Assessment of exopolysaccharide production by Lactobacillus delbrueckii subsp. bulgaricus ropy strain in different substrate media

Abstract The aim of this research was to determine the optimal medium for Exopolysaccharides (EPS) production by a Lactobacillus delbrueckii subsp. bulgaricus ropy strain isolated from a locally produced commercial fermented milk, in reconstituted skim milk (RSM) 10% (w/v), milk whey (MW), and soy milk whey (SMW), under optimal growth conditions for this strain. Milk whey was made by coagulating fresh milk using papaya latex 3% (v/v); soy milk whey was obtained from tofu household industry. The chemical composition of the substrate media was determined by proximate analysis, and sterilization was accomplished in an autoclave at 121°C for 15 min. Culture media were inoculated with 1% (v/v) of a starter culture of L. delbrueckii subsp. bulgaricus and then incubated at 30°C for 16 hr. EPS production, lactic acid content, cell counting, and pH were determined after the media were cooled at 5°C. Findings showed that on the basis of the growth characteristics of L. delbrueckii subsp. bulgaricus, the best medium for EPS production was RSM 10% (258.60 ± 26.86 mg/L) compared to the milk whey (69.60 ± 9.48 mg/L) and soy milk whey (49.80 ± 9.04 mg/L).

The composition of the culture media greatly affects the ability of microorganisms to metabolize specific nutrients in order to produce energy for their growth and reproduction needs. Lactobacillus bulgaricus subsp. bulgaricus ropy strain can produce polysaccharides making the yogurt thicker and finer in texture (Ali et al., 2019). For its optimal growth, L. delbrueckii subsp. bulgaricus, as a species of LAB, requires several types of nutrients. Some types of media such as milk or whey contain enough nutrients needed by LAB and are often used to produce important bacterial metabolites such as lactic acid, alcohols, acetic acid, and other value-added substances of economic interest.
Considering that L. delbrueckii subsp. bulgaricus is normally present in fresh milk and that reconstituted milk (10% b/v RSM) resembles fresh milk, it was selected as a test medium. Some researchers use nonfat milk in order to maintain L. delbrueckii subsp. bulgaricus.
On the other hand, milk whey (MW) is a waste that separates after curdling in cheese production and still contains enough nutrients to support microbial growth (Briczinski & Roberts, 2002). Furthermore, SMW is also a liquid waste that separates from the soy milk lumps during the manufacture of soybean tofu and contains nutrients needed for microbial growth. In addition, the nutritional composition of soy milk used in tofu production is almost the same as in mammalian milk.
For the optimization of this process, it is very important to select the medium that best stimulates and enhances the production of EPS by the L. delbrueckii subsp. bulgaricus ropy strain.

| Bacterial strains and fermentation conditions
Lactobacillus delbrueckii subsp. bulgaricus ropy strain was obtained from the collection of the Milk Processing Biotechnology Laboratory in the Faculty of Animal Husbandry in Hasanuddin University, Indonesia, which was isolated from a locally commercial ropy yoghurt and routinely propagated in 10% Reconstituted Skim Milk medium (RSM), twice a week (Malaka et al., 2016). The bacterial strain (0.1% inoculums w/v) was grown in sterile 10% (w/v) RSM (sterilized at 121°C, 15 min), and incubate data temperature of 37°C for 16 hr in order to enhance optimal cell growth.
Three batch fermentations without pH control were carried out in 100-ml Erlenmeyer flasks containing 10% RSM, MW, and SMW media, respectively. The 10% (w/v) RSM media was prepared by dissolving 10 g of powdered skim milk in 90 ml of distilled water. Milk whey was obtained from a small-scale Dangke cheese processing factory, and SMW was obtained from residual effluents of a smallscale tofu processing factory. Chemical compositions of all three media were determined by proximate analysis. Media were sterilized at 121°C for 15 min, and after cooling to 30°C, they were inoculated with the starter culture of L. delbrueckii subsp. bulgaricus (0.1% w/v inoculum) and then incubated at 30°C for 16 hr in order to study EPS production. Samples were stored at 5°C before testing.

| Growth measurement
Bacterial growth of L. delbrueckii subsp. bulgaricus was determined by measuring lactic acid production and pH decrease in the media.
Lactic acid was determined by weighing 2.00 ± 0.05 g of culture media and titrating with 0.1 N NaOH using phenolphthalein as indicator.
The concentration of lactic acid was calculated using the formula (Marshall, 1993): Counting of L. delbrueckii subsp. bulgaricus live cells was done by the dilution method using 0.9% NaCl solution and Bromocresol purple agar (BCPA) as growth medium. Serial dilutions of each sample were poured into petri dishes in duplicate (duplo) and then incubated at 37°C for 48 hr. Results were expressed as colony forming units (CFU/ml). pH measurements were carried out using a pH meter (Hanna pH meter Instrument 302 Portugal).

| EPS production
EPS isolation was done from a cell-free supernatant (centrifugation at 4,032 g). The supernatant was heated in a water bath at 65°C for 30 min (LTLT; low temperature long time pasteurization) to activate the enzyme able to degrade the polymer; afterward, papain enzyme was added as much as 1% to coagulate casein. Samples were then centrifuged at 6,000 rpm, and the supernatant was stored at 5°C for 24 hr. The polymer was precipitated with 95% cold ethanol in a 1:1 ratio (v/v), stored at 5°C for 24 hr, and then centrifuged at 7, 168 g. The precipitate was washed with deionized water three times to eliminate the remaining sugars from the culture medium, then freeze-dried, and weighed as dried EPS powder. EPS production was expressed in mg/L units (modified from Cerning et al., 1992;Mozzi, Giori, Oliver, & Valdez, 1994).

| Data analysis
Data obtained from the results of this study were processed by analysis of variance in a completely randomized design (CRD) with % Lactic Acid = (ml NaOH × N NaOH) × 0.09 Sample weight (g) × 100% three treatments and five replications (Gaspersz, 1994;Hinkelmann & Kempthorne, 2007) using the SPSS program. Mathematical models were applied as follows: where Y ij = Value of observation in the experimental unit that receives treatment i (media type factor). µ = The average all of observations. α = Effect of media type on the percentage of lactic acid, pH, EPS and total L. delbrueckii subsp. bulgaricus. ε ij = Effect of random error on observation of the j which received treatment i.
When data analysis showed that there was a significant effect of the treatment, it was then evaluated by the smallest significant difference test (LSD). The significant difference was set at the 95% confidence level. Removal of several types of amino acids can inhibit the growth of L. delbrueckii subsp. bulgaricus; the effect is especially remarkable with asparagine, glutamine, and threonine, although aspartic acid, glutamic acid, or glycine only slightly affect the growth of these bacteria (Hebert, Raya, & Giori, 2004).
Major ionic species in milk include bicarbonate, chloride, citrate, calcium, magnesium, potassium, and sodium. All minerals are distributed between the soluble phase and the colloidal phase. The distribution of calcium, citrate, magnesium, and phosphate in the colloidal phase and soluble phase and their interaction with milk protein is very important for the stability of milk and milk products (Varnam & Sutherland, 1994). Minerals are needed in microbial metabolism as electron acceptors for the metabolism of glucose and other sugars.
Milk whey is a liquid residue obtained from cheese and casein production. Whey solids consist of 80%-90% protein and 10%-20% of soluble milk constituents such as lactose, vitamins, and minerals (Bylund, 1995). The lack of casein protein in milk whey arises from the fact that casein precipitates as curd, and the protein contained in whey corresponds to α-lactalbumin and β-lactoglobulin which are proteins needed to enhance the body's resistance to diseases.

| Effect of media type on pH
The effect of media type on changes in pH values is shown in Figure 1.
Among the three types of media, it can be seen that the 10% RSM media showed the fastest decrease in pH (lowest pH value) which corresponded to an average value of 5.096, while the lowest pH decrease (corresponding to the final highest pH value) at MW media  Figure 2 shows the relationship among the three types of media as regards of lactic acid production. The highest production rate was found in 10% RSM, which ranged from 0.824% to 0.930%, with an average final concentration of 0.884%. As for MW and SMW, lactic acid production was very close, reaching an average of 0.602%.

| Effect of media type on lactic acid production
Lactic acid production in MW and SMW media was lower because a reduced carbohydrate content in comparison with the 10% RSM medium.
During lactic acid fermentation, lactose undergoes a hydrolysis step to produce monosaccharides in the glycolysis pathway (Berg, Tymoczko, & Stryer, 2002). In some cases, disaccharides breakdown through phosphohydrolase, and one sugar molecule turns into a free monosaccharide and the other one becomes a monosaccharide phosphate. Lactobacillus bulgaricus as one type of BAL generally produces a system of lactose phosphotransferase where lactose will enter the cytoplasm as lactose phosphate which will join phospho-β-d-galactosidase to produce glucose and galactose-6-phosphate. Glucose phosphorylase is metabolized by glucokinase via the glycolytic pathway. Galactose-6-phosphate is metabolized via the tagatose-6-phosphate pathway (Neves, Pool, Kok, Kuipers, & Santos, 2005).
The low production of lactic acid in whey is due to the very limited amount of lactose in MW. Likewise, lactic acid in SMW is produced from a type of carbohydrate different to that of lactose type.
In this study, substrates used for EPS production by L. shows that there is a close relationship between cell growth, lactic acid production, and EPS production, which is associated with the presence of certain nutrients, especially carbohydrates as the main medium and protein as a catalyst and minerals as cofactors.

| Effect of media types on the growth of L. delbrueckii subsp. bulgaricus
The effect of the type of media on bacterial growth, determined by the number of live bacteria, is shown in Figure 3. Maximum growth after incubation at 30°C for 16 hr reached an average of 11.508 (log CFU) in 10% RSM media. In whey media, the amount of L. delbrueckii subsp. bulgaricus reached 10.835 (log CFU/ml) and in the SMW medium it was 10.982 (log CFU/ml). These numbers indicate that these bacteria have reached the stationary growth phase.
Microbial growth begins with the decomposition of the substrate constituted by proteins, minerals, and carbohydrates. The oxaloacetate decarboxylase that produces pyruvate will join through the diacetyl/ acetoin pathway. Lactobacillus bulgaricus as homofermentative LAB, on the pyruvate pathway, will produce lactic acid and more ATP is formed; thus, the glucose utilization becomes more efficient and stimulates the growth rate (Papagianni, 2012). This can be seen in the 10% RSM medium, which presents the most efficient growth. Growth depends on the ability of the cell to form new protoplasm from nutrients found in the growth substrate. It begins with an increase in cell mass and the number of ribosomes, duplication of ribosomes, synthesis of new cell walls and plasma membranes, separation of two chromosomes, septal formation, and cell division (Nair, 2010). Carbon derived from the carbohydrate substrate in the media is needed by bacteria as an energy source. Amino acids are used as growth factors for the formation of nucleic acids and minerals. They are needed as electron acceptors and as coenzymes in cell metabolism (de Vladar, 2012).
Based on the variance analysis, the type of media accounted for a very significant effect (p < .01) on the total number of live bacteria of L. delbrueckii subsp. bulgaricus. The Smallest Significant Difference Test (SSD) indicated that the bacterial counts in the three types of media were significantly different, although closer between 10% RSM and SMW. Under conditions of insufficient availability of carbohydrates as a carbon source for cell growth and metabolism, these bacteria can use other carbon sources present in the media, such as proteins, acetic acid, or even carbonic acid. Although the bacteria counts were large, the decrease in pH and the production of lactic acid is rather dependent on the presence of lactose or other carbohydrates contained in the media.
Therefore, even though the number of bacteria was rather large, pH value does not necessarily decrease to 4.5 or lactic acid concentration increase to 0.8%-1.6%, although L. delbrueckii subsp.
bulgaricus reaches this pH value and lactic acid concentration in the milk medium.
Based on substrate conversion, the best medium where substrate is converted into cells was observed in whey medium. This shows that in relation to EPS production, appropriate conditions are required to support the synthesis besides the need of a main carbon source for polymerization during the synthesis of polysaccharides.

| Effect of media type on EPS production
Variable fingerprint analysis shows that the type of media had a very significant effect (p < .01) on EPS production. The LSD test showed that the 10% RSM media was significantly very different (p < .01) in comparison with the MW medium and the SMW medium.  (Chervaux et al. (2000).
In substrate conversion of whey media (  (Ates, 2015). The synthesis of EPS in nutrient media shows that the polymer is continuously excreted shortly after growth and during the cell division. The specific enzymes that play a role in the biosynthesis of EPS are the glycosyltransferase. This enzyme is needed to assemble repetitive units used in long chain determination, during the polymerization of nucleotide sugars as unique monomers precursors for EPS. The gene for EPS production by L. bulgaricus is produced as a doubling reaction through chromosomes (Zeidan et al., 2017).
Exopolysaccharides, in the form of heteropolysaccharides, are synthesized with precursor polymerases formed in cytoplasmic cells.
Nucleotide sugars play an important role in carrying isoprenoid lipids located in the cytoplasm. The synthesized cell components compete with EPS formation during different growth phases (Madhuri & Prabhakar, 2014), what can be noticed by the occurrence of lactic acid production which causes a decrease in pH, the occurrence of cell growth and EPS production during the incubation process of the growth media. This explains why the conversion of the substrate from each media is different according to the different chemical compositions.  (Ramos et al., 2001). Deficiency of nitrogen, phosphorus or sulfur sources in media containing carbohydrates increases EPS production.
Glucose-6P is a key metabolite at the branching point between EPS glycolysis and biosynthesis. If metabolism follows the glycolysis pathway, glucose metabolism will lead to the formation of lactic acid which causes a decrease in pH (Boels, Kleerebezem, & Vos, 2003). Rimada and Abraham (2003) found that kefir granules in whey deproteinization media had higher EPS production compared to milk media which in the whey media was 247 mg/L and in milk media was 218 mg/L. This is due to the differences in the type of starter used.
This result is supported by its growth characteristics which show that in RSM medium the microbial activity is maximal (cell count 11.5030 ± 0.0073 log CFU/ml, lactic acid 0.884 ± 0.004%) compared to MW media (cell count 10.833 ± 0.005 log CFU/ml, lactic acid 0.602 ± 0.004%) and SMW (11.19 ± 0.4011 log CFU/ml, lactic acid 0.602 ± 0.008%). Milk whey can be used for EPS production with enrichment of culture media, and SMW can be used as a medium for microbial and LAB cell production.

ACK N OWLED G M ENTS
We would like to thank Hasanuddin University for providing internal grant research funding, as well as the Institute for Research and Community Service Hasanuddin University in helping monitor and evaluate this research.

CO N FLI C T O F I NTE R E S T
The authors declare that they do not have any conflict of interest.

E TH I C A L A PPROVA L
This study does not involve any human or animal testing.