Abstract: In the present study lactic acid bacterial strains previously identified as the main fermenting microorganisms of the Sudanese garris and yeast strains previously identified as the dominating yeast strains of the Sudanese robe, were used to prepare fermented milk products in High Temperature Short Time (HTST) pasteurized cows milk. The lactic acid bacteria included Lactobacillus plantarum, Lactobacillus paracasei subsp. paracasei and Lactobacillus fermentum while the yeast strains included Pichia membranefaciens and Candida famata. The strains were inoculated into HTST milk and incubated at 25°C. The changes in proximate chemical composition, organic acids and viable microbial counts were estimated. The LAB reached maximum counts after 18 h, while the yeast attained maximum populations after 24 h. The study showed that a stable fermented milk product with high nutritive value could be produced using a mixed culture of lactic acid bacteria and yeast and that a fermentation period of 24 h at 25°C would be sufficient.
INTRODUCTION
In fermentation, the raw materials are converted by microorganisms (bacteria, yeasts and moulds) to products that have acceptable qualities of food (Ray and Daeschel, 1992). In natural fermentation, the conditions are set so that the desirable microorganisms grow preferentially and produce metabolic byproducts which give the unique characteristics of the product. When the yield is unstable and where the desired microorganisms might not grow, a controlled fermentation is used. In a controlled fermentation the fermentative microorganisms are isolated and characterized then maintained for use as starter culture. Starter cultures are added to the raw materials in large numbers and incubated under optimal conditions.
Combinations of different bacterial strains belonging to the genera Lactobacillus, Streptococcus and Bifidobacterium, have been used traditionally in fermented dairy products to promote human health (Dunnel et al., 1999).
Fermented milks are products prepared from milks, whole, partially or fully skimmed, concentrated or milk substituted from partially or fully skimmed dried milk, homogenized or un-homogenized, pasteurized or sterilized and fermented by means of specific microorganisms. Milk from eight species of domesticated mammals (cow, buffalo, sheep, goat, horse, camel, yak and zebu) has been used to make traditional fermented milk products throughout the world (Kroger et al., 1989). The characterization of the microorganisms responsible for the fermentation led to the isolation of starter cultures, which could be produced, on a large scale to supply factories involved in the manufacture of these products (Cogan and Accolas, 1996).
Lactic Acid Bacteria (LAB) and yeasts have been reported to be the predominant microorganisms in most of the African indigenous fermented food products (Steinkraus, 1996; Halm and OLson, 1996; Olasupo and Azeez, 1992; El-Hadi, 2001).
Sudan has a very long tradition in producing a great variety of fermented dairy products. Possibly more than 50% of annual milk production is converted into fermented dairy products (Dirar, 1993). These fermented milks are widely popular and consumed by a larger section of the human population. Of the many fermented milks of Sudan, a few of them are important like robe, garris, mish, zabady and others. In general, these milk products are resultant of natural fermentation. The quality and safety of final product is dependent upon various influencing factors, especially the hygienic status of raw milk used and the presence of lactic acid bacteria and yeasts in good numbers and their ability to grow and causing in desirable fermentation. The present study aims at the preparation of quality consistent and safe traditional fermented milk products using different culture combinations of lactic acid bacteria and yeasts, investigation of the possible interactions between the yeasts and LAB through studying final populations in the fermented milks as well as determination of the nutritive value of these products.
MATERIALS AND METHODS
Preparation of Starter Cultures and Inoculation into HTST Milk
The cultures of lactic acid bacteria (Lactobacillus plantarum,
Lactobacillus paracasei subsp. paracasei and Lactobacillus fermentum)
which had been previously isolated and identified as the main fermenting LAB
in garris (El-Hadi et al., 2004) and yeast strains Pichia membranefaciens
and Candida famata which had been previously identified as the main yeast
strain in robe product (El-Hadi, 2001) were used. The LAB cultures had
been stored at -80°C in sterile tubes containing MRS broth. LAB was cultivated
by streaking on MRS agar (Merck) and incubated anaerobically using Gas Pack
at 30°C for 24 h. A colony was picked from each pure culture plate, grown
successively in MRS broth and centrifuged. The pellet was washed in peptone
physiological salt solution centrifuged again and redistributed in peptone physiological
salt solution.
Pure cultures of yeasts Pichia membranefaciens and Candida famata were cultivated by streaking on Potato Dextrose Agar (PDA), incubated at 30°C for 48 h then centrifuged and washed. Yeast cultures had been stored on potato dextrose agar slants at 4°C until required. Cell suspensions of the individual cultures of LAB and yeast were prepared from 36 h culture broths of Lactobacillus MRS and potato dextrose media, respectively. Appropriate dilutions of the cell suspension were prepared in sterile 0.85% saline, to obtain desirable levels in the product preparation. The inoculated milk samples were aseptically transferred into screw-capped glass bottles.
Preparation of the Fermented Milk Products
Normal, previously pasteurized (HTST) and cooled cow’s milk was used.
The milk was standardized to 3.0% (v/v) fat, homogenized and poured into sterile
jars. Each jar was inoculated 0.1% (v/v) of the individual culture combinations
as follows: (I) Lactobacillus plantarum, Lactobacillus paracasei subsp.
paracasei and Lactobacillus fermentum (product-A); (ii) Lacotbacillus
plantarum, Lactobacillus paracasei subsp. paracasei, Lactobacillus
fermentum and Pichia membranefaciens yeast (product -B); and (iii)
Lacotbacillus plantarum, Lactobacillus paracasei subsp. paracasei,
Lactobacillus fermentum (LAB) and Candida famata yeast (product-C).
The culture combinations were then thoroughly mixed. Inoculated milk samples
were incubated at 25°C for 18-20 h. The fermented product was then dispensed
in sterile 200 mL capacity glass bottles.
Viable Microbial Counts
The Lactic acid bacterial strains were enumerated (pour plate) on MRS agar
(Merck, Darmstadt, Germany) and incubated at 30°C for 48 h. Serial dilutions
in both cases were done in quarter strength Ringers' solution (Oxoid, Unipath,
England). The yeast counts was carried out by (spread plate) on Potato Dextrose
Agar (PDA) and incubated at 25°C for 72 h. Characteristic colonies appearing
on the respective selective agar media were counted, multiplied by the dilution
factor and expressed as log10 colony forming units per milliliter
(log10 cfu mL-1). The viable counts of yeasts and LAB
were carried out according to (IDF, 1990).
Proximate Chemical Composition
Proximate chemical composition of fermented milk products were carried out
to determine the contents of moisture, protein, ash, fat, lactose, Total Soluble
Solids (TSS) pH and Titrable Acidity (TA) according to the AOAC (1990).
Determination of Organic Acids
Organic acids were determined according to the modified method of Marsili
et al. (1981). A 1.00 g sample was added to 0.2 mL 0.5 M H2SO4
and 8 mL acetonitrile and mixed for 30 min. After centrifugation and filtration,
the samples were analyzed using an Aminex HPX-87H HPLC column, held at 37°C,
connected to a Perkin-Elmer HPLC (Perkin-Elmer, Norwark, CT, USA). H2SO4
(15 mM), at a flow of 0.4 mL min-1, was used as the mobile phase.
Organic acids were identified according to their retention times compared with
standard solutions of the following acids: citric, orotic, pyruvic, succinnic,
lactic, formic, acetic and uric acid (Sigma). The analysis was externally calibrated
using mixed standard solutions in deionized water, prepared as for the samples.
RESULTS AND DISCUSSION
When these characteristics are compared with those of fresh milk samples, it was clearly seen that there was a noticeable decrease in pH values of the fermented milks (which ranged between 4.02±0.9 to 4.26±0.1 and an increase in titrable acidity (expressed as lactic acid %) which ranged between 1.75±0.05 and 1.95±0.15% Table 1. The greatest change occurred by the first culture combination (product A). The pH after 48 h in all the fermentation types was generally in close agreement to those of pH of robe (4.0) and garris (3.9) as determined by El-Hadi (2001) and El-Hadi et al. (2004), respectively. In addition, there was a marked drop in total soluble solids contents which ranged between 7.5±0.2 to 7.8±0.2 as well as lactose which ranged between 3.82±0.2 to 3.43±0.1%. The protein and ash contents were higher when compared with those of fresh milk. These changes are obviously due to fermentation. The fat content of the fermented milk samples were more or less similar in all fermented milk samples. The moisture percentage ranged between 90.4±1.4 to 91.2±1.6 %. These values are in close agreement to the moisture content of Gariss as reported by Mirghani (1994) who determined a value of 92.6%. Average values of ash and protein percentage were found to be comparable to Yoghurt and Gariss (Mirghani, 1994). Most of the chemical parameters were in close agreement with the data in the literature for many African fermented dairy products (Mutukumira, 1996; Steinkraus, 1996; Keller and Jordan, 1990; Isono et al., 1994).
| Table 1: | Chemical composition of Rob samples obtained from different sources |
| A, B and C are the fermented milk product samples | |
| Table 2: | Organic acids (ppm) of various fermented milk products after 48 h fermentation at 26°C |
| N:D Not Detected | |
| Table 3: | The viable LAB counts in HTST milk after 48 h fermentation with selected combinations of yeasts and LAB strains LAB count in combination with (log10 cfu mL-1) |
| Table 4: | The viable yeast counts in HTST milk after 48 h fermentation with selected combinations of yeast and LAB strains yeast count in combination with (log10 cfu mL-1) |
The organic acids detected during fermentation included citrate, orotate, pyruvate, succinate, lactate, formate, ureate and acetate (Table 2). Lactic, formic and uric acids were not detected in the pasteurized milk but detected in all fermented milk types. The concentration of the organic acids content in all fermented milk samples was due to fermentation. The concentrations of succinic, orotic and citric acids were high at the end of the fermentation. The concentrations of most of these acids were more or less similar in all fermented milk products. The increase in lactate in the fermented milk products was due to the consumption of lactose by the lactic acid bacteria and converting it to lactic acid. The presence of lactic and acetic acids could play a significant role in improving the shelf-life of the fermented products as they have a direct antimicrobial effects (Davidson, 1997). The antagonism is believed to result from the action of acids on the bacterial cytoplasmic membrane, which interferes with the maintenance of membrane potential and inhibits active transport (Blom and Mortvedt, 1991).
The counts of LAB strains in the fermented milk products were in the range 8.14 -9.51 log10 cfu mL-1 (Table 3). Both of L. plantarum and L.paracasei subsp. paracasei grew approximately two log cycle while L. fermentum grew approximately one log cycle and the viable counts were nearly similar in both single and culture combination. The viable counts of the three Lactobacillus strains were not changed by co-inoculation with yeast. The final viable counts of L. paracasei subsp. paracasei and L. fermentum were comparatively higher in culture combination with Pichia membranefaciens than in single culture. On the other hand L. plantarum had higher viable counts in culture combination with Candida famata. These LAB counts are comparable to those encountered with many other dairy starters (Teuber, 1995). The lower potentialities of L. fermentum to grow in the HTST milk could be due to presence of wild strains of lactic acid bacteria in the fermented milk. However, the higher counts of L. fermentum founded co-culture with Pichia membranefaciens suggest that the yeast stimulated growth of the LAB. The yeast could achieve this by providing essential metabolites such as pyruvate, amino acids and vitamins.
The final populations of yeast cultures were in the range of 6.45-7.12 log10 cfu mL-1 Table 4. Candida famata in co-culture with L. plantarum had a lower final population than in the original inoculum, suggesting that their viability may have been reduced at the end of the fermentation. Yeast growth in milk products is attributed to the ability of the yeasts to utilise milk constituents, such as proteins, fat, lactose and citrate (Fleet, 1990). Other reports also attribute this growth in part to symbiosis with other microflora in the mixed culture (Koroleva, 1988). Pichia membranefaciens grew well in the HTST milk but showed faster growth in culture combination with L. paracasei subsp. paracasei attaining maximum populations of about 7.12 log10 cfu mL-1 after 48 h. When in co-culture, it attained the maximum populations after 48 h.
CONCLUSION
The study showed that fermentation of HTST milk improved its nutritional value by increasing most of the chemical components. In addition, the study showed that Lactobacillus plantarum, Lactobacillus paracasae and Pichia membranefaciens could grow mutually in culture combination and could have potential as a mixed starter culture. The higher final populations of some LAB strains in culture combination with yeasts, was indication of a beneficial effect of the yeasts on the LAB. However, in some cases co-inoculation did not result in a major effect on the final numbers of either the yeast or the LAB. Further study is needed to investigating the genetic blue-prints of the starter cultures used in the present study as that will provide new applications which can be beneficial to the processor and consumer.
ACKNOWLEDGMENTS
The authors express their sincere thanks to the staff of the Department of Bisystems Engineering, Faculty of Agriculture, Kobe University and technicians of the Department of Food Science and Technology, Faculty of Engineering and Technology, University of Gezira, Sudan for their valuable assistance in implementing the present study.