MEAT SCIENCE

Meat Science 69 (2005) 371380 www.elsevier.com/locate/meatsci

Antioxidant and antibacterial activities of natural extracts: application in beef meatballs

ndez-Lo pez J. Ferna
a

a,*

rez-Alvarez a, V. Kuri , N. Zhi b, L. Aleson-Carbonell a, J.A. Pe

a Agroalimentaria, Escuela Polite cnica Superior de Orihuela (EPSO), Universidad Miguel Herna ndez, Ctra. a Departamento de Tecnolog Beniel Km 3,2 (03312) Orihuela, Alicante, Spain b Agri Food Centre/SBC, Seale-Hayne, University of Plymouth, Newton Abbot, UK Received 9 February 2004; received in revised form 6 August 2004; accepted 16 August 2004

Abstract The antioxidant and antibacterial eect of rosemary, orange and lemon extracts was investigated in cooked Swedish-style meatballs. Activity in a lard system was established for all the extracts and further determination of the development of rancidity as thiobarbituric acid reactive substances consistently showed that about 50% of the rancidity can be controlled by the citrus preparations. Two of the rosemary extracts (water soluble and oil soluble) were more eective with practically complete elimination of rancidity (TBA values) after a period of 12 days. Rosemary extract activity against lactic acid bacteria and Listeria but not Brochothrix thermosphacta was demonstrated in an agar diusion test, but in the product only lactic acid bacteria counts were slightly reduced. Sensory analysis results, particularly aroma and acceptability scores, indicated the signicant advantages in using rosemary and citrus extracts in rancidity-susceptible meat products. 2004 Elsevier Ltd. All rights reserved.
Keywords: Rosemary; Garlic; Antioxidant; Antimicrobial; Citrus; Spoilage; Lactic acid bacteria

1. Introduction The appearance of foods is one of the major determinants of its appeal to consumers and consequently, sales of the product. Lipid oxidation and bacterial contamination are the main factors that determine food quality loss and shelf-life reduction. Therefore, delaying lipid oxidation and preventing bacterial cross-contamination are highly relevant to food processors. The growth of microorganisms in meat products may cause spoilage or foodborne diseases. Oxidative processes in meat lead to the degradation of lipids and proteins which, in turn, contribute to the deterioration in avour, texture and colour of displayed meat products (Decker, Chan, Livisay, Buttereld, & Faustman, 1995). Several authors
*

Corresponding author. Tel.: +34 966749656; fax: +34 966749677. ndez-Lo pez). E-mail address: j.fernandez@umh.es (J. Ferna

have postulated that in meat and meat products pigment and lipid oxidation are interrelated (Anton, Salgues, & Renerre, 1993; Yin & Faustman, 1993). Genot, Borel, Metro, Gandemer, and Renerre (1991) concluded that O 2 can initiate lipid peroxidation, leading to the formation of prooxidant substances capable of reacting with oxymyoglobin (OMb) and resulting in metmyogobin (MMb) formation. Anton et al. (1993) postulated that OMb could be oxidized not only by lipid-oxy radicals but by other pro-oxidant radicals generated by O 2 . Several investigators reported that the susceptibility of myoglobin to autoxidation is the main factor in explaining colour stability in meat and meat products (Renerre, Anton, & Gatellier, 1992). Although synthetic additives have been widely used in the meat industry to inhibit both, the process of lipid oxidation and microbial growth, the trend is to decrease their use because of the growing concern among

0309-1740/$ - see front matter 2004 Elsevier Ltd. All rights reserved. doi:10.1016/j.meatsci.2004.08.004

372

ndez-Lo pez et al. / Meat Science 69 (2005) 371380 J. Ferna

consumers about such chemical additives (Chastain, Hufman, Hsieh, & Cordray, 1982; Chen, Pearson, Gray, Fooladi, & Ku, 1984). Consequently, search for natural additives, especially of plant origin, has notably increased in recent years (Lo liger, 1991). Compounds obtained from natural sources such as grains, oilseeds, spices, fruit and vegetables have been investigated (Chen, Muramoto, Yamauchi, & Huang, 1996). Therefore, the development and application of natural products with both antioxidants and antibacterial activities in meat products may be necessary and useful to prolong their storage shelf life and potential for preventing food diseases. Rosemary (Rosmarinus ocinalis L.) has been reported to contain certain compounds including, rosmanol, rosmariquinone, rosmaridiphenol and carnosol, which may be up to four times as eective as butylated hydroxy anisole (BHA) and equal to butylated hydroxy toluene (BHT) as antioxidants (Houlihan, Ho, & Chang, 1984, 1985; Nakatani & Inatani, 1984). Moreover, several authors reported that some compounds present in rosemary extracts could have antibacterial activity (Cuvelier, Richard, & Berset, 1996). Del Campo, Amiot, and Nguyen-The (2000) reported that the compounds responsible for the antibacterial action seemed presumably to be the phenolic di-terpenoids, which are the main compounds of the apolar fraction of the rosemary extracts. Garlic extracts have also been shown to have antioxidant activity in dierent in vitro models. The antioxidant activity of Allium plants has mainly been attributed to a variety of sulphur-containing compounds and their precursors (Kim, Kubota, & Kobayashi, 1997; Lampe, 1999; Nuutila, Puupponen-Pimia , Aarni, & Oksman-Caldentey, 2003). These compounds have been also reported as responsible for their in vitro antibacterial activity (Harris, Cottrell, & Plummer, 2001; Tsao & Yin, 2001). Citrus fruits are an important source of bioactive compounds (avonoids and vitamin C). The main avonoids found in citrus species are hesperidine, narirutin, naringin and eriocitrin (Mouly, Arzouyan, Gaydou, & Estienne, 1994; Schieber, Stintzing, & Carle, 2001). Ascorbic acid, a well known natural antioxidant, together with natural avonoids are also attracting more and more attention not only due to their antioxidant properties, but as anti-carcinogenic and anti-inammatory agents because of their lipid anti-peroxidation effects (Elangovan, Sekar, & Govindasamy, 1994; n, Frutos, Pe rez-Alvarez, Mart nez-Sa nchez, & Mart o, 2002). Del R As can be seen, several studies have demonstrated the antibacterial and/or antioxidant properties of these plants, mainly using in vitro assays. Moreover, some researchers reported that there is a relationship between the chemical structures of the most abundant

compounds in the plants and their above mentioned functional properties (Dean & Svoboda, 1989; Farag, Daw, Hewedi, & El-Baroty, 1989). Because these compounds are likely to interact with their environment, and the composition could vary according to regional conditions, the investigation of their activity in a range of food systems is still needed to successfully apply them into meat products. The objectives of this study were to investigate (I) the antioxidant activities of commercially available natural extracts, (II) the ability of these natural extracts for inhibiting the growth of 11 foodborne bacterial contaminants and (III) their ability to extend the storage shelf life of a cooked meat product.

2. Materials and methods 2.1. Natural extracts Rosemary and garlic extracts: Three dierent commercial rosemary extracts and one of garlic were provided by Kalsec (Kalsec Mildenhall, UK): Herbalox seasoning Type HT-O which is an rosemary oil miscible extract, Duralox oxidation management Blend CN-2 which is a rosemary water miscible extract, Herbalox seasoning Type W which is a rosemary oil and water miscible extract, and Aquaresin garlic which is a garlic water miscible extract. Citrus extracts (orange and lemon): these extracts were obtained from by-products of lemon and orange juice industries. As these by-products have a high water content (80%) which would make their application in the food industry dicult, it was necessary to process them to obtain a dry powder of aprox. 7% moisture. ndez-Gine s, Ferna nThe process described by Ferna pez, Sayas-Barbera , Sendra, and Pe rez-Alvarez dez-Lo (2003) included some built-in precautions to minimize the losses of the associated bioactive compounds; avonoids and ascorbic acid. 2.2. Bacterial cultures The eleven foodborne bacteria used as test organisms were selected as follows: Listeria innocua 4202, Listeria monocytogenes 5105, Lactobacillus sake 550, Leuconostoc mesenteroides subsp. mesenteroides 824, Leuconostoc mesenteroides subsp. dextranicum 882, Leuconostoc carnosum 558, Lactobacillus curvatus 860 Brochothrix thermosphacta CRA 7883, Brochothrix thermosphacta CRA 7884, Brochothrix thermosphacta CRA 3235, and Lactococcus lactis FMRD 492. All isolates from vacuum packed spoiled meat products (Borch & Molin, 1988; Davies et al., 1999) except the last one (FMRD 492). FMRD strain was obtained from the Agricultural and Food Research Centre (DARDNI, Belfast, UK) and

 ndez-Lo pez et al. / Meat Science 69 (2005) 371380 J. Ferna

373

CRA strains are from Campden and Chorleywood (Food Research Association, Chipping Campden, UK). All of them were selected as markers for meat products deterioration. 2.3. Determination of antioxidant eect in natural extracts (Rancimat method) The American Oil Chemists Society (AOCS) air oxidation method (AOM-AOCS Cd 12b-92) was used to determine the antioxidant eect of each extract (La ubli & Bruttel, 1986). All experiments were performed with a 679 Rancimat (Metrohm, Herisau, Switzerland). Each extract (at the percentages showed in Table 1) was added to 2.5 g of melted commercial pork lard (Co-op, UK) into the reaction vessel. Then, they were inserted into the heating block for 10 min to preheat the sample (120 C). The air supply and the absorption vessels were connected and recording of the conductivity curves started. Lard (without extract added) was used as control sample. The results are expressed as stability index (SI), calculated as a ratio of the induction time of the treatment sample and the induction time of the control. Three replications of this experiment were made. 2.4. Determination of antibacterial eect in natural extracts Antimicrobial activity of natural extracts against 11 bacteria was tested using the agar diusion method (Daeschel, 1992; Kuri, 1998). Stock cultures of all tested bacteria were grown in nutrient broth (Oxoid Unipath Ltd., Basingtoke, UK) for 18 h. Final cell concentrations were standardized until 107108 cfu/ml using the McFarland solutions standards (NCCLS, 1990). One millilitre of this inoculum was added to each plate containing soft brain heart infusion (BHI) agar (Oxoid Unipath Ltd., Basingtoke, UK). When the agar was solidied, 4 wells (6 mm diameter) were formed in each plate. Fifty microlitre of each extract was applied into each well. The control sample was prepared with the same amount of sterile distilled water instead of extract. After a diusion time of 1215 min at room temperature, the plates inoculated with Brochothrix spp. were
Table 1 Type and amount of each natural extract used in this study Treatment Control Rosemary oil extract (OR) Rosemary water extract (WR) Rosemary oil and water extract (OWR) Garlic extract (GR) Lemon ber (LF) Orange ber (OF) Extract Herbalox Type HTO Duralox Herbalox Type W Aquaresin garlic Lemon extract Orange extract Percentage (%, w/w) 0.10 0.15 0.25 0.05 5.00 5.00

incubated at 22 C for 48 h. All lactic acid bacteria (LAB) were incubated at 30 C for 48 h and Listeria spp. at 37 C for 48 h. After this time, the inhibition diameter was measured in 3 directions and the average tabulated or used to calculate activity units. Three replications of this experiment were made. 2.5. Meatball manufacture 2.5.1. Product formulation Swedish-style meatballs were manufactured according to a conventional formula: 78% minced beef (20% fat content), 14.5% ake potatoes, 5% water and 2.5% salt. Sunower oil (1.5%) was used as a carrier for the extracts. A set of 6 treatment samples diering only by the extract added were prepared (Table 1). Garlic extract was not used in meatballs manufacture due to their pro-oxidant activity showed with rancimat method and to their strong aroma showed in previous assays. Rosemary extracts were used at the concentrations recommended by the supplier for its use in meat product. Citrus extracts were used at the concentration sug ndez-Gine s et al., gested by previous studies (Ferna 2003; Fernandez-Lopez et al., 2004). 2.5.2. Product processing The products were prepared in a pilot plant resembling to commercial processing conditions. All ingredients were homogenized in a bowl mixer with a spiral dough hook (Silverson, Birmingham, UK) during 5 min. For each treatment, the corresponding extract was added at the concentrations shown in Table 1, and then mixed again for 5 min. Meatballs were formed by hand (15 g, 2025 mm) and then subjected to a twostage cooking process. First the meatballs were ash fried into sunower oil at 190 C for 30 s. The objective of this stage was to seal the surface of the ball and produce the characteristic browned look. They were then thoroughly cooked in a forced draught oven (Zanussi, Italy) at 250 C during 4 min to reach an internal temperature of 72 C in the center of the meatball. The temperature was monitored using an Omega digital thermometer (Omega Engineering, Inc., Stamford, CT) with a chromelalumel (Omega K) thermocouple probe positioned in the geometric center of the product samples. When the endpoint temperature was achieved, the samples were immediately placed in a chiller (2 5 C) to reach a product temperature below 12 C. Three replications of this experiment were made. 2.5.3. Storage conditions After reaching the packing temperature the samples were placed into plastic containers, sealed with one layer of a semi-permeable lm (Polyvinylidene chloride, Freshcling, Plymouth, UK), and stored in darkness at

374

ndez-Lo pez et al. / Meat Science 69 (2005) 371380 J. Ferna

8 1 C for 12 days to follow an accelerated shelf life determination protocol (IFST, 1993). Sampling and storage conditions records from each treatment took place at 1, 3, 6, 9 and 12 days (storage time) and every sample was analysed promptly as follows. 2.6. Meatballs analysis 2.6.1. Colour measurements Meat colour was measured using a Minolta colorimeter CR-200 (Minolta Camera Co., Osaka, Japan) with illuminant D65, 2 observer, Diuse/O mode, 8 mm aperture of the instrument for illumination and 8 mm for measurement. The colorimeter was standardized with a white tile (L* = 98.14, a* = 0.23 and b* = 1.89). Colour was described as coordinates: lightness (L*), redness (a*, red-green) and yellowness (b*, yellow-blue). Nine replicate measurements were taken for each sample, following the guidelines for colour measurements from American Meat Science Association (Hunt et al., 1991). 2.6.2. TBA values Lipid oxidation was assessed in triplicate by the 2-thiobarbituric acid (TBA) method of Tarladgis, Watts, and Younathan (1960) with minor modications. A 10 g sample was blended with 50 ml distilled water for 2 min and then transferred to a distillation tube. The cup used for blending was washed with an additional 47.5 ml distilled water, which was added to the same distillation tube with 2.5 ml 4N HCl and a few drops of antifoam agent silicone o/w (Fisher Scientic, Loughborough, UK). The mixture was distilled and 50 ml distillate was collected. Five ml of 0.02 M 2-thiobarbituric acid in 90% acetic acid (TBA reagent) was added to a vial containing 5 ml of the distillate and mixed well. The vials were capped and heated in a boiling water bath for 30 min to develop the chromogen and cooled to room temperature. The absorbance was measured at 538 nm, against a blank prepared with 5 ml distilled water and 5 ml TBA-reagent, using a PV 8625 spectrophotometer (Philips, UK). Thiobarbituric acid-reactive substances (TBARS) were calculated from a standard curve (850 nmol) of malondialdehyde (MA), freshly prepared by acidication of TEP (1,1,3,3-tetraethoxy propane). Reagents were obtained from Sigma (UK). The TBA numbers were calculated as mg MA/kg sample. 2.6.3. Microbiological analysis A composite sample (10 g) was formed with portions of at least 3 meatballs and homogeneized with sterile 1.5% peptone water, in a Stomacher 400 (Colworth, London, UK) for 1 min. Aliquotes were serial diluted in peptone water and plated out following standard methodologies (Gerhardt et al., 1994). Lactic acid bacteria (LAB) counts were determined on MRS Agar (pH

5.6), with the plates incubated under anaerobic conditions (Gas generating kit anaerobic system, Oxoid Unipath Ltd., Basingtoke, Hampshire, UK) at 30 C for 2 days. Psychrotrophic microbial counts were determined on Plate Count Agar (PCA) with plates incubated at 7 C for 10 days. Coliform organisms were enumerated on MacConkey Agar (MCA) plates incubated at 30 C for 48 h. Moulds and yeasts were determined on RBC (Rose bengal cloramphenicol) agar plates incubated at 22 C for 5 days. Culture media were from Oxoid (Oxoid Unipath Ltd., Basingtoke, Hampshire, UK). Results were expressed as log10 cfu/ml. 2.6.4. Sensory evaluation Sensory evaluation followed standard guidelines ISO 6658-1995 and ISO 6564-1985 (British Standards, 1986a, 1986b). A preliminary descriptive evaluation was performed on commercial Swedish meatballs (Crown Brands International, Ltd, Swindon, UK) and samples made as described above to identify and dene principal sensory attributes and nd consensus. Each meatball sample was evaluated by a trained 4-member panel. The same panel evaluated samples at each storage time (1, 6, 9 and 13 days storage). The sensory questionnaires measured intensity on a 7-point balanced semantic scale (weak to strong) for the following attributes colour (croma), brightness, surface-slime, aroma (rancidity), aroma (putrefaction), aroma (o odour), aroma (acid), aroma (sour) and overall acceptance. 2.7. Statistical analysis Conventional statistical methods were used to calculate means and standard deviations. Statistical analysis (ANOVA) was applied to determine signicant dierences (P < 0.05). To discover where there were signicant dierences between the levels of the main factor, contrasts (Tukey test) between means were made (A & Azen, 1979). For the antioxidant and antimicrobial properties of natural extracts ANOVA with two factors (extract with seven levels: control, OR, WR, OWR, GR, OF and LF, and replication with three levels) were applied. For the meatball analysis, ANOVA with three factors (treatment with six levels: control, OR, WR, OWR, OF and LF, storage time with four levels: 1, 6, 9 and 12 days, and replication with three levels) were applied for each parameter. Statgraphics Plus 5.1 for Windows (Statgraphics Co., Tulsa, OK, USA) was used for statistical analysis.

3. Results and discussions 3.1. Antioxidant activity of natural extracts Antioxidant activities of each natural extract expressed as stability index (SI) are shown in Table 2.

 ndez-Lo pez et al. / Meat Science 69 (2005) 371380 J. Ferna Table 2 Stability index of each natural extract determined by the Rancimat method Extract Control (pork lard without extract) Duralox (WR) Herbalox Type HT-O (OR) Herbalox Type W (OWR) Lemon extract (LF) Orange extract (OF) Aquaresin garlic (GR) Stability indexx 1.00a 1.82b 1.78b 1.46c 1.19d 1.30e 0.69f

375

af Means within a column with dierent letters are signicantly different (P < 0.05). x Stability index = induction time (conductivity curves) of the sample/induction time of the control.

Citrus extracts (LF, OF) showed antioxidant activity but lower (P < 0.05) than rosemary extracts (OR, WR, OWR). The SI of orange extract (OF) was higher (P < 0.05) than lemon extract (LF). The antioxidant activity of phenolic compounds from plants is well known (Pokorny, 1991; Shahidi, 2000). This activity has been mainly attributed to avonoids and ascorbic acid in citrus fruits (hesperidin, neohesperidin and eriocitrin) and to carnosol and rosmarinic acid in rosemary (Schwarz et al., 2001). All of these polyphenolic compounds have the ability to act as antioxidants by a free radical scavenging mechanism and also through their known ability to chelate transition metals n et al., 2002). (inactivation of iron ions) (Mart 3.2. Antibacterial activity of natural extracts The antibacterial activity of natural extracts against the spoilage bacterial strains tested is shown in Table 3. The water (control) was ineective. Lemon extracts (LF), orange extracts (OF) and garlic extracts (GR) were only active against one, two and three bacterial strain/species, respectively. On the other hand, rosemary extracts (OR, WR, OWR) were active against all bacteria tested. From rosemary extracts studied, the oil miscible extract (OR) showed the highest inhibitive eect against each bacterium. Water miscible extract (WR) showed higher antibacterial eect than water and oil miscible extract (OWR) for all bacteria tested except for Brochothrix spp. For all rosemary extracts (OR, WR, OWR), Brochothrix spp. were the most sensitive bacteria. Some reports found that the most apolar phenolic compounds from rosemary extracts are presumably responsible of their antibacterial activity (Del Campo et al., 2000; Karamanoli, Vokou, Menkissoglu, & Constantinidou, 2000). Davidson (1993) reported that gram-positive bacteria are generally more susceptible to nonpolar phenolic compounds than gram-negative microorganisms. This could explain some of the

The only extract that did not show any antioxidant properties was garlic extract (GR), which showed a SI lower than that of the control, which could indicate pro-oxidant activity linked to the garlic product. The antioxidant activity of Allium plants has been attributed to a group of sulphur-containing compounds, from which allicin (diallyl thiosulphinate) appeared to be the main component (Kim et al., 1997). Animal tests have shown that in low concentrations, allicin can be responsible for the antioxidative properties of garlic, although it can also act as a pro-oxidant in high concentrations (Lawson, 1998). Also, controversial results about antioxidant properties of garlic extracts have been found to be dependent on the sample extraction method. These dierences have been attributed to the spectrum of compounds extracted with the dierent solvents used (Nuutila et al., 2003; Yin & Cheng, 1998). The extracts with the highest (P < 0.05) antioxidant activity were those obtained from rosemary (OR, WR, OWR). There were no signicant dierences (P > 0.05) between the oil extract of rosemary (OR) and the water miscible rosemary extract (WR), which showed the highest SI. However, the oil and water miscible rosemary extract (OWR) had a lower SI.

Table 3 Inhibition eects of natural extracts tested against spoilage bacteria Spoilage bacterial strains Diameter of the zones of inhibition in mm (6 mm well) OR Br. thermosphacta CRA 7883 Br. thermosphacta CRA 7884 Br. thermosphacta CRA 3235 L. innocua 4202 L. monocytogenes 5105 Lb. sake 550 Lc. mesenteroides subsp mesenteroides 824 Lc. mesenteroides subsp dextranicum 882 Lb. carnosum 558 Lb. curvatus 860 Lb. lactis FMRD 492 , no inhibition. 28.1 25.5 26.4 20.5 25 21.4 21 21 26.2 23.1 21.8 WR 19.5 19.5 24 17.4 19.5 13.4 16.6 15.4 23 16.2 18.2 OWR 23.8 23.8 25.1 14.2 15.4 11.19 13.8 14.2 14.3 15.4 13 GR 8.5 10.5 10.5 OF 7.0 7.5 LF 8.5

376

ndez-Lo pez et al. / Meat Science 69 (2005) 371380 J. Ferna

dierences in the antibacterial activity between the rosemary extracts studied (OR, WR, OWR), if we assume that the oil extract (OR) is richer in nonpolar phenolic compounds than the others. Frazier (1980) reported that citrus bioavonoids also had antimicrobial properties. These compounds have reportedly wide-ranging antimicrobial properties eective against a broad range of human pathogens, fungi and food spoilage organisms (Cho, Seo, Choi, & Joo, 1990; Harich, 1997; Morgan, Andersen, & Hankinson, 1971). 3.3. Antibacterial and antioxidant properties of natural extracts in meatballs The eect of the natural extracts on lipid oxidation of cooked meatballs during storage is shown in Fig. 1. The analysis of variance for the TBARS data indicates that the TBA values were signicantly aected (P < 0.05) by both the storage period and the extract treatments. Initial (day 1) TBA values for all extract samples were signicantly lower than those for the control (P < 0.05). These results suggest that these antioxidants retarded lipid oxidation during and immediately after cooking. These results agree with that reported by ndez-Lo pez Ahn, Gru n, and Fernando (2002) and Ferna et al. (2003) for other natural antioxidants applied to cooked beef. Sato and Hegarty (1971) reported that non-heme iron was the active catalyst in cooked meats. Chen et al. (1984) also reported that iron was released from heme pigments during cooking and proposed that the resultant increase in non-heme iron was responsible for lipid oxidation. At the end of storage time (day 12) all treatments resulted in signicantly lower (P < 0.05) TBA values when compared to the control, which indicates that all the tested natural extracts added to meatballs showed antioxidant properties. The product samples with rosemary
4 TBA (mg MA/kg sample ) 3,5 3 2,5 2 1,5 1 1
Control OR

extracts (OR, WR, OWR) showed the lowest (P < 0.05) TBA values at each time of storage. Only treatments with orange extracts (OF) and rosemary water miscible extracts (WR) maintained (P > 0.05) the initial TBA values during the 12 day storage period, and no dierences were found between them (P < 0.05). Samples treated with rosemary oil and water miscible extracts (OWR) had slightly increased TBA values only during the rst 6 days of storage and became stable after that period. The products with added lemon extracts (LF) reached higher (P < 0.05) TBA values than those with orange extracts (OF) by the end of storage. Consistently, this difference in antioxidant properties between orange and lemon extracts (OF and LF) has also been detected by using the Rancimat method, as explained before. All colour coordinates showed dierences (P < 0.05) between treatments and storage days, except yellowness which only showed dierences between treatments (Table 4). In all samples, lightness increased with storage time (P < 0.05), and the highest values of L* were obtained in control samples. Some authors reported that this increase could be related to the increase in MMb formation (Anton et al., 1993; Genot et al., 1991). These results suggests that the presence of antioxidant compounds in the natural extracts could retard MMb formation in meatballs and so L* values decreased. For this reason, it could be expected that treatments with orange extracts (OF) and rosemary water miscible extracts (WR) would have lowest lightness values because they had the highest antioxidant capacity (Table 2), but in this study, treatments with lemon extracts (LF) and orange extracts (OF) showed the lowest L* values. This could be explained by the increased water retention associated with hygroscopic materials, and because these extracts were prepared as a dry powder, they have absorbed free water within the product, subsequently decreasing lightness values. This relation between free water and lightness in meat and meat products has been ndez-Lo pez, Pe rezreported by several authors (Ferna , 2000; Kauman, Warner, Alvarez, & Aranda-Catala & Russell, 1991).
Table 4 Eect of natural extracts on colour parameters of meatballs during storage at 8 C Treatment Control OR WR OWR LF OF
a,b

L day 1 42.26ax 42.32ax 42.48ax 42.25ax 39.46bx 39.28bx

L day 12 46.97ay 44.73by 44.61by 44.89by 42.08cy 41.57cy

a day 1 9.52ax 9.53ax 9.39ax 9.33ax 10.22bx 10.37bx

a day 12 7.13ay 8.33by 8.42by 8.45by 8.83by 8.92by

b day 1 8.88ax 8.78ax 8.50ax 9.11ax 10.30bx 10.29bx

b day 12 9.14ax 9.07ax 8.78ax 9.22ax 10.34ax 10.33ax

6
WR OWR

9
LF OF

12 Time (days)

Fig. 1. Rancidity (TBA) evolution in meatballs with dierent natural extracts added, during storage time. For sample denomination see Table 1.

For sample denomination see Table 1. Means within a column with dierent letters are signicantly different (P < 0.05). x,y Means within a row with dierent letters are signicantly dierent (P < 0.05).

 ndez-Lo pez et al. / Meat Science 69 (2005) 371380 J. Ferna

377

In all samples redness decreased as the storage time progressed (P < 0.05) but red colour (a* values) of the control sample faded very rapidly. Phillips, Mancini, Sun, Lynch, and Faustman (2001) reported a decrease in a* values of cooked ground beef linked to storage time. This is not surprising as meat which has been stored longer would be expected to have predominantly either OMb or MMb, as opposed to deoxymyoglobin (DMb), which in turn would predispose the meat to a faster browning rate (Hunt, Sorheim, & Slinde, 1995, 1999). At the end of storage (day 12) a* values of the control samples were lower (P < 0.05) than day 1, and lower that of antioxidant treatments. Several authors have studied the eect of dierent antioxidants on the colour of meat and meat products (Higgins, Kerry, Buckley, & Morrisey, 1998; Lee, Hendricks, & Cornforth, 1998) and have reported that meat oxidation decreases a* values. Therefore, it was expected that samples with citrus extracts (LF and OF), which were the treatments with lower antioxidant activity (Table 2), would have the lowest a* values. Probably, this decrease in redness associated with the oxidation process is being compensated by the presence of red component pigments such as carotene in these citrus extracts. To conrm this, a signicant eect of carotenes increasing a* values by about one unit was observed after preparation of the samples (day 1) (Table 4). If adjustments were made by subtracting the contribution of added pigments, the results would be consistent with previous observations. In all samples yellowness values were not modied (P < 0.05) by storage time. Therefore, the dierences in b* values observed between treatments incorporating citrus extracts (LF, OF) and the others, can be attributed to the presence of pigments in the citrus extracts and not to the oxidation processes. In this work, the presence of coliforms, moulds and yeasts, and psychrotrophic microorganisms was not detected in any cooked meatball samples, regardless of storage time. This product was prepared following safe practices resembling commercial aseptic conditions, with a heat treatment that guaranteed pasteurization eectively inactivating vegetative cells and then stored under conditions that would limit the growth bacterial groups. Cross contamination from the environment (i.e. airborne or food handlers) or the survival of spores or resistant cells was possible, as it is in commercial operations. Lactic acid bacteria were detected at low levels in the control and samples with rosemary water and oil miscible extracts (OWR) from day 1 (data not shown). Some spore formers and heat resistant strains which have been linked with spoilage of meats (including VP and MAP) are likely to contribute to LAB counts (Borch, Kant-Muermans, & Blixt, 1996; Kuri, 1998). After 12 days storage, the growth of LAB to levels of 103 log cfu/g was similar (P > 0.05) for control

samples and treatments with rosemary extracts (OR, WR and OWR) (Table 5). These bacterial groups were not detected in samples from any treatment with citrus extracts (LF and OF) during storage time. Despite the presence of lactic acid bacteria, there was no evidence of strong lactic fermentation in any product, as conrmed by very low (<15 mM) lactic acid as determined by HPLC (data not shown), which also indicated the absence of signicant amounts of sugars. Therefore, some bacteria may be present but their growth on the product is controlled under storage conditions. While there was in vitro antilisterial activity when the rosemary extracts (OR, WR, OWR) were tested using the agar diusion test, this was not clear when used in the meatballs at the concentrations tested. Pandit and Shelef (1994) found rosemary extracts to have an eect against L. monocytogenes growth in a pork liver sausage. As can be seen, the antibacterial activity obtained for rosemary extracts (OR, WR, OWR) (Table 3) have not been manifested during the storage of cooked meatballs with added extracts. This could be explained by the dilution of the rosemary extracts necessary for its use in meat products, but there is also a possibility of reduction of the eectiveness of the antimicrobial extract due to physical interactions with the food matrix, which has been observed before for nisin in meat (Davies et al., 1999; Kuri, Collins, & Madden, 1998). Doyle (1999) notes a reduced eect of plant extracts related to fat in meat. Also, it is important to observe that citrus extracts (LF and OF) which showed the lowest antibacterial activity (Table 3) in the previous assay, appear to be more eective than the others to control LAB growth during the storage of meatballs. This is likely to be related with the lowered water activity within the product that would be the result of adding the citrus fruit preparations as a dry powder which contain bre with high water absorption, as reported by Lario et al. (2003). Reduced water activity could reduce microbial growth but the presence of fermentable hydrocolloids in meat products may result in the development of o avours (Kuri, Ponce, & Guerrero, 1994).

Table 5 Lactic acid bacteria (LAB) counts (log cfu/g) in meatballs with added natural extracts, during storage time Time (days) 1 6 9 12 Control 2.00ax 2.69bx 2.77bx 3.30cx OR n.d.ay 2.84bx 3.00bx 3.00bx WR n.d.ay n.d.ay 2.84bx 3.30cx OWR 2.30ax 2.69ax 2.69ax 3.30bx LF n.d.ay n.d.ay n.d.ay n.d.ay OF n.d.ay n.d.ay n.d.ay n.d.ay

For sample denomination see Table 1. n.d.: not detected (<1.69; detection limit). ac Means within a column with dierent letters are signicantly different (P < 0.05). x,y Means within a row with dierent letters are signicantly dierent (P < 0.05).

378

ndez-Lo pez et al. / Meat Science 69 (2005) 371380 J. Ferna

From the eight sensory attributes evaluated, only three (aroma rancidity, surface lime and overall acceptance) provided useful information on the changes due to treatment and storage day, while for brightness, citrus aroma, putrefaction and sour aroma no dierences (P > 0.05) were detected for any of the factors. The products were fairly stable for a week, with some dierences becoming apparent between day 6 and day 9. In Fig. 2 it can be observed that at the end of the storage period (day 12), the dierences in rancidity perception between treatments (P < 0.05) are consistent with dierences measured as TBA values (Fig. 1). One exception is the lower score for orange extracts (OF) in comparison to lemon extracts (LF). This correlation between the sensory and the chemical methods is less evident at day 6, because at this point, changes in TBA values seem to show dierences before panellist could perceive the increased rancidity. While these results suggest that the panellist may not be sensitive enough to detect dierences of TBA values below a threshold, sensory scores provided better discrimination at day 9 in comparison to TBA values. Interestingly, samples with citrus extracts (LF, OF) were scored at equivalent terminal levels from day 9. Another factor to consider is the presence of

o avours in lemon extracts (LF), which could contribute to perceived rancidity, but are not identiable by the TBA reaction. Overall acceptability scores for each treatment (Fig. 2) showed an inverse relation with TBA values and rancidity perception (Figs. 1 and 2, respectively). Product surface slime was not detected in any treatment during the rst days of storage, but in control samples it was detected at day 9, increasing by day 12, when it was also detected in samples from treatments with rosemary extracts (OR, WR and OWR). No perception of surface slime was detected in citrus treatments (OF, LF) during storage time (results not shown). These observations agree with LAB enumeration results (Table 5). The only attribute that showed dierences (P < 0.05) between treatments but not between storage days was aroma o odour, which suggests that it was mainly due to the type of extract added and not to the deterioration of meatballs during storage. So, this o odour perception can be attributed to the peculiar composition of each extract, which is detected at its incorporation and was not modied during storage. The highest (P < 0.05) scores for o odours were obtained for citrus treatments (LF, OF), the lowest for control samples and there were not dierences (P > 0.05) in o odours perception between rosemary treatments (OR, WR, OWR). Therefore, it can be observed that this o odour perception may not to have any inuence upon the overall acceptability of the products.

Rancidity (aroma)

4. Conclusions The use of rosemary extracts has proved to be eective as an antioxidant in cooked swedish-style meatballs. The application of rosemary and citrus extracts improved the acceptability of the product. The application of rosemary preparations could be useful to control the development of rancidity and o avours, while orange and lemon serve the same purpose, but may have some additional eects such as water binding that need to be managed. Some antibacterial activity was detected, but additional measures would be needed to control spoilage. The eect of antibacterial interaction with fat and the physical availability and compatibility of the fatty and lean phases and the active compounds need to be established in order to control better the deterioration of meat products.

1 1 6 9

(a)
7

Control

OR

WR

OW R

LF

OF

12 Time (days)

Overall acceptability

1 1 6 9 12 T ime (days)

Acknowledgement Funding for work at the University of Plymouth was enabled by a grant by the Seale Hayne Educational Trust. JFL visit to U.P. was funded by the Generalitat

(b)

Control

OR

WR

OW R

LF

OF

Fig. 2. Evolution of sensory scores for perception of rancidity aroma (a) and overall acceptability (b) of meatballs during storage time. For sample denomination see Table 1.

 ndez-Lo pez et al. / Meat Science 69 (2005) 371380 J. Ferna

379

Valenciana and LACS was within a Socrates exchange programme. Linda Thomas (Aplin & Barrett Ltd., Danisco-Cultor, Beaminster, UK) kindly provided some of the bacterial strains. References
A, A. A., & Azen, S. P. (1979). Statistical analysis. A computer oriented approach. London: Academic. Ahn, J., Gru n, I. U., & Fernando, L. N. (2002). Antioxidant properties of natural plant extracts containing polyphenolic compounds in cooked ground beef. Journal of Food Science, 67, 13641369. ` tude des relations Anton, M., Salgues, C., & Renerre, M. (1993). E oxidatives entre les lipides membranaires et la myoglobine in vitro. Science Alimentarie, 13, 261274. Borch, E., Kant-Muermans, M. L., & Blixt, Y. (1996). Bacterial spoilage of meat and cured meat products. International Journal of Food Microbiology, 33, 103120. Borch, E., & Molin, G. (1988). Numerical taxonomy of psychrotrophic lactic acid bacteria from prepacked meat and meat products. Antonie van Leeuwenhoek, 54, 301323. British Standards BS 5929: Part 4. (1986a). British standard methods for sensory analysis of foods. Part 4. Flavour prole methods. British Standards Institution, London. British Standards BS 5929: Part 1. (1986b). British standard methods for sensory analysis of foods. Part 1. General guide to methodology. British Standards Institution, London. Chastain, M. F., Hufman, D. L., Hsieh, W. H., & Cordray, J. C. (1982). Antioxidants in restructured beef/pork steaks. Journal of Food Science, 47, 17791782. Chen, H. M., Muramoto, K., Yamauchi, F., & Huang, C. L. (1996). Natural antioxidants from rosemary and sage. Journal of Food Science, 42, 11021104. Chen, C. C., Pearson, A. M., Gray, J. I., Fooladi, M. H, & Ku, P. (1984). Some factors inuencing the nonheme iron content of meat and its implications in oxidation. Journal of Food Science, 49, 581584. Cho, S. H., Seo, I. W., Choi, J. D., & Joo, I. S. (1990). Antimicrobial and antioxidant activity of grapefruit seed extract on shery products. Bull Korean Fish Society, 23, 289296. Cuvelier, M. E., Richard, H., & Berset, C. (1996). Antioxidative activity and phenolic composition of pilot-plant and commercial extracts of sage and rosemary. Journal of the American Oil Chemists Society, 73, 645652. Daeschel, M. A. (1992). Procedures to detect antimicrobial activities of microorganisms. In B. Ray & M. A. Daeschel (Eds.), Food preservatives of microbial origin (pp. 155176). Boca Raton, FL: CRC Press. Davidson, P. M. (1993). Parabens and phenolic compounds. In P. M. Davidson & A. L. Branen (Eds.), Antimicrobials in foods (pp. 263306). New York: Marcel Dekker. Davies, E. A., Milne, C. F., Bevis, H. E., Potter, R. W., Harris, J. M., Williams, G. C., et al. (1999). Eective use of nisin to control lactic acid bacterial spoilage in vacuum-packed bologna-type sausages. Journal of Food Protection, 62(9), 10041010. Dean, S. G., & Svoboda, K. P. (1989). Antimicrobial activity of summer savory (Satureja hortensis L.) essential oil and its constituents. Journal of Horticultural Science, 64, 205210. Decker, E. A., Chan, W. K. M., Livisay, S. A., Buttereld, D. A., & Faustman, C. (1995). Interactions between carnosine and the dierent redox states of myoglobin. Journal of Food Science, 60, 12011204. Del Campo, J., Amiot, M. J., & Nguyen-The, C. (2000). Antimicrobial eect of rosemary extracts. Journal of Food Protection, 63, 13591368.

Doyle, M. E. (1999). Use of other preservatives to control Listeria in meat. American Meat Institute. Elangovan, V., Sekar, N., & Govindasamy, S. (1994). Chemoprotective potential of dietary bipoavonoids against 20-methylcholanthrene-induced tumorigenesis. Cancer Letters, 87, 107113. Farag, R. S., Daw, Z. Y., Hewedi, F. M., & El-Baroty, G. S. A. (1989). Antimicrobial activity of some Egyptian spice essential oils. Journal of Food Protection, 52, 665667. ndez-Gine s, J. M., Ferna ndez-Lo pez, J., Sayas-Barbera , E., Ferna rez-Alvarez, J. A. (2003). Eects of storage Sendra, E., & Pe conditions on quality characteristics of bologna sausages made with citrus ber. Journal of Food Science, 68, 710715. ndez-Lo pez, J., Ferna ndez-Gine s, J. M., Aleson-Carbonell, L., Ferna , E., Sendra, E., & Pe rez-Alvarez, J. A. (2004). Sayas-Barbera Functional compounds from citrus by-products and their application in meat products. Trends in Food Science and Technology, 15, 176185. ndez-Lo pez, J., Pe rez-Alvarez, J. A., & Aranda-Catala , V. Ferna (2000). Eect of mincing degree on colour properties in pork meat. Color Research & Application, 25, 376380. ndez-Lo pez, J., Sevilla, L., Sayas-Barbera , M. E., Navarro, C., Ferna n, F., & Pe rez-Alvarez, J. A. (2003). Evaluation of the Mar antioxidant potential of hyssop (Hyssopus ocinalis L.) and rosemary (Rosmarinus ocinalis L.) extract in cooked pork meat. Journal of Food Science, 68, 660664. Frazier, S. F. (1980). Antimicrobial composition of matter from naturally occurring avonoid glycosides, US Pat. 4,238,483. Genot, C., Borel, M. N., Metro, B., Gandemer, G., & Renerre, M. (1991). Enhancement of myoglobin autoxidation induced by phospholipids extracted from beef muscles of various metabolic types. In Proceedings of the 37th International Congress of Meat Science and Technology (pp. 356359), 16 September, Kulmbach, Germany. Gerhardt, P., Murray, R. G. E., Costilow, R. N., Nester, E. W., Wood, W. A., & Phillips, G. B. (1994). Manual of methods for general bacteriology. Washington: American Society for Microbiology. Harich, J. (1997). Antimicrobial grapefruits extracts. US Pat. 5,631,001. Harris, J. C., Cottrell, S. L., & Plummer, S. (2001). Antimicrobial properties of Allium sativum (garlic). Applied Microbiological Biotechnology, 57, 282286. Higgins, F. M., Kerry, J. P., Buckley, D. J., & Morrisey, P. A. (1998). Eect of dietary a-tocopheryl acetate supplementation on a-tocopherol distribution in raw turkey muscles and its eect on the storage stability of cooked turkey meat. Meat Science, 37, 373383. Houlihan, C. M., Ho, C. T., & Chang, S. S. (1984). Elucidation of the chemical structure of a novel antioxidant, rosmaridiphenol, isolated from rosemary. Journal of the American Oil Chemists Society, 61, 10361039. Houlihan, C. M., Ho, C. T., & Chang, S. S. (1985). The structure of rosmariquinone- A new antioxidant isolated from Rosmarinus ocinalis L. Journal of the American Oil Chemists Society, 62, 9698. Hunt, M. C., Acton, J. C., Benedict, R. C., Calkins, C. R., Cornforth, D. P., Jeremiah, L. E., et al. (1991). Guidelines for meat color evaluation. Chicago: American Meat Science Association and National Live Stock and Meat Board. Hunt, M. C., Sorheim, O., & Slinde, E. (1995). Eects of myoglobin form on internal cooked color development in ground beef. In Proceedings of the 41st International Congress of Meat Science Technology (pp. 394395), 2025 August, San Antonio, USA. Hunt, M. C., Sorheim, O., & Slinde, E. (1999). Color and heat denaturation of myoglobin forms in ground beef. Journal of Food Science, 64, 847851. IFST. (1993). Shelf life of foods Guidelines for its determination and prediction. London: Institute of Food Science and Technology.

380

ndez-Lo pez et al. / Meat Science 69 (2005) 371380 J. Ferna Mouly, P. P., Arzouyan, C. R., Gaydou, E. M., & Estienne, J. M. (1994). Dierentiation of citrus juices by factorial discriminant analysis using liquid chromatography of avonone glycosides. Journal of Agricultural and Food Chemistry, 42, 7079. Nakatani, N., & Inatani, R. (1984). Two antioxidant diterpenes from rosemary (Rosmarinus ocinalis L.) and a new revised structure for rosmanol. Agricultural Biological Chemistry, 48, 20812085. NCCLS. Global Consensus Standardization for Health Technologies (1990). Quality assurance for commercially prepared microbiological culture media. Approved standard document M22-A, Vol. 10, No. 14. Nuutila, A. M., Puupponen-Pimia , R., Aarni, M., & OksmanCaldentey, K. M. (2003). Comparison of antioxidant activities of onion and garlic extracts by inhibition of lipid peroxidation and radical scavenging activity. Food Chemistry, 81, 485493. Pandit, V. A., & Shelef, L. A. (1994). Sensitivity of Listeria monocytogenes to rosemary (Rosmarinus ocinalis L). Food Microbiology, 11, 5763. Phillips, A. L., Mancini, R., Sun, Q., Lynch, M. P., & Faustman, C. (2001). Eect of erythorvic acid on cooked colour in ground beef. Meat Science, 57, 3134. Pokorny, J. (1991). Natural antioxidant for food use. Trends in Food Science and Technology, September, 223226. Renerre, M., Anton, M., & Gatellier, P. (1992). Autoxidation of puried myoglobin from two bovine muscles. Meat Science, 32, 331342. Sato, K., & Hegarty, G. R. (1971). Warmed over avor in cooked meats. Journal of Food Science, 36, 10981102. Schieber, A., Stintzing, F. C., & Carle, R. (2001). By-products of plant food processing as a source of functional compounds recent developments. Trends in Food Science and Technology, 12, 401413. Schwarz, K., Bertelsen, G., Nissen, L. R., Gardner, P. T., Heinonen, M. I., Hopia, A., et al. (2001). Investigation of plant extracts for the protection of processed foods against lipid oxidation. Comparison of antioxidant assays based on radical scavenging, lipid oxidation and analysis of the principal antioxidant compounds. European Food Research Technology, 212, 319328. Shahidi, F. (2000). Natural phenolic antioxidants and their food applications. Lipid Technology, 12, 8084. Tarladgis, B. G., Watts, B. M., & Younathan, M. T. (1960). A distillation method for the quantitative determination of malonaldehyde in rancid foods. Journal of the American Oil Chemists Society, 37, 4448. Tsao, S. M., & Yin, M. C. (2001). In-vitro antimicrobial activity of 4 dialyl sulphides occurring naturally in garlic and Chinese leek oils. Journal of Medical Microbiology, 50, 646649. Yin, M. C., & Cheng, W. (1998). Antioxidant activity of several Allium members. Journal of Agricultural and Food Chemistry, 46, 40974101. Yin, M. C., & Faustman, C. (1993). Inuence of temperature, pH and phospholipid composition upon the stability of myoglobin and phospholipid: a liposome model. Journal of Agricultural and Food Chemistry, 41, 853857.

Karamanoli, K., Vokou, D., Menkissoglu, U., & Constantinidou, H. I. (2000). Bacterial colonization of phyllosphere of mediterranean aromatic plants. Journal of Chemical Ecology, 26, 20352048. Kauman, R. G., Warner, R. D., & Russell, R. L. (1991). Variation in lightness and exudation among ten porcine muscles. In Proceedings of the 37th International Congress of Meat Science Technology (pp. 135138), Kulmbach, Germany. Kim, S. M., Kubota, K., & Kobayashi, A. (1997). Antioxidative activity of sulphur-containing avour compounds in garlic. Bioscience Biotechnology Biochemistry, 61, 14821485. Kuri, V. (1998). Lactic Acid Bacteria and Salmonellae from Mexican pork products: Characterization and antagonism. Ph.D. Thesis. The Queens University of Belfast. Kuri, V., Collins, M. A., & Madden, R. H. (1998). Stability and antibacterial activity of bacteriocin preparations in pork model. In Proceedings of the 44th International Congress of Meat Science and Technology (pp. 342343), 30 August to 4 September 1998, Barcelona, Spain. Kuri, V., Ponce, E., & Guerrero, I. (1994). Physicochemical characteristics of fermented sausage extended with carbohydrate materials. Journal of Muscle Foods, 5, 119135. Lampe, J. W. (1999). Health eects of vegetables and fruit: assessing mechanisms of action in human experimental studies. American Journal of Clinical Nutrition, 70, 475S490. a, J., Sayas-Barbera , E., Ferna ndez-Lo pez, Lario, Y., Sendra, E., Garc rez-Alvarez, J. A. (2003). Preparation of high dietary ber J., & Pe powder from lemon juice by-products. In New Functional Ingredients and Foods Abstract Book (P1-G20), 911 April 2003, Copenhagen, Denmark. La ubli, M. W., & Bruttel, P. A. (1986). Determination of the oxidative stability of fats and oils: comparison between the Active Oxygen Method (AOCS Cd 1257) and the Rancimat Method. Journal of the American Oil Chemists Society, 63, 792795. Lawson, L. D. (1998). Garlic: a review of its medicinal eects and indicated active compounds. In L.D. Lawson, R. Bauer (Eds.), Phytomedicines of Europe: chemistry and biological activity (ACS Symposium series No. 691 (pp. 176209). Washington, DC: American Chemical Society. Lee, B. J., Hendricks, D. G., & Cornforth, D. P. (1998). Antioxidant eect of carnosine and phytic acid in a model beef system. Journal of Food Science, 63, 394398. Lo liger, J. (1991). The use of antioxidants in foods. In I. O. Aruoma & B. Halliwell (Eds.), Free radical and food additives (pp. 121131). London: Taylor & Francis. n, F. R., Frutos, M. J., Pe rez-Alvarez, J. A., Mart nez-Sa nchez, Mart o, J. A. (2002). Flavonoids as nutraceuticals: structural F., & Del R related antioxidant properties and their role on ascorbic acid preservation. In Atta-Ur-Rahman (Ed.), Studies in natural products chemistry (Vol. 26, pp. 324389). Amsterdam: Elsevier Science. Morgan, D. R., Andersen, D. L., & Hankinson, C. L. (1971). Process of inhibiting staling of milk prior to sterilization, US Pat. 3,615,717.