https://orcid.org/0000-0003-0213-8858
Figures
Abstract
Conventional culture media such as de Man, Rogosa and Sharpe are essential for the growth of lactic acid bacteria and the production of metabolites used in the food and pharmaceutical industries. However, their high cost limits their application, particularly in low and middle-income countries. This study aimed to develop a cost-effective and efficient culture medium based on agro-industrial and environmental by-products, pineapple peels, sugarcane molasses, and black soldier fly larvae cake. After physico-chemical analysis of the by-products, excluding sugarcane molasses, a statistical mixture design was used to determine the optimal proportions for supporting lactic acid bacteria growth and bacteriocin production. Growth and metabolite production were monitored via plate count and agar well diffusion assays respectively. The strains tested belonged to the genera Lactobacillus, Bifidobacterium, Streptococcus, Lactococcus, and Bacillus as an out taxa group. The larvae cake showed high protein (48.47 ± 0.14%) and amino acid (17 types) content, while pineapple peels and molasses were rich in carbohydrates (89.52 ± 0.16% and 86.86 ± 0.07%). Based on regression models, the compromise formulation was defined as 55.15% larvae cake hydrolysate, 19.85% pineapple peel hydrolysate, and 25.00% sugarcane molasses. This medium highly supported lactic acid bacteria growth (9.43–9.86 log CFU/mL) compared to MRS and M17 (9.20–9.69 log CFU/mL), with Lactobacillus strains performing better. It also supported bacteriocin activity (11.0–14.5 mm inhibition zones), higher or similar to MRS and M17, with Lactococcus lactis subsp. lactis MA2 exhibiting the strongest effect. These results highlight the potential of this formulation as a sustainable, low-cost alternative for microbiology and biotechnology, particularly in resource-limited settings. The determination of its formula will allow its manufacture once the proximate compositions of the ingredients are known, regardless their origin. Future investigations will focus on optimisation of culture conditions, powder form formulation, and cost evaluation.
Citation: Tchabou Tientcheu ML, Kaktcham PM, Foko Kouam EM, Tchamani Piame L, Djodjeu Kamega LC, Aarzoo, et al. (2025) Development of a low-cost culture medium from industrial and environmental by-products for sustainable cultivation of Lactic Acid Bacteria. PLoS One 20(12): e0337684. https://doi.org/10.1371/journal.pone.0337684
Editor: Guadalupe Virginia Nevárez-Moorillón, Universidad Autonoma de Chihuahua, MEXICO
Received: June 30, 2025; Accepted: November 12, 2025; Published: December 1, 2025
Copyright: © 2025 Tchabou Tientcheu et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are within the paper and its Supporting Information files.
Funding: The author(s) received no specific funding for this work.
Competing interests: The authors have declared that no competing interests exist.
Introduction
In an industrial context undergoing a transition towards more sustainable practices, beneficial microorganisms are attracting increasing interest due to their versatility and safety [1] Among them, lactic acid bacteria (LAB) play a pivotal role in the agri-food, pharmaceutical, cosmetic and animal production sectors, owing to their technological, functional (probiotic) and antimicrobial properties [2]. Beyond their industrial potential, LAB are widely studied in laboratory settings. Their use in lab-scale studies or for biomass and beneficial metabolites productions (for manufacturing starter cultures and probiotic products) at industrial level requires nutrient-rich and complete culture media that meet their needs in terms of carbon and nitrogen sources as well as growth factors. In this respect, they are typically cultivated on conventional media such as MRS [3] and M17 which contains animal peptones, yeast extract, and meat extract. However, these media are expensive and thus represent a major limitation for large-scale production or use in laboratories of resource-limited settings [4]. Indeed, several studies have shown that the high production cost of LAB and their beneficial metabolites is directly related to the nitrogen source, with protein and yeast extract being the most expensive components [5]. In response to these challenges, several studies have explored the use of alternative substrates derived from agro-industrial and environmental wastes/by-products. Culture media have been formulated using materials such as whey [6] organic wastes [7] sheep hooves [8,9], and whey permeate [10] all demonstrating varying degrees of potential. The use of these substrates aligns with sustainable development goals of valorising low-cost residual materials while reducing the environmental impact of microbiological processes. But, these formulations often require supplementation with conventional growth factors (vitamins, minerals) or nitrogen sources (yeast extract, protein hydrolysates) that are still expensive [11], prompting the quest for cheaper substrates which alone or in combination can meet the carbon, nitrogen and growth factor needs and support the growth and metabolites production by LAB. Kaktcham et al. [12] reported that media formulated solely from fish waste and sugarcane molasses could efficiently support bacteriocin production by Lactococcus lactis subsp. lactis 2MT. Similarly, Valle Vargas et al. [13] demonstrated that media formulated from whey, sugarcane molasses, and palm kernel cake could support the growth of fish probiotics Lactococcus lactis A12, Priestia megaterium M4, and Priestia sp. M10, while Acosta-Piantini et al. [14] showed that hydrolysed sugarcane molasses could serve as an economical substrate for mass production of Lacticaseibacillus paracasei. However, such studies are still scarce not only, but also the combination of multiple unexplored wastes/by-products that are released in our environment and contribute to pollution can also be valorised in the formulation of low-cost culture media. Such an approach could provide a balanced source of carbon, nitrogen, and growth factors, thereby enhancing both LAB growth and metabolite production. Three local agro-industrial and environmental by-products that have not yet been used in the formulation of culture media alone or in combination were selected in the present study. The black soldier fly larvae (Hermetia illucens) cake, a protein-rich residue from the biofuel and animal feed industries, represents an emerging biotechnological resource in Africa. It contains high-quality proteins, lipids, minerals (Fe, Zn, Mg), and bioactive compounds which can stimulate microbial growth and metabolism [15]. Pineapple peels, abundantly generated by fruit juice industries and local markets, are rich in fermentable sugars (glucose, fructose, sucrose), organic acids, and B-group vitamins, as well as phenolic compounds that may enhance bacterial stress resistance [16]. Sugarcane molasses, a by-product of sugar refineries such as SOSUCAM in Cameroon, provides readily assimilable carbohydrates, minerals (Ca, K, Mg), and trace elements essential for enzymatic functions in LAB [17]. Their combination may allow the development of economical and sustainable culture media in line with the principles of the circular bioeconomy [18].
This study was therefore aimed to formulate from agro-industrial and environmental by-products a low-cost culture medium able to sustain the growth and beneficial metabolites production of a wide genera of lactic acid bacteria.
Materials and methods
Bacterial strains, media, and growth conditions
Ten bacterial strains with technological and probiotic interests were used in this study (Table 1), including nine lactic acid bacteria belonging to genera Lactobacillus, Lacticaseibacillus, Lactiplantibacillus, Streptococcus, Lactococcus and Bifidobacterium. Bacillus subtilis was also used and considered as an out taxa group.
From frozen stocks, the LAB strains were revived by two successive subcultures of 18 h each [19] in MRS broth (HIMEDIA, India) for lactobacilli and in M17 broth (HIMEDIA, India) for lactococci, pediococci and streptococci, before being used for subsequent experiments. The pathogenic bacteria were revived as stated above but in BHI broth (TITAN BIOTECH LTD, India).
Sampling, physico-chemical analysis, and preparation of hydrolysates of by-products
Pineapple peels were collected from vendors at Market B in the city of Dschang (Menoua Division). Sugarcane molasses was obtained from the SOSUCAM sugar company in Mbanjock, located in the Centre Region of Cameroon, stored at 4°C in tightly sealed glass bottles to prevent microbial contamination and moisture absorption. Before use, the solution was diluted to 4% (w/v) with distilled water, thoroughly mixed, and filtered through Whatman paper No.1. Black soldier fly larvae cake (BC) was sourced from biofuel producers at the Faculty of Agronomy and Agricultural Sciences (FAAS). University of Dschang (West Region of Cameroon), stored in hermetically sealed plastic bags inside a desiccator to maintain dryness until use. All samples were transported to the Research Unit for Biochemistry, Food Science and Nutrition (URBPMAN), where they were processed. Pineapple peels were oven-dried at 45°C for 48 h, ground using a blender (Silver Crest, China), and sieved through a 200-micron mesh. The resulting powder was stored in a desiccator until use.
Physico-chemical analysis of the by-products
The physico-chemical analysis of pineapple peels, BC and sugarcane molasses was conducted according to IUPAC methods [20] for fats and sugars, AOAC methods [21] for proteins and ash. The Minerals (Ca, P, K, Na, Zn, Mg, Mn and Fe) were quantified using inductively coupled plasma optical emission spectrometry (Optima 8000 ICP-OES, Perkin Elmer, USA) as described by Savic et al. [22]. Sugar profiles were determined by high-performance liquid chromatography (HPLC) following the protocols of Reuter et al. [23]. Briefly, after filtration through a 0.45 µm membrane filter, liquid samples were directly injected into a high-performance liquid chromatography (HPLC) system; the analysis was performed using an Aminex HPX-87P column (300 × 7.8 mm) (Bio-Rad, USA), maintained at 80°C, with ultrapure water as the mobile phase in isocratic mode at a flow rate of 0.6 mL/min. Detection was carried out using a refractive index detector (RID). Each sample was injected at a volume of 20 µL. Sugar identification was based on retention time comparison with authentic standards (glucose, fructose, sucrose, lactose, maltose), and quantification was achieved using external calibration curves prepared with standard concentrations ranging from 0.1 to 10 g/L. Results were expressed in g/100 g of sample.
The amino acid composition was determined according to the method described by Savych et al. [24] with slight modifications. Samples were first hydrolysed using hydrochloric acid (HCl, 6N) at 110°C for 24 h under a nitrogen atmosphere to release bound amino acids. After hydrolysis, the samples were cooled, filtered, evaporated to dryness under vacuum, and reconstituted in a borate buffer. Amino acids were then derivatised using o-phthalaldehyde (OPA) in combination with 2-mercaptoethanol, enabling fluorescence detection. The analysis was carried out using high-performance liquid chromatography (HPLC) on a C18 reverse-phase column, employing a gradient elution with a mobile phase consisting of phosphate buffer and an organic solvent mixture (acetonitrile/methanol/water). Fluorescence detection was performed at 340 nm excitation and 455 nm emission. Amino acids were identified by comparing retention times with commercial standards, and quantification was performed using external calibration. Results were expressed as mg of amino acid per 100 g of dry sample.
Preparation of pineapple peels hydrolysate
The method described by Pumiput et al. [25] was used with slight modifications to prepare pineapple peel hydrolysate. A 4 g portion of the substrate was steam-treated in an autoclave (Sonoclav, Germany) at 121°C (15 psi) for 20 min. After cooling, distilled water was added to adjust the final volume to 100 mL. The mixture was then boiled at 100°C for 30 min in a water bath (HH-W420 Coslab, India). The resulting boiled mixture was sieved through cotton cloth and acid hydrolysis was thereafter performed by autoclaving the filtrate at 121°C (15 psi) for 30 min with HCl (1% v/v of 6N solution). The pH of the hydrolysate was adjusted to 6.5 using calcium carbonate (98%), followed by filtration through Whatman No. 1 filter paper.
Preparation of black soldier fly larvae cake hydrolysate
Acid hydrolysis of the BC was performed based on conditions selected from the literature and preliminary laboratory work (clearness of the solution and amino acids profile of hydrolysate). Thus, 8 g of BC was hydrolysed in 99 mL of distilled water with 1 mL HCl (6N) at 160.5°C for 17.5 min in an oven (Memmert, Germany). After hydrolysis, the samples were centrifuged, and the supernatants were neutralised using NaOH (6N).
Development of the low-cost culture medium
Mixing design.
In search of the optimal proportions of ingredients that can support the growth and metabolite production by the LAB strains, a simplex centroid design was implemented. Specifically, a three-factor extreme peak mixing plan (BC hydrolysate, pineapple peel hydrolysate and 4% sugarcane molasses solution) was used to define the proportions of each mixture component [26]. Preliminary work and the literature were used to establish the extremes. Table 2 below shows the experimental range of factors.
Once the ranges of the different factors were determined, the calculations for converting the coded variables into real variables for a 3-factor mixing design were carried out, and the results are reported in Table 3 below, with nine experiments carried out. All these experiments were repeated three times.
Assessment of the performances of media formulated on Biomass production
Nine media were formulated according to the experimental matrix and inoculated with 1% (v/v) of a standardised (McFarland 4) suspension of each strain and incubated at 30°C or 37°C for 18 h, in aerobic or anaerobic conditions, depending on the strain. Viable cell counts were determined using the plate count method in MRS and M17 agar at 30 or 37°C for 48 h. The bacterial growth was estimated by difference between the counts at final and initial times of incubation and expressed in log CFU/mL. MRS and M17 broth media were used as control.
Assessment of the performances of media formulated on bacteriocin production
The agar well diffusion method described by Kaktcham et al. [27] was used. After 18 h of incubation, cultures were centrifuged at 10.000 × g for 10 minutes at 4°C. The supernatants were neutralised to pH 6.5 using NaOH (1N), filtered through a 0.22 µm membrane and heat-treated at 100°C for 5 min (to destroy H2O2). Wells of 6 mm diameter were punched into BHI or MRS agar plates previously seeded on the soft second layer with an indicator strain (S. aureus NCDC 109, E. coli ATCC 11775 and P. acidilactici) and 100 µL of treated supernatant was added to each well. After 24 h of incubation at 30 or 37°C. Inhibition zones ≥ 2 mm were considered indicative of bacteriocin activity. MRS and M17 broth media were used as controls.
Model fitting of mixture design and definition of the compromise medium
The experimental data obtained from the simplex centroid mixture design with three components were used and modeled using a commonly applied linear model:
The coefficients β1, β2, and β3 represent the linear effects of each component on the response, whereas β12, β13, and β23 indicate the interactions between pairs of components. Positive coefficients suggest stimulation of growth or metabolite production, while negative coefficients indicate inhibition.
Model fitting was performed using the least squares method with the aid of Minitab 18 software. Response surface plots were generated based on the fitted model to visualize the combined effect of the different proportions on the dependent variable. Analysis of variance (ANOVA) was carried out to assess the validity of the model. A significance level of 5% (p < 0.05) was used to determine whether the mixture components have a statistically significant effect on the measured response. For the model to be validated, the following criteria should be respected: coefficient of determination (R²) ≥ 0.75; absolute mean deviation analysis of the (AMDA) =0 and the bias factor 0.75<Bf<1.25. Statistical analyses enabled the optimisation of the formulation by targeting specific responses to be maximised or minimised. The software proposed optimal conditions under which the experimental results closely matched the predicted values. The compromise in a mixture design was achieved through a desirability function approach (>0.75), which combines multiple response variables into a single objective function. This method allows a balanced optimisation based on predefined priorities.
Establishment of the formula of the compromise medium
To ensure the reproducibility of the culture medium formulation using the same categories of agro-industrial by-products regardless of their geographical origin or batch-specific composition the final medium formula was established based on the physico-chemical composition of the tested hydrolysates and by-products (hydrolysate of BC, hydrolysate of pineapple peels and 4% sugarcane molasses solution). The fresh matter equivalent to the volume of each proportion of the ingredients was obtained by multiplying the density by this volume. Then, the dry matter content of the volume of each proportion (X, Y or Z) of the ingredients was obtained by multiplying the % dry matter of the ingredient by the fresh matter equivalent to this volume. Thereafter, the quantity of protein for instance (aiX) in each volume of ingredient was obtained by multiplying the % protein in the dry matter by the dry matter (X) of the ingredient volume. Given that the total volume of the medium is 100 mL, and to report the values as concentrations, the final quantities of proteins, lipids and carbohydrates were multiplied by 10. Hence the formula:
Where: X = dry matter content (g) in 55.1508 mL of BC hydrolysate; Y = dry matter content (g) in 19.849 mL of pineapple peel hydrolysate; Z = dry matter content (g) in 25.00 mL of 4% sugarcane molasses solution. aᵢ, bᵢ, cᵢ represent the protein, lipids, and carbohydrates contents (%DM) in hydrolysates and sugarcane molasses solution; P, L and C represent total proteins, lipids, and carbohydrates contents (g/L) of the formulated compromise culture medium.
Statistical analysis
Statistical analysis was conducted using Minitab 18 and SPSS version 22 softwares. Minitab was used to design the experimental and mixture designs and to analyze response data. Desirability functions were generated to determine optimal formulations for bacterial growth and bacteriocin production. Model performance was evaluated using the coefficient of determination (R²) as previously stated. Elsewhere, the results of bacterial growth and bacteriocin production obtained from the nine runs were expressed as means ± standard deviation. Analysis of variance (ANOVA) was applied to assess significant differences between means at a 5% significance level (p < 0.05), followed by Duncan’s multiple range test for pairwise comparisons when the means were different, using SPSS (version 22).
Results
Physico-chemical analysis of the by-products
Table 4 below and S1 Table present the macronutrient and mineral compositions of the three by-products assayed. BC had the highest protein content (48.47 ± 1.4%) and the lowest carbohydrate content (26.27 ± 1.17%). In contrary, pineapple peels and sugarcane molasses were rich in sugars (89.52 ± 0.16% and 86.86 ± 0.07% respectively) and contained lower protein levels (4.89 ± 0.05% and 7.02 ± 0.01%). These by-products were also found to be sources of minerals, with potassium (K) being the most abundant (2.872 ± 0.05 mg/100g) in sugarcane molasses, followed by calcium (Ca) and magnesium (Mg) in BC (1.252 ± 0.14 mg/100g and 0.437 ± 0.03 mg/100g respectively). Manganese (Mn), an essential trace element for lactic acid bacteria, was also detected in small amounts (0.0171 ± 0.24 mg/100g).
The amino acid profile of the three by-products (S2 and S3Tables) shows that BC contained all 17 amino acids for which standards were available, with the significantly (p < 0.05) highest concentrations of histidine, glycine, aspartic acid, alanine and leucine. In contrast, sugarcane molasses and pineapple peel contained only 8 and 6 amino acids respectively and displayed significantly (p < 0.05) higher concentrations of lysine (1.99 ± 0.17 and 1.73 ± 0.01 µmol/mL respectively), with the significantly highest (p < 0.05) value recorded in pineapple peel.
The sugar profiles of the by-products are shown in Table 5 and S4 Table. All three substrates contained mainly sucrose, glucose, and fructose, but not maltose and galactose. Molasses had the significantly (p < 0.05) higher sucrose content, followed by pineapple peels. The latter was rather the significantly richest in glucose and fructose, followed by sugarcane molasses. The BC does not contain glucose.
Performances of the low-cost culture media formulated
Experimental trials and biomass production.
Based on the experimental matrix provided by the mixture design, nine media were formulated and used to assess strains’ biomass and bacteriocin productions as responses. The results of colony count after 48 h of incubation are presented in Table 6 and S5 Table. The tested strains could grow correctly in formulated media and some of them even matched or outperformed their growth on the reference commercial media MRS and M17. For Lactiplantibacillus plantarum 5602, the bacterial count ranged from 9.04 ± 0.03 to 9.36 ± 0.10 log CFU/mL. The significantly (p < 0.05) highest value was obtained with trial 5 and was significantly (p < 0.05) higher than the count obtained in MRS (9.20 ± 0.24 log CFU/mL). For Lacticaseibacillus rhamnosus 347, the bacterial count ranged from 9.38 ± 0.27 to 9.69 ± 0.28 log CFU/mL. The significantly (p < 0.05) highest value was obtained with trial 9 and was equal to the count obtained in MRS (9.69 ± 0.15 log CFU/mL). Concerning Lactobacillus gasseri 5359, the bacterial count range was 9.36 ± 0.21 to 9.68 ± 0.23 log CFU/mL, the significantly (p < 0.05) highest values obtained with trials 2 and 5, also significantly (p < 0.05) higher than the value obtained with MRS (9.59 ± 0.20 log CFU/mL). With reference to Lactobacillus acidophilus 291, the growth range was 9.52 ± 0.30 to 9.73 ± 0.40 log CFU/mL, the significantly (p < 0.05) highest value, also significantly (p < 0.05) higher than that of MRS (9.62 ± 0.13 log CFU/mL), obtained with trials 2 and 5. As to Lactobacillus delbrueckii subsp. bulgaricus 293, the growth range was 9.52 ± 0.12 to 9.67 ± 0.21 log CFU/mL, the significantly (p < 0.05) highest value, also significantly (p < 0.05) higher than that of MRS (9.55 ± 0.25 log CFU/mL), obtained with trials 2 and 5. Considering Streptococcus thermophilus 295, the growth ranged 9.33 ± 0.19 to 9.64 ± 0.21 log CFU/mL, with the statistically highest values significantly (p < 0.05) higher than that in M17 (9.31 ± 0.11 log CFU/mL) and obtained with trials 2 and 5. As for Lactococcus lactis subsp. lactis MA2, the growth ranged 9.31 ± 0.22 to 9.48 ± 0.43 log CFU/mL, with the statistically highest value significantly (p < 0.05) higher than that in M17 (9.12 ± 0.59 log CFU/mL) and obtained with trial 5. With respect to and Lactococcus lactis subsp. lactis MF5, the growth range was 9.40 ± 0.31 to 9.59 ± 0.21 log CFU/mL, with the statistically highest value significantly (p < 0.05) higher than that in M17 (9.49 ± 0.28 log CFU/mL) and obtained with trial 2. Regarding Bifidobacterium bifidum 231, the biomass range was 9.52 ± 0.01 to 9.65 ± 0.17 log CFU/mL, the significantly (p < 0.05) highest values, also higher (p < 0.05) than that in MRS (9.53 ± 0.29 log CFU/mL), obtained with trials 1 and 6. With reference to Bacillus subtilis 215, the biomass range was 9.52 ± 0.04 to 9.63 ± 0.21 log CFU/mL, the significantly (p < 0.05) highest value, also higher (p < 0.05) than that in MRS (9.54 ± 0.00 log CFU/mL), obtained with trials 5. These results confirm the strong growth-supporting potential of the formulated alternative media, particularly for probiotics and technologically relevant strains. Trial 5 medium proved to highly support the growth of 70% of the strain, compared to 50% with trial 2.
Table 7 presents the growth ranges (CFU/mL), maximum values, and comparison with the control for each lactic acid bacteria strain. Strains showing higher growth than the control included Lactiplantibacillus plantarum 5602, Lactobacillus acidophilus 291, Lactobacillus gasseri 5359, Lactococcus lactis subsp. lactis MA2, Bifidobacterium bifidum 231, Lactobacillus delbrueckii subsp. bulgaricus 293 and Streptococcus thermophilus 295. Strains with growth comparable to the control were Lacticaseibacillus rhamnosus 347, Lactococcus lactis subsp. lactis MF5 and Bacillus subtilis 215. These results helped to identify the most favorable substrates for biomass production and guided the formulation of optimised culture media.
Proposed mathematical models for biomass production and contribution of factors
After analysing the effect of each proportion as well as the interactions of the by-products (proportions of BC hydrolysate, pineapple peel hydrolysate and sugarcane molasses solution) on the response (biomass), the mathematical model for each strain was predicted using the following regression equations:
Biomass of Lactiplantibacillus plantarum 5602
Biomass of Lacticaseibacillus rhamnosus 347
Biomass of Lactobacillus acidophillus 291
Biomass of Lactabacillus gasseri 5359
Biomass of Lactobacillus delbrueckii subsp. bulgaricus 293
Biomass of Streptococcus thermophilus 295
Biomass of Lactococcus lactis subsp. lactis MA2
Biomass of Lactococcus lactis subsp. lactis MF5
Biomass of Bifidobacterium bifidum 231
Biomass of Biomass of Bacillus subtilis 215
where: Y = Biomass (log CFU/mL); x1 = Black soldier fly larvae cake hydrolysate (% v/v); x2 = Pineapple peel hydrolysate (% v/v); x3 = Sugarcane molasses (% v/v).
The analysis of the regression equations highlights the respective contributions of linear effects and interactions among the media constituents on the cell viability of the tested strains. For all the strains, the SM (x3), the interaction of BCH and PPH (x1x2), as well as the interaction of PPH and SM (x2x3) showed positive effects on biomass production, except for Lacticaseibacillus rhamnosus 347. In addition, the linear effect of BCH (x1) was also positive on biomass production of Lactobacillus gasseri 5359, Streptococcus thermophilus 295, Bifidobacterium bifidum 231, Bacillus subtilis 215, Lactobacillus delbrueckii subsp. bulgaricus 293, and Lactococcus lactis subsp. lactis MA2 and MF5. This indicates that BCH and SM stimulated biomass production of the strains, while PPH did not contributed to this enhanced production. In addition to these positive linear effects, the positive interactions (x1x2 and x2x3,) observed for these by-products suggests that they act in synergy to enhance bacterial growth.
Validation of mathematical models for biomass production
The regression models established for predicting biomass production of lactic acid bacteria showed a good fit between experimental and predicted data. Determination coefficients (R²) were all greater than 0.83, while bias factors (Bf) ranged between 0.75 and 1.25, confirming the adequacy and reliability of the models. These findings validate the accuracy of the mixture design approach used to predict biomass responses. The detailed validation parameters are presented in S6 Table.
Iso-response curves for biomass production
Fig 1 below shows the different iso-response curves for each measured biomass, illustrating the optimal growth zones for the various strains: Lactiplantibacillus plantarum 5602(a), Lacticaseibacillus rhamnosus 347(b), Lactobacillus acidophilus 291(c), Lactobacillus gasseri 5359(d), Lactobacillus delbrueckii subsp. bulgaricus 293 (e), Streptococcus thermophilus 295 (f); Lactococcus lactis subsp. lactis MA2 (g), Lactococcus lactis subsp. lactis MF5 (h), Bifidobacterium bifidum 231(i) and Bacillus subtilis 215(j), according to the different proportions of BC hydrolysate, pineapple peel hydrolysate and a 4% sugarcane molasses solution
Lb plantarum Lactiplantibacillus plantarum 5602(a), Lb rhamnosus: Lacticaseibacillus rhamnosus 347(b), Lb acidophilus: Lactobacillus acidophilus 291(c), Lb gasseri: Lactobacillus gasseri 5359(d), Lb b: Lactobacillus delbrueckii subsp. bulgaricus 293 (e), S.t: Streptococcus thermophilus 295 (f); Lc. MA2: Lactococcus lactis subsp. lactis MA2 (g), Lc. MF5: Lactococcus lactis subsp. lactis MF5 (h), B. bifidum: Bifidobacterium bifidum 231(i) and B. subtilis: Bacillus subtilis 215 (j). BCH: black soldier fly larvae cake hydrolysate (% v/v); PPH: pineapple peel hydrolysate (% v/v); SM: 4% sugarcane molasses (%v/v).
The iso-response curves presented in (Fig 1) illustrate the variation in biomass production for the different bacterial strains tested, as a function of the relative proportions of three agro-industrial by-products: black soldier fly larvae cake hydrolysate (BCH), pineapple peel hydrolysate (PPH), and a 4% sugarcane molasses solution (SM).
Each ternary diagram displays contour lines representing equal levels of biomass response, allowing the identification of regions where optimal or suboptimal growth occurs for each strain. For Lactiplantibacillus plantarum 5602, the highest biomass values are located in the central region of the triangle, where the three components are balanced. This suggests a synergistic effect between carbon sources (molasses, pineapple) and nitrogen sources (BCH). with reference to Lacticaseibacillus rhamnosus 347, optimal biomass is concentrated toward the BCH-SM axis, indicating that Lacticaseibacillus rhamnosus 347 benefits from the combined presence of proteins and sugars, while PPH alone appears less effective. When it comes to Lactobacillus acidophilus 291, growth is maximal in the region rich in molasses, with moderate contributions from BCH. This suggests a strong preference for readily fermentable sugars, especially glucose and sucrose. As for Lactobacillus gasseri 5359, the optimum zone is skewed toward BCH-rich compositions, indicating that this strain is highly dependent on amino acids and peptides provided by the larvae hydrolysate. With regard to Lactobacillus delbrueckii subsp. bulgaricus 293, this strain reaches peak biomass in molasses-dominated mixtures, consistent with its known metabolic profile which favours high-energy carbon sources. The biomass of Streptococcus thermophilus 295 is maximised at intermediate concentrations of SM and BCH, with PPH having a limited effect, whilst the combination of sugars and moderate nitrogen supports this thermophilic strain. The optimal growth zone of Lactococcus lactis subsp. lactis MA2, lies along the PPH-SM axis, showing a strong affinity for fruit-derived sugars and molasses, what may reflect the strain’s adaptation to sugar-rich matrices. Concerning Lactococcus lactis subsp. lactis MF5, biomass increases toward the centre, suggesting balanced contributions from all three ingredients; the widely spaced iso-response lines indicate high tolerance to formulation shifts. Regarding Bifidobacterium bifidum 231, the highest biomass is found near BCH and SM, suggesting a dependency on both protein hydrolysates and fermentable sugars, while PPH has little impact. (j) The peak biomass of Bacillus subtilis 215 production occurs in BCH-rich regions, consistent with its proteolytic capacity and lesser dependence on simple sugars. Across all strains, the optimal growth zones are generally located in combinations involving BCH and SM, with PPH playing a supporting role or contributing partially. This reflects the importance of providing both nitrogen (from BCH) and carbon (from SM and PPH) sources in adequate proportions.
Compromise optimal mixture and validation of the predicted optimal biomass values of the strains
As previously noticed, optimal biomass of the strains was not obtained with one mixture (trial), but four of them (trials 1, 2, 5 and 6). Reason why, it was of utmost importance to define a compromise optimal mixture (optimal proportions of by-products) for obtaining optimal growth of all the strains at once. In this respect, the individual optimal mixtures of the strains were analysed using MINITAB 18 to define the optimal proportions of each by-product required to formulate the compromise optimal medium. Table 8 and S7 Table show that the optimal growth of all the strains requires a compromise medium with 55.1508% (v/v) BCH 19.849% (v/v) PPH and 25% (v/v) of a 4% SM solution.
Based on the compromise mixture, optimal biomass values were predicted by the MINITAB 18 software, and for their confirmation, further growth experiments of each strain in the compromise medium were carried out and the experimental optimal values obtained were compared to the predicted ones, as shown in Table 8. Overall, the experimental optimal biomass values obtained do not differ significantly (p<0.05) from the predicted ones, with a desirability of 0.95, thus validating the biomass values predicted by the software.
Bacteriocin production in media formulated based on experimental matrix
Table 9 below and S8 Table show the results of the antimicrobial activity tests of the neutralised cell-free supernatants of bacteriocin-producing strains against indicator strains. The inhibition zone diameters recorded for Bifidobacterium bifidum 231 against E. coli ATCC 11775 for the various trials ranged from 8.00 ± 0.42 to 11.50 ± 0.17 mm, with the significantly (p<0.05) highest value, also significantly higher than that obtained in MRS (11.00 ± 0.37 mm) obtained with trial 5. With regard to Lactiplantibacillus plantarum 5602, the recorded inhibition zone diameters against Staphylococcus aureus NCDC 109 for the various trials ranged from 7.00 ± 0.00 to 11.50 ± 0.40 mm, with the significantly (p<0.05) highest value, also significantly (p<0.05) higher than that obtained in MRS (11.00 ± 0.00 mm) obtained with trial 5. Concerning Lacticaseibacillus rhamnosus 347, the recorded inhibition zone diameters against Staphylococcus aureus NCDC 109 for the various trials ranged from 7.00 ± 0.07 to 11.50 ± 0.37 mm, with the significantly(p<0.05) highest value equal to that obtained in MRS (11.50 ± 0.27 mm) obtained with trial 5. As for Lactococcus lactis subsp. lactis MA2, the inhibition zone diameters recorded against Pediococcus acidilactici for the various trials ranged from 9.00 ± 0.07 to 14.00 ± 0.90 mm, with the significantly (p<0.05) highest value equal to that obtained in MRS (14.00 ± 0.40 mm) obtained with trial 5. With reference to Lactococcus lactis subsp. lactis MF5, the inhibition zone diameters recorded against Pediococcus acidilactici for the various trials ranged from 10 ± 0.07 to 14.00 ± 0.07 mm, with the significantly (p<0.05) highest value equal to that obtained in MRS (14.00 ± 0.00 mm) obtained with trial 2.
Table 10 presents the antimicrobial activities of bacteriocin-producing strains cultivated in the formulated low-cost media compared to the control (MRS/M17). Overall, the formulated media supported comparable or increased antimicrobial activity relatively to the control. Strains showing increased activity (p < 0.05) than in the control are Lactiplantibacillus plantarum 5602 and Bifidobacterium bifidum 231. Strains with comparable activity to that in the control are Lactococcus lactis subsp. lactis MF5, Lactococcus lactis subsp. lactis MA2 and Lacticaseibacillus rhamnosus 347.
Proposed mathematical models for Bacteriocin production and contribution of factors
After analysing the effect of each proportion as well as the interactions of the by-products (proportions of BC hydrolysate, pineapple peel hydrolysate and sugarcane molasses solution) on the response (bacteriocin production), the mathematical model for each strain was predicted using the following regression equations:
Antimicrobial activity of Lactiplantibacillus plantarum 5602
Antimicrobial activity of Lacticaseibacillus rhamnosus 347
Antimicrobial activity of Lactococcus lactis subsp. lactis MA2
Antimicrobial activity of Lactococcus lactis subsp. lactis MF5
Antimicrobial activity of Bifidobacterium bifidum 231
where: Y = Biomass (log CFU/mL); x1 = Black soldier fly larvae cake hydrolysate (% v/v); x2 = Pineapple peel hydrolysate (% v/v); x3 = Sugarcane molasses (% v/v).
The analysis of the regression equations highlights the respective contributions of linear effects and interactions among the media constituents on the antimicrobial activity of the tested strains. The linear effect of SM(x3), along with its interaction of BC hydrolysate and pineapple peel hydrolysate (x1x2) and pineapple peel and molasses hydrolysates (x2x3) contributed positively to the antimicrobial activity of all tested strains, namely Lactiplantibacillus plantarum 5602, Lacticaseibacillus rhamnosus 347, Lactococcus lactis subsp. lactis MA2 and MF5 and Bifidobacterium bifidum 231, highlighting the significance of these factors in optimising antimicrobial activity. In contrary, the linear effects of BCH (x1) and PPH (x2) were negative on bacteriocin production.
Validation of mathematical models for bacteriocin production
The coefficient of determination (R2) for the proposed models ranged from 85.82% to 96.26%, the absolute mean deviation analysis (AMDA) ranged from 0.00 to 0.02 and the bias factor (Bf) range was 0.99 to 1.05 (S9 Table). All the values of these parameters fall within the acceptable reference range to validate a model (R2 are ≥ 0.75; AMDA = 0; 0.75<Bf<1.25).
Iso-response curves for different responses for bacteriocin production
The iso-response curves corresponding to each measured response illustrate the optimal inhibition zones for the different strains, namely Lactiplantibacillus plantarum 5602(a), Lacticaseibacillus rhamnosus 347(b); Lactococcus lactis subsp. lactis MA2 (c), Lactococcus lactis subsp. lactis MF5 (d) Bifidobacterium bifidum 231(e). these results are presented as a function of varying proportions from black soldier fly larvae cakes hydrolysate (BCH), pineapple peel hydrolysate (PPH) and 4% sugarcane molasses (SM) solution, as shown in (Fig 2):
Lb plantarum Lactiplantibacillus plantarum 5602 (a); Lb rhamnosus: Lacticaseibacillus rhamnosus 347 (b); Lc. MA2: Lactococcus lactis subsp. lactis MA2 (c); Lc. MF5: Lactococcus lactis subsp. lactis MF5 (d) and B. bifidum: Bifidobacterium bifidum (e) 231. BCH: black soldier fly larvae cake hydrolysate (% v/v); PPH: pineapple peel hydrolysate (% v/v); SM: 4% sugarcane molasses (%v/v).
It reveals that each ternary diagram includes contour lines representing equal inhibition levels, making it possible to identify the conditions that promote either strong or weak antagonistic effects for each bacterial strain. For Lactiplantibacillus plantarum 5602 (a), the most pronounced inhibitory effect is observed in the central region of the triangle, where the three components are balanced. This suggests a synergistic interaction between carbon-rich (PPH, SM) and nitrogen-rich (BCH) substrates in enhancing bacteriocin production or other inhibitory metabolites. In contrast, Lacticaseibacillus rhamnosus 347 (b) showed maximal antimicrobial performance along the BCH-SM axis, indicating that this strain benefits particularly from the combination of protein and sugar sources. The use of PPH alone appears to be less effective in stimulating inhibitory activity. As for Lactococcus lactis subsp. lactis MA2 (c), optimal inhibition zones are located along the PPH-SM axis, reflecting a strong response to fruit-derived sugars and molasses, and possibly an adaptation of this strain to sugar-enriched environments for bacteriocin synthesis. Regarding Lactococcus lactis subsp. lactis MF5 (d), the wide space between contour lines indicates a broad tolerance to changes in formulation, suggesting stable antimicrobial performance across different compositions. Finally, Bifidobacterium bifidum 231 (e) exhibits optimal inhibition between PPH-SM, highlighting the key role of PPH and SM derived sugars in enhancing its antimicrobial potential.
Validation of predicted and experimental values of optimal bacteriocin production based on the compromise mixture
The optimum values predicted by the MINITAB 18 software and experimental values for each trial of the medium formulated under optimum conditions are presented in Table 11 below as well as S10 Table. Briefly, the values obtained after experimentation do not differ significantly (P<0.05) from the predicted results with a desirability of 0.95, thus validating the optimal conditions defined by the software.
Equation to establish the formula of the compromise culture medium
The formula of the compromise medium was defined based on the macronutrients composition of the by-products (BCH, PPH and 4% SM). The culture medium formula is based on the macronutrient composition of the various hydrolysates and by-product preparations. Table 12 below and S11 Table show the macronutrient composition of the various by-product hydrolysates.
Based on the formula in equations (1), (2) and (3), the formula of the medium was defined as:
Where: X = dry matter content (g) in 55.1508 mL of black soldier fly larvae cake hydrolysate; Y = dry matter content (g) in 19.849 mL of pineapple peel hydrolysate; Z = dry matter content (g) in 25.00 mL of 4% sugarcane molasses
Using this formula, the final macronutrient contents (g/L) in the medium was calculated and found to be: 14.99 g/L of proteins, 0.80g/L of lipids, and 24.67 g/L of carbohydrates.
Discussion
Determination of the physico-chemical composition of the by-products
The physico-chemical analysis of pineapple peels revealed contents of 0.75% lipids, 4.89% protein, and 89.52% carbohydrates. These values are significantly higher than those reported by Santos et al. [16], who observed 0.5%, 3%, and 60% respectively, in Brazilian samples. Similarly, Thatchajaree et al. [28] reported 2.53% lipids, 4.20% protein, and 73.30% carbohydrates in dried pineapple peels from Thailand. Such variations are likely due to factors including geographical origin, agro-ecological conditions, and post-harvest processing techniques [29]. Beyond their carbohydrate richness, pineapple peels also displayed a noteworthy amino acid profile. The analysis revealed a predominance of lysine, followed by proline, aspartic acid, glycine, alanine, histidine, serine, glutamic acid, and cysteine. These amino acids, particularly lysine, are essential for the growth of several lactic acid bacteria (LAB), such as Lactobacillus acidophilus and Bifidobacterium bifidum, which are auxotrophic for this compound. The presence of both sugars and amino acids in this by-product makes it a dual-purpose ingredient in culture medium formulation. As for sugarcane molasses, the results indicated 0% lipids, 7.02% protein, and a high total sugar content of 86.86%. These values differ from those reported by Feumba et al. [29], who found 1% lipids, 6% protein, and 65% sugars. Such discrepancies may stem from differences in sugar extraction processes, as well as pedo-climatic and cultivation factors [30]. The sugar profile was dominated by sucrose (579.38 ± 2.40 g/100 g), followed by glucose (195.79 ± 0.10 g/100 g) and fructose (112.47 ± 0.5 g/100 g). These fermentable sugars are widely recognised as readily assimilable carbon sources for LAB such as Lb. plantarum, Lb. acidophilus, and B. bifidum. Thus, molasses serves as a key source of fermentable carbon in the tested formulations. Simultaneously, the BC showed 15% lipids, 48% protein, and 26.27% total sugars. These findings vary slightly from those of Ahmad et al. [15] who reported lipid contents ranging from 10 to 27.65%, protein levels at 55%, and sugars between 10 and 20%. Similarly, Christian et al. [31] reported 25.69 ± 0.12% lipids and 48.20% proteins. Such differences are likely attributable to the larvae’s diet, rearing conditions (temperature, humidity), and physiological stage at harvest [32]. The amino acid profile of this ingredient revealed the presence of 17 amino acids, dominated by histidine, glutamic acid, aspartic acid, glycine, and alanine. These results are consistent with those of Christian et al. [31] who identified 18 amino acids with a similar composition. These compounds play an essential role in protein synthesis and metabolic regulation in various probiotic strains, notably Lb. gasseri, S. thermophilus, and Lb. bulgaricus, which exhibit multiple amino acid requirements. To optimise the formulation, a simplex-centroid mixture design was applied. This experimental approach allowed the modeling of combined ingredient effects on the biological response (biomass production). The resulting iso-response curves highlighted positive interactions among the tested components. Notably, the central region of the mixture triangle corresponding to balanced proportions was frequently associated with optimal responses. This suggests a synergistic effect, in which the blend of simple sugars (from molasses and pineapple peels) and amino acids (mainly from larvae cake) provides a nutritionally favorable environment for LAB growth. This observation aligns with findings by Oonincx et al. [33] who used a mixture design to optimise the growth of Latilactobacillus sakei using various food-derived nitrogen sources. Their study demonstrated that specific combinations yielded better outcomes than any individual component alone. Similar conclusions were drawn by Galante et al. [34] and Cárdenas et al. [35] who confirmed the relevance of this method for optimising agro-industrial by-product-based media, particularly by capturing complex nutritional interactions. The combined use of pineapple peels, sugarcane molasses, and BC provides a complementary supply of fermentable sugars, proteins, and essential amino acids, meeting the nutritional requirements of lactic acid and probiotic bacteria. The mixture design approach enabled the identification of optimal proportions and demonstrated significant nutritional synergy, validating the potential of these agro-industrial by-products as effective, sustainable alternatives to conventional culture media
Development of the by-products-based low-cost culture medium
Performances of the formulated medium on biomass production.
In this study, nine strains of lactic acid bacteria (LAB) and one non-LAB strain with probiotic and technological interest were evaluated for their growth on a culture medium formulated from agro-industrial by-products. After 48 hours of incubation, high cell viabilities were observed, demonstrating that the formulated medium, enriched by enzymatic hydrolysis of by-products, provides an optimised environment for bacterial growth. In some cases, growth was even higher than that obtained with conventional commercial media such as MRS and M17. The high biomass levels reached (between 9.04 and 9.73 log CFU/mL) highlight the effectiveness of combining hydrolysed by-products. The enzymatic hydrolysis of pineapple peels and BC released fermentable sugars, peptides, proteins, amino acids, polyphenols, and minerals, thus increasing nutrient bioavailability for the bacteria. Previous studies have reported that enzymatic or acid hydrolysis improves the digestibility and accessibility of substrates, thereby promoting microbial growth [36]. Notably, strains such as Lacticaseibacillus rhamnosus and Lactiplantibacillus plantarum have demonstrated efficient metabolism of pineapple peel hydrolysates [37]. Furthermore, diluted molasses has been identified as an excellent substrate for growth enhancement in LAB [38]. The positive linear effects of BC hydrolysate and molasses could be explained by their nutritional contents, as they provide carbon, nitrogen and minerals requirements for bacterial growth. The positive interactions between BC hydrolysate and molasses on the one hand, and pineapple peel hydrolysate and molasses on the other hand suggest a complementary of synergism in the nutrients provided by each source. These patterns agree with previous observations that the synergistic balance between nitrogen-rich and carbohydrate-rich by-products improves LAB performance in mixed-substrate media [39]. The mixture design method was successfully applied in recent microbial culture medium formulations. This approach not only enables a cost-effective alternative to commercial media but also fits within the principles of a circular economy by valorising agro-industrial by-products. It simplifies laboratory and industrial protocols while respecting the strict nutritional requirements of lactic acid bacteria [40].
Performances of the formulated medium on bacteriocin production.
Bacteriocins are characterised by strong auxotrophic requirements due to their limited ability to synthesise certain vitamins and amino acids. Their growth is only possible in rich and complex culture media containing mainly: a nitrogen source (peptides, amino acids), a carbon source, vitamins, and minerals. These nutrients must be provided in optimal concentrations [41]. Five bacteriocin-producing lactic acid bacteria strains were studied, the antimicrobial activity test of the bacteriocin-producing strains showed inhibition diameters of 11 mm for Lactiplantibacillus plantarum 5602, 11.5 mm for Lacticaseibacillus rhamnosus 347, 14 mm for Lactococcus lactis subsp. lactis MA2, 14 mm for Lactococcus lactis subsp. lactis MF5, and 11 mm for Bifidobacterium bifidum 231. All strains exhibited equal or higher activity in the formulated medium compared to MRS/M17 media. This could be attributed to their ability to efficiently metabolise the sugars and amino acids present in the hydrolysates of pineapple peels and BC. These results are consistent with those of Dabrowska et al. [11] who demonstrated that the addition of hydrolysed proteins to a culture medium improves bacterial growth and bacteriocin production by providing a more accessible nitrogen source and, in some cases, increased bioactivity. Similarly, Abbasiliasi et al. [42] showed that some probiotic strains enhance biomass production yield when hydrolysates are added to the formulated medium. The good performance of the formulated medium can also be explained by its richness in essential amino acids such as lysine, leucine, and valine mainly derived from BC, which is recognised for its high protein content and well-balanced amino acid profile [43,44]. Pineapple peels and sugarcane molasses additionally provide fermentable sugars and minerals, creating a complete and highly bioavailable nutritional environment. These components are essential for cell growth and the biosynthesis of antimicrobial compounds such as bacteriocins. The results of this study thus confirm the relevance of these agro-industrial by-products in formulating low-cost culture media for bacterial biomass and bacteriocin production. Bacteriocin production depends on the cell density and metabolic activity of lactic acid bacteria. It has been shown that bacteriocin synthesis is often correlated with bacterial growth, particularly during the exponential phase when cells are metabolically active [41]. The availability of nitrogen and other nutrients such as proteins and sugars influenced both growth and production of antimicrobial metabolites [45].
Compromise medium and its formula
The compromise medium formulated contains 55.15 mL of BC hydrolysate, 19.84 mL of pineapple peel hydrolysate, and 25 mL of 4% sugarcane molasses solution per 100 mL of final medium. This medium offers a balanced composition of essential macronutrients, notably proteins, lipids, and carbohydrates. It is particularly rich in simple sugars such as glucose and fructose, which are rapidly metabolised by lactic acid bacteria, promoting an initial fast growth phase. Moreover, the presence of more complex sugars like sucrose allows a gradual release of carbohydrate sources, ensuring a sustained energy supply during fermentation. This sequence of sugar utilisation is supported by Corsetti et al. [46] who highlighted that lactic acid bacteria preferentially consume simple sugars before progressively metabolising complex sugars, thus optimising their growth and metabolic activity. The lipids present in this medium, although at low concentrations, play a crucial role in membrane structure and protection against environmental stress. According to Srinivas et al. [47] lipids contribute to cell membrane stability and enhance the resistance of lactic acid bacteria to stress conditions such as freeze-drying and other adverse environments, thereby improving the viability and productivity of the cultures. Proteins derived mainly from the BC provide a significant source of amino acids. This substrate contains 17 essential and non-essential amino acids, offering a comprehensive nutrient profile for lactic acid bacteria. The availability of these amino acids is critical since growth factor requirements vary considerably among bacterial strains. Indeed, Zhao et al. [48] demonstrated that lactic acid bacteria differ in their capacity to produce and utilise extracellular amino acids, implying that each strain has specific nutritional needs. This variability underscores the importance of a custom-formulated culture medium capable of meeting the specific preferences of microorganisms to optimise their growth and bacteriocin production. Thus, the medium formulated from agro-industrial by-products not only provides a complete nutritional composition but also presents a resource utilisation dynamic adapted to the metabolic requirements of lactic acid bacteria. This formula results in good bacterial growth and bacteriocin production performances while valorising local, economical, and sustainable resources.
Conclusion
This study was aimed to formulate a low-cost culture medium able to sustain the growth and beneficial metabolites production of a wide genera of lactic acid bacteria, from agro-industrial and environmental by-products.
Based on their nutrient contents, black soldier fly larvae cake, pineapple peels and sugarcane molasses were found to be good nitrogen and carbon sources for LAB growth and bacteriocin production. The use of experimental mixing design made it possible to obtain optimal proportions of the ingredients in the medium as 55.15% larvae cake hydrolysate, 19.85% pineapple peel hydrolysate, and 25.00% of 4% sugarcane molasses solution. A compromise medium that optimally sustain the growth and bacteriocin production of bacterial strains from various genera and species, as well as with various purposes (technological properties, probiotics and bacteriocin production) was derived. Its formula was also defined, so as to ease its preparation using these ingredients, but from different origins. Interestingly, the formulated compromise medium performed more to similar than MRS or M17 for strains cultivation and bacteriocin production. By valorising industrial by-products, this study fits into a circular economy and sustainable development model. It opens promising perspectives for the use of this medium at lab-scale for research, or at pilot and industrial-scales (agri-food, pharmaceutical, and cosmetic) for the production of the biomass of lactic starters or probiotics, and bacteriocins.
Supporting information
S1 Table. Physico-chemical composition of BC, pineapple peel and sugarcane molasses (Dataset).
https://doi.org/10.1371/journal.pone.0337684.s001
(PDF)
S2 Table. Amino acids profiles of the various by-products (BC, pineapple peel and sugarcane molasses).
https://doi.org/10.1371/journal.pone.0337684.s002
(PDF)
S3 Table. Amino acids profiles of the by-products (BC, pineapple peel, sugarcane molasses) (Dataset).
https://doi.org/10.1371/journal.pone.0337684.s003
(PDF)
S4 Table. Sugar profiles of the various by-products (BC, pineapple peels, sugarcane molasses) (Dataset).
https://doi.org/10.1371/journal.pone.0337684.s004
(PDF)
S5 Table. Strains’ biomass production in the media formulated based on experimental matrix (Dataset).
https://doi.org/10.1371/journal.pone.0337684.s005
(PDF)
S6 Table. Validation of mathematical models for biomass production.
https://doi.org/10.1371/journal.pone.0337684.s006
(PDF)
S7 Table. By-products proportions of the compromise medium and validation of biomass production model (Dataset).
https://doi.org/10.1371/journal.pone.0337684.s007
(PDF)
S8 Table. Strains’ bacteriocin production in the media formulated based on experimental matrix (Dataset).
https://doi.org/10.1371/journal.pone.0337684.s008
(PDF)
S9 Table. Validation of mathematical models for bacteriocin production.
https://doi.org/10.1371/journal.pone.0337684.s009
(PDF)
S10 Table. By-products proportions of the compromise medium and validation of bacteriocin production model (Dataset).
https://doi.org/10.1371/journal.pone.0337684.s010
(PDF)
S11 Table. Macronutrients composition of BC hydrolysate, pineapple peel hydrolysate and sugarcane molasses (Dataset).
https://doi.org/10.1371/journal.pone.0337684.s011
(PDF)
Acknowledgments
The authors thank the CV Raman Fellowship for African Researches, for supporting a research stay in India during which analyses of amino acid profile, sugar profile, and mineral content of the by-products were performed.
References
- 1. Choi KR, Jiao S, Lee SY. Metabolic engineering strategies toward production of biofuels. Curr Opin Chem Biol. 2020;59:1–14. pmid:32298980
- 2. Timmis K, Ramos J, Verstraete W, de Vos W, Ramos A, Rappuoli R. Scientists’ call to action: Microbes, planetary health, and the SDGs. Nat Rev Microbiol. 2023;21(11):624–6.
- 3. De Man JC, Rogosa M, Sharpe ME. The medium for the cultivation of lactobacilli. J Appl Bacteriol. 1960;23(1):130–5.
- 4. Krzywonos M, Eberhard T. High density process to cultivate Lactobacillus plantarum biomass using wheat stillage and sugar beet molasses. Electron J Biotechnol. 2011;14(2).
- 5. Canon F, Maillard M-B, Henry G, Thierry A, Gagnaire V. Positive Interactions between Lactic Acid Bacteria Promoted by Nitrogen-Based Nutritional Dependencies. Appl Environ Microbiol. 2021;87(20):e0105521. pmid:34347516
- 6. Linares-Morales JR, Salmerón-Ochoa I, Rivera-Chavira BE, Gutiérrez-Méndez N, Pérez-Vega SB, Nevárez-Moorillón GV. Influence of Culture Media Formulated with Agroindustrial Wastes on the Antimicrobial Activity of Lactic Acid Bacteria. J Microbiol Biotechnol. 2022;32(1):64–71. pmid:34675139
- 7. Pandit D, Bhunia S, Bitla G, Gupta H. Optimization of MRS (De Man, Rogosa and Sharpe) media with an organic waste substitute to assess the potential of Lactobacillus in ink degradation. Int J Aquat Sci. 2021;8(5):566–74.
- 8.
Keltboun A. Valorization of slaughterhouse by-products for their use as substrates for the formation of culture media for certain lactobacilli. PhD thesis, Badji Mokhtar University.2018.
- 9. Nochebuena-Pelcastre X, Álvarez-Contreras AK, Hernández-Robles MF, Natividad-Bonifacio I, Parada-Fabián JC, Quiñones-Ramirez EI, et al. Development of a low pollution medium for the cultivation of lactic acid bacteria. Heliyon. 2023;9(12):e22609. pmid:38094054
- 10. Doneva M, Loginovska K, Dyankova S, Nacheva I, Metodieva P, Ninova-Nikolova N. Application of whey protein hydrolysates as a fermentation medium with strains of lactic acid bacteria. J Chem Technol Metall. 2024;59(1):15–22.
- 11. Dąbrowska A, Babij K, Szołtysik M, Chrzanowska J. Viability and growth promotion of starter and probiotic bacteria in yogurt supplemented with whey protein hydrolysate during refrigerated storage. Postepy Hig Med Dosw (Online). 2017;71(0):952–9. pmid:29176008
- 12. Kaktcham PM, Foko Kouam EM, Tchabou Tientcheu ML, Temgoua J-B, Wacher C, Zambou Ngoufack F, et al. Nisin-producing Lactococcus lactis subsp. lactis 2MT isolated from freshwater Nile tilapia in Cameroon: Bacteriocin screening, characterization, and optimization in a low-cost medium. LWT. 2019;107:272–9.
- 13. Valle Vargas MF, Villamil Diaz LM, Ruiz Pardo RY, Quintanilla Carvajal MX. Design of an agro-industrial by-products-based media for the production of probiotic bacteria for fish nutrition. Sci Rep. 2024;14(1):17955. pmid:39095475
- 14. Acosta-Piantini E, Rodríguez-Díez E, Chavarri M, López-de-Armentia I, Villaran MC, Lombraña JI. Preparation of Hydrolyzed Sugarcane Molasses as a Low-Cost Medium for the Mass Production of Probiotic Lactobacillus paracasei ssp. paracasei F19. Separations. 2023;10(1):33.
- 15. Ahmad I, Ullah M, Alkafafy M, Ahmed N, Mahmoud SF, Sohail K, et al. Identification of the economics, composition, and supplementation of maggot meal in broiler production. Saudi J Biol Sci. 2022;29(6):103277. pmid:35401021
- 16. Santos A, Almeida F, Souza R, Silva D. Physico-chemical characterisation of Brazilian pineapple peels: potential use as a functional ingredient. Food Sci Technol. 2020;40(1):180–7.
- 17. Carlozzi P, Touloupakis E, Di Lorenzo T, Giovannelli A, Seggiani M, Cinelli P, et al. Whey and molasses as inexpensive raw materials for parallel production of biohydrogen and polyesters via a two-stage bioprocess: New routes towards a circular bioeconomy. J Biotechnol. 2019;303:37–45. pmid:31351109
- 18. Ligarda-Samanez CA, Huamán-Carrión ML, Calsina-Ponce WC, Cruz GD la, Calderón Huamaní DF, Cabel-Moscoso DJ, et al. Technological innovations and circular economy in the valorization of agri-food by-products: advances, challenges and perspectives. Foods. 2025;14(11):1950. pmid:40509479
- 19. Zambou NF, Sieladie D, Fonteh A, Kaktcham PM, El –Soda M. Characterization and selection of lactobacilli with probiotic properties from raw cow milk of “Bororo” cattle breeders in Dschang (Cameroon). Curr Res J Biol Sci. 2012;4(2):143–52.
- 20. International Union of Pure and Applied Chemistry IUPAC. Method of analysis of fats. Int Compend Health Legis. 1979;46(2):241.
- 21.
Association of Official Analytical Chemists (AOAC). Official Methods of Analysis. 13th ed. Washington, DC: Dr. William Horwitz; 1980.
- 22. Savic S, Petrovic S, Petronijevic M, Cvetanovic A, Petronijevic Z. Determination of mineral content in spices by ICP-OES. Adv Technol. 2019;8(1):27–32.
- 23.
Ng CM, Reuter WM. Analysis of sugars in honey using the Perkin Elmer Altus HPLC system with RI detection. 2015.
- 24. Savych A, Polonets O, Morozova L, Syrovatko K, Recun T. HPLC-FLD analysis of amino acids content in Chrysanthemum morifolium. PHAR. 2022;69(2):337–43.
- 25. Pumiput P, Chuntranuluck S, Kitpreechavanich V, Punsuvon V, Vaithanomsat P. Production process of hydrolysate from steam explosion of oil palm trunk for xylitol fermentation. Agric Nat Resour. 2008;42(1):73–8.
- 26.
Goupy J, Creighton L. The method of experimental designs: optimizing the choice of tests and the interpretation of results. 3rd ed: Dunod Paris; 2006.
- 27. Kaktcham PM, Zambou NF, Tchouanguep FM, El-Soda M, Choudhary MI. Antimicrobial and Safety Properties of Lactobacilli Isolated from two Cameroonian Traditional Fermented Foods. Sci Pharm. 2012;80(1):189–203. pmid:22396914
- 28. Thatchajaree S, Khamhaengpol A, Suttajit M. Nutritional composition of dried pineapple peel powder and its application in food formulation. Int J Food Sci Nutr. 2024;75(1):120–7.
- 29. Feumba R, Mejiato P, Tatsadjieu L. Influence of agro-ecological conditions on the physico-chemical composition of pineapple peels in Cameroon. J Food Res. 2016;5(5):45–52.
- 30. Lemoufouet J, Ngassoum M, Tatsadjieu L. Physico-chemical characterisation of Cameroonian sugarcane molasses and evaluation of its potential for microbial growth. Afr J Biotechnol. 2016;15(5):112–8.
- 31. Christian M, Tchotchoua M, Ndjouenkeu R. Variability of physico-chemical parameters of sugarcane and its by-products in relation to environmental and agronomic factors. Agric Sci. 2023;14(1):34–42.
- 32. Nor Fatin N, Abu Hassan M, Lani M. Amino acid profile of black soldier fly larvae fed with different organic waste substrates. Environ Sci Pollut Res. 2022;29(24):35000–11.
- 33. Oonincx D, van Huis A, van Loon J. Nutritional quality of insect larvae fed with organic waste stream. PLoS One. 2015;10(12):e0144601.
- 34. Galante L, Fernández M, Benito M, Córdoba J. Optimisation of food-derived nitrogen sources for the growth of Lactobacillus sakei using a mixture design approach. Sci Rep. 2022;12:1835.
- 35. Cárdenas J, Peña A, Poutou R. Optimization of culture media for lactic acid bacteria using alternative substrates: A review. Biocatal Agric Biotechnol. 2020;24:101548.
- 36. Mokhtari A, Gharachorloo M, Madadlou A. Media formulation for lactic acid bacteria: mixture design modelling and experimental verification. LWT Food Sci Technol. 2021;148:111737.
- 37. Zhang X, Jiang Y, Liu H. Enzymatic hydrolysis improves substrate digestibility and microbial growth in fermentation processes. Biochem Eng J. 2020;161:107682.
- 38. Nkan L, Ndjouenkeu R, Nwafor O. Effect of pineapple peel hydrolysate on the growth of Lacticaseibacillus rhamnosus. J Appl Microbiol. 2014;117(3):772–81.
- 39. Gonzalez-del Rosario I, Hernández-Martínez JM, Bolaños-Reynoso E, Rosas-Mendoza ES, Méndez-Contreras JM. Kinetic parameters of Lactobacillus acidophilus growth in the lactic fermentation of non-centrifugal cane sugar agroindustry wastes. Renew Energ Biomass Sustain. 2024;6(1):43–9.
- 40. Nurkhamidah S, Wahyudi A, Saputra H. Utilization of diluted molasses as an alternative substrate for lactic acid bacteria growth. Int J Microbiol. 2018.
- 41. Singh R, Kumar S, Sharma P. Optimization of culture medium components for enhanced microbial biomass production using mixture design. Biotechnol Rep. 2021;29:e00562.
- 42. Abbasiliasi S, Tan JS, Tengku Ibrahim TA, Bashokouh F, Ramakrishnan NR, Mustafa S, et al. Fermentation factors influencing the production of bacteriocins by lactic acid bacteria: a review. RSC Adv. 2017;7(47):29395–420.
- 43. Kareba M, Zúñiga M, Monedero V. Improvement of biomass production in Lactobacillus plantarum using protein hydrolysates: Optimization by response surface methodology. J Appl Microbiol. 2018;124(6):1461–71.
- 44. Barragan-Fonseca KB, Dicke M, van Loon JJA. Nutritional value of the black soldier fly (Hermetia illucens L.) and its suitability as animal feed – a review. JIFF. 2017;3(2):105–20.
- 45. van Huis A. Insects as food and feed, a new emerging agricultural sector: a review. JIFF. 2020;6(1):27–44.
- 46.
Corsetti A, De Marco A. Food microbiology: Fundamentals and frontiers in lactic acid bacteria metabolism. Lactic Acid Bacteria: Microbiological and Functional Aspects. 5th ed. CRC Press. 2017.
- 47. Costa S, Summa D, Radice M, Vertuani S, Manfredini S, Tamburini E. Lactic acid production by Lactobacillus casei using a sequence of seasonally available fruit wastes as sustainable carbon sources. Front Bioeng Biotechnol. 2024;12:1447278. pmid:39157446
- 48. Gao X, Kong J, Zhu H, Mao B, Cui S, Zhao J. Lactobacillus, Bifidobacterium and Lactococcus response to environmental stress: Mechanisms and application of cross-protection to improve resistance against freeze-drying. J Appl Microbiol. 2022;132(2):802–21. pmid:34365708