Bacterial and Physicochemical Dynamics During the Vermicomposting of Bovine Manure: A Comparative Analysis of the Eisenia fetida Gut and Compost Matrix

Abstract

Vermicomposting is a sustainable biotechnological process that transforms organic waste through the synergistic activity of earthworms, such as Eisenia fetida, and their associated microbiota. This study evaluated bacterial and physicochemical dynamics during the vermicomposting of bovine manure by analyzing the microbial composition of the substrate and the gut of E. fetida at three time points (weeks 0, 6, and 12). The V3–V4 region of the 16S rRNA gene was sequenced, and microbial diversity was characterized using QIIME2. Significant differences in alpha diversity (observed features, Shannon index, and phylogenetic diversity) and beta diversity indicated active microbial succession. Proteobacteria, Bacteroidota, and Actinobacteriota were the dominant phyla, with abundances varying across habitats and over time. A significant enrichment of Proteobacteria, Bacteroidota, and the genera Chryseolinea, Flavobacterium, and Sphingomonas was observed in the manure treatments. In contrast, Actinobacteriota, Firmicutes, and the genera Methylobacter, Brevibacillus, Enhygromyxa, and Bacillus, among others, were distinctive of the gut samples and contributed to their dissimilarity from the manure treatments. Simultaneously, the physicochemical parameters indicated progressive substrate stabilization and nutrient enrichment. Notably, the organic matter and total organic carbon contents decreased (from 79.47% to 47.80% and from 46.10% to 27.73%, respectively), whereas the total nitrogen content increased (from 1.70% to 2.23%); these effects reduced the C/N ratio, which is a recognized indicator of maturity, from 27.13 to 12.40. The macronutrient contents also increased, with final values of 1.41% for phosphorus, 1.50% for potassium, 0.89% for magnesium, and 2.81% for calcium. These results demonstrate that vermicomposting modifies microbial communities and enhances substrate quality, supporting its use as a biofertilizer for sustainable agriculture, soil restoration, and agrochemical reduction.

1. Introduction

Vermicomposting is an effective biotechnological strategy for the sustainable management of organic waste, in which earthworms—mainly Eisenia fetida—transform organic matter through digestion and synergistic action with their associated microbial communities [1,2,3]. During this process, both the external environment of the substrate (e.g., manure) and the internal environment of the earthworm (gut) serve as dynamic ecological niches where complex microbial succession occurs [4,5]. The vermicomposting process, which converts organic waste into nutrient-rich vermicompost via earthworms, is strongly influenced by interactions among microbial communities residing in both the gut and the surrounding matrix [6,7,8]. Understanding changes in bacterial composition and function during cow dung vermicomposting is essential for optimizing the process and improving the final product’s quality [9,10,11]. In this context, physicochemical analyses play crucial roles in the monitoring and evaluation of vermicomposting progress and effectiveness. Parameters such as pH, electrical conductivity (EC), moisture (M), organic matter (OM), total organic carbon (TOC), and nutrient concentrations (e.g., N, P, K, Ca, and Mg) provide essential information about substrate stability, maturity, and nutrient bioavailability [12,13,14]. These indicators reflect the biochemical transformations that are driven by microbial metabolism and earthworm activity, and they are essential for validating the agronomic quality of the final product [15,16,17]. For example, reductions in the C/N ratio and organic matter content, in addition to increases in total nitrogen and available nutrient contents, are commonly used benchmarks to determine vermicompost maturity and safety for soil application [18,19]. Moreover, physicochemical shifts in the substrate directly influence microbial selection pressures, affecting the composition and functional potential of microbial communities throughout the process [20,21,22].

Earthworms, considered ecosystem engineers, play critical roles in aerating, conditioning, and fragmenting the substrate, significantly altering microbial activity and biodegradation potential [11,23,24]. The guts of hosts such as Eudrilus eugeniae, Perionyx excavatus, and Polypheretima elongata harbor a diverse array of microorganisms, including bacteria, fungi, and protozoa, that contribute to the decomposition of complex organic matter [25,26,27]. In particular, dynamic changes in bacterial communities within the earthworm gut (e.g., Eisenia fetida) result from complex interactions involving digestive enzymes, mucus rich in antimicrobial peptides, and even antibiotics produced by certain bacterial strains [28,29]. These factors strongly influence the depolymerization of complex organic molecules and the overall transformation of the substrate [13,14,15]. Consequently, the bacterial communities in vermicompost are not static but evolve as the process progresses, reflecting changes in substrate chemistry and microbial metabolism [30,31]. Notably, gut transit in earthworms significantly modifies the microbial community structure, often reducing the abundance of pathogenic bacteria while increasing the abundance of beneficial microorganisms [32,33]. Among the beneficial taxa are genera such as Bacillus, Azotobacter, Rhizobium, and Actinomycetes, which play key roles in nutrient cycling, plant growth promotion, and pathogen suppression [11,18,34,35]. The dynamic interaction between the gut microbiome of earthworms and the bacterial communities in the vermicompost matrix is fundamental to this process [36,37]. Understanding this interaction can provide valuable insights for optimizing vermicomposting systems, increasing nutrient availability, and reducing pathogens in the final product [38,39]. Previous studies have shown that the gut environment of earthworms provides a unique ecological niche that selectively enriches certain bacterial groups, resulting in distinct microbial compositions compared with those of adjacent soils [40,41,42]. Conditions in the gut, such as a neutral pH, high moisture content, and elevated organic compound concentrations, create an ideal setting for microbial growth and activity [43,44]. Importantly, earthworms stimulate microbial activity and increase microbial populations [45,46], enhancing organic matter decomposition and nutrient release in plant-available forms [47,48]. Specifically, the gut of earthworms has the highest concentrations of total nitrogen, water-soluble organic matter, nitrite, ammonium, and iron, with a decreasing gradient from the foregut to the hindgut [39,46,49].

Despite significant advances in the study of vermicomposting, critical knowledge gaps remain about the parallel evolution and interaction of microbial communities in the earthworm gut and composting substrates throughout the process. Most previous studies have focused on characterizing these environments in isolation or at single time points, limiting our understanding of the ecological and functional mechanisms that drive microbial succession over time. Comparative analyses that simultaneously examine the microbial dynamics of the earthworm gut and the surrounding organic matrix across different composting stages are lacking. Moreover, few studies have integrated microbial and physicochemical perspectives to comprehensively evaluate vermicomposting performance and maturation. This integrated approach is essential for elucidating the biological filtering, ecological selection, and functional enrichment processes that shape microbial communities and determine compost quality. Therefore, in this study, bacterial dynamics at the taxonomic and diversity levels were assessed in manure and E. fetida gut samples at different stages of vermicomposting. In addition, key physicochemical parameters (pH, EC, moisture, OM, TOC, total nitrogen, C/N ratio, P, K, Mg, and Ca) were analyzed to identify patterns of microbial succession and substrate transformation, as well as their implications for sustainable, agrochemical-free agriculture and soil restoration.

2. Materials and Methods

2.1. Sample Collection

The experiment was conducted at the Instituto Tecnológico Superior de Lerdo using manure from adult dairy cows (Holstein Friesian breed) collected at the Eucalyptus Farm in the Ejido 13 de Marzo, Gómez Palacio, Durango, México (25°42′29.808″ N, 103°29′45.614″ W; Figure 1). The manure was pre-composted by mixing, homogenizing, and keeping it moist to lower the pH. After this pretreatment, the experiment was initiated.

Three plastic containers (40 cm long, 25 cm wide and 33 cm high) were fitted with 1/2-inch PVC pipes to collect leachate (Figure 2). These containers were treated as independent biological replicates. Each container received 4 kg of manure and 50 g of adult E. fetida earthworms (100 days old), sourced from the worm farm at the Instituto Tecnológico Superior de Lerdo. The temperature was maintained at 25 °C, and the humidity was 80%. Irrigation was automated via sprinklers controlled by soil moisture sensors connected to an Arduino Uno board, with real-time data displayed on an LED screen. The experiment lasted 12 weeks. At each of the three sampling points (week 0, 6, and 12), manure and gut samples were collected independently from each container. Then, 30 g of pre-composted manure (ET1), mid-vermicomposting stage (ET2) and vermicompost (ET3) were collected in sterile Falcon tubes and stored at −80 °C for DNA extraction. For the gut samples, 20 randomly selected pre-inoculation stage (LT1), mid-vermicomposting stage (LT2), and vermicompost (LT3) earthworms were rinsed thoroughly with sterile cold saline solution (0.9% NaCl), euthanized in 70% ethanol, and aseptically dissected on a sterile tray. The entire gut was extracted, and the gut contents were scraped into sterile Eppendorf tubes (~0.5 mL) and stored at −80 °C. This design provided three biological replicates per treatment per sampling time.

2.2. Laboratory Work

DNA was extracted from all the samples using the Zymo Research™ DNA Zymobiomics Kit, Irvine, CA, USA. DNA quantity and quality were assessed with a ThermoScientific^® Nanodrop spectrophotometer (Waltham, MA, USA). The V3–V4 region of the 16S rRNA gene was amplified using primers described by [50]: S-D-Bact-0341-b-S-17, 5′-CCTACGGGNGGCWGCAG-3′ and S-D-Bact-0785-a-A-21, 5′-GACTACHVGGGTATCTAATCC-3′. This produced ~460 bp amplicons, which were then synthesized with Illumina “overhang” adapters [51], generating ~550 bp products. The PCRs were followed as per the Illumina’s [51] protocol and included 12.5 μL of MyTaq^TM Ready Mix 1X (Bioline^®, Meridian Bioscience, Inc., Cincinnati, OH, USA), 1 μL of each primer (10 µM), 5 μL of DNA (50 ng total), and 5.5 μL of molecular-grade water. Thermal cycling was carried out in a Labnet Multigene™ Gradient PCR thermocycler: initial denaturation at 95 °C for 3 min; 25 cycles of 95 °C for 30 sec, 55 °C for 30 sec, and 72 °C for 30 sec; and a final extension at 72 °C for 5 min. The amplicon size (~550 bp) was verified on a Bioanalyzer DNA 1000 (Agilent Technologies, Inc., Santa Clara, CA, USA) chip. Amplicons were purified with 0.8× Agencourt^® AMPure^® XP (Beckman Coulter, Inc., Brea, CA, USA) beads. Indexing was performed using the Nextera kit (Illumina, Inc., San Diego, CA, USA) [52] with a 10-cycle PCR (same conditions as above), followed by a second purification step using 1.2× AMPure^® XP (Beckman Coulter, Inc., Brea, CA, USA) beads. The final library size (~630 bp) was confirmed with a Bioanalyzer chip. Libraries were quantified, normalized to equimolar concentrations, and sequenced via Illumina MiSeq^® (Illumina, Inc., San Diego, CA, USA) with 2 × 250 bp paired-end reads, according to the 16S metagenomics protocol.

2.3. Bioinformatics Analysis

The sequencing data were processed via Quantitative Insights into Microbial Ecology (QIIME2) on a Linux Ubuntu platform [53]. Low-quality reads and chimeras were removed, and amplicon sequence variants (ASVs) were generated via the DADA2 algorithm [54]. Taxonomic classification was performed via the sklearn classifier against the Greengenes2 database [55] across all taxonomic levels. Good’s coverage index was calculated in RStudio ver. 2025.05.1-513 for each sample to assess whether sequencing depth was sufficient, with values between 0.95 and 1.00 generally considered indicative of adequate coverage. The relative abundances of the dominant taxa at the phylum, family, and genus levels were visualized with heatmaps in Morpheus (Broad Institute). Venn diagrams (InteractiVenn) were used to assess shared and unique ASVs across treatments. To statistically compare bacterial communities, rarefaction was applied to standardize sequence counts across samples. Alpha diversity metrics (observed features, the Shannon index, Pielou’s evenness, and Faith’s phylogenetic diversity) were calculated, and differences among treatments were evaluated via the Kruskal–Wallis and Dunn tests (p < 0.05). Significant results were visualized as bar plots in GraphPad Prism v8.0.2. Beta diversity was assessed via Jaccard, Bray–Curtis, unweighted UniFrac, and weighted UniFrac distances [56,57,58]. Permutational multivariate analysis of variance (PERMANOVA) tests were applied (p < 0.05) to beta metrics to determine statistical significance. Significant beta metrics were visualized via principal coordinate analysis (PCoA) plots via Emperor [59]. SIMPER analysis [60] was performed at the phylum, family, and genus levels in PAST v5.1 to identify the taxa contributing most to group dissimilarity. Kruskal–Wallis tests (p < 0.05) were applied to taxa contributing >0.1%. Differentially abundant taxa were visualized in CLUSTVIS heatmaps [61].

2.4. Physicochemical Analysis

The physicochemical parameters (pH, EC, M, OM, TOC, total nitrogen, carbon/nitrogen ratio, P, K, Mg, and Ca) were measured at three time points (ET1, ET2, and ET3) in one-kilogram samples. pH, EC, and total nitrogen were analyzed at the Environmental Laboratory of the Instituto Tecnológico Superior de Lerdo, in Lerdo, Durango, Mexico. pH and EC of 1:10 water-to-sample suspensions were measured. The mixtures were stirred at 230 rpm for 30 min and allowed to settle for one hour prior to measurement. A pocket pH/conductivity meter (model HI98129, HANNA Instruments^®, Smithfield, RI, USA) was used, as described previously [62]. TOC was determined using the Kjeldahl method. The carbon-to-nitrogen (C/N) ratio was calculated to evaluate the balance between carbon and nitrogen, which influences decomposition rates and nutritional quality. Determinations of M, OM, TOC, P, K, Mg, and Ca contents were conducted at the National Laboratory of Water, Soil, Plant, and Environmental Analysis Services (Laboratorio Nacional de Servicios de Análisis de Agua, Suelos, Plantas y Medio Ambiente) at the Centro Nacional de Investigación Disciplinaria en Relación Agua, Suelo, Planta, Atmósfera (CENID-RASPA), located in Gomez Palacio, Durango, Mexico, following the procedures outlined in NOM-021-RECNAT-2000 [63]. For each physicochemical parameter, the mean and standard deviation were calculated.

3. Results

The average number of reads obtained for ET1, ET2, and ET3 was 39,807.6, 166,567, and 149,680.6, respectively. For LT1, LT2, and LT3, the averages were 39,554.3, 145,601.5, and 160,544, respectively. After DADA2 processing, the average number of non-chimeric sequences was as follows: ET1—13,199 (33.06%), ET2—81,530.3 (48.94%), ET3—66,870.6 (44.38%), LT1—8088.6 (20.17%), LT2—66,815.5 (45.85%), and LT3—75,744.5 (47.18%) (Table S1). Good’s coverage values ranged from 0.951 to 0.999 across all samples (see Table S1, last column), indicating that the sequencing depth was adequate to reflect most of the bacterial diversity present in each sample. The bacterial taxon richness at each taxonomic level per treatment is summarized in

Table 1. Number of bacterial taxa identified at each taxonomic level across treatments.

Treatment	Phylum	Class	Order	Family	Genus	Species
ET1	31	59	162	264	438	523
ET2	41	94	269	482	963	1224
ET3	51	111	294	517	924	843
LT1	31	55	150	238	340	378
LT2	45	95	268	487	891	1089
LT3	41	95	264	486	881	1073

and detailed in Table S2.

The dominant phyla in ET1, ET2, and ET3 were Proteobacteria and Bacteroidota, with mean abundances of 41.52%, 35.96%, and 48.32%, respectively, for Proteobacteria and 25.89%, 12.00%, and 21.48%, respectively, for Bacteroidota (Figure 3). In the worm gut samples, Proteobacteria and Actinobacteriota were most abundant in LT1 (37.69%, 22.96%), Actinobacteriota and Proteobacteria were most abundant in LT2 (25.05%, 22.04%), and Actinobacteriota and Firmicutes_D were most abundant in LT3 (23.99%, 21.52%) (Figure 4).

At the family level, the most abundant families in ET1 were Polyangiaceae, Xanthomonadaceae_616009, and Cyclobacteriaceae_900466 (8.40%, 7.00%, and 6.16%, respectively). In ET2, Xanthomonadaceae_616009, Flavobacteriaceae, and Polyangiaceae predominated (10.36%, 5.57%, and 4.51%, respectively). In ET3, Flavobacteriaceae (8.91%) and Cyclobacteriaceae_900466 (7.00%) were the most prevalent (Figure 5). In LT1, the dominant families were Aeromonadaceae (9.51%), Rhodobacteraceae (8.92%), and Pseudomonadaceae (6.00%). In LT2, the most abundant were Microbacteriaceae, Peptostreptococcaceae_256921, and Planococcaceae (4.49%, 4.44%, and 4.39%, respectively). In LT3, Peptostreptococcaceae_256921, Planococcaceae, and Brevibacillaceae dominated (4.71%, 3.51%, and 3.17%, respectively) (Figure 6).

At the genus level, the microbial composition exhibited dynamic shifts throughout the vermicomposting process. ET1 was dominated by Luteimonas_C_615545 (8.34%), Lysobacter_A_615544 (5.51%), and Bacillus_O (3.49%). In ET2, the most abundant bacteria were Flavobacterium (4.77%), Sphingomonas_L_486704 (4.12%), and Cellvibrio (4.11%). ET3 had relatively high abundances of Chryseolinea (8.84%) and Arenimonas_613591 (5.47%) (Figure 7). For LT1, Aeromonas (9.51%) and Pseudomonas_K (8.92%) were the most abundant. In LT2, Methylobacter_A_601880, Enhygromyxa, and Brevibacillus_D (4.49%, 4.15%, 3.82%) dominated, whereas in LT3, Brevibacillus_D was the most abundant genus (4.31%) (Figure 8).

All six treatments (ET1, ET2, ET3, LT1, LT2, and LT3) shared 25 phyla, including Proteobacteria, Actinobacteriota, Bacteroidota, Firmicutes_D, and Chloroflexota, among others (Figure 9a). Unique phyla were identified in LT1 (Synergistota), LT2 (Firmicutes_H), and ET3 (Omnitrophota, Eisenbacteria, Margulisbacteria). At the family level, 109 families were shared among all the treatments, whereas unique families were observed in LT1, LT2, LT3, ET2, and ET3 (32, 21, 27, 29, and 60, respectively) (Figure 9b). At the genus level, 126 genera (8.3%) were common across all the samples, whereas LT1, LT2, LT3, ET2, and ET3 had 61, 69, 79, 146, and 117 unique genera, respectively. ET1 did not exhibit any unique taxa at the phylum, family, or genus levels (Figure 9c).

Significant differences were observed in three alpha diversity metrics: observed features (H = 12.65, p = 0.026; LT1_mean = 324.7, LT2_mean = 1272, LT3_mean = 1204, ET1_mean = 524.3, ET2_mean = 1212, ET3_mean = 1123), the Shannon index (H = 12.76, p = 0.025; LT1_mean = 6.9, LT2_mean= 8.9, LT3_mean = 8.9, ET1_mean= 7.8, ET2_mean= 8.9, ET3_mean = 8.5), and Faith’s phylogenetic diversity (H = 13.60, p = 0.018; LT1_mean = 32.4, LT2_mean = 83.0, LT3_mean = 72.7, ET1_mean = 46.5, ET2_mean = 82.0, ET3_mean = 87.1) (Figure 3). Pielou evenness showed no significant differences (H = 10.27, p = 0.067; LT1_mean = 0.83, LT2_mean = 0.87, LT3_mean = 0.87, ET1_mean = 0.86, ET2_mean = 0.87, ET3_mean = 0.84). The results are presented in Figure 10.

Beta diversity metrics also revealed significant differences (PERMANOVA p = 0.001 for all): Jaccard index pseudo-F = 3.07; Bray–Curtis pseudo-F = 6.16; unweighted UniFrac pseudo-F = 4.28; and weighted UniFrac pseudo-F = 10.33. Principal coordinate analysis (PCoA) plots clearly revealed separation among the treatments for all the indices (Figure 11).

SIMPER analysis identified 40 phyla contributing to treatment dissimilarities. Proteobacteria contributed the most (29.53%), being most abundant in ET2 and least abundant in ET1. Actinobacteriota contributed 14.68% of the total bacteria enriched in LT3 and scarce in ET1. Bacteroidota contributed to 12.87% of the total bacteria, abundant in ET2 and rare in LT2 (Figure 12). At the family level, 31 taxa significantly contributed to the dissimilarity. Sphingomonadaceae had the highest contribution (5.65%), most abundant in ET2 and least abundant in LT1. Other contributors included Cyclobacteriaceae_900466 (3.14%) and Xanthomonadaceae_616009 (2.96%) (Figure 13). Among the 31 genera contributing significantly to group dissimilarity, Chryseolinea (2.01%) was most abundant in ET2 and nearly absent in LT1. Flavobacterium (1.82%) was prominent in ET2 but rare in ET3. Sphingomonas_L_486704 (1.37%) was enriched in ET2 and depleted in all LT samples. Aeromonas was enriched exclusively in LT1, whereas in LT2 and LT3 other genera (e.g., Methylobacter, Brevibacillus, Enhygromyxa, Bacillus, among others) became significantly dominant (Figure 14).

Various physicochemical parameters were evaluated at three stages of the process of bovine manure vermicomposting: ET1 (pre-composted manure), ET2 (6 weeks), and ET3 (12 weeks). The results revealed changes that were consistent with the transformation and stabilization of the organic material. The substrate pH progressively decreased throughout the process, starting from alkaline values in ET1 (mean = 8.07) and reaching near-neutral values in ET3 (mean = 7.28). EC also decreased from an initial mean of 1.97 dS/m in ET1 to 1.38 dS/m in ET3, suggesting a reduction in the concentration of soluble salts. The moisture increased considerably after the process began, increasing from 63.33% in ET1 to values above 80% in ET2 and ET3, with notable stability between weeks 6 and 12. OM decreased significantly, from 79.47% in ET1 to 64.47% in ET2, and further decreased to 47.80% in ET3. This decrease was attributed to the automated irrigation system, which provided better control and contributed to the stabilization of this parameter. Similarly, TOC decreased from a mean of 46.10% in ET1 to 37.40% in ET2 and then to 27.73% in ET3. The total nitrogen content continuously increased as the process progressed, with mean values increasing from 1.70% in ET1 to 1.89% in ET2 and reaching 2.23% in ET3. This trend, together with the reduction in organic carbon contents, resulted in a marked decrease in the C/N ratio from a mean of 27.13 in ET1 to 19.82 in ET2 and finally to 12.40 in ET3. This final value indicates adequate vermicompost maturity. With respect to essential macronutrients, P increased slightly, from a mean of 1.14% in ET1 to 1.41% in ET3. The potassium (K) content increased considerably, with mean values of 0.74% in ET1, 1.22% in ET2, and 1.50% in ET3. The magnesium (Mg) content also increased, from 0.25% in ET1 to 0.89% in ET3. Finally, the calcium (Ca) content showed the most notable increase, increasing from 0.85% in ET1 to 1.35% in ET2 and 2.81% in ET3. Overall, these results demonstrate the efficiency of vermicomposting in transforming bovine manure, promoting substrate stabilization, reducing phytotoxic compound levels, and substantially enriching the final product with essential plant nutrients. The complete data for each parameter evaluated are presented in

Table 2. Physicochemical parameters during bovine manure vermicomposting process.

Stage	pH	EC (dS/m)	M (%)	OM (%)	TOC (%)	TN (%)	C/N Ratio	P (%)	K (%)	Mg (%)	Ca (%)
ET1	8.07 ± 0.04	1.97 ± 0.03	63.33 ± 1.52	79.47 ± 1.62	46.10 ± 0.94	1.70 ± 0.05	27.13 ± 0.40	1.14 ± 0.08	0.740 ± 0.01	0.25 ± 0.03	0.85 ± 0.03
ET2	7.66 ± 0.04	1.37 ± 0.01	80.13 ± 0.04	64.47 ± 2.21	37.40 ± 1.25	1.890 ± 0.04	19.82 ± 0.81	1.26 ± 0.01	1.22 ± 0.006	0.450 ± 0.02	1.35 ± 0.03
ET3	7.28 ± 0.05	1.38 ± 0.03	80.12 ± 0.03	47.80 ± 1.45	27.73 ± 0.84	2.23 ± 0.06	12.40 ± 0.02	1.41 ± 0.01	1.50 ± 0.05	0.89 ± 0.06	2.81 ± 0.02

EC = electrical conductivity; M = moisture; OM = organic matter; TOC = total organic carbon; TN = total nitrogen; C/N = carbon/nitrogen ratio; P = phosphorus; K = potassium; Mg = magnesium; Ca = calcium. Values represent the mean of three replicates per treatment ± standard deviation.

.

4. Discussion

This study provides a comprehensive view of the microbial communities present in cow manure and their transformation into vermicompost (ET1, ET2, and ET3), as well as into the gut of E. fetida earthworms (LT1, LT2, and LT3) during the vermicomposting process. Significant variations in taxonomic diversity, microbial composition, and phylogenetic richness were observed between the two habitats, reflecting clear ecological differentiation. Both sequence read counts and the proportion of non-chimeric sequences were greater in manure samples (ET2 and ET3) and in mid- and late-stage earthworm gut samples (LT2 and LT3), suggesting that microbial DNA quantity is influenced by digestion stages during vermicomposting [39,41,64]. These findings align with previous studies showing progressive microbial enrichment in the gut of earthworms as a function of diet and compost maturity [9,29,41]. Recent research has reinforced this trend, indicating that gut transit promotes the succession of specialized microbial communities with key roles in organic matter degradation and stabilization [16,65,66]. At the taxonomic level, Proteobacteria, Bacteroidota, and Actinobacteriota were the dominant phyla, with notable differences in their relative abundances. Proteobacteria dominated manure samples (especially ET2), whereas Actinobacteriota were more prevalent in LT2 and LT3, suggesting a possible selective enrichment in the earthworm gut [13,27,67,68,69]. This pattern agrees with the findings by Aira et al. [9], who linked Actinobacteriota to the degradation of complex organic compounds and adaptation to nutrient-rich environments. Furthermore, aerobic composting has been shown to favor Actinobacteriota and Firmicutes, which are maintained or enriched throughout vermicomposting [70,71].

Alpha diversity analyses revealed significant increases in observed features, Shannon index, and Faith’s phylogenetic diversity over time. The lowest values were consistently recorded in LT1 (pre-inoculation gut samples), reflecting limited bacterial complexity at the beginning of the process. In contrast, LT2, LT3, and ET2 exhibited significantly higher diversity values (p < 0.05), suggesting an enrichment of microbial communities as the vermicomposting progressed. These patterns indicate that both gut transit and substrate transformation contribute to increased bacterial richness and phylogenetic breadth as organic matter stabilizes [5,72,73]. Moreover, previous studies have shown that composting with carbon-rich additives enhances microbial richness and diversity, particularly during intermediate and final stages [36,74]. Additionally, beta diversity analyses (Jaccard, Bray–Curtis, weighted and unweighted UniFrac) also revealed significant structural differences among communities. This supports earlier findings that worm gut transit substantially modifies microbial assemblages through both physiological and mechanical selection [7,65,75]. As a result, the microbial profiles in vermicompost are distinct and shaped by earthworm digestion processes [42].

SIMPER analysis revealed that Proteobacteria contributed the most to the community dissimilarity (29.53%), being particularly enriched in ET2, likely due to their capacity to utilize readily available nutrients [29,71]. Chryseolinea and Sphingomonas, enriched in ET2 and ET3, have been associated with the degradation of organic material in composting and vermicomposting systems [76,77]. Meanwhile, microbial shifts observed in the earthworm gut, such as the enrichment of Actinobacteriota and Firmicutes and the decline of Bacteroidota, are likely modulated by physiological mechanisms inherent to the gut environment of E. fetida. Several studies have shown that the gut epithelium secretes a variety of digestive enzymes and antimicrobial peptides (AMPs), such as lysenin and fetidin, which exert selective inhibitory effects on specific bacterial taxa [21,22]. In addition, the mucus layer lining the gut acts both as a protective barrier and as a nutrient source, favoring colonization by symbiotic or tolerant bacteria while limiting pathogen proliferation [40]. The gut also presents steep gradients in oxygen availability, redox potential, and pH from the foregut to the hindgut, thus generating distinct microhabitats that function as ecological filters [41,78]. These conditions impose strong selective pressures that promote the survival of facultative anaerobes and spore-forming genera such as Bacillus and Brevibacillus, which were enriched in LT3. Collectively, these gut-derived factors contribute to a biological filtration and microbial sorting process that significantly influences community structure and likely enhances the functional potential of the resulting vermicompost.

Interestingly, no unique taxa were detected in ET1 at any analyzed level, suggesting a more homogeneous and potentially less functionally specialized community at the initial stage. This likely reflects the high compositional uniformity and limited microbial specialization of the pre-composted manure. ET1 was characterized by high organic matter content (79.47%), a high C/N ratio (27.13), and an alkaline pH (8.07), conditions unfavorable for microbial differentiation and niche specialization. These physicochemical traits may have delayed the onset of microbial succession and the establishment of functionally enriched communities, which became more evident in ET2 and ET3. ET2, ET3, and LT2 exhibited greater numbers of unique families and genera, reflecting the influence of earthworm activity on environmental filtering and microbial succession [7,9]. The identification of unique taxa in LT2 and LT3 supports the idea, proposed in previous studies, that earthworms act as biological filters, promoting specialized microbial communities adapted to gut conditions [39,79,80]. Overall, these results confirm that earthworm digestion generates structurally and functionally distinct microbial communities. Beyond microbial dynamics, these enriched communities confer ecological benefits. Vermicompost not only has a high nutrient content but also harbors functionally enriched microbiomes with potential applications as biofertilizers and biostimulants [46,81,82]. Its use in agriculture has been associated with improved soil fertility, better physical properties, and enhanced microbial activity, reducing the need for synthetic agrochemicals and supporting agroecological practices [6,8,46,83]. Moreover, vermicomposting holds promise for the bioremediation of contaminated soils [84,85,86], as microbial communities in vermicompost have demonstrated the ability to degrade organic pollutants, immobilize heavy metals, and restore soils affected by industrial activities or chemical overuse [87,88,89,90].

During the process of bovine manure vermicomposting, physicochemical changes that were consistent with the transformation and stabilization of organic material were observed. The pH decreased from alkaline values (8.07 in ET1) to near-neutral levels (7.28 in ET3), coinciding with acidification due to the formation of humic acids, as described by Nobili et al. [91] and Wang et al. [39]. The electrical conductivity decreased from 1.97 to 1.38 dS/m, reflecting reduced salinity and a lower risk of phytotoxicity [92]. The moisture content remained at approximately 80%, maintained by an automatic irrigation system, which favored microbial activity and provided ideal conditions for earthworm development [93,94]. The reductions in organic matter content (from >79% to <48%) and organic carbon content (from ~46% to ~27%) indicate high microbial and enzymatic activity associated with the degradation of complex organic compounds [95]. These findings are consistent with those of Khan et al. [96], who reported losses of 35–45% of organic matter during the vermicomposting of manure mixed with lignocellulosic residues, suggesting an efficient mineralization process. Moreover, the total nitrogen content increased from 1.70% to over 2.23%, which was attributed to both the relative concentration due to carbon loss and the release of nitrogen compounds through microbial and earthworm activity. Similar increases have been reported in recent studies [97,98], which documented significant improvements in the nitrogen content in vermicomposting systems that were enriched with agricultural waste after 60 days of processing. With respect to essential macronutrients, progressive increases in the levels of phosphorus (from 1.14% to 1.41%), potassium (from 0.74% to 1.50%), magnesium (from 0.25% to 0.89%), and calcium (from 0.85% to 2.81%) were detected. These increases, which were documented by various authors, demonstrate that vermicomposting promotes the release of mineral nutrients from the organic fraction, increasing their availability to plants [99,100]. Collectively, these changes significantly improve the fertilizer value of the final product, positioning vermicompost as a viable and sustainable organic amendment.

Comparative analyses with other earthworm species further contextualize these findings. Eisenia andrei has been shown to outperform E. fetida in terms of pathogen reduction and microbial selection under continuous vermireactor conditions [4,11], whereas P. excavatus achieves faster compost stabilization and higher nitrogen content in controlled environments, albeit with greater sensitivity to humidity and temperature fluctuations [1]. In contrast, Aporrectodea caliginosa, which is a soil-dwelling species, promotes lower microbial diversity and slower decomposition, highlighting the specific influence of earthworm species on microbial succession [22]. These differences emphasize how earthworm physiology and gut microbiota interactions shape compost quality and process efficiency [20,21]. Such interspecific differences are important considerations for optimizing vermicomposting systems according to specific environmental or agronomic goals.

Compared to previous vermicomposting studies, which often focus on either the compost matrix or the earthworm gut at isolated time points [9,29,67], this study presents a fine-scale, parallel time-series analysis that captures microbial dynamics in both habitats across three distinct composting stages (0, 6, and 12 weeks). This longitudinal approach allowed us to identify transient and persistent microbial signatures, offering insights into the temporal succession and potential functional roles of specific taxa. Notably, the selective enrichment of genera such as Methylobacter, Enhygromyxa, and Brevibacillus, among others, in the gut during intermediate and late stages (alongside the depletion of potential pathogens and opportunists) suggests a functional filtering mechanism that has not been previously quantified in a time-resolved manner. While earlier studies described taxonomic differences or general trends [13,40,41], few have simultaneously evaluated microbial succession in both ecological compartments with a resolution that reveals habitat-specific microbial enrichment. Therefore, this study contributes to the refinement of the earthworm-driven microbial selection model, and complements prior functional inferences by proposing a link between specific taxa and the physicochemical evolution of the substrate. By bridging microbial ecology and compost maturity indicators, our findings provide a deeper mechanistic understanding of how earthworms shape compost quality, not just through bioturbation and fragmentation, but via selective microbial recruitment and the enhancement of beneficial taxa. These insights hold relevance for the optimization of vermicomposting systems and the development of targeted biofertilizer strategies.

While this study provides valuable insights into the taxonomic and diversity-based dynamics of microbial communities during vermicomposting, several aspects warrant further investigation. Although no direct correlation analysis between the gut and matrix microbiota was conducted, the observed temporal patterns suggest the selective recruitment and modulation of gut communities in response to substrate microbial dynamics. These associations should be further explored through network and co-occurrence analyses to better understand microbial interactions and functional convergence between habitats. Moreover, direct measurements of gut physicochemical parameters (e.g., pH, redox potential, mucus composition) are recommended to clarify the mechanisms driving microbial selection and recruitment during gut transit. Although predictive functional profiling was not performed in this study, the integration of bioinformatic tools such as PICRUSt2 or FAPROTAX could complement taxonomic findings by inferring potential metabolic functions associated with key bacterial taxa. These approaches allow for the prediction of genes involved in processes such as organic matter degradation, nutrient cycling, and pathogen suppression, thereby offering a more comprehensive understanding of the ecological roles of microbial communities throughout the vermicomposting process. Incorporating such functional prediction analyses will be essential for elucidating the microbial-driven mechanisms underlying compost maturation and quality.

5. Conclusions

This study reveals that E. fetida and its gut microbiota play a central role in shaping the microbial succession and functional enrichment that occur during bovine manure vermicomposting. By simultaneously analyzing microbial and physicochemical dynamics across multiple time points and habitats, we demonstrate that earthworm digestion not only restructures microbial communities but also enhances nutrient availability and substrate maturity. These findings contribute to a deeper understanding of the ecological mechanisms underlying vermicomposting and highlight the gut as a selective environment that promotes beneficial microbial consortia. The evidence presented supports the use of vermicompost as a sustainable biofertilizer and positions earthworm-driven composting as a valuable strategy for organic waste transformation, soil restoration, and agroecological resilience.

Future research should explore the agronomic effects of vermicompost through pot and field trials, as well as the physicochemical conditions within the earthworm gut, to better understand the mechanisms of microbial selection. Overall, the enriched microbial consortia and nutrient profile of the final product highlight vermicomposting as a biotechnological strategy with strong potential for sustainable agriculture, soil restoration, and reduction in chemical inputs.