Optimization of the Production Process of a Fermented Mango-Based Beverage with Lactiplantibacillus plantarum (Lp6 and Lp32)

Abstract

This study aimed to develop a fermented mango-based beverage using Lactiplantibacillus plantarum strains Lp6 and Lp32, focusing on enhancing its functional properties, ensuring microbiological safety, improving nutritional value, and achieving sensory acceptability. A central composite design (CCD) was employed to assess the effects of two factors (fermentation time and inoculum concentration) on several response variables: viable cell concentration (CC), total phenolic compounds (TPCs), total flavonoid compounds (TFCs), and concentrations of L-lactic acid and D-lactic acid. The optimized formulation was achieved using L. plantarum Lp6, with an inoculum concentration of 9.89 Log (7.76 × 10⁹) CFU/mL and a fermentation time of 20.47 h. Under these conditions, the beverage reached the highest values for CC, TPC, TF, and L-lactic acid while minimizing the production of D-lactic acid. Following optimization, the fermented beverage underwent further characterization, including physicochemical analysis, microbiological evaluation, proximate composition analysis, and sensory evaluation. The final product exhibited a viable cell count of 13.01 Log (10.23 × 10¹²) CFU/mL, demonstrated functional potential, complied with microbiological safety standards, and showed adequate nutritional content. Sensory analysis revealed high consumer acceptability, attributed to its distinctive mango aroma and flavor. These findings highlight the potential of this fermented mango-based beverage as a novel functional food with promising market appeal.

1. Introduction

The increasing consumer demand for health-oriented products has driven the development of functional beverages, which offer additional health benefits due to the presence of bioactive compounds [1,2,3,4]. According to Mordor Intelligence, the fermented foods and beverages market is expected to grow at a compound annual growth rate (CAGR) of 6.43% between 2025 and 2030 [5]. These beverages are often enriched with bioactive ingredients such as probiotics, which enhance their functional and nutritional properties [6,7]. Probiotics are defined as live microorganisms that, when administered in adequate amounts, confer health benefits to the host. Numerous studies have demonstrated that probiotics play a crucial role in maintaining intestinal microbiota balance, enhancing digestive function, modulating immune responses, and reducing systemic inflammation [8,9,10].

Lactic acid bacteria (LAB) are widely used as starter cultures in probiotic fermentations due to their beneficial physiological properties. These microorganisms metabolize available carbohydrates, primarily producing lactic acid as the main end-product. This organic acid not only acts as a preservative, by lowering the pH and inhibiting spoilage organisms, but also plays a key role in developing the distinctive flavor of fermented foods [11,12]. The efficiency of lactic acid production during fermentation is influenced by multiple factors, including temperature, pH, nutrient availability, the nature of the fermentation matrix, fermentation duration, inoculum concentration, and, most importantly, the specific LAB genus and strain used [1,10,11]. During fermentation, LAB produces two enantiomers of lactic acid: D-lactic acid and L-lactic acid. Although both forms are generated, elevated concentrations of D-lactic acid relative to L-lactic acid can inhibit its metabolism, leading to accumulation and potentially resulting in D-lactic acidosis, as well as an unpleasant aroma [13,14].

Beyond its role in fermentation, LAB confers additional advantages when incorporated into functional beverages. These include enhancements to the nutritional profile; improvements in sensory attributes such as texture, aroma, and flavor; and an extension of product shelf life. These properties are crucial for enhancing consumer acceptance and improving the market competitiveness of fermented functional products [3,15,16,17]. A Global Functional Foods Market Size & Outlook report predicts an 8.6% increase in the marketing of functional foods [18], driven by the growing belief and research showing that these foods and their nutritional components can improve consumer health [19].

Traditionally, dairy products have been the primary vehicles for delivering probiotics. However, the rising prevalence of lactose intolerance, the growing popularity of vegan diets, and the demand for novel foods with improved sensory appeal have led to a shift toward plant-based alternatives, particularly fruit- and vegetable-based matrices, as suitable carriers for beneficial microorganisms [15,20,21,22]. This trend reflects evolving dietary preferences and underscores the need to explore new fermentable substrates that can yield functional and health-promoting products with attractive sensory characteristics [3,4,22].

In this context, the Ataulfo mango (Mangifera indica L.) stands out as a promising substrate, not only because of its appealing sensory qualities, such as texture, acidity, flavor, and aroma, but also due to its rich nutritional composition, which includes carotenoids, dietary fiber, polyphenols, minerals, and vitamins [23,24,25]. Its high content of natural sugars further enhances its suitability for supporting the growth and metabolic activity of lactic acid bacteria [24,26,27]. This nutritional and physicochemical profile makes the Ataulfo mango particularly well-suited for fermentation with specific, robust strains of lactic acid bacteria that offer proven functional benefits. Among these, Lactiplantibacillus plantarum is one of the most studied and widely used strains, owing to its GRAS (Generally Recognized as Safe) status [28], metabolic versatility, and adaptability to diverse plant-based substrates [17,20,21,22,29,30]. Therefore, this study hypothesized that the development of a fermented mango-based beverage using the Lp6 and Lp32 strains with acceptable quality is feasible. On this basis, we aimed to develop a mango-based fermented beverage using L. plantarum strains Lp6 and Lp32 to produce a product that exhibits enhanced functional properties, microbiological stability, and safety, as well as high sensory acceptability.

2. Materials and Methods

2.1. Plant Material

Commercially ripe Ataulfo mango fruits (°Brix 16–17; pH 4.2–4.5; Figure 1) were harvested in Chiapas, Mexico, in 2020. The pulp was manually extracted, vacuum-packed in food-grade plastic bags, and stored at −20 °C until further use.

2.2. Mango Beverage Preparation

Preparation of a mango-based beverage for fermentation was performed according to Aviles-Rivera et al. [31]. Frozen mango pulp was thawed and homogenized in a household blender (Oster Model BLST4126R, Newell Brands, Mexico) until a smooth paste was obtained. The beverage was prepared by mixing the mango paste with water at a 1:1 (v/v) ratio, resulting in a pH of 4.29 ± 0.3. Total soluble solids (TSSs) were adjusted (sugar Zulka^®) to 13 ± 0.2°Brix at the final concentration of the sugars, which was as follows: 9 g of sucrose, 1.5 g of fructose, and 0.7 g of glucose per 200 mL. The beverage was then pasteurized at 85 °C for 5 min using a water bath (Thermo Scientific, Precision CIR 35, Waltham, MA, USA), immediately cooled in an ice bath, and stored under refrigeration (4 ± 1 °C) until inoculation with L. plantarum strains Lp6 and Lp32.

2.3. Inoculum

The strains were reactivated following the method described by Aviles-Rivera et al. [31]. L. plantarum strains Lp6 and Lp32 were cultured in MRS broth (de Man, Rogosa, and Sharpe; Condalab, St. Forja, Madrid, Spain) through three consecutive subcultures of 12, 8, and 10 h, respectively, at 37 °C under anaerobic conditions. Cells from the last subculture were harvested by centrifugation (4500× g, 10 min, 4 °C), washed twice with phosphate-buffered saline (PBS; 0.1 M, pH 7.2), and resuspended to achieve the inoculum concentrations defined by the response surface methodology design described in Section 2.6.1.

2.4. Fermented Mango Beverage

A 200 mL aliquot of the mango beverage was inoculated (1% v/v) with either L. plantarum Lp6 or Lp32 and incubated at 37 °C for varying fermentation times.

Fermented beverages were developed using a central composite design (CCD) within a response surface methodology (RSM), considering inoculum concentration and fermentation time as the independent variables. After fermentation, samples were stored at −20 °C and later analyzed for bacterial viability, functional properties, and acidification. The evaluated response variables included CC, TPC, and FTC, as well as concentrations of L- and D-lactic acid.

2.4.1. Viable Cell Concentration (CC)

Viable L. plantarum cells were quantified using the pour plate method on MRS agar (BD Difco^®), following the procedure described by Sanders [32]. Serial dilutions were prepared and plated, followed by incubation at 37 °C for 48 h. Colony-forming units (CFUs) were counted, and results were expressed as log₁₀ CFU/mL.

2.4.2. Total Phenolic Compounds (TPCs)

The total phenolic compounds content was determined using the Folin–Ciocalteu colorimetric method, as described by Swain and Hillis [33]. A 10 μL aliquot of the fermented mango beverage (diluted 1:1 with distilled water) was mixed with 230 μL of distilled water and 10 μL of 2 N Folin–Ciocalteu reagent. After 3 min, 25 μL of 4 N sodium carbonate solution was added, and the mixture was incubated at room temperature for 2 h. Absorbance was measured at 750 nm using a Synergy HT microplate reader (BioTek Instruments, Winooski, VT, USA), with distilled water as the blank. Results were calculated using a gallic acid calibration curve expressed as milligrams of gallic acid equivalents (mg GAE) per 100 mL (Equation (1)).

$$TPC=(Standardcurveconcentration\times Volume\times 100)/Sample).$$

(1)

2.4.3. Total Flavonoid Compounds (TFCs)

Total flavonoid content was determined according to the method described by Chang et al. [34]. A 10 μL measure of the fermented mango-based beverage was mixed with 250 μL of distilled water, followed by the addition of 10 μL of 10% (w/v) aluminum chloride (AlCl₃) and 10 μL of 1 M potassium acetate. The reaction mixture was incubated at 25 °C for 30 min, and the absorbance was measured at 415 nm using a microplate reader. Quantification was performed using a quercetin standard curve, and results are expressed as milligrams of quercetin (mg Q) per 100 mL of sample (Equation (2)).

$$TFC=(Standardcurveconcentration\times Volume\times 100)/Sample).$$

(2)

2.4.4. L- and D-Lactic Acid

Lactic acid isomers were quantified using the method described by Xu et al. [35], with slight modifications. Fermented samples were centrifuged at 4500× g × 10 min, and the supernatant was filtered through a 0.25 µm nylon membrane and diluted 1:10 with Milli-Q water. A 20 µL aliquot was injected into a high-performance liquid chromatography (HPLC) system equipped with UV-VIS detection (Varian 9050, Spectralab Scientific Inc., Mississauga, ON, Canada) and a Chirex 3126 (D)-penicillamine chiral column (150 × 4.6 mm). The mobile phase was a 2 mM copper (II) sulfate solution (CuSO₄·5H₂O), delivered at 1 mL/min under isocratic conditions. The column temperature was maintained at 40 °C, and detection was performed at 254 nm. Quantification of L- and D-lactic acids was based on an external standard calibration curve (6–650 µg/mL) constructed from certified reference standards.

2.5. Characterization of the Optimal Fermented Beverage

The fermented mango beverage produced under optimal processing conditions was characterized in terms of stability, safety, composition, and overall acceptability through physicochemical, microbiological, proximate, and sensory analyses.

2.5.1. Physicochemical Analysis

Physicochemical parameters, including pH, TSS, color, and sugar content, were measured as described by Aviles-Rivera et al. [31]. pH was determined using a digital pH meter (Thermo Scientific, Orion Versa Star, MA, USA) by immersing the electrode in 10 mL of the sample, following the AOAC guidelines [36]. TSS was determined at 20 °C using a digital refractometer (Mettler Toledo RE40D, Mexico City, Mexico) and reported in °Brix. Color was assessed using a portable spectrophotometer (Konica Minolta CM-700d, Ramsey, NJ, USA), with triplicate measurements of the sample surface. Color coordinates were recorded in the CIELAB color space: L* (lightness), a* (red to green), and b* (yellow to blue).

Sucrose, D-fructose, and D-glucose concentrations were measured using the enzymatic-spectrophotometric method developed by the Megazyme kit K-SUFRG [37]. D-glucose was measured before and after enzymatic hydrolysis of sucrose with β-fructosidase (invertase; EC 3.2.1.26). D-fructose was determined by subtracting the D-glucose contribution, using phosphoglucose isomerase (PGI; EC 5.3.1.9). Absorbance was measured at 340 nm with a Cary 60 UV-Vis spectrophotometer (Agilent Technologies, Santa Clara, CA, USA), and the results are expressed as grams per 100 mL of fermented mango-based beverage [31].

2.5.2. Microbiological Analysis

Microbiological safety was assessed by the Official Mexican Standard for Food Testing Methods (NOM-210-SSA1-2014) [38]. A 25 mL aliquot was homogenized (2 min) with 225 mL of phosphate-buffered solution. Serial dilutions were prepared for the enumeration of total coliforms and Escherichia coli. E. coli was quantified using standard biochemical tests (ATCC 25922), and the results are expressed as the most probable number per milliliter (MPN/mL).

For Salmonella spp. detection, 25 mL of sample was incubated in 225 mL of buffered peptone water at 37 °C for 18 h (pre-enrichment). Then, 0.1 mL was transferred to 10 mL of Rappaport–Vassiliadis broth and incubated at 41.5 °C for 24 h. Thereafter, 1 mL was inoculated into 10 mL of Müller–Kauffmann tetrathionate-novobiocin broth and incubated at 36 °C for 24 h. Selective plating was performed on bismuth sulfite agar, xylose lysine deoxycholate agar, and Hektoen enteric agar (36 °C, 24 h). Presumptive Salmonella colonies were identified biochemically using the VITEK^® 2 GN card system (bioMérieux), and serological confirmation was conducted using somatic (O) and flagellar (H) polyvalent antisera. PCR (Eppendorf^® Mastercyclet^®, Hamburgo, Alemania) was used for molecular confirmation. Results are reported qualitatively (presence/absence in 25 mL).

Fungal and yeast counts were conducted following the guidelines established in the Mexican Official Standard NOM-111-SSA1-1994 [39]. A 10 mL sample was homogenized for 2 min with 90 mL of phosphate-buffered solution, serially diluted, and plated on chloramphenicol yeast glucose agar (CYGA medium, Becton Dickinson, Queretaro, Mexico). The plates were incubated at 25 °C for 5 days. Results are reported as CFU/mL [40].

2.5.3. Proximal Analysis

Moisture (Method 920.39), protein (Method 988.05, N × 6.25), dietary fiber (Megazyme International, Wicklow, Ireland), ash (Method 942.05), and fat (Method 920.39) contents were determined according to the AOAC methods [41]. Moisture was determined by drying 2 mL of the beverage at 70–80 °C for 24 h in a Yamato oven (model DKN602C, Santa Clara, CA, USA), and ash content was assessed by incineration at 550 °C for 12 h in a Thermolyne muffle (model FD1530M, Waltham, MA, USA). Protein content was determined by the Kjeldahl method, based on quantifying total nitrogen, the value of which was multiplied by a conversion factor of 6.25. Fat was extracted using a Soxhlet apparatus (125 mL PYREX^®, Temple, AZ, USA) with anhydrous petroleum ether and quantified by weight difference. Total dietary fiber was determined according to the AOAC method 991.43.

2.5.4. Sensory Analysis

The sensory evaluation protocol was approved by the Institutional Ethics Committee of the Center for Research in Food and Development (CIAD), under reference number CEI-013-2/2024. A total of 112 untrained panelists (men and women over 18 years of age), including students, faculty, researchers, administrative staff from CIAD, and members of the general public, participated in the study. The sensory evaluation was conducted using a mango-based beverage fermented with L. plantarum Lp6, prepared under optimized conditions of 9.89 Log (7.76 × 10⁹) CFU/mL inoculum and a fermentation time of 20.47 h. Each participant received 10 mL of the fermented beverage, served in biodegradable plastic cups at a temperature between 4 °C and 8 °C. To neutralize residual flavors between samples, participants were provided with purified water and classic Habaneras whole wheat cookies (Gamesa^®, a subsidiary of PepsiCo, Inc, Monterrey, Mexico). Sensory attributes, color, smell, taste, and overall acceptability were evaluated using a 7-point hedonic scale (1 = “dislike very much” to 7 = “like very much”).

2.6. Statistical Analysis

2.6.1. Optimization

For each strain (Lp6 and Lp32), optimal conditions for producing the fermented mango-based beverage were determined using RSM, applying a CCD with two independent factors: inoculum concentration (IC), ranging from 5.2 to 10.8 Log CFU/mL, and fermentation time (FT), ranging from 0 and 50 h [31]. The natural and coded levels of these independent variables are shown in

Table 1. Natural and coded levels of the independent variables.

Variable	Natural Levels
Inoculum concentration (log CFU/mL)	5.2	6.0	8.0	10.0	10.8
Fermentation time (h)	0	6	24	42	50
Coded levels	−1.414	−1	0	1	1.414

.

Polynomial regression models were fitted for each response variable of the fermented mango-based beverage, and statistical significance was evaluated through analysis of variance (ANOVA) and regression analysis. Model coefficients were considered significant at p < 0.05. Simultaneous optimization of the response variables was carried out using contour plot superposition and the desirability function approach. All statistical analyses were performed using Minitab [42].

2.6.2. Characterization

Statistical analysis of the beverage’s physicochemical, microbiological, and proximate composition was conducted using descriptive statistics. Results are presented as means, standard errors, and confidence intervals. All analyses were performed using Minitab [42].

2.6.3. Sensory Evaluation

For the statistical analysis of the sensory data, contingency tables were created based on consumer evaluations of quality attributes (color, smell, flavor, and overall acceptability). Correspondence analysis was applied to identify the most frequently selected attributes and their associations within the Lp6 fermented mango-based beverage. The 7-point hedonic scale was grouped into five categories, very low (1), low (2–3), neutral (4), high (5–6), and very high (7), following the classification proposed by Greenacre [43].

3. Results and Discussion

3.1. Analysis of Quality Variables

Table 2. Results of the response variables by treatment of the fermented mango-based beverage with Lp6.

Factors			Response Variable
* IC	FT	CC	TPCs	TFCs	L-Lactic Acid	D-Lactic Acid
6	6	5.65	41.67	4.53	2.6	1.7
10	6	10.43	44.60	4.50	46.8	5.9
6	42	10.71	43.13	4.43	265.9	92.5
10	42	11.70	49.87	4.07	250.1	76.7
5.2	24	9.08	41.80	4.40	188.6	45.0
10.8	24	14.45	46.40	4.13	194.5	55.2
8	0	6.64	41.67	4.33	2.40	1.10
8	50	11.25	51.67	4.43	258.3	87.4
8	24	13.71	45.20	4.33	232.0	62.6
8	24	12.73	45.13	8.57	214.7	58.0
8	24	13.67	43.07	6.90	194.8	73.4
8	24	13.69	43.73	8.27	238.5	63.9
8	24	12.64	41.60	7.60	219.2	75.0
Mean		11.26	44.58	5.42	177.60	53.72
SE Mean		0.76	0.88	0.48	26.40	8.79

* IC = inoculum concentration (log CFU/mL); FT = fermentation time (h); CC= viable cell concentration (log CFU/mL); TPCs = total phenolic compounds (mg GAE/100 mL); TFCs: total flavonoid compounds (mg Q/100 mL); L-lactic acid (mg/100 mL); D-lactic acid (mg/100 mL)—SE standard deviation.

and

Table 3. Results of the response variables by treatment of the fermented mango-based beverage with Lp32.

Factors		Response Variable
* IC	FT	CC	TPCs	TFCs	L-Lactic Acid	D-Lactic Acid
6	6	5.98	37.53	4.00	75.29	28.47
10	6	10.47	40.60	3.70	50.42	30.10
6	42	10.31	39.60	4.00	312.82	312.82
10	42	12.35	45.13	3.70	461.95	438.18
5.2	24	9.76	41.27	4.53	234.48	177.30
10.8	24	14.83	48.07	4.00	238.07	168.64
8	0	6.32	37.07	4.07	2.44	1.89
8	50	11.64	47.60	3.77	588.14	513.57
8	24	13.48	45.13	4.27	262.55	231.95
8	24	14.63	44.67	7.67	303.86	243.97
8	24	13.63	41.60	6.53	289.89	214.72
8	24	13.52	46.67	6.67	263.58	287.65
8	24	14.67	42.73	7.27	285.78	279.72
Mean		11.66	42.90	4.94	259.20	209.90
SE Mean		0.83	1.01	0.42	43.90	42.90

* IC = inoculum concentration (log CFU/mL); FT = fermentation time (h); CC= viable cell concentration (log CFU/mL); TPCs = total phenolic compounds (mg GAE/100 mL); TFCs = total flavonoid compounds (mg Q/100 mL); L-lactic acid (mg/100 mL); D-lactic acid (mg/100 mL)—SE standard deviation.

present the response variables of fermented beverages with L. plantarum strains Lp6 and Lp32. Both formulations reached a viable cell concentration of over 11 Log (1 × 10¹⁰) CFU/mL, indicating the successful adaptation of the strains to the base mango beverage matrix [10,11]. This bacterial load is considered adequate to deliver potential health benefits in fermented functional beverages [44,45].

The average TPC and TFC concentrations in fermented beverages with Lp6 and Lp32 were 43.7 mg GAE/100 mL and 5.2 mg QE/100 mL, respectively. The type and concentration of phenolic compounds in fermented plant-based matrices can be influenced by microbial enzymatic activities, such as β-glucosidases and decarboxylases, that act on the polyphenols naturally present in fruits and vegetables [46,47]. The relatively lower flavonoid levels may be attributed to inoculum concentration and fermentation time, as extended fermentation can lead to partial degradation of flavonoids or their transformation into other secondary metabolites [48,49].

The production of L-lactic and D-lactic acid isomers in fermented products can enhance or impair their functional and sensory attributes, depending on their concentration and relative proportion. In this study, the mango beverage fermented with Lactiplantibacillus plantarum Lp6 exhibited an L-lactic/D-lactic acid ratio of 3.3, whereas the beverage fermented with Lp32 showed a lower ratio of 1.2. Notably, the Lp32 formulation accumulated over 200 mg/L of both isomers (Table 3), which adversely affected the sensory characteristics of the final product. Excessive D-lactic acid accumulation in fermented foods has been linked to potential health risks, particularly in vulnerable populations. Elevated levels of this isomer may interfere with L-lactic acid metabolism and have been associated with D-lactic acidosis, especially in individuals with underlying medical conditions such as short bowel syndrome [13,14,50,51,52]. Additionally, an imbalance favoring D-lactic acid can negatively influence the product’s flavor profile, stability, and overall consumer acceptability [35,50]. In contrast, the predominance of the L-lactic acid isomer in the Lp6 fermented beverage indicates a more favorable metabolic profile and improved sensory attributes. Therefore, Lp6 was identified as the most suitable strain for producing fermented mango beverages. Based on these findings, viable cell concentration, total phenolic compounds, total flavonoid compounds, and L-lactic acid were selected as response variables for optimization using the Lp6 strain.

3.2. Optimization of Lp6 Fermented Mango Beverage

A second-order polynomial model was fitted to each response variable, excluding non-significant terms (p > 0.05), and predictive equations were generated. For the fermented mango-based beverage with the Lp6 strain, the statistically significant regression coefficients (p < 0.05), along with their standard errors and determination coefficients (R²), are summarized in

Table 4. Statistically significant regression coefficients (p < 0.05) and standard errors for each response variable of the Lp6 fermented mango-based beverage.

Response Variables	Terms	Coefficients	p-Value	R2 (%)
Viable cell concentration (log CFU/mL);	IC	2.275	0.000	97.15
	FT	2.063	0.000
	IC IC	−1.82	0.005
	FT FT	−4.713	0.000
	IC FT	−1.841	0.017
Total phenolic compounds (mg GAE/100 mL)	IC	2.848	0.004	82.91
	FT	3.754	0.001
	FT FT	2.60	0.047
Total flavonoid compounds (mg Q/100 mL)	IC	−0.136	0.828	66.88
	FT	−0.179	0.788
	IC IC	−2.796	0.016
	FT FT	−2.659	0.019
* L-Lactic acid (mg/100 mL)	IC	6.47	0.0533	97.05
	FT	141.18	0.000
	IC IC	−35.9	0.044
	FT FT	−104.0	0.000
* D-lactic acid (mg/100 mL)	IC	0.48	0.908	95.65
	FT	48.72	0.000
	IC IC	−17.25	0.022
	FT FT	−25.51	0.003

IC = inoculum concentration (log CFU/mL); FT = fermentation time (h). * Two isomeric forms: L-lactic acid and D-lactic acid.

.

Based on the statistically significant terms for each response variable in the Lp6 fermented mango-based beverage (Table 4), simplified polynomial equations (Equations (3)–(7)) were derived from the response surface models. Only variables with a statistically significant contribution (p < 0.05) were retained, while non-significant terms (p > 0.05) were excluded to enhance the predictive accuracy and interpretation of the models for optimizing the beverage elaboration process.

CC = −20.01 + 5.183 IC + 0.6699 FT − 0.2321 IC IC − 0.007541 FT FT − 0.0263 IC FT

(3)

TPC = 34.6 + 1.017 IC − 0.0577 FT+ 0.00416 FT FT

(4)

TFC = −17.80 + 5.66 IC + 0.2056 FT − 0.357 IC IC − 0.00425 FT FT

(5)

L-Lactic acid = −330 + 75.5 IC + 13.97 FT − 4.58 IC IC − 0.1664 FT FT

(6)

D-lactic acid = −147.7 + 35.4 IC + 3.990 FT − 2.201 IC IC − 0.04082 FT FT

(7)

According to the overlay analysis, the target ranges (maximum and minimum) for each response variable were defined as follows: for CC, the range was set between 13 and 14 Log CFU/mL to ensure levels above the required to confer health benefits [9,10,15]. For L-lactic acid, the optimal concentration was established between 170 and 190 mg/100 mL, while for D-lactic acid, a lower range of 50 to 70 mg/100 mL was selected, aligning with levels reported in other fermented beverages that do not negatively impact sensory quality [20,35,47,53]. Regarding TPC, an optimal range of 45 to 46 mg GAE/100 mL was defined, and for TFC, 6 to 7 mg Q/100 mL, as relevant to the functional quality of fermented beverages [20,49,54].

Figure 2 presents the contour plot, which shows the optimal region for all evaluated quality parameters. The white area represents the simultaneous optimization zone, where the best combination of inoculum concentration and fermentation time is achieved. The optimal formulation corresponded to an inoculum concentration of 9.89 log (7.76 × 10⁹) CFU/mL and a fermentation time of 20.47 h, resulting in optimal levels of cell concentration, phenolic compounds, flavonoids, and both lactic acid isomers. These optimal conditions were further confirmed by a desirability function analysis, which validated their consistency across all five response variables.

3.3. Validation of Optimal Conditions

To validate the optimization results, mango-based beverages were fermented using L. plantarum Lp6 strain under the identified optimal conditions (IC: 9.89 log (7.76 × 10⁹) CFU/mL; FT: 20.47 h) (Figure 2). The quality parameters and their corresponding 95% confidence intervals were evaluated. The experimental values closely matched those predicted by the model (Figure 2 and

Table 5. Predicted and replicated values in the fermented mango-based beverage with Lp6.

	Response Variables
Values	* CC	TPCs	TFCs	L-Lactic Acid	D-Lactic Acid
Predicted	13.78	45.21	5.7	185.29	51.58
Replicated	13.0	43.4	6.2	182.8	41.53
Confidence interval of 95%	(12.03, 13.99)	(41.34, 45.43)	(5.7, 6.69)	(156.1, 209.7)	(35.83, 47.18)

* CC= viable cell concentration (Log CFU/mL); TPCs = total phenolic compounds (mg FA/100 mL); TFCs = total flavonoid compounds (mg QE/100 mL); L-lactic acid (mg/100 mL); D-lactic acid (mg/100 mL).

), confirming the model’s accuracy and predictive reliability.

The findings of this study confirm the feasibility of producing a fermented mango-based beverage using L. plantarum Lp6, achieving a viable cell concentration of 13.01 log (10.23 × 10¹²) CFU/mL. This value substantially exceeds the minimum threshold of 6–7 log CFU/mL recommended for probiotic foods to ensure their functional efficacy [9,10]. The high microbial viability observed suggests that the beverage has the potential to confer health benefits to consumers while also highlighting the suitability of mango as a fermentation substrate that supports the growth and metabolic activity of the Lp6 strain. Nevertheless, it is essential to note that functional effects are strain-specific and may vary depending on the microorganism used [15].

Mwanzia et al. [26] reported that an apple mango beverage fermented with Lactobacillus rhamnosus GR-1 reached a viable cell count of 9 log (7.76 × 10⁹) CFU/mL, a result comparable to that of Cele et al. [20], who observed similar concentrations in beverages fermented with L. plantarum using the Sabre, Peach, and Tommy Atkins mango varieties. The differences between the present study and previous studies may be attributed to variations in experimental conditions, including inoculum concentration, fermentation time, and strain-specific adaptation to the available carbohydrate. Additionally, factors such as the production of exopolysaccharides and antimicrobial metabolites could have enhanced bacterial growth and contributed to the microbiological stability and safety of the final product [55].

The optimized fermented mango beverage provides functional components, including a TPC of 43 mg GAE/100 mL and a TFC of 6.2 mg Q/100 mL (Table 5). These values underscore the importance of controlling fermentation parameters to enhance the bioavailability of these compounds while maintaining viable populations of beneficial bacteria. As noted by Ruiz et al. [49], the bioconversion of bioactive compounds during fermentation is a complex process influenced by multiple factors, including the selected microbial strain, pH, temperature, fermentation time, and the substrate’s physicochemical properties. It has been reported that foods with TPC content are related to an increase in the ability to protect cells from damage caused by free radicals [29,56].

Consuming 200 mL of the fermented mango-based beverage provides approximately 86 mg of gallic acid equivalents (GAEs), representing 15% of the minimum recommended daily intake of polyphenols, which ranges from 584 to 1786 mg/day; primary dietary sources include non-alcoholic beverages such as coffee, tea, fruit juices, fruits, vegetables, and fermented products [57]. Regarding TFC, the beverage supplies 12.4 mg of quercetin (Q) per 200 mL, which falls within the recommended intake range of 20–240 mg/day for individuals seeking health benefits. However, excessive flavonoid intake may diminish or even counteract their beneficial effects [58].

In contrast, Chun et al. [59] estimated that the average daily intake of phenolic compounds in the American diet is approximately 450 mg (expressed as sinapic acid equivalents), with an average flavonoid intake of 103 mg catechin equivalents, primarily derived from fruits and vegetables. Similarly, Corrêa et al. [60] reported average intakes of 314 mg/day for phenolic compounds and 138.92 mg/day for flavonoids, identifying non-alcoholic beverages as the primary dietary contributors.

The fermented mango-based beverage with the L. plantarum Lp6 strain exhibited a favorable ratio of L-lactic/D-lactic acid isomers, with a lower concentration of D-lactic acid (Table 5), contributing to improved product quality and stability [35]. According to the Mexican Official Standard NOM-F-420-S-1982 [61], the acceptable acidity range in milk, expressed as lactic acid, is between 1.4 and 1.7 g/L (140–170 mg/100 mL), a value comparable to that found in this fermented mango beverage. Conversely, Chen et al. [62] reported lactic acid concentrations as high as 571 mg/100 mL in a papaya beverage fermented with L. plantarum, resulting in markedly more acidic sensory properties.

3.4. Characterization of the Optimal Beverage

3.4.1. Physiochemistry and Proximal

Table 6. Quality characteristics of optimal fermented mango-based beverage with Lp6.

Nutritional Quality		Physicochemical Characteristics
Moisture	* 86.68 ± 0.02	pH	3.57 ± 0.19
Protein	0.64 ± 0.12	TSS (°Brix)	12.81 ± 0.2
Fat	0.31 ± 0.04	Color beverage:
Ash	0.49 ± 0.05	Luminosity	47.4 ± 0.21
Total carbohydrates	11.40 ± 0.7	** value a*	5.4 ± 0.11
Glucose	0.34 ± 0.05	** value b*	24.8 ± 0.25
Fructose	1.27 ± 0.01	Chromaticity	25.4 ± 0.27
Sucrose	8.70 ± 0.1	°Hue	77.0 ± 0.15
Fiber dietary	1.09 ± 0.3

* Mean and standard deviation of three replicates. TSS: total soluble solids. ** value a*= green to red; ** value b*= yellow to blue.

summarizes the quality parameters of the optimized mango-based beverage fermented with L. plantarum Lp6. The beverage exhibited a pH of 3.6, characteristic of an acidic product resulting from fermentation, during which the bacteria metabolize available sugars to produce and accumulate organic acids, primarily lactic acid [12,20,63]. The TSS content was 12.8 °Brix, representing fermentable substrates that support the growth and metabolic activity of L. plantarum [31]. Among the individual sugars quantified, sucrose was the most abundant (8.7 g/100 mL), while glucose and fructose were present at lower concentrations (Table 6). This pattern may reflect a preferential utilization of monosaccharides by L. plantarum Lp6 as primary energy sources, in agreement with previous findings [12,64,65].

According to NOM-086-SSA1-1994 [66], the standard serving size for beverages in Mexico is 200 mL. According to this reference, the commercial fermented strawberry beverage Lifeway provides 9 g of protein, 17 g of sugars, 1.6 g of fat, and no dietary fiber per serving. In comparison, the fermented beverage Yakult^® contains 2.5 g of protein, 30 g of sugars, and is free of fat and dietary fiber. In contrast, the fermented mango beverage formulated with L. plantarum Lp6 contains 1.3 g of protein, 2.0 g of dietary fiber, and 22 g of sugars per 200 mL serving. This composition suggests that incorporating mango pulp and lactic acid bacteria yields a product with a notably higher dietary fiber content than other fermented beverages currently available on the market. Furthermore, the presence of polyphenols and carotenoids from mango enhances the beverage’s potential as a nutraceutical product.

Color is a fundamental quality attribute in food products, as consumers frequently associate it with freshness and overall acceptability, often expecting processed items to resemble their fresh counterparts [67]. The fermented mango-based beverage exhibited an appealing orange hue, as measured using the MINOLTA color scale [68] and presented in Table 6, with values of lightness (L*) = 47, chroma (C*) = 25, and hue angle (°Hue) = 77. These values reflect a color comparable to that of ripe mango, although less intense than fresh Ataulfo mango pulp, which typically displays higher lightness (L* = 60), greater chromaticity (C* = 63), and a lower hue angle (°Hue = 63) [69]. The color differences observed may be attributed to the pasteurization process applied to the beverage, as thermal treatments are known to impact sensory properties, including potential color degradation or transformation [20,70].

3.4.2. Microbiological

The fermented mango beverage produced with the Lp6 strain exhibited satisfactory microbiological quality, attributable to the combined effects of pasteurization and fermentation with lactic acid bacteria. No Escherichia coli, total coliforms, yeasts, or molds were detected in the analyzed samples. Additionally, Salmonella spp. was not detected in the beverage, meeting the microbiological safety requirements established for non-alcoholic beverages by the Mexican Official Standard NOM-218-SSA1-2011 [71].

3.4.3. Sensory

A total of 112 consumers, 47% women and 53% men, participated in the sensory evaluation of the fermented mango beverage with the Lp6 strain. In terms of age distribution, 58% of participants were young adults (18–26 years old), 37% were adults (27–59 years old), and 5% were older adults (≥60 years old). In its optimized formulation, the sensory evaluation results showed that the fermented mango-based beverage with Lp6 received scores above 6.0 for overall acceptability, color, aroma, and flavor. Furthermore, Figure 3 shows no differences in gender. These results indicate favorable consumer reception, with participants expressing a high to very high level of preference and liking (

Table 7. Contingency table of attributes and evaluation levels of fermented mango-based beverage with Lp6.

Variables	Values
	Low	Neutral	High	Very High	All
Acceptability	1	4	64	43	112
Color	0	1	45	66	112
Smell	1	4	43	64	112
Flavor	1	6	59	46	112
All	3	15	211	219	448

).

The use of evaluation levels and the assignment of values to attributes were determined through a correspondence analysis applied to a contingency table involving two variables: sensory attributes and hedonic scale evaluation levels (Table 7). This analysis enables the identification of frequency heterogeneity within the table and, simultaneously, highlights the categories of both variables that contribute to this heterogeneity, that is, the association between sensory attributes and hedonic scale evaluation levels.

In the correspondence analysis of the sensory attributes of the beverage, a positive relationship was observed between flavor and overall acceptability (R² = 0.741, p < 0.05) and between smell and overall acceptability (R² = 0.38, p < 0.05). The distribution of the points indicates that higher ratings for flavor and smell correspond to increased overall acceptability of the mango beverage. While color and general appearance are important in the development of new beverages, the gustatory (flavor and taste) and olfactory (aroma and smell) properties play a more crucial role in consumer acceptance [72].

The correspondence analysis based on the data in Table 7 is depicted in Figure 4. Two components accounted for 99.1% of the total inertia (heterogeneity) in the dataset, with Component 1 alone explaining 86.5%. The attributes of color, smell, flavor, and overall acceptability (represented by the red dots) were assessed at various levels: low, neutral, high, and very high (represented by the blue squares). Among these, the “very high” level was the most frequently selected, while the “low” level was the least chosen. Regarding attribute associations, color and smell exhibited a positive correlation, with both receiving high ratings, suggesting that most participants appreciated the color and aroma of the beverage. Similarly, overall acceptability and flavor were positively correlated with high ratings, suggesting that these attributes played a significant role in driving consumer preference. In contrast, low and neutral ratings were negatively correlated with all evaluated attributes, reinforcing the observation that participants generally gave favorable assessments to the fermented mango-based beverage.

A study by Liu et al. [70] reported that during the fermentation of mango juice with L. plantarum NCU116, L. acidophilus NCU402, and L. casei NCU215, the concentration of p-isopropenyl toluene increased, contributing to a richer fruity and sweet aroma. This finding may help explain why, in the present study, the interplay between the mango’s natural acidity and sweetness contributed to enhanced consumer preference.

4. Conclusions

The optimal conditions for producing the fermented mango-based beverage were identified as an inoculum concentration of 9.89 log (7.76 × 10⁹) CFU/mL and a fermentation time of 20.47 h using the Lactiplantibacillus plantarum Lp6 strain, thereby supporting the hypothesis for this strain. On the contrary, the hypothesis for the Lp32 strain was discarded. Under these parameters, the beverage reached maximum levels of viable cells, total phenolic compounds, total flavonoids, and L-lactic acid, while minimizing the concentration of D-lactic acid. The applied mathematical models were statistically significant, confirming their reliability in describing the relationships between the independent variables and the measured responses.

Additionally, a microbiologically safe fermented mango-based beverage with high sensory acceptability was successfully developed. The sensory attributes of color, aroma, and flavor showed a positive correlation with overall product acceptance. These results demonstrated that, under the optimized process conditions, the combination of mango as a substrate and the Lp6 strain yields a fermented mango-based beverage with potential functional properties, microbiological safety, and strong consumer appeal.

5. Recommendations

As the final stage in the development of the fermented mango-based beverage, it is essential to conduct shelf-life evaluations, cost-effectiveness analyses, and market feasibility studies. Additionally, scaling up the optimized process for industrial production should be considered. Furthermore, future research should focus on evaluating the beverage’s functional properties, particularly its antioxidant capacity, potential prebiotic effects, and probiotic viability during gastrointestinal transit. Such studies would provide scientific evidence to support health claims and further position the product as a functional food.

6. Limitations

This study has certain limitations, primarily related to the exclusive use of the Ataulfo mango cultivar, the specific maturity stage selected, and the particular microbial strains employed during the fermentation process. These factors may influence the generalizability of the findings to other cultivars, ripeness stages, or fermentation conditions.