAE, and the organic extract (SCOE) did not
induce detectable level of TNF over 72 h (results not shown).
When HPBL were cultured in the presence of shark cartilage
acid extracts (Solgar SCAE) significant amounts of TNF were
produced in vitro comparable to that induced by LPS, a strong
inducer of TNF (Fig. 3). In serum a level of b20 pg/ml of
TNF is considered normal [24]. No cytokine was detected in
unstimulated cultures containing medium alone. TNF could be
detected in culture supernatants as early as 4 h. By 48 h detect-
able levels were low. A time-course study of TNF release
revealed that peak levels of TNF were observed at 8 and 24 h
Fig. 1. Cytokine response of cultured human leukocytes stimulated
with different extract preparations from three commercial sources of
shark cartilage. The production of TNF by HPBL cultures stimulated
for 24 h with SCAE, SCSS and SCPBS extracts prepared from three
different brands of commercial shark cartilage was measured. To make
reliable comparisons between brands and preparations, the level of
TNF in culture supernatants was related to mg extract protein. Error
bars represent standard error above the mean. The level of TNF
induced in cultures stimulated with extracts was significantly higher
() than that induced in cultures treated with medium alone ( p b 0.005).
 L. Merly et al. / International Immunopharmacology 7 (2007) 383391
Fig. 2. TNF response of human leukocytes stimulated by three different
extracts prepared from shark (Solgar) and bovine cartilage (BC). Cultures
were stimulated with acid (AE), salt soluble (SS) and phosphate buffered
saline (PBS) extracts for 24 h. Protein concentration of extracts were SC-
AE 0.360 mg protein/ml, BC-AE 0.930 mg protein/ml; SC-SS 0.400 mg
protein/ml, BC-SS 0.960 mg protein/ml; and SC-PBS 0.520 mg protein/
ml, BC-PBS 1.16 mg protein/ml. Culture medium alone and LPS (5 g/
ml) served as control stimulants. Error bars represent standard error
above the mean. The level of TNF induced in cultures treated with SC
extracts was significantly higher () than that induced in cultures treated
with BC extracts or medium alone ( p b 0.01).
(Fig. 3). A similar pattern of cytokine release was produced with
LPS over a 48 h period.
The cytokine-inducing ability of SCAE (Solgar) was also
compared to that of collagen and chondroitin sulfate, two com-
ponents of cartilage. No cytokine induction was noted in response
to stimulation with chondroitin sulphate (Fig. 4). Collagen did
induce cells to produce TNF, however, the level of cytokine
produced was significantly less than that induced by SCAE. The
level of TNF produced appeared to be related to the concen-
tration of the stimulant, suggesting a dose-related response.
3.2.2. Stimulation of IFN production
HPBL cultured with SCAE produced significant amounts
of IFN in vitro (Fig. 5). Most IFN was produced in response
Fig. 3. Time-course of TNF induction in leukocytes stimulated with
shark cartilage (Solgar). HPBL cultures were stimulated with LPS
(5 g/ml), SCAE (0.450 mg protein/ml) or culture medium alone at
37 C, 5% CO2. Error bars represent standard error above the mean.
SCAE and LPS induced significantly high levels () of TNF com-
pared to treatment with medium alone ( p b 0.005).
387
Fig. 4. TNF response of leukocytes to stimulation by SCAE, collagen
and chondroitin sulfate. The level of TNF produced was measured at
24 h following stimulation. Culture medium and LPS were control
stimulants. Error bars represent standard error above the mean. SCAE
and LPS induced significantly higher levels () of TNF than did
treatment with medium alone ( p b 0.005). Collagen and chondroitin
sulfate did not induce significant levels of TNF except for collagen at
high concentration, 300 mg/ml ( p b 0.05).
to SCAE by the 3rd day of culture, whereas PHA (a potent
mitogen) induced comparable level of IFN by the 2nd day.
However, when SCAE and PHA were combined to stimulate
HPBL cultures simultaneously, there was a significant increase
in the level of IFN produced. Furthermore, this increased
activity was reached by the second day. Results indicate that
SCAE enhanced the PHA response by inducing a higher level
of IFN. The normal serum range of IFN is b 20 pg/ml [25].
Fig. 5. Induction of IFN in HPBL cultures stimulated with PHA, SCAE
(Solgar), and a combination of both PHA and SCAE. Cultures were
stimulated with PHA (5 g/ml), SCAE (0.367 mg protein/ml) and a
combination of PHA and SCAE. Unstimulated cultures were grown in
culture medium alone. Error bars represent standard error above the mean.
SCAE induced significantly higher level () of IFN at 72 h when
compared to cultures treated with medium alone ( p b 0.008). PHA
induced a significantly higher level of IFN at 48 h, while the combination
of SCAE and PHA induced higher levels of IFN at bother 48 and 72 h
when compared to cultures treated with medium alone ( pb 0.008).
388 L. Merly et al. / International Immunopharmacology 7 (2007) 383391
Fig. 6. IL-1 production by leukocytes stimulated with SCAE (Solgar).
HPBL were stimulated with SCAE (0.396 mg protein/ml), LPS (5 g/ml)
or culture medium at 37 C. Error bars represent standard error above the
mean. The level of IL-1 induced by SCAE was significantly high () at
12 and 24 h, while LPS induced significant levels of cytokine over 96 h
( p b 0.001).
3.2.3. Stimulation of IL-1 production
The production of IL-1 in response to cartilage extracts
was examined (Fig. 6). Results showed that SCAE (Solgar)
induced IL-1 early in culture, reaching peak level by 12 h and
declining to insignificant level by 72 h. The IL-1 response to
LPS, however, was maintained at a significantly high level
through 72 h showing a significant decline only at 96 h.
3.2.4. Stimulation of IL-4 production
When HPBL were cultured in the presence of shark cartil-
age extract (Solgar), there was no physiologically significant
production of IL-4 over 48 h (Fig. 7), a response similar to that
obtained with unstimulated cultures (medium alone). Positive
control cultures stimulated with Con A, secreted a significant
level (40 pg/ml) of IL-4.
Fig. 7. IL-4 response of leukocytes stimulated with SCAE (Solgar).
HPBL cultures were stimulated with Con A (5 g/ml), or SCAE
(0.360 mg protein/ml) or culture medium. Error bars represent standard
error above the mean. SCAE did not induce levels of IL-4 greater than
that induced in cultures treated with medium alone, while Con A
induced significantly higher levels () of IL-4 than unstimulated (me-
dium alone) cultures ( p b 0.005).
Fig. 8. Induction of IL-8 in leukocytes. HPBL cultures were stimulated
with SCAE (0.396 mg protein/ml), LPS (5 g/ml), PHA (5 g/ml),
Con A (5 g/ml) or culture medium. Error bars represent standard error
above the mean. SCAE and LPS induced significantly higher levels
() of IL-8 than medium alone at 24, 48 and 72 h ( p b 0.001).
3.2.5. Stimulation of IL-8 production
The release of IL-8 in response to cartilage stimulation was
compared to chemokine induction following stimulation of
cultures by mitogens, such as LPS, PHA and Con A (Fig. 8).
Significant levels of IL-8 were induced by SCAE within 24 h
and levels were sustained over a period of 72 h. The produc-
tion of IL-8 in response to LPS however, reached peak levels at
48 h and by 72 h showed a gradual decline. Physiologically
significant level of IL-8 was not produced by cultures stimu-
lated with culture medium alone, PHA, or Con A.
3.3. Susceptibility to proteolytic digestion
The effect of proteolytic digestion on the TNF-inducing
ability of shark and bovine cartilage was examined by using
Fig. 9. The effect of proteolytic digestion on TNF induction. Shark
(Solgar) and bovine cartilage extracts were treated with trypsin and
chymotrypsin and used to stimulate HPBL cultures for 24 h. Error bars
represent standard error above the mean. In HPBL cultures stimulated
with protease-treated extracts, significantly less () TNF was produ-
ced than in cultures stimulated with untreated extracts derived from
both shark and bovine cartilage ( p b 0.005).
 L. Merly et al. / International Immunopharmacology 7 (2007) 383391
extracts treated with trypsin and chymotrypsin (digestive
enzymes) prior to stimulating leukocyte cultures (Fig. 9).
Following treatment, SCAE (Solgar) lost 80% activity suggest-
ing a protein moiety is associated with the cytokine-inducing
activity. A similar effect was noted with bovine cartilage extract.
4. Discussion
The results of this study provide evidence that cartil-
age extracts induce the production of several different
cytokines and/or chemokines by human peripheral blood
leukocytes. The most well-characterized cytokine re-
sponse of HPBL to cartilage stimulation, thus far, is
TNF production. When different commercial brands of
shark cartilage were tested for their ability to induce
TNF, there was a significant variation in the level of
cytokine induced, although each brand induced a detect-
able level of TNF. When extracts were prepared from
the amount of commercial shark cartilage equivalent to
the manufacturer's recommended daily dose (i.e., appro-
ximately a three-fold increase in starting material), a 20%
increase in the level of TNF was noted compared to the
level induced with extract prepared from 2 g of SC (data
not shown).
For all brands of cartilage tested, the most cytokine-
inducing activity was associated with the acid extracts of
cartilage. Acid extracts simulate the acidic environment
of the stomach. For a dietary supplement to be biolo-
gically effective at the time of absorption it must be acid
resistant. Considering that the acid extract of shark
cartilage was the most effective inducer of a cytokine
response, the acidity of the stomach may very well play
a role in the in vivo release of the active component(s)
from crude cartilage preparations taken orally.
Manufacturers are reticent to provide information on
the source of the original shark material such as geo-
graphic location, species identity, age/sex of animal,
tissue from which material was derived, and/or the pro-
cess used to obtain material. These unknown differences
most likely result in the variation in cytokine response
seen with different commercial sources of shark cartilage.
While both shark cartilage and bovine cartilage ex-
tracts induced the production of TNF, shark cartilage
induced significantly higher levels of TNF, with the
acid extract inducing highest levels, of all extracts tested.
For bovine cartilage, the salt-soluble extract induced a
higher level of TNF than the acid extract. Taken toge-
ther, the results suggest that although both types of
cartilage contain cytokine-inducing activity, the active
component(s) present in each might not be similar.
Alternatively, the difference might reflect the nature of
the starting material used to prepare the extracts.
389
Both shark and bovine cartilage extracts, when trea-
ted with proteases, trypsin and chymotrypsin, lost con-
siderable ability to stimulate TNF production. Results
suggest that the active component(s) in cartilage is pro-
teinaceous. Since not all activity was abolished follo-
wing enzyme treatment, this leads us to speculate that
the extracts contain an additional active component(s)
which is resistant to the proteolytic activity of trypsin
and chymotrypsin. Given the complexity of the gut
environment, the effect of proteases may not be as
profound as that seen in vitro, and residual activity may
be sufficient to stimulate the epithelial lining of the gut
to produce a variety of immune messengers which in
turn could have a systemic effect given the rich blood
supply to the gut. When speculating on in vivo condi-
tions, the presence of microbial enzymes must also be
considered which could be a factor in determining the
potential effectiveness of shark cartilage as a dietary
supplement.
Cartilage is a complex material composed of several
proteinaceous components, including collagen and chon-
droitin sulphate. Chondroitin sulfate is least likely to be
responsible for the cytokine-inducing effect of cartilage.
Collagen alone was able to induce a cytokine response but
not at the level seen with SCAE. It is unlikely that colla-
gen present in commercial preparations of shark cartilage
contributes significantly to the level of in vitro stimulation
noted considering the large amount of collagen required
(0.36 mg/ml SCAE versus 300 mg/ml collagen).
An important aspect of this investigation was the
time-course study of the induction of cytokines in res-
ponse to cartilage stimulation, specifically TNF. The
biphasic response to SCAE stimulation may reflect
TNF produced by two or more different cell popula-
tions given that the cytokine is a pleiotropic cytokine
[26]. Macrophages might be the cell type to initially
respond to direct stimulation by cartilage extract [27],
while other cell types (e.g., lymphocytes) account for
peak activity noted at 24 h, which may be an indirect
effect in response to additional cytokine messengers
released by the cell type initially responding to cartilage
stimulation [28]. Alternatively, the decrease in level of
TNF detected in supernatants at 12 h may not represent
a decline in secretion but rather a binding of the free
cytokine by up-regulated cytokine receptors on activat-
ed cells or by soluble TNF receptors released into
culture supernatants by responding cells.
Shark cartilage extract induced the production of
IFN at levels comparable to that obtained in positive
control cultures stimulated with PHA, a known mitogen.
Shark cartilage extract appeared to behave like a mito-
gen in stimulating the production of IFN after 72 h.
390 L. Merly et al. / International Immunopharmacology 7 (2007) 383391
When cultures were stimulated with PHA alone, IFN
level peaked at 48 h. However, when PHA and SCAE
were combined as stimulants, the cytokine response
was enhanced with a significantly higher level (almost
800 pg/ml) of IFN detected at 48 h. These results
suggest that an active component(s) in shark cartilage
behaves like a mitogen stimulating leukocytes to pro-
duce potent mediators, and its presence appears to en-
hance the activity of PHA. An alternative interpretation
might be the reverse, that is, that PHA induces an earlier
and enhanced response to SCAE. What is undetermined
is whether the enhanced response is due to one or more
cell type(s) responding.
The induction of proinflammatory cytokines, TNF,
IFN, and IL-1 suggests that cartilage preferentially
induces Th1 type cytokines. The production of IL-4
(b5 pg/ml) was considered to be physiologically in-
significant (normal serum range b30 pg/ml). In contrast,
the chemotactic chemokine, IL-8, was produced by
stimulated leukocytes at higher (20,000 pg/ml) than
normal levels encountered in serum (010 pg/ml). Esta-
blishing normal ranges for cytokine serum levels
derived from immune assays of serum/plasma of healthy
individuals can be misleading, since recognized varia-
tion exists in cytokine levels in normal individuals to the
extent that cytokine levels can often be undetectable by
ELISA. It is, therefore, essential to include normal con-
trols in each individual experiment to take into account
possible variation [29]. The cytokine induction profile
for shark cartilage, therefore, includes three potent Th1
type cytokines, TNF, IFN and IL-1, two of which
are characteristically involved in inflammatory res-
ponses and an inflammatory chemokine, IL-8. Shark
cartilage did not induce a significant level of IL-4, thus it
not only preferentially stimulates a Th1 type response
but appears to indirectly inhibit the development of a
Th2 response through the action of IFN which is an
inhibitor of Th2 cell population expansion [30,31].
From the cytokine profile induced by shark cartilage
one can conclude that activation, proliferation, and dif-
ferentiation of B cells (i.e., a humoral immune response)
is not an integral part of the immune response to SCAE
stimulation since IL-4, a key cytokine in the activation
of B cells, was not induced to physiologically significant
levels by SCAE [32]. Based on the differences in peak
production of cytokines over time we can speculate on
the sequence of cellular events. An early event is most
likely to be monocyte/macrophage activation by SCAE
inducing the autocrine and paracrine effects of TNF,
followed by Th cell activation, preferably Th1, in res-
ponse to TNF and IL-1 co-stimulation with IFN
inhibiting proliferation of Th2 cells.
We have shown that cartilage induces various inflam-
matory cytokines and a potent chemokine in immune
cells. Thus, if through intestinal absorption the active
component(s) in shark cartilage were to reach systemic
circulation and/or target sites in the body, immune
regulation could be significantly influenced. Conse-
quently, the inflammatory cytokine profile induced by
cartilage might be deleterious to the health of indivi-
duals suffering from chronic diseases where the under-
lying pathology is associated with inflammation and
where up regulation of Th1 type response is undesirable.
For such individuals shark cartilage, taken as an unre-
gulated dietary supplement, is counter indicated. Alter-
natively, this type of immune response, in which IL-4
production is restricted, could be beneficial for hyper-
sensitive individuals whose allergic state is often asso-
ciated with production of IgE. In the absence of Th2
stimulation, the IgE response would be down regulated
to the benefit of the individual.
Acknowledgments
We thank Barbara Anderson for performing phlebot-
omy, Betzabel Gonzalez for her technical contributions,
and John Makemson and Charles Bigger for their
comments on the manuscript. LM was supported by an
MBRS student fellowship and this study was supported
in part by a summer research award to LM under the
Comparative Immunology Initiative (R25 GM061347).
Facilities for the project were provided by FIU Com-
parative Immunology Institute and this is contribution
FIU-CII-003 from the Institute. The authors received no
funding from and have no association with the manu-
facturers and/or distributors of the commercial brands of
shark cartilage used in this study.
References
[1] Eisenberg DM, Davis RB, Ettner SL, Appel S, Wilkey S, et al. Trends
in alternative medicine use in the United States, 19901997
results of a follow-up national survey. JAMA 1998;280(18):
156975.
[2] Kennedy J. Herb and supplement use in the US adult population.
Clin Ther 2005;27(11):184758.
[3] Tindle HA, Davis RB, Phillips RS, Eisenberg DM. Trends in use
of complementary and alternative medicine by US adults: 1997
2002. Altern Ther Health Med 2005;11(1):429.
[4] Oneschuk D, Fennell L, Hanson J, Bruera E. The use of comple-
mentary medications by cancer patients attending an outpatient pain
and symptom clinic. J Palliat Care 1998;14(4):216.
[5] Suarez-Almazor M, Kendall CJ, Dorgan M. Surfing the Net-
Information on the World Wide Web for persons with arthritis:
patient empowerment or patient deceit? J Rheumatol 2001;28
(1):18591.
 L. Merly et al. / International Immunopharmacology 7 (2007) 383391
[6] Brem H, Folkman J. Inhibition of tumor angiogenesis mediated
by cartilage. J Exp Med 1975;141:42739.
[7] Lee A, Langer R. Shark cartilage contains inhibitors of tumor
angiogenesis. Science 1983;221:11857.
[8] Oikawa T, Ashino-Fuse H, Shimamura M, Koide U, Iwaguchi T. A
novel angiogenic inhibitor derived from Japanese shark cartilage
(I). Extraction and estimation of inhibitory activities toward tumor
and embryonic angiogenesis. Cancer Lett 1990;51(3):1816.
[9] Gingras D, Renaud A, Mousseau N, Beliveau R. Shark cartilage
extracts as antiangiogenic agents: smart drinks or bitter pills?
Cancer Metastasis Rev 2000;19:836.
[10] Gonzalez RP, Leyva A, Moraes MO. Shark cartilage as source of
antiangiogenic compounds: from basic to clinical research. Biol
Pharm Bull 2001;24(10):1097101.
[11] Loprinzi CL, Levitt R, Barton DL, Sloan JA, Atherton PJ, et al.
Evaluation of Shark Cartilage in patients with advanced cancer: A
North Central Cancer Treatment Group Trial. Wiley Interscience;
2005. Published online.
[12] Felzenszwalb I, Peleilo de Mattos JC, Bernardo-Filho M,
Caldeira-de-Araujo A. Shark cartilage-containing preparation:
protection against reactive oxygen species. Food Chem Toxicol
1998;36:107984.
[13] Conde CMS, Albano F, Bouskela E, Felzenszwalb I, Svensjo E.
Inhibition of ischemia/reperfusion induced plasma leakage by
alpha-tocopherol, trolox, and a shark cartilage preparation with
anti-oxidant properties. Nutr Res 2001;21:136371.
[14] Chen JS, Chang CM, Wu JC, Wang SM. Shark cartilage extract
interferes with cell adhesion and induces reorganization of focal
adhesions in cultured endothelial cells. J Cell Biochem 2000;78:
41728.
[15] Ratel D, Glazier G, Provencal M, Boivin D, Beaulieu E, et al.
Direct-acting fibrinolytic enzymes in shark cartilage extract:
potential therapeutic role in vascular disorders. Thromb Res
2005;115(12):14352.
[16] Feyzi R, Hassan Z, Mostafaie A. Modulation of CD4+ and CD8+
tumor infiltrating lymphocytes by a fraction isolated from shark
cartilage: shark cartilage modulates anti-tumor immunity. Int
Immunopharmacol 2003;3:9216.
[17] Kralovec J, Guan Y, Metera K, Carr RI. Immunomodulating
principles from shark cartilage Part I. Isolation and biological
assessment in vitro. Int Immunopharmacol 2003;3:65769.
[18] Hassan Z, Feyzi R, Sheikhian A, Bargahi A, Mostafaie A, et al.
Low molecular weight fraction of shark cartilage can modulate
391
immune responses and abolish angiogenesis. Int Immunophar-
macol 2005;5(6):96170.
[19] Romagnani S. Immunologic influences on allergy and the Th1/
Th2 balance. J Allergy Clin Immunol 2004;113(3):395400.
[20] Muraille E, Leo O. Revisiting the Th1/Th2 paradigm. Scand J
Immunol 1998;47(1):19.
[21] Kidd P. Th1/Th2 balance: the hypothesis, its limitations, and
implications for health and disease. Altern Med Rev 2003;8(3):
22346.
[22] Smith PK, Krohn RI, Hermanson GT, Mallia AK, Gartner FH, et al.
Measurement of protein using bicinchoninic acid. Anal Biochem
1985;150:7685.
[23] Nandan R, Brown DR. An improved in vitro pyrogen test: to
detect picograms of endotoxin contamination in intravenous
fluids using limulus amoebocyte lysate. J Lab Clin Med 1977;89
(4):9108.
[24] Aggarwal BB, Reddy S. Tumor necrosis factor. In: Nicola
NA, editor. Guidebook to cytokines and their receptors. A
Sambrook and Tooze Publication of Oxford University Press;
1994. p. 1034.
[25] Gray PW. Interferon Gamma. In: Nicola NA, editor. Guidebook
to cytokines and their receptors. A Sambrook and Tooze
Publication of Oxford University Press; 1994. p. 1189.
[26] Tracey KJ, Cerami A. Tumor necrosis factor: a pleiotropic cytokine
and therapeutic target. Annu Rev Med 1994;45:491503.
[27] Mosser DM. The many faces of macrophage activation. J Leukoc
Biol 2003;73(2):20912.
[28] Wang J, Wakeham J, Harkness R, Xing Z. Macrophages are a
significant source of type 1 cytokines during mycobacterial
infection. J Clin Invest 1999;103(7):10239.
[29] Whiteside TL. Cytokine measurements and interpretation of
cytokine assays in human disease. J Clin Immunol 1994;14(6):
32739.
[30] Billiau A, Heremaus H, Vermeire K, Matthys P. Immunomod-
ulatory properties of interferon gamma: an update. Ann NY Acad
Sci 1998;856:2232.
[31] Maggi E, Parronchi P, Manetti R, Simonelli C, Piccinni MP, et al.
Reciprocal regulatory effects of IFN-gamma and IL-4 on the in
vitro development of human Th1 and Th2 clones. J Immunol
1992;148(7):21427.
[32] Lund FE, Garvey BA, Randall TD, Harris DP. Regulatory roles
for cytokine-producing B cells in infection and autoimmune
disease. Curr Dir Autoimmun 2005;8:2554.