Preparation and maintenance of lactic starter

A lactic acid bacteria culture, Lactobacillus acidophilus NCDC 13, was obtained from the National Collection of Dairy Cultures (NCDC) of the National Dairy Research Institute, Karnal (India) and was utilised in the study. Whey (paneer whey) with 6% total solids, 0.25% fat and pH 5.5 was combined with skimmed milk (0.15% fat, 9% total solids; pH 6.7) after filtering at a 60:40 (v/v) ratio to obtain a whey-skim milk mixture (WSM). The blend was maintained at a pH of 6.8 using KOH (25% w/v aqueous solution). Pearl millet bought from the local market was cleaned, washed and soaked in water at a 1:3 (w/v) ratio for 24 h at ambient temperature. The germinated pearl millet grains were dried at 55°C–60°C for 8 h, followed by grinding and sieving through a 52-mesh sieve to obtain germinated pearl millet flour (GPMF). Liquid barley malt extract (LBME) with a specific gravity of 1.398, 80.5% solids (refractometer), pH 5.72, 5.17% protein (dry basis) and 1.453% ash was procured from Malt Company (India) Limited, Gurgaon, India. All other chemicals were purchased from standard manufacturers.

Preparation of lactic fermented composite substrate

A multipurpose composite fermented substrate was made by the incorporation of GPMF (4.73%, w/v) and LBME (3.27%, w/v) into WSM (60:40, v/v); the blend was heat-treated at 95°C for 10 min, followed by cooling at 37°C. The growth characteristics of L. acidophilus NCDC 13 (1% inoculum, corresponding to 10⁶ CFU/mL) were studied in the substrate at 37°C at 2 h intervals until the death phase occurred. The enumeration of lactic culture was performed in salicin (0.5% w/v) modified MRS medium (HiMedia, Mumbai) using the pour-plate approach reported by Haughbty et al. (Reference Haughbty, Maturin, Koening and Marshall1993), with peptone water (0.9% NaCl, 0.1% peptone) as diluent. A mathematical model, as per the provided equation, was utilised to represent cell growth to evaluate its suitability.

(1)\begin{equation} \frac{X = ({X_0}{X_{max}}e{(^{{\mu _{max}}t)}})}{({X_{max}} - {X_0} + {X_0}{e^{({\mu _{max}}t)}}})\end{equation}

where X = cell concentration at time t, X ₀ = initial cell count, X _max = maximum attainable cell count, μ_max = maximum specific growth rate as projected by the model and t = time. This equation signifies both the exponential (X < X _max) and the stationary stage (X = X _max) of growth. The term (1-X/X _{ma x}) indicates that microorganisms hinder their own growth due to competition for the resources of carbon and nitrogen obtained from the fermentation substrate or due to the accumulation of inhibitory components. The kinetic parameters of Eqn (1) were estimated using the advanced non-linear curve fitting function of the OriginPro 8.0 software package (OriginLab Corporation, USA). Bacterial growth curves were generated from the experimental data as well as predicted data obtained by using the logistic equation. The suitability of the selected equation for predicting microbial growth was validated based on the coefficient of determination (Adj. R ² > 0.95), reduced chi-square (χ ² < 0.5) values and random nature of residuals (Singh et al., Reference Singh, Mishra and Saha2011).

Stability of lactobacilli in the composite substrate

The stability of the lactic culture was estimated in the fermented substrate under refrigerated conditions. The lactobacilli count was determined at 4-day intervals over a 30-days until the count declined to 6 log CFU/mL. The decline in the lactic organism was correlated with time and expressed through an appropriate mathematical model.

In vivo animal bioassay

Swiss albino male mice (six per group; total 18) with an average weight of 17–18 g were kept in the Small Animal House of ICAR–NDRI (Karnal) in polypropylene cages with proper ventilation and were provided with water (ad libitum). A synthetic diet (Table 1) was provided at 5 g per day to each animal. Animals were kept at 24°C ± 2°C with a 12 h light/12 h dark cycle at 60 ± 5% relative humidity. All animals were given a basal diet for 5 days for adjustment, after which they were classified into various groups. For faecal microorganisms analysis, the group was classified into three groups: Group 1 (fed on a basal diet), Group 2 (fed on an unfermented substrate) and Group 3 (fed the group was on a fermented substrate) and feeding was done for 21 days. The animals were divided into two groups for faecal enzymes analysis, Group 1 (fed on unfermented substrate) and Group 2 (fed on fermented substrate) and feeding period was divided into three stages: baseline 1 – pre-feeding period, basal diet (15 days); baseline 2 – Feeding period, respective diets (15 days); baseline 3 – post-feeding period, basal diet (15 days). Approved study protocols, as well as maintenance of animals, were practiced according to the rules and regulations of the animal ethical committee of the National Dairy Research Institute (Registration No.: 1705/GO/Re/SL/13/CPCSEA).

Table 1.

Composition of the synthetic diet (g/kg of diet)

Faecal samples were freshly collected from the rectum by rectal stimulation (Lee et al., Reference Lee, Park, Jang, Baek, Kim, Huh, Choi, Chung, Kim, Lee and Ha2011) and immediately transferred into aseptic bags and stored at 4°C for analysis immediately afterward. The effect of lactic bacteria on faecal enzyme β-glucuronidase (fβG) activity, which supports the transformation of pro-carcinogens to probable carcinogens, was analysed as described by Marteau et al. (Reference Marteau, Pochart, Flourie, Pellier, Santos, Desjeux and Rambande1990) with modification. The activity of β-glucuronidase was assessed based on the detection of the released concentration of p-nitrophenol from the p-nitrophenyl-β-D-glucuronide substrate by the enzyme, by comparing the intensity of the yellow colour of the released compound spectrophotometrically. The faecal sample (1 g) was added to 30 mL buffer (0.1 mol/l Na₂HPO₄, 0.15 moL/l NaCl, pH 7.4) and combined for 5 min. A 2 mL sample was mixed with 1 mL of a solution (0.13 moL/l) of the p-nitrophenyl β-D-glucuronide (PNDG) substrate made in the above-said buffer for reaction at 37°C. The reaction was stopped after 5 min by adding 0.5 mL of a cold 1 M Na₂CO₃/l solution. The absorbance of the supernatant was observed at 405 nm after centrifugation for 10 min at 4000 g and 4°C. The enzyme activity was expressed in terms of absorbance in μmol substrate/g of faeces.

Enumeration of faecal microorganisms (lactobacilli and coliforms) was carried out by the pour-plate technique at weekly intervals for 21 days. Faeces samples (1 g) of were thoroughly homogenised in 99 mL of 0.1% peptone water, and serial dilutions were prepared up to 10⁹/g of sample. Lactobacilli and coliform count were taken using Lactobacillus MRS Agar and Violet Red Bile Agar (VRBA) (Himedia, Mumbai), respectively. The plates were incubated at 37°C for 48 h and 37°C for 24 h for lactobacilli and coliform count, respectively and the results were expressed in log CFU/g of faecal sample (Jood et al., Reference Jood, Khetarpaul and Goyal2012). For faecal enzyme and faecal microbial profiles, a sample from one animal was subjected to three replicate analyses.

Conversion of the substrate to fermented beverage and its characterisation

The fermented composite beverage was prepared by the incorporation of pectin stabiliser (0.64%, w/v), salt, spices (cumin and black pepper), with 30% (v/v) potable water into the fermented substrate. The product was analysed as follows: pH by pH meter (Labindia, Maharashtra, India; AOAC (Association of Official Analytical Chemists), 1995), viscosity by Viscostar plus viscometer (Fungilab, Barcelona, Spain), colour parameter using a colorFlex EZ spectrophotometer (Hunter Lab, Reston, VA); compositional parameters like fat, protein, moisture, ash, total dietary fibre were estimated as per standard protocol (AOAC (Association of Official Analytical Chemists), 1995); starch by anthrone-sulphuric acid procedure (Raghuramulu et al., Reference Raghuramulu, Nair and Kalyanasundaram2003); phytic acid was obtained by phytic acid estimation kit, Megazyme International Limited, Wicklow, Ireland (Ganguly et al., Reference Ganguly, Sabikhi and Singh2022); polyphenol was estimated by Prussian blue spectrophotometric method (Ganguly et al., Reference Ganguly, Sabikhi and Singh2022); mineral content was analysed using Atomic Absorption Spectroscopy (Hitachi Z-5000 Series, Tokyo, Japan) in sample extracts obtained through the dry-ashing technique (Ganguly et al., Reference Ganguly, Sabikhi and Singh2022); sensory attributes of product on a 9-point hedonic scale, where 1 = dislike extremely, 2 = dislike very much, 3 = dislike moderately, 4 = dislike slightly, 5 = neither like nor dislike, 6 = like slightly, 7 = like moderately, 8 = like very much and 9 = like extremely (Stone et al., Reference Stone, Sidel, Oliver, Woosley and Singleton1974) by a panel of 20 scientists and students from the Dairy Technology Division. The total energy was computed using the following formula:

Energy value (kcal) = % Protein × 4 + % Carbohydrate × 4 + % Fat × 9.

Mathematical and statistical analysis

The advanced non-linear curve fitting function of the OriginPro software package version 8.0 (Origin Lab Corporation, Northampton, MA) was used to model the bacterial growth and decay curve with time. The experimental data were subjected to analysis of variance (ANOVA) using SPSS 16.0 Software (SPSS Inc., Chicago, IL). The comparison of the mean values was performed using Duncan’s multiple range test. Statistical significance was taken at a 95% level of significance.

Mathematical modelling of the growth curve of a Lactobacillus organism in the composite substrate

Lactobacillus acidophilus (NCDC 13) grew well in the composite substrate, as observed in our previous study (Ganguly and Sabikhi, Reference Ganguly and Sabikhi2012). Fig. 1 represents the growth characteristics of the NCDC 13 strain in a composite substrate. An exponential growth stage was observed in 12.00–13.00 for the organism in the optimised substrate. The stationary stage was observed after incubation for 14 h. Cell population stayed constant (13.64 ± 0.02 and 13.52 ± 0.04 log CFU/mL) during the stationary stage (14.00–15.00). The cell densities (12.72 ± 0.04 log CFU/mL) decreased significantly (P ≤ 0.01) after the completion of the stationary phase. Since fermentations were conducted without pH control, organic acid production from metabolic activity led to a reduction in pH of the medium. The respective cell count and pH observed after the exponential stage were 13.1 ± 0.05 log CFU/mL and 4.5–4.6. The criteria considered for the determination of the end of fermentation were pH of 4.5–4.6.

Figure 1.

Growth characteristics of Lactobacillus acidophilus NCDC-13 in composite substrate. Changes in Lactobacilli count (log cfu/mL) and pH. A – Lag phase, B – Exponential phase, C – Stationary phase, D – Death phase.

Table 2A lists the kinetic parameters of the logistic equation estimated from the Origin software. The maximum specific growth rate obtained by the logistic equation was 0.18 h⁻¹ for L. acidophilus (NCDC 13) in the specific substrate of the study. The suitability in predicting the growth curve by logistic equations was validated by the coefficient of determination (Adj. R ² −0.95) and reduced χ ² (χ ² − 0.311), respectively. Both of these parameters (R ² > 0.95 and χ ² < 0.5) indicated the suitability of the model for the prediction of data. Additionally, the residuals obtained were random, which is desirable. Based on these observations, the suitability of the logistic equation for modelling the growth curve in the composite substrate was established. Charalampopoulos et al. (Reference Charalampopoulos, Pandiella and Webb2003) also used logistic equations to describe the growth of four probiotic strains of Lactobacillus sp. in cereal substrate. They observed a maximum specific growth rate for L. acidophilus in malt, barley and wheat medium of 0.19, 0.18 and 0.15 (h⁻¹), respectively.

Table 2.

Numerical values of model parameters of bacterial growth and decay curve

where X ₀ – initial cell concentration (Log cfu/ml), X _max – maximum attainable cell concentration (Log cfu/ml) and μ _max – maximum specific growth rate (h⁻¹). A and B – model parameters in decay curve equivalent to initial cell concentration (Log cfu/mL) and decay rate (1/h).

Stability of lactobacilli organisms in the composite substrate during storage

The lactobacilli count decreased from 12.70 ± 0.05 to 6.36 ± 0.35 log CFU/mL by the end of the 30 days of storage period, which was statistically highly significant (P < 0.01) (Fig. 2). Although a ∼6 log count reduction was observed after 30 days of storage, the substrate contained more than 10⁶ CFU/mL organisms and could be considered effective. The decreasing trend of lactobacilli organisms was described by Eqn (2), an exponential model applied to explain the change of cell viability with time.

(2)\begin{equation}y = a{e^{bx}}\end{equation}

Figure 2.

Changes in lactobacilli count (log CFU/mL) during storage in composite substrate.

where y – Lactobacilli count in log CFU/mL at a time (x); a and b – model constants. The kinetic parameters of Eqn. (2) were projected from the origin software and are depicted in Table 2B. The decay rate of the organism in the substrate was estimated at 0.02 (h⁻¹). The suitability for predicting the decay curve by an exponential equation was validated by a coefficient of determination (Adj. R ² – 0.99) and reduced χ ² (χ ² – 0.033). Both these parameters (R ² > 0.95, χ ² < 0.5) and the random nature of residuals indicated the suitability of the model for the prediction of data.

Effect of fermented composite substrate on faecal enzymes

The influence of feeding fermented and unfermented substrate to mice on β-glucuronidase enzyme is depicted in Fig. 3. During the pre-feeding stage (baseline 1), no distinct trend in enzyme activity was observed. During the feeding stage (baseline 2), similar enzyme activity on day 0 was observed for both the control and experimental groups, followed by a decrease in both groups (depicted by the absorbance values). The decrease was more prominent in the group fed with fermented substrates. A highly significant (P < 0.01) decline in enzyme activity was noticed in the fermented group after 15 days of feeding, ∼ 18.5% (unfermented) and 42.9% (fermented) (Fig. 4). In the third stage (baseline 3), discontinuation of feeding resulted in an increase in enzyme activity in both groups, but an increase was more prominent in unfermented group.

Figure 3.

Faecal β-glucuronidase activities as μmol PNDG metabolised/h/g faeces for mice before feeding, during feeding and after the withdrawal of respective substrate. UF – unfermented substrate; F – fermented substrate, average of 6 mice and 3 trials; baseline 1 – pre-feeding period; baseline 2 – Feeding period; baseline 3 – post-feeding period.

Figure 4.

Effect of feeding of unfermented and fermented substrate on faecal enzyme (β-glucuronidase activity) in mice. A – Unfermented, B – Fermented. Before: before the feeding experiment started, After: at the end of 15 days feeding.

Consumption of fermented dairy and food products has been associated with a reduced risk of colon cancer (Liang et al., Reference Liang, Song, Hu, Wu, Li, Dong, Liang and Wang2022; Emami et al., Reference Emami, Salehi, Hassanzadeh, Mansourian, Mohammadzadeh and Maghool2022; Piccioni et al., Reference Piccioni, Marcello, Marcello, Veronica, Annunziata, Antonio, Francesco and Giuseppe2023). Sabikhi and Mathur (Reference Sabikhi and Mathur2004) observed a decrease in β-glucuronidase activity as a result of feeding functional Edam cheese containing Bifidobacterium bifidum ATCC 15696. Kushal et al. (Reference Kushal, Anand and Chander2006) observed that the supplementation of diet with Lactobacillus acidophilus NCDC-13 and Bifidobacterium bifidum NCDC-255 or coculture of these two strains as free and entrapped cells resulted in reduced faecal enzyme activity when compared with the pre-study level. Lacticaseibacillus rhamnosus GG or Lactobacillus acidophilus administration resulted in a significant decline in the activity of β-glucuronidase in rats treated with 1,2-dimethylhydrazine dihydrochloride (DMH) (Verma and Shukla, Reference Verma and Shukla2013). Similarly, another research work reported inhibition of the activity of β-glucosidase and β-glucuronidase due to administration of L. acidophilus KFRI342 using the DMH model (Chang et al., Reference Chang, Shim, Cha, Reaney and Chee2012). Furthermore, B. longum was observed to decrease these two faecal enzyme activities in normal rats (Lee et al., Reference Lee, Park, Jang, Baek, Kim, Huh, Choi, Chung, Kim, Lee and Ha2011). Several clinical trials have reported similar inhibitory consequences of administering beneficial lactic organisms on these carcinogenic enzymes in human subjects (Lee et al., Reference Lee, Rheem, Yun, Ahn, Joung, Lee, Oh, Chun, Rheem, Yea, Lim, Cha and Kim2013).

The possible mechanism for reducing faecal enzyme activity might be the implantation of beneficial Lactobacilli spp. in the gut, which is suggested to suppress the pathogenic and oncogenic microbial communities in the colon. Bifidobacteria and lactobacilli generally have lower activities of these harmful enzymes, whereas β-glucuronidase is produced in high amounts by Enterobacteria and Clostridia spp (Wollowski et al., Reference Wollowski, Rechkemmer and Pool-Zobel2001). The reduction in enzyme activity in the unfermented group might be due to the presence of fibres and non-digestible carbohydrates in the substrate, which imparted beneficial effects by decreasing the coliform count and supporting the growth of beneficial microbes in the colon (Fu et al., Reference Fu, Zheng, Gao and Xu2022).

Characterization of fermented beverage

The fermented beverage had a sour–spicy taste. The pH and viscosity of the product were 4.2 and 11.5 cp, respectively, with an L* value of 53.47. The average total solids, fat, protein, ash, starch and fibre contents of the beverage were 11, 0.3, 2, 0.61, 1 and 1.4%, respectively. The total energy content of the product was 37.63 kcal per 100 mL of product. The average phytic acid and polyphenol contents were 15.28 and 8.28 mg per 100 mL of product, respectively. The Ca, Fe, Mg and Zn levels of 100 mL of product were 33.28, 0.16, 0.90 and 0.30, mg, respectively. The sensory score, with a mean score of colour and appearance 7.7, flavour 7.5, consistency 7.5 and overall acceptability 7.5, revealed that the product was sensory acceptable to the panelists (Table 3).

Table 3.

Characteristics of fermented whey-cereal-based composite beverage

Mean ± SE, n = 3.