Résumé

Lfrik marocain traditionnel est produit par fermentation spontanée du lait camelin cru et entier à la température ambiante. Dans ce travail, Lfrik pasteurisé a été élaboré en utilisant du lait camelin pasteurisé et inoculé avec des souches de Lactococcus et de Leuconostoc obtenues à partir des levains lactiques du commerce (Flora Danica et CHN11) selon les caractéristiques du lait fermenté. L’évolution de quelques caractéristiques biochimiques et microbiologiques au cours de la fermentation a été étudiée. L’acidification du lait camelin fermenté et du lait bovin a été comparée dans huit échantillons pour chacun. La fermentation était plus lente dans le lait camelin que dans le lait bovin dont l’acidité était deux fois plus élevée. Cependant, l’augmentation de l’azote non protéique a été plus rapide pendant la fermentation du lait camelin et la concentration en acide benzoïque naturel après fermentation a été plus élevée dans le lait de chamelle. De plus, la qualité organoleptique de Lfrik pasteurisé a été testée et s’est révélée similaire à celle de Lfrik traditionnel.


Mots-clés: Lfrik, lait de chamelle, caractéristiques, fermentation contrôlée, qualité sensorielle

INTRODUCTION

Milk fermentation is a process that extends its shelf life, improves its taste and enhances its digestibility (Shiby and Mishra, 2013).

In the traditional fermented milk process, raw milk is allowed to ferment naturally at ambient temperature and without prior heat treatment until it turns sour. Due to the spontaneous nature of this fermentation, this traditional method results in a product with varying taste and flavor (Farah et al., 1989). Thus, nowadays milk fermentation is a carefully controlled microbial process for which selected cultures have been developed (Shiby and Mishra, 2013). These starter cultures have been formulated by using micro-organisms which impart special desired characteristics to the fermented product (Marshall and Law, 1984). In addition, they must be able to multiply in a few hours in order to produce enough lactic acid and aroma compounds for the complete conversion of milk to fermented dairy products, that depends not only on nutritional factors but also on other environmental parameters such as temperature (Breheny et al., 1975; Lawrence and Tomas, 1979; Ross 1980) and associative growth (Driessen et al., 1982). In the case of camel milk, few studies were previously published worldwide on its controlled fermentation process (Farah et al., 1990; Attia, 2001; Abdel Rahman, 2009).

In Morocco, fermented camel milk “Lfrik ” is traditionally made by the herders and/or small dairy shops “Mahlabas” using raw camel milk without prior heat treatment and allowing it to spontaneously ferment in goat skin bags. In a recent study, our team reported that raw camel milk was contaminated with high levels of coliforms bacteria presenting, therefore, a potential health risk for its consumers (Ismaili et al., 2016). To our knowledge, no previous published work was done on controlled fermentation of camel milk produced in Morocco. Therefore, the purpose of the present study was to produce a fermented product similar to traditional Lfrik but with improved hygienic quality using pasteurized camel milk and selected commercial starter cultures. Biochemical and microbiological properties during fermentation and in fermented products as well as the sensory characteristics were studied and compared to those of fermented cow milk produced in the same conditions.

MATERIAL AND METHODS

Milk and Lfrik samples

Eight samples of raw camel milk, raw cow milk and traditional Lfrik were obtained from traditional dairy shops in Laayoune city and at private farm located in Ain Aouda, close to Rabat. Milk samples were transported refrigerated in coolers to laboratory and kept refrigerated until use for analysis, within 1 hour to ONSSA (Office Nationale de Sécurité Sanitaire des Aliments) laboratory in Laayoune and within 1 day to IAV laboratory in Rabat.

Upon arrival, raw milk samples and Lfrik were analyzed for their pH, acidity and total aerobic flora (TFA).

Pasteurization and fermentation

Raw camel milk and cow milk samples were batch pasteurized at 90°C for 15 min in a water bath and immediately cooled to about 30°C.

Pasteurized camel milk samples were then directly inoculated using one of two commercial freeze-dried mesophilic lactic cultures (Flora Danica and CHN-11, CHR Hansen, Denmark). The inoculation rate was as suggested by the producer of the culture package. The two commercial cultures are composed of a mixture of Leuconostoc, Lactococcus lactis biovar diacetylactis and Lactococcus lactis subsp. cremoris.

Incubation of inoculated samples was at 20○C or 30○C for 24 h. During the fermentation process, monitoring of pH and titratable acidity (TA) was carried out every 3 hours and TFA every 6 hours. Samples inoculated with “Flora Danica” were used to monitor content changes of lactose, benzoic acid and Non Protein Nitrogen (NPN) at different intervals during incubation.

Pasteurized cow milk samples were inoculated with “Flora Danica” under the same conditions as camel milk. These samples were used for comparison purpose.

Biochemical and microbial Analyses

The pH of milk samples was measured using a digital pH meter (CyberScan pH 1500, Eutech Instruments) and their TA was determined based on the AOAC method (1990).

TFA counts were determined using PCA medium (Plate Count Agar, Difco laboratories, Detroit, Michigan, USA) incubated at 30°C for 3 days (IDF, 1987a).

The AOAC method 991.21(AOAC, 2010) as described by Barbano et al. (1991) was used to determine NPN content in fermented milk samples. Protein precipitation was done by the addition of trichloroacetic acid (TCA) solution. Precipitated milk protein was removed by filtration and NPN in the filtrate was determined by Kjeldahl method.

Benzoic acid extraction and quantification were done according to the IDF method (1987b). A volume of 20 μl of extracts and standard solution were injected in a High-Performance Liquid Chromatography (HPLC) system with a UV detector (Agilent 1100 Series, Germany). A C18 column (250 x 4.6 mm, i.d.) with 5 microns particle size (IBM Instrument Inc.) was used. The mobile phase was composed of 20 % acetonitrile and 80 % sodium acetate solution (0.005 M adjusted to pH 4.4 with acetic acid). The flow rate of the mobile phase was set at 0.8 ml/min and the benzoic acid detection at 227 nm.

Determination of lactose in camel and cow milk samples at different intervals during the fermentation process was done by HPLC according to the method reported by Xinmin et al. (2008). Separation was performed using water–acetonitrile (20/80, v/v) mobile phase maintained at a flow rate of 1 ml/min.

Sensory evaluation

Sensory evaluation of fermented milk samples was done in two trials. A first trial was conducted in Laayoune city using a panel of 63 consumers of traditional Lfrik and a set of 4 samples: fermented camel milk with “Flora Danica”, Fermented camel milk with CHN-11, fermented cow milk with “Flora Danica” and traditional Lfrik as a control. The Lfrik sample was purchased the same day of the trial from a “Mahlaba”.

The second trial was carried out in the city of Rabat with a panel of 52 non-consumers of camel milk and Lfrik. Samples used in this trial were: Fermented camel milk with “Flora Danica”, fermented camel milk with “CHN-11” and fermented cow milk with “Flora Danica”.

The fermented milk samples used in both trials were incubated at room temperature (18 ± 2°C) for about 12 h. In both trials, Samples were coded with three digit numbers and served at room temperature (20 ± 2°C) in white plastic cups. The order of samples presentation was balanced to account for first order and carry-defects.

Hedonic evaluation of samples was done using a 9 points balanced verbal hedonic scale (dislike extremely to like extremely) (Meilgaard et al., 1999). The panelists were also asked to rate the intensity of acidity, bitterness, viscosity and fat sensation in fermented camel and cow milk samples using a 5 points rating scale (very low “1”, low “2”, medium “3”, high “4”, very high “5”). Mineral water, at ambient temperature, was provided to panelists for rinsing out the mouth between samples.

The instructions given to participants during the sensory evaluation session were done in writing and orally in Arabic and French.

Statistical analyses

Statistical analysis of data was performed with Minitab program using analysis of variance (ANOVA) and statistically significant differences between means were determined using Tukey test at a level of significance α = 0.05.

The analysis of overall sensory liking and sensory attributes intensity scores was performed with excel using XLSTAT statistical software.

RESULTS AND DISCUSSION

Milk samples fermentation monitoring

Figures 1 and 2 show the pH and TA values at different intervals during the incubation of inoculated camel milk and cow milk samples at 20 and 30°C. At both temperatures, camel milk samples inoculated with “Flora Danica” showed faster pH decrease and higher increase in TA values compared to those inoculated with “CHN-11”. In addition, the fermentation rate was faster and more intense in inoculated cow milk than camel milk samples. In cow milk samples, TA values after 12 h incubation were approximately twice higher than those of camel milk either at 20°C or 30°C. TA values reached respectively 0.81 and 0.86 % lactic acid at 20○C and 30○C while that of camel milk inoculated either with “Flora Danica” or “CHN-11” did not exceed 0.42 % lactic acid at 20○C and 0.51 % lactic acid at 30○C. Similar findings were reported by Attia et al. (2001).

This slow camel milk fermentation rate can be explained by the presence of natural microbial inhibitors (lysosyme, lactoferrin, lactoperoxydase and immunoglobulins) that are in higher levels in camel milk than in cow milk (Yagil, 1982; Barbour et al., 1984; El-Agamy et al., 1992; Attia et al., 2001). This is supported by the microbial growth rate observed during the incubation of milk samples (Figure 3). After 12 h of incubation, TFA in fermented cow milk reached values of 1.0 108 Cfu.mL-1, while those of fermented camel milk were 1.0 103 Cfu.mL-1.

In addition, more lactose use during the fermentation process was observed in cow milk than in camel milk samples. After 12 h incubation at 20°C, lactose in cow and camel products were respectively about 80 % and 60 % less than initial values (Figure 4).

Evolution of NPN and benzoic acid contents during fermentation

Evolution during incubation of NPN and benzoic acid contents were carried out in camel and cow milks inoculated with the lactic starter “Flora Danica” at 20 and 30○C.

The increase of NPN content during fermentation in camel milk was faster than in cow milk (Figure 5a). After 12 h incubation, NPN content of fermented camel milk was 1.74 and 1.94 mg.Kg-1 at 20 and 30°C, respectively. However, NPN content in cow milk after 12 h incubation was around 1.0 mg.Kg-1. The observed larger content of NPN in fermented camel milk than in fermented cow milk is probably due to more proteolytic activity in camel milk. Several authors explained this increase in NPN during milk fermentation by an increase in free amino acids (IDF, 1983; Alm, 1983; Saidi and Warthesen, 1993).

Benzoic acid production during fermentation was low and had identical trend in milk samples throughout the fermentation up to 9 h and then increased significantly from about 2 ppm to reach respectively values of 11 and 15 ppm in cow and camel milk samples at 12 h of fermentation (Figure 5b). Similar results of benzoic acid of fermented cow milk were obtained by Nishimoto et al. (1969) and Saidi and Faid (2001) who reported respectively an average of about 11 ppm and a range of 5-18. It should be noticed that benzoic acid plays an important role as an antifungal agent in fermented products (Chipley, 1993; Davidson, 2001).

Sensory evaluation

In the first sensory trial carried out in the city of Laayoune and where a panel of regular consumers of Lfrik was used, overall liking results (Table 1) showed that the fermented camel milk samples prepared using the commercial starters had similar scores (p>0.05) as the control (traditional fermented camel milk). Average scores in a scale of 9 points were respectively 7.6, 7.5 and 7.2 for Lfrik, camel milk fermented with “Flora Danica” and camel milk fermented with “CHN-11”. In this trial, cow milk fermented with “Flora Danica” had the lowest overall liking score and was significantly less liked (p<0.05) than the fermented camel milk.

Concerning the other sensory attributes (acidity, viscosity, bitterness and sensation of fat) of fermented milk samples, a significant difference (p<0.05) was observed for the viscosity attribute. However, Tukey test analysis showed no significant difference between traditional Lfrik and fermented camel milk with “Flora Danica” (Table 1).

Organoleptic quality of the pasteurized Lfrik tested by regular consumers of Lfrik was found to be similar to that of the traditional Lfrik.

In the second trial done in the city of Rabat and using a panel of non-consumers of camel milk and Lfrik, overall liking scores showed that fermented cow milk was more liked than fermented camel milk samples (Table 2). However, the difference of overall liking scores was not significant (p > 0.05) between fermented cow milk and fermented camel milk prepared using “Flora Danica”. In this trial, a significant difference (p < 0.05) was observed only for the acidity.

It should also be noted that overall liking scores were higher with the first panel than in the second one. This can be explained by the difference in the raw milk samples used for the preparation of different fermented batches and the extent of fermentation process between the two trials. This can be supported by the higher values of acidity in the first trial than in the second trial.

CONCLUSION

Preparation of Lfrik from pasteurized camel milk using commercial lactic starters and incubated at room temperatures (20-30○C) for about 12 h gave products with similar pH and acidity as the traditional Lfrik.

The fermentation process was slower in camel milk than in cow milk. After 12 h of fermentation, the acidity reached respectively about 0.4 and 0.8% lactic acid in camel and cow milks. In addition, NPN and benzoic acid contents were higher in obtained fermented camel milks than in fermented cow milk.

Sensory evaluation showed that fermented camel milk using commercial lactic starters had similar characteristics and overall liking scores than the traditional Lfrik. Fermented camel milk samples prepared using “Flora Danica” were more liked than those prepared using “CHN-11”.

The use of pasteurized camel milk in the preparation of its fermented products allowed a significant improvement of their hygienic quality.

REFERENCES

Abdel Rahman I.E., Dirar H.A., Osman M.A. (2009). Microbiological and biochemical changes and sensory evaluation of camel milk fermented by selected bacterial starter cultures. African Journal of Food Science, 3: 398-405.

Alm L. (1983). The effect of fermentation on proteins in milk. Kieler Milchwirtschaftliche Forschungsberichte, 35: 329-332.

AOAC (1990). Official methods of analysis. (12th ed.). Association of official Analytical Chemists. Washington DC, USA: AOAC International.

AOAC (2010). Official Methods of Analysis (17th), Association of official Analytical Chemists, Gaithersburg, MD: AOAC International.

Attia H., Kherouatou N., Dhouib A. (2001). Dromedary milk lactic acid fermentation: microbiological and rheological characteristics. Journal of Industrial Microbiology & Biotechnology, 26: 263-270.

Barbano D.M., Lynch J.M., Fleming J.R. (1991). Direct and indirect determination of true protein content of milk by Kjeldahl analysis, Collaborative study. J. AOAC Int., 74: 281-288.

Barbour E., Nabbut N., Frerichs W., Al-Nakhli H. (1984). Inhibition of pathogenic bacteria by camel’s milk: relation to whey lysosyme and stage of lactation. Journal of Food Protection, 47: 838- 840.

Breheny S., Kanasaki M., Hillier A.J., Jago G.R. (1975). Effect of temperature on the growth and acid production of lactic acid bacteria. Australian Journal of Dairy Technology, 30: 145.

Chipley J.R. (1993). Sodium benzoate and benzoic acid, In Antimicrobials in Foods. 2nd Edition, ed. P.M. Davidson & A.L. Branen. Marcel Dekker, New York, 11-35.

Davidson M.P. (2001). Chemical preservatives and natural antimicrobial compounds. M.P. Doyle, L.R. Beuchat, T.J. Montville (Eds.), Food microbiology: Fundamentals and frontiers, ASM Press, Washington, 593-627

Driessen F.M., Kingma F. Stadhouders J. (1982). Evidence that Lactobacillus bulgaricus in yogurt is stimulated by carbon dioxide produced by Streptococcus thermophilus. Netherlands Milk and Dairy Journal, 36: 135-144

El-Agamy E., Ruppanner R., Ismail A., Champagne C.P., Assaf R. (1992). Antibacterial and antiviral activity of camel milk protective proteins. Journal of Dairy Research, 59: 169-175.

Farah Z.I., Streiff T. and Bachmann M.R. (1989): Manufacture and characterization of camel milk butter; Milchwissenschaft, 44: 412 - 414

Farah Z., Streiff T., Bachmann M.R. (1990), Preparation and consumer acceptability tests of fermented camel milk. Kenya Journal of Dairy Research, 57: 281-283.

IDF (1983). Cultured dairy foods in human nutrition. International Dairy Federation, Doc 159, Part 2: l8-26. Brussels, Belgium.

IDF (1987a). Milk and milk products, enumeration of microorganisms. International Dairy Federation. Standard 100A. Brussels, Belgium.

IDF (1987b). Determination of benzoic and sorbic acid content in milk, dried milk, yogurt and other fermented milks. International Dairy Federation. International Provisional Standard, 139. Brussels, Belgium.

Ismaili A.M., Saidi B., Zahar, M., Hamama A., Ezzaier R. (2016). Composition and microbial quality of raw camel milk produced in Morocco. Journal of the Saudi Society of Agricultural Sciences. http://dx.doi.org/10.1016/j.jssas.2016.12.001.

Lawrence R.C., Thomas T. D. (1979). The fermentation of milk by lactic acid bacteria, In Microbial Technology, Society for General Microbiology Symposium 29. A. T. Bull, D. C. Ellwood, and C. Ratledge, ed. Soc. Gen. Microbiol. Ltd., Great Britain.

Marshall V.M.E., Law B.A. (1984). The physiology and growth of dairy lactic-acid bacteria. Advances in the microbiology and biochemistry of cheese and fermented milk, 67-98.

Meilgaard M.C., Civille G.V., Carr B.T. (1999). Sensory evaluation techniques (3rd ed.). Boca Raton, Florida: Academic Press.

Nishimoto T., Uyeta M., Taue S. (1969). Precursor of benzoic acid in fermented milk. Journal of the Food Hygienic Society of Japan, 10:410-413.

Ross G.D. (1980). Observations on the effect of inoculum pH on the growth and acid production of lactic streptococci in milk. Australian Journal of Dairy Technology, 35: 147-149.

Saidi B., Warthesen J.J. (1993). Heat and fermentation effects on total nonprotein nitrogen and urea in milk. Journal of Food Science, 58: 548-551.

Saidi B., Faid M. (2001). Effect of hippurate addition on benzoate production during lben fermentation. Revue Marocaine des Sciences Agronomiques et Vétérinaires, 21:151-156.

Shiby V.K., Mishra H.N. (2013). Fermented milks and milk products as functional foods - A review. Critical reviews in Food Science and Nutrition, 53:482-496.

Xinmin W., Ruili Z., Zhihua L., Yuanhong W., Tingfu J. (2008). Determination of glucosamine and lactose in milk-based formulae by high-performance liquid chromatography. Journal of food composition and analysis, 21: 255-258.

Yagil R. (1982). Camels and camel milk. FAO Animal Production and Health paper. FAO, Rome, p. 26.