Microbial Succession and Functional Metabolite Formation During SCOBY Fermentation of Pomelo Peel Substrates

Abstract

This study investigated the fermentation kinetics, microbial community succession, and potential functional metabolite formation in Symbiotic culture of bacteria and yeast (SCOBY)-mediated fermentation using pomelo peel substrates. Pomelo peel substrates were prepared using 1% and 6% (w/w) SCOBY combined with 10 g and 25 g pomelo peel and fermented at 30 °C for 25 days. The results showed that higher SCOBY inoculum significantly accelerated acid production, resulting in a rapid decrease in pH and an increase in titratable acidity. Total soluble solids continuously decreased due to microbial utilization of sugars. The highest lactic acid bacteria count (6.04 log CFU/mL) and total viable count (7.23 log CFU/mL) were observed in S6-P25 at day 25. Bioactive compound analysis revealed that total flavonoid content reached its maximum in S6-P25 at day 20 (15.34 ± 0.70 mg RE/g dry weight, DW), while the highest total phenolic content was found in S1-P25 (151.5 ± 1.29 mg GAE/g DW), suggesting that a lower SCOBY level may favor polyphenol production. Antioxidant activity (DPPH and TEAC) increased with fermentation time and was highest in S6-P25. Microbiome analysis demonstrated that Firmicutes was the dominant phylum, with Apilactobacillus ozensis accounting for 99% of the relative abundance, indicating strong microbial selection and its potential role in acid production and fermentation ability. This microbial structure was consistent with the improved fermentation performance and enhanced bioactive properties observed in the pomelo peel substrates. These findings highlight SCOBY fermentation as a promising biotechnological strategy for converting citrus processing by-products into fermented ingredients for food applications.

1. Introduction

Fermentation is an effective biotechnological approach for improving the nutritional value, functional properties, and shelf life of food substrates through microbial metabolism. In recent years, kombucha fermentation, which is driven by a SCOBY, has attracted increasing attention due to its ability to produce organic acids, bioactive compounds, and microbial metabolites with potential health benefits [1,2]. The metabolic activities of acetic acid bacteria, lactic acid bacteria, and yeasts in SCOBY create a dynamic fermentation system that influences substrate composition, antioxidant capacity, and microbial community succession [3,4]. Therefore, modulation of fermentation conditions, particularly inoculum size and substrate composition, plays a critical role in determining the fermentation performance and functional characteristics of the final product [5].

At the same time, the valorization of agro-industrial by-products has become an important strategy for achieving sustainable food production. Fruit processing generates large amounts of peel residues, which are often discarded despite being rich in dietary fiber, polyphenols, flavonoids, and other bioactive compounds [6,7]. Pomelo (Citrus grandis (L.) Osbeck) peel is a typical example of such underutilized biomass. It contains abundant phenolic compounds, essential oils, and antioxidant components and has been reported to exhibit antimicrobial, anti-inflammatory, and radical scavenging activities [8,9]. However, the direct utilization of pomelo peel in food systems is limited by its coarse texture, bitter taste, and low bioavailability of bioactive compounds [10]. Fermentation has been proposed as a potential method to enhance the release and biotransformation of bound phenolics and flavonoids, thereby improving their functional properties and potential applications [11,12].

Recent studies have demonstrated that SCOBY fermentation can be applied to various plant-based substrates, including fruit juices, herbal infusions, and food processing by-products, to produce functional fermented beverages and ingredients [13,14]. During fermentation, microbial enzymes hydrolyze complex macromolecules and convert phenolic compounds into more bioavailable forms, leading to enhanced antioxidant activity [15,16]. In addition, the microbial community structure plays a key role in fermentation kinetics, organic acid production, and metabolite formation [17]. High-throughput sequencing technologies have made it possible to elucidate the microbial succession and core microbiome responsible for fermentation performance and product functionality [18,19].

Inoculum size is one of the most important factors affecting fermentation dynamics. A higher inoculum level can accelerate acid production, rapidly reduce pH, and promote microbial growth, whereas a lower inoculum level may slow down substrate degradation and allow the accumulation of certain functional compounds [20,21]. Although several studies have investigated kombucha fermentation using different substrates, systematic studies evaluating the combined effects of SCOBY inoculum level and fruit peel addition on microbial community structure, physicochemical characteristics, and functional properties remain limited.

Therefore, this study aimed to investigate the effect of SCOBY inoculum level and pomelo peel as substrate on fermentation kinetics, microbial community succession, and the formation of bioactive compounds during fermentation. In particular, full-length 16S rRNA sequencing was applied to determine the structure and succession of the core microbiome associated with SCOBY fermentation. By linking microbial community dynamics with physicochemical changes and functional metabolite production, this study provides new insights into microbiome-driven biotransformation processes and highlights the potential of SCOBY fermentation as a sustainable strategy for the valorization of citrus processing by-products.

2. Materials and Methods

2.1. Raw Materials and SCOBY Starter Culture

Fresh pomelo (Citrus grandis (L.) Osbeck) fruits were obtained from Bali District, New Taipei City, Taiwan. The peels were exocarp (flavedo) and mesocarp (albedo), cut into cubes (1 × 1 × 1 cm), and stored at −20 °C until required for use. The proximate composition of the pomelo peel was determined as follows: moisture (7.22%), ash (4.13%), crude fat (2.31%), and crude protein (3.93%).

The symbiotic culture of bacteria and yeast (SCOBY) was obtained from Biozyme Biotechnology Co., Ltd. (New Taipei City, Taiwan). The SCOBY was originally isolated from a mixed fruit and vegetable fermentation broth prepared from more than 20 different plant materials, including pineapple, grapefruit, lemon, pomelo, kumquat, grape, komatsuna, napa cabbage, cabbage, pak choi, spinach, broccoli, sweet potato leaves, lettuce, cucumber, watermelon, sponge gourd, and pear. Fresh plant materials were processed and fermented in a sucrose solution at ambient temperature. The resulting cellulose pellicle was collected and subcultured to obtain a stable SCOBY.

2.2. Preparation of Fermentation Substrates

Fermentation substrates were supplemented with a 5% (w/w) sucrose solution, which was selected based on previously reported kombucha fermentation conditions [17]. Four experimental groups were established with different SCOBY inoculum levels and pomelo peel additions: S1-P10 (1%, w/w SCOBY + 10 g pomelo peel), S1-P25 (1%, w/w SCOBY + 25 g pomelo peel), S6-P10 (6%, w/w SCOBY + 10 g pomelo peel), and S6-P25 (6%, w/w SCOBY + 25 g pomelo peel). All components were thoroughly mixed in a water bath shaker for 10 min prior to fermentation to ensure homogeneous distribution of the substrates and inoculum.

2.3. Fermentation Setup and Sampling Procedures

Static aerobic fermentation was conducted at 30 °C for 25 days according to previously reported kombucha fermentation conditions [20,22]. Samples were collected at predetermined time points (0, 5, 10, 15, 20, and 25 days) to monitor physicochemical, microbiological, and functional changes during fermentation. After sampling, the fermented pomelo substrates were homogenized using a high-speed blender (Osterizer, Sunbeam-Oster Company, Boca Raton, FL, USA) to obtain uniform samples for subsequent analyses.

2.4. pH Measurement

The pH was measured using a calibrated pH meter (Thermo Electron Orion 1111101, Thermo Electron Corporation, Waltham, MA, USA; Schott Instrument, Mainz, Germany) at room temperature (~25 °C). The electrode was rinsed with distilled water and gently blotted dry between measurements. All determinations were performed in triplicate.

2.5. Total Soluble Solids (°Brix)

Total soluble solids of the fermentation broth were measured using a digital refractometer (TECPEL BX-90, Tecpel Co., Ltd., Taipei, Taiwan; range 0–32 °Brix, accuracy ± 0.01). Approximately 2–3 drops of sample were placed on the prism surface, and stabilized readings were recorded.

2.6. Ethanol Concentration

Ethanol content was measured using an automatic alcohol analyzer (AL80, Dong Sheng Chemical Instrument Co., Taichung, Taiwan). Samples were filtered through a 0.45 µm membrane prior to analysis. As the fermentation broth contained soluble sugars and other dissolved solids, the alcohol analyzer was used only for qualitative determination to confirm that the ethanol content remained below the regulatory limit for non-alcoholic beverages (<0.5% v/v).

2.7. Titratable Acidity Analysis

Titratable acidity (TA) was determined by titrating 10 mL of sample with 0.1 mol/L NaOH using phenolphthalein as an indicator according to previously reported kombucha fermentation analysis methods [5]. Briefly, 2–3 drops of phenolphthalein solution were added, and the mixture was titrated from a 50 mL burette under continuous stirring until a faint pink color persisted for at least 30 s. The volume of NaOH consumed was recorded. TA was expressed as citric acid equivalents and calculated using the following equation:

TA (%) = [volume of NaOH (mL) × NaOH concentration (N) × 0.06 × 100]/sample volume (mL)

2.8. Color Measurement Analysis

Color changes in the samples were evaluated by determining the differences in L*, a*, and b* values (ΔL*, Δa*, Δb*) between fresh and dried samples using a colorimeter (X-Rite 60, X-Rite Color Technology Co., Ltd., Grand Rapids, MI, USA). Here, L* represents lightness (0 = black, 100 = white), a* represents the red–green axis (−a* = green, +a* = red), and b* represents the yellow–blue axis (−b* = blue, +b* = yellow). For the pomelo peel fermentation broth, color parameters were measured using a colorimeter (X-Rite 60, X-Rite, Inc., Grand Rapids, MI, USA) equipped with a standard illuminant D65 and a 10° observer angle. The instrument was calibrated with a white standard plate prior to measurement, and each sample was analyzed in triplicate with the mean values reported.

2.9. Total Phenolic Content (TPC)

Total phenolic content (TPC) was determined using the Folin–Ciocalteu colorimetric method with slight modifications [23]. Briefly, 10 μL of the sample extract was added to a well of a 96-well microplate, followed by 100 μL of Folin–Ciocalteu reagent. Subsequently, 80 μL of 1 M sodium carbonate (Na₂CO₃) solution was added. The mixture was gently mixed and incubated at room temperature for 15 min. The absorbance was measured at 765 nm using a microplate reader. Gallic acid was used as the standard for calibration, and the results were expressed as mg gallic acid equivalents (GAE) per g sample.

2.10. Total Flavonoid Content (TFC)

Total flavonoid content (TFC) was determined using the aluminum chloride colorimetric method with slight modifications [24]. Briefly, 100 μL of the sample extract was mixed with 30 μL of 5% sodium nitrite (NaNO₂) solution and incubated at room temperature for 1 min. Then, 60 μL of 10% aluminum chloride (AlCl₃) solution was added and allowed to react for 5 min. Afterward, 200 μL of 1 M sodium hydroxide (NaOH) solution was added, and the mixture was thoroughly mixed. A total of 180 μL of the final reaction solution was transferred to a 96-well microplate, and the absorbance was measured at 510 nm using a microplate reader. Rutin was used as the standard compound, and the results were expressed as mg rutin equivalents (RE) per g sample.

2.11. DPPH Radical Scavenging Activity

DPPH radical scavenging activity was determined according to a previously reported method with slight modifications [25]. Briefly, 0.1 mL of the sample extract was mixed with 3.9 mL of 0.1 mM DPPH solution prepared in methanol. The mixture was incubated in the dark at room temperature for 30 min. The absorbance was measured at 517 nm using a UV–Vis spectrophotometer. The radical scavenging activity was calculated as the percentage of inhibition using the following equation:

DPPH scavenging activity (%) = [(A control − A sample)/A control] × 100

2.12. Trolox Equivalent Antioxidant Capacity (TEAC)

The antioxidant capacity of the samples was also evaluated using the ABTS radical cation decolorization assay with slight modifications [26]. The ABTS•⁺ solution was prepared by mixing 7 mM ABTS with 2.45 mM potassium persulfate and allowing the mixture to react in the dark for 12–16 h at room temperature. Before analysis, the ABTS solution was diluted with ethanol to obtain an absorbance of 0.70 ± 0.02 at 734 nm. Subsequently, 0.1 mL of the sample extract was mixed with 3.9 mL of the diluted ABTS solution. After 6 min of reaction at room temperature, the absorbance was measured at 734 nm using a spectrophotometer. Trolox was used as the standard compound, and the results were expressed as µmol Trolox equivalents (TE) per g sample.

2.13. Microbiological Analysis

Lactic acid bacteria (LAB) were enumerated using MicroFast™ Lactic Acid Bacteria Count Plates (LR1312) (MicroFast Co., Ltd., Taichung, Taiwan). Serially diluted samples were aseptically inoculated (1 mL) onto the center of each rehydrated plate and uniformly distributed using the dedicated spreader. The plates were incubated at 30 °C for 48 h, and the results were expressed as log colony-forming units per milliliter (log CFU/mL).

Total viable counts (TVC) were determined using Compact Dry TCR plates (Nissui Pharmaceutical Co., Ltd., Tokyo, Japan). Appropriate serial dilutions were prepared with sterile diluent, and 1 mL of each dilution was inoculated onto the plates. The plates were incubated in an inverted position at 30 °C for 48 h. Plates containing 20–200 colonies were selected for enumeration, and the results were expressed as log CFU/mL.

Yeast and mold counts (YMC) were enumerated using Compact Dry YMR plates (Nissui Pharmaceutical Co., Ltd., Tokyo, Japan). One milliliter of appropriately diluted sample was inoculated onto the center of each plate and allowed to spread automatically. The plates were incubated at 25 ± 1 °C for 72 h, and colonies were counted from the reverse side. The results were expressed as log CFU/mL. All microbiological analyses were performed in duplicate.

2.14. Microbial Community Succession During SCOBY Fermentation

2.14.1. Full-Length 16S rRNA Gene Sequencing

The samples obtained from the fermentation described in Section 2.3, comprising both the liquid broth and suspended pomelo peel particles from each treatment, were collected, homogenized, and stored at −20 °C until DNA extraction. Genomic DNA was extracted using the QIAamp PowerFecal DNA Kit (Qiagen, Hilden, Germany), with optimized steps to enhance polysaccharide removal. DNA concentration and purity were quantified using a Qubit 4.0 Fluorometer (Thermo Fisher Scientific, Waltham, MA, USA) and subsequently adjusted to 1 ng/µL for PCR amplification. Full-length 16S rRNA genes (V1–V9 regions) were amplified using barcoded primers following the PacBio protocol [27]. PCR was performed in 25 µL reaction mixtures containing 2 ng of genomic DNA and KAPA HiFi HotStart ReadyMix (Roche, Basel, Switzerland) under the following conditions: 95 °C for 3 min; 25 cycles of 95 °C for 30 s, 57 °C for 30 s, and 72 °C for 60 s; and a final extension at 72 °C for 5 min. PCR products (~1500 bp) were confirmed by 1% agarose gel electrophoresis and purified using AMPure PB Beads (PacBio, Menlo Park, CA, USA).

2.14.2. SMRTbell Library Preparation and Sequencing

Purified amplicons were pooled and subjected to SMRTbell library preparation according to the PacBio standard workflow [28]. Libraries were purified using AMPure PB Beads, followed by DNA damage repair and adaptor ligation. Library quality and fragment size distribution were evaluated using a Qubit 4.0 Fluorometer and an Agilent 2100 Bioanalyzer (Agilent Technologies, Santa Clara, CA, USA). Sequencing was performed on the PacBio platform to generate circular consensus sequencing (CCS) reads.

2.14.3. Bioinformatic and Statistical Analyses

Raw reads were quality-filtered, and CCS reads were generated from subreads with ≥3 passes. High-quality HiFi reads (RQ > 30) were denoised into amplicon sequence variants (ASVs) using DADA2 (v1.26) [29]. Taxonomic classification was performed against the SILVA database (v138) [30].

Alpha diversity indices, including Chao1, Shannon, and Simpson, were calculated using the vegan package in R. Beta diversity was assessed based on Bray–Curtis dissimilarity, and principal coordinates analysis (PCoA) was performed for visualization. Differences in alpha diversity were evaluated using the Kruskal–Wallis test, while differences in beta diversity were assessed using permutational multivariate analysis of variance (PERMANOVA). Differentially abundant taxa among groups were identified using LEfSe (LDA score > 2.0) and DESeq2 (adjusted p < 0.05), as described in previous studies [31].

2.15. Statistical Analysis

All experiments were conducted in triplicate, and the results are expressed as mean ± standard deviation (SD). Statistical analyses were performed using SAS software (version 9.4, SAS Institute Inc., Cary, NC, USA) and JMP software (version 16, SAS Institute Inc., Cary, NC, USA). Significant differences among groups were determined by one-way analysis of variance (ANOVA) followed by Duncan’s multiple range test, with significance defined at p < 0.05.

3. Results and Discussion

3.1. Changes in pH and Titratable Acidity (TA)

The pH and titratable acidity (TA) of the SCOBY fermentation substrates exhibited clear trends associated with both fermentation time and treatment conditions (Figure 1), consistent with previous studies showing that increased microbial activity accelerates acidification in kombucha-like fermentations [22,32]. In the 1% SCOBY groups, S1-P10 displayed an initial pH of 4.11–4.13 that declined to 3.05–3.10 by day 25, while S1-P25 started at 4.20–4.23 and decreased to 3.35–3.38, indicating that higher pomelo peel addition slightly elevated the initial pH but did not substantially alter the overall acidification trajectory, which aligns with reports that substrate composition can influence initial pH but has limited impact on long-term fermentation dynamics [33]. In contrast, doubling the SCOBY concentration to 6% markedly enhanced acidification, with S6-P10 decreasing from 3.98–4.00 to 2.79–2.81 and S6-P25 from 4.07–4.14 to 3.02–3.05 over the same period, supporting findings that higher inoculum levels increase metabolic rates and organic acid production in symbiotic cultures of bacteria and yeast [34]. Correspondingly, TA increased in all treatments, from 0.01–0.02% to 0.06–0.08% in S1-P10, 0.03–0.04% to 0.07–0.08% in S1-P25, 0.03–0.04% to 0.13–0.16% in S6-P10, and 0.04–0.06% to 0.20–0.22% in S6-P25, indicating an inverse relationship between pH and TA as observed in similar kombucha fermentation systems [31]. These results suggest that SCOBY concentration is a primary factor controlling acidification kinetics, while substrate variation such as pomelo peel addition modulates initial fermentation conditions, consistent with previous literature on the effect of phenolic rich additives on microbial metabolism [21]. These findings help optimize initial conditions and provide guidance for improving acidity and sensory properties in functional fermented beverages.

3.2. Changes in Total Soluble Solids (TSS) and Ethanol Concentration

Total soluble solids (TSS) showed distinct fermentation kinetics depending on SCOBY concentration (

Table 1. Changes in total soluble solids (°Brix) during fermentation of SCOBY substrates prepared with different SCOBY concentrations and pomelo peel additions.

Sample	Fermentation (Days)
	0	5	10	15	20	25
S1-P10	15.1 ± 0.10 ab	15.1 ± 0.06 ab	15.0 ± 0.06 ab	15.0 ± 0.06 b	15.0 ± 0.06 b	14.8 ± 0.06 bc
S1-P25	15.3 ± 0.12 a	15.0 ± 0.00 ab	15.0 ± 0.06 b	15.0 ± 0.06 b	14.9 ± 0.06 b	15.0 ± 0.12 b
S6-P10	14.9 ± 0.06 b	14.9 ± 0.15 b	14.9 ± 0.10 bc	14.8 ± 0.06 b	14.9 ± 0.06 b	14.6 ± 0.12 cd
S6-P25	15.0 ± 0.15 ab	14.9 ± 0.12 b	14.9 ± 0.06 de	14.4 ± 0.12 ef	14.3 ± 0.15 ef	14.1 ± 0.12 f

Values are expressed as the mean ± standard deviation (SD). Different superscript letters (a–f) within the same row indicate significant differences (p < 0.05) according to Tukey’s HSD test.

). In the 1% SCOBY groups, TSS decreased only slightly from approximately 15.0–15.4 to 14.8–15.1 °Brix after 25 days, indicating limited sugar utilization. In contrast, the 6% SCOBY groups, particularly S6-P25, exhibited a more pronounced reduction to 14.0–14.2 °Brix, suggesting enhanced microbial metabolic activity and faster substrate consumption. This trend is consistent with previous kombucha studies reporting that higher inoculum levels accelerate sucrose hydrolysis and monosaccharide utilization due to increased yeast invertase activity and bacterial metabolism [17,35]. The relatively small decrease in TSS in the 1% SCOBY groups indicates a slower fermentation rate and incomplete sugar conversion.

Ethanol was detected in all fermentation treatments; however, the concentration remained below 0.5% (v/v) throughout the fermentation period. This level falls within the regulatory limit for non-alcoholic beverages. In kombucha fermentation, ethanol produced by yeasts is typically oxidized to organic acids by acetic acid bacteria under aerobic conditions, preventing its accumulation in the final product [20,22]. Pomelo peel addition had no substantial effect on ethanol levels but slightly enhanced the reduction in TSS in the high-SCOBY treatment, suggesting that additional nutrients released from the peel may stimulate microbial metabolism.

3.3. Changes in L*a*b* Color Parameters

The L*a*b* color parameters of the SCOBY fermentation substrates were markedly influenced by fermentation time, SCOBY inoculum level, and pomelo peel supplementation (Figure 2). The L* value of S1-P10 decreased from approximately 22.7 at day 0 to 19.5–20.7 at day 25, indicating progressive darkening, whereas a more pronounced reduction was observed in S6-P25, where L* declined from about 31.3 to 24.1–25.4, suggesting intensified pigment formation under higher microbial activity. Similar decreases in lightness during kombucha fermentation have been attributed to polyphenol oxidation, melanoidin formation, and increased turbidity caused by microbial metabolites and bacterial cellulose production [36]. The a* values remained negative in S1-P10 (−0.52 to −0.06), indicating limited red pigment development, while pomelo peel-supplemented samples exhibited positive values throughout fermentation, with S6-P25 increasing from approximately 1.32 to 3.02–3.08, reflecting the formation of reddish-brown compounds. This shift is likely associated with the release and transformation of citrus carotenoids and flavonoids from the peel matrix and their subsequent oxidation or polymerization during fermentation [23]. The b* values decreased in S1-P10 from about 7.9 to 6.0–6.7, but remained high or increased in S6-P10 and S6-P25, reaching 15.2–16.2 and 15.4–16.9 at day 25, respectively, indicating enhanced accumulation of yellow–brown pigments. Such behavior has been reported in fruit-based kombucha, where phenolic bioconversion and acidic conditions promote the formation of complex brown pigments and chromatic intensification [15].

3.4. Bioactive Compounds and Antioxidant Activity

The evolution of bioactive compounds and antioxidant capacity during fermentation was strongly affected by SCOBY inoculum level and pomelo peel supplementation (Figure 3). The total phenolic content (TPC) increased markedly in all treatments, with S1-P10 rising from approximately 28.8–32.6 to 97.7–100.9 mg GAE/g DW at day 20, followed by a slight decline at day 25, whereas S1-P25 exhibited substantially higher values, reaching 149.0–151.5 mg GAE/g DW at day 20 and remaining above 141 mg GAE/g DW at the end of fermentation. A similar trend was observed for S6-P10, where TPC peaked at 114.6–120.9 mg GAE/g DW on day 20, while S6-P25 showed a gradual increase from about 69.9 to 125.8–130.8 mg GAE/g DW. This enhancement is commonly attributed to microbial enzymatic hydrolysis of bound phenolics, acid-induced cell wall degradation, and bioconversion of complex polyphenols into smaller and more extractable molecules during kombucha fermentation [37,38]. The total flavonoid content (TFC) also increased during fermentation, with S1-P10 rising from 4.10–5.23 to 7.48–7.68 mg RE/g DW and S6-P25 reaching the highest values (10.99–11.85 mg RE/g DW at day 20), confirming the release and transformation of citrus-derived flavonoids under intensified microbial metabolism [23]. The antioxidant activity exhibited trends consistent with phenolic accumulation. DPPH radical scavenging activity in S1-P10 increased from about 30–31% to approximately 43–45%, whereas significantly higher activity was observed in S6-P25, which increased from 49.8–51.6% to 70.6–73.0% at day 25. Likewise, TEAC values increased from 16.7–20.4 to 32.9–38.0 μmol TE/g DW in S1-P10 and remained at relatively high levels in S1-P25 and S6-P10 (approximately 40–45 μmol TE/g DW), indicating enhanced electron-donating capacity. The strong correlation between phenolic content and antioxidant activity suggests that phenolic biotransformation plays a dominant role in the improvement of functional properties, as widely reported in kombucha and fruit-based fermented beverages [37,39]. Moreover, the higher SCOBY inoculum accelerated metabolic activity, leading to faster acidification and more efficient enzymatic conversion, which promoted the liberation of phenolics and flavonoids from the pomelo peel matrix. The slight decline in TPC observed after prolonged fermentation may be associated with oxidative degradation or polymerization of phenolic compounds into insoluble forms [40].

3.5. Changes in Total Viable Count (TVC), Lactic Acid Bacteria (LAB), and Yeast and Mold Count (YMC)

The microbial dynamics of SCOBY fermentation substrates prepared with different SCOBY concentrations and pomelo peel additions are presented in terms of total viable count (TVC), lactic acid bacteria (LAB), and yeast and mold populations (Figure 4). The initial total viable count (TVC) ranged from 4.93 to 5.43 log CFU/mL, indicating a comparable microbial baseline among treatments. A rapid increase was observed during the first 10 days, reaching approximately 7.07–7.25 log CFU/mL, indicating active microbial proliferation. Nevertheless, the relatively stable TSS values suggest that the consumption of soluble sugars may have been balanced by the release of soluble compounds from pomelo peel, thereby maintaining the overall TSS level. Thereafter, the TVC remained relatively stable, suggesting the establishment of a balanced microbial consortium. At day 25, higher values were observed in S1-P25 (7.40 log CFU/mL) and S6-P25 (7.23 log CFU/mL) compared with the P10 treatments, indicating that pomelo peel supplementation had a stronger effect on bacterial growth than increasing the SCOBY inoculum from 1% to 6%. This result is consistent with previous studies reporting that fruit or plant by-product supplementation enhances microbial proliferation in kombucha fermentation by providing additional carbon sources, micronutrients, and growth-promoting compounds, leading to bacterial populations on the order of 10⁷–10⁸ CFU/mL [20].

The lactic acid bacteria (LAB) population increased from an initial level of 3.54–3.95 log CFU/mL to around 6.0 log CFU/mL by day 10 and remained relatively constant thereafter. The higher LAB counts observed in the P25 treatments (6.12 log CFU/mL in S1-P25 and 6.04 log CFU/mL in S6-P25 at day 25) suggest that pomelo peel provided favorable substrates such as soluble carbohydrates and pectin-derived oligosaccharides that may exert prebiotic-like effects. The stabilization of LAB in the later fermentation stage is likely associated with the accumulation of organic acids and the progressive decline in pH, which limits excessive microbial growth while maintaining a metabolically active population. Similar LAB levels (approximately 10⁵–10⁶ CFU/mL) have been reported in kombucha and kombucha-like systems during the maturation phase of fermentation [39].

In contrast, the yeast and mold count (YMC) showed a continuous decline throughout fermentation. The initial population (2.60–2.65 log CFU/mL) decreased markedly after 5 days and became non-detectable after 10–15 days in most treatments. This reduction can be attributed to the increasing acidity, the production of antimicrobial metabolites, and the competitive dominance of acetic acid bacteria and LAB, which create an unfavorable environment for yeast and mold survival. A similar succession pattern has been described in kombucha fermentation, where yeasts are active in the early stage but are progressively suppressed by bacterial metabolism and acid accumulation in the later stage [41].

3.6. Microbial Community Structure and Succession Revealed by Heat Tree Analysis

The heat tree analysis revealed clear shifts in microbial community structure between the initial and late fermentation stages as well as between different SCOBY inoculum levels under the same pomelo peel supplementation (Figure 5). At day 0, both S1-P25 and S6-P25 exhibited relatively complex microbial communities dominated by Proteobacteria, Firmicutes, Bacteroidota, and Actinobacteriota, indicating that the initial microbiota mainly originated from the SCOBY matrix and raw materials. However, a higher relative abundance of Lactobacillales and Lactobacillaceae, particularly Acetilactobacillus jinshanensis, was already observed in S6-P25, suggesting that increasing the inoculum size accelerated the early establishment of lactic acid bacteria as key functional members. This phenomenon has been reported in kombucha-like fermentations, where a higher inoculum shortens the microbial adaptation phase and promotes the rapid dominance of acid-tolerant taxa [37,41]. After 25 days of fermentation, the microbial community became markedly simplified and functionally oriented. In both treatments, Firmicutes emerged as the dominant phylum, with a pronounced enrichment of Lactobacillales, indicating the formation of an acid-adapted microbial ecosystem. The persistence and enlargement of nodes affiliated with Acetilactobacillus in S6-P25 suggest a metabolically active LAB population closely associated with carbohydrate metabolism and organic acid production. In contrast, several taxa belonging to Proteobacteria and Bacteroidota that were present at day 0 were reduced or disappeared at day 25, which can be attributed to the progressive decline in pH and the accumulation of antimicrobial metabolites during fermentation. Similar microbial succession patterns, characterized by the transition from a diverse microbiota to a LAB- and acetic acid bacteria-dominated consortium, are typical for kombucha fermentation systems [38]. Notably, S6-P25 exhibited a more compact and LAB-centered microbial network than S1-P25 at day 25, indicating that the higher SCOBY concentration enhanced microbial selection toward functionally relevant taxa and improved community stability. Such structural simplification is often associated with increased metabolic efficiency and fermentation robustness [42]. Furthermore, the enrichment of LAB-related taxa corresponded well with the enhanced bioactive compound content and antioxidant activity observed under the same condition, suggesting a strong linkage between microbial community assembly and functional metabolite production. LAB is known to release bound phenolic compounds through enzymatic hydrolysis and to synthesize bioactive metabolites during sugar metabolism, thereby contributing to the improved functional properties of fermented substrates [23].

3.7. Phylogenetic Tree Analysis of the Core Microbiome in SCOBY Fermentation

The phylogenetic tree of the core microbiome in SCOBY fermentation substrates revealed a clear shift in microbial structure as fermentation progressed and as a function of SCOBY inoculum level (Figure 6). At day 0, both S1-P25 and S6-P25 exhibited relatively complex bacterial communities dominated by members of Proteobacteria, Firmicutes, Actinobacteriota, and Bacteroidota, indicating that the initial fermentation matrix provided diverse ecological niches for multiple microbial groups. Within these phyla, lactic acid bacteria such as Apilactobacillus ozensis and Acetilactobacillus jinshanensis were already present, suggesting their origin from the SCOBY inoculum and their rapid adaptation to the sucrose-rich environment. In addition, the detection of Bifidobacterium pseudolongum subsp. globosum and Phocaeicola plebeius implies the contribution of pomelo peel as a source of plant-derived polysaccharides that support saccharolytic bacteria [40].

As fermentation proceeded to day 25, a marked reduction in community complexity was observed, accompanied by the strong enrichment of acid-tolerant taxa. In S1-P25, the microbial structure became largely dominated by Acetobacter okinawensis, whereas S6-P25 showed a simplified consortium mainly composed of Streptococcus thermophilus. This transition reflects the progressive acidification of the fermentation system and the accumulation of organic acids, which selectively favor acid-resistant and metabolically specialized species while suppressing less tolerant microorganisms. Similar microbial succession patterns have been reported in kombucha and other symbiotic fermentations, where early-stage diversity is gradually replaced by a stable core microbiota dominated by acetic acid bacteria and lactic acid bacteria under low pH conditions [43]. Microbiome analysis demonstrated that Firmicutes was the dominant phylum, with Acetilactobacillus jinshanensis accounting for 95.36% of the relative abundance, indicating strong microbial selection and its potential role in acid production and fermentation stability. Moreover, the higher initial SCOBY concentration (6%) appeared to accelerate microbial selection, leading to a more rapid establishment of a simplified but stable community structure at the end of fermentation. This observation supports the concept that inoculum size influences not only microbial growth kinetics but also ecological competition and niche occupation [44]. Collectively, the present results suggest that pomelo peel supplementation provides a suitable substrate for the development of a functional SCOBY-associated microbiome, while fermentation time and SCOBY concentration jointly shape the transition from a diverse initial consortium to a specialized and acid-tolerant core microbiota, which is essential for the stable production of fermentation metabolites and the functional properties of the final product.

3.8. Species-Level Microbial Succession During Fermentation

Species-level analysis clearly demonstrated a pronounced ecological succession between day 0 and day 25 under different SCOBY inoculum levels (Figure 7). At day 0, treatment S1-P25 and S6-P25 were both dominated by Sinocapsa zengkensis, accounting for 99.47% and 94.80% of the relative abundance, respectively. Minor taxa detected in S6-P25 at day 0 included Cutibacterium acnes (1.95%), Staphylococcus capitis (0.63%), and Mucisphaera calidilacus (0.42%), while other species remained below 0.35%. The overwhelming predominance of S. zengkensis at the initial stage suggests that substrate-associated or environmental microorganisms were prevalent before significant acidification occurred, which is consistent with reports describing diverse early-stage microbiota in kombucha fermentation systems [45]. After 25 days of fermentation, a dramatic shift in community structure was observed. In treatment S1-P25, Apilactobacillus ozensis DSM 23829 (=JCM 17196) increased to 99.97% relative abundance, while in treatment S6-P25, it accounted for 99.06%. Concurrently, S. zengkensis was completely eliminated (0.00%) in both treatments. Only trace levels of Acetobacter okinawensis JCM 25146 (0.92% in S6-P25 at day 25) and “Other” taxa (0.03–0.02%) were detected. This near monodominance of A. ozensis indicates strong acid tolerance and competitive fitness under prolonged fermentation conditions. The ecological simplification from a diverse initial microbiota to a Firmicutes-dominated community aligns with previous studies showing that sustained acid production selectively favors lactic acid bacteria while suppressing non-acid-resistant taxa [46,47]. Interestingly, Acetilactobacillus jinshanensis, detected at low abundance at day 0 (0.22% in S1-P25 and 0.34% in S6-P25), was no longer detectable at day 25, suggesting competitive displacement during acid-driven succession. The comparable dominance levels of A. ozensis in both S1 and S6 treatments at day 25 indicate that, although inoculum size influences early fermentation kinetics, prolonged acidic conditions ultimately converge toward a highly specialized and stable community structure. Such ecological convergence has been reported in extended acidic fermentations, where strong environmental filtering results in reduced diversity but enhanced functional stability [45].

Overall, these results indicate a clear transition from a diverse initial microbiota to a highly specialized, acid-adapted community dominated by lactic acid bacteria. This shift is closely associated with enhanced bioactive compound release and antioxidant activity, suggesting a strong link between microbial succession and functional metabolite production.

4. Conclusions

This study evaluated the effects of SCOBY inoculum size and Pomelo peel addition on 25-day fermentation kinetics, microbial succession, physicochemical properties, and antioxidant activity. Higher inoculum (6% w/w) accelerated pH decline and increased titratable acidity, while lower inoculum (1%) with high peel supplementation enhanced total phenolic accumulation. Pomelo peel promoted lactic acid bacteria growth, flavonoid release, and antioxidant capacity, likely via enzymatic hydrolysis and acid-mediated cell wall degradation. 16S rRNA sequencing revealed a shift from diverse initial microbiota to an acid-adapted Firmicutes-dominated community, with Apilactobacillus ozensis as the predominant species under high inoculum. These results indicate that inoculum size not only influences fermentation rate but also shapes microbial community assembly and functional stability. The integration of fermentation kinetics and microbiome analysis provides valuable insights for optimizing kombucha-like fermentation systems using agricultural by-products.