Randomized, Double‐Blind, Placebo‐Controlled Trial of the Probiotic Lactiplantibacillus plantarum Probio87 for Gut–Vaginal Microbiota Modulation in Healthy Women

ABSTRACT

The gut–vaginal axis represents a key frontier in women's health, where probiotics can influence both ecosystems through microbial and metabolic crosstalk. This randomized, double‐blind, placebo‐controlled clinical trial evaluated the effects of Lactiplantibacillus plantarum Probio87 (9 log CFU/sachet) administered orally for 12 weeks on gut and vaginal microbiota modulation. Healthy women aged 18–80 years without urogenital infections, pregnancy, recent vaginal/hormonal therapies, chronic illness medication, and recent pelvic or gynecological surgery were enrolled. 112 participants were randomized to probiotic (n = 58) or placebo (n = 54). Vaginal and fecal samples collected at baseline and week 12 underwent 16S rRNA sequencing for taxonomic and diversity analyses. No adverse effects were reported throughout the study. Compared with placebo, the probiotic group showed a significantly greater reduction in Nugent score (p = 0.008), indicating improved vaginal health. Microbiota analyses revealed that Probio87 maintained vaginal microbiota by suppressing BV‐associated taxa while preserving beneficial Lactobacillales. In the gut, Probio87 enhanced microbial richness and evenness, with notable increases in butyrate‐producing genera, indicating a predicted rise in short‐chain fatty acid‐producing potential and greater ecological resilience. These findings demonstrate that oral supplementation with Probio87 safely promotes balanced gut and vaginal microbiota, supporting eubiosis and integrative female health.

Oral Lactiplantibacillus plantarum Probio87 supports the gut‐vaginal axis in healthy women. After 12 weeks, it lowered Nugent score, reduced harmful vaginal bacteria, and maintained beneficial Lactobacillales. It also improved gut diversity and increased SCFA‐producing bacteria, supporting a balanced microbiota and overall women's health.

1. Introduction

Probiotics are live microorganisms that, when administered in sufficient quantities, confer health benefits to the host by supporting intestinal microbial balance and bolstering the body's defense mechanisms against gastrointestinal disorders [1]. Beyond gut health, probiotic strains have demonstrated systemic functional benefits, including immunomodulation [2], metabolic regulation [3], and bioactive compound production [4]. Currently, their use has expanded into obstetrics and gynecology, where they modulate microbial communities and improve outcomes in conditions such as bacterial vaginosis and recurrent yeast infections. Understanding the vaginal microbiota is therefore central to evaluating probiotic benefits.

The vaginal microbiota plays an important role in the maintaining health and preventing infection. Factors such as hormonal changes (particularly estrogen), vaginal pH, and glycogen content can influence the vaginal environment. These factors affect the ability of lactobacilli to adhere to epithelial cells and colonize the vagina. While Lactobacillus‐dominated communities protect against infections, they do not completely exclude other microbial species. In this context, pathogens such as Gardnerella vaginalis, which often linked to BV, may coexist with commensal Lactobacillus species in subclinical infections [5].

Vaginitis is one of the most common gynecological issues worldwide, often asymptomatic and driven by vaginal dysbiosis. Its prevalence has been reported to range from 5% to 50% among different study populations across major continents such as the United States of America, Europe and South Asia [6]. Meanwhile, the CDC reported that bacterial vaginosis (BV) affects 21.2 million women aged 14 to 49, and is a common cause of vaginal symptoms while increasing risks of acquisition of sexually transmitted infections, HIV, and with adverse pregnancy outcomes. Bacterial vaginosis and yeast vaginitis, the major urogenital infections that affect an estimated 1 billion women in the world annually [7]. This substantial disease burden underscores the need for safer, microbiome‐supportive approaches.

Considering that innate immunity plays an important role in the switch to BV from a healthy state, probiotic administration in premenopausal women has been shown to induce substantial gene‐expression changes that enhanced antimicrobial defense in the vaginal. Probiotics are preferred compared to antibiotics used for BV treatment due to infection recurrence and drug resistance. Recent research by Younus and team on the existence of drug‐resistance pathogenic Gardnerella vaginalis clones in Malaysia highlights the need for close monitoring, especially among women of childbearing age [8]. It is a main concern that overuse of antibiotics could result in the development of antibiotic‐resistant bacteria. Even in healthy women, hormonal fluctuations throughout different life stages (menstruation, pregnancy, and menopause) can disrupt the microbial balance, resulting in symptoms such as dryness, irritation, and an increased risk of infections. Overall, physiological changes throughout a woman's life significantly impact the vaginal microenvironment. These shifts necessitate adaptive management strategies, not only to prevent the onset of diseases, but also act as a precautionary measure to maintain vaginal health. Collectively, these biological and epidemiological insights provide a strong rationale for exploring safe, microbiome‐supportive interventions such as probiotics.

These insights highlighted the need to evaluate probiotic interventions not only in diseased populations but also in healthy women as a preventive measure. With its established gastrointestinal resilience, antimicrobial activity, and adherence properties, Lactiplantibacillus plantarum Probio87 represents a promising candidate for gut–vaginal microbiome modulation. Therefore, this study aimed to investigate the effects of oral administration of L. plantarum Probio87 on gut‐vaginal microbiota modulation in generally healthy women. The primary clinical endpoint was the change in Nugent score over 12 weeks, while microbiota composition and diversity analyses were evaluated as secondary endpoints. By targeting a healthy cohort, this trial highlights the role of Probio87 in supporting microbial equilibrium along the gut–vaginal axis, and demonstrates its potential as a safe probiotic for preventive women's health.

2. Materials and Methods

2.1. Probiotic and Placebo Formulation

The probiotic strain, Lactiplantibacillus plantarum Probio87 meets established probiotic criteria. It is resistant to the acid and bile conditions of the upper gastrointestinal tract. It does not exhibit antibiotic resistance as required of the European Food Safety Authority (EFSA). The strain is also able to adhere to mucin, and capability to utilize prebiotics such as fructooligosaccharide (FOS) and galactooligosaccharide (GOS). Besides, it exhibits a carbon metabolism profile adhering to the general patterns of Lactobacilli, and possesses antimicrobial properties against human pathogens. Each sachet of the probiotic product contained 9 log CFU of viable L. plantarum Probio87 blended with a carrier matrix, whereas the placebo sachets contained only the carrier without live microorganisms. Both probiotic and placebo powders were identical in appearance, light yellow, free‐flowing, and packaged in identical sachets, to ensure blinding integrity. All products were stored below 25°C in a dry environment and protected from direct sunlight, following the manufacturer's recommendations. Participants were instructed to dissolve one sachet in room‐temperature water and consume it once daily for 12 weeks.

2.2. Participant Recruitment and Eligibility Criteria

Written informed consent was obtained from all women prior to enrolment. Participants were recruited from the International Islamic University Malaysia (IIUM) campus and nearby communities through outreach activities and public announcements. Eligible participants were generally healthy women aged 18–80 years, clinically free from urogenital infections, and presenting a Nugent score below 7.0. All participants were required to be willing and able to comply with the study protocol and attend follow‐up visits throughout the 12‐week intervention period. Excluding criteria included pregnant, use of vaginal suppositories, oral medications for vaginal disorders, or hormonal therapy within 4 weeks before the start of the study. Long‐term medication of more than 6 months for chronic illness was also exclusionary. Woman who had recently used vaginal estrogen products (creams, rings, or tablets), vaginal moisturizers, lubricants, homeopathic preparations, or spermicide agents within 4 weeks prior to enrolment were also excluded. In addition, individuals with a history of pelvic or gynecological surgery within the preceding 6 months were not eligible to participate.

2.3. Study Design, Randomization, and Ethical Compliance

This was a randomized, double‐blind, placebo‐controlled study. Eligible participants, screened according to the predefined inclusion and exclusion criteria, were randomly assigned in a 1:1 ratio to receive either the probiotic or placebo intervention. Randomization was performed using a computer‐generated allocation list prepared by an independent statistician who had no direct contact with the participants. Each subject was assigned a unique code corresponding to either treatment arm. The allocation sequence was concealed from all investigators and study personnel until data collection and analysis were completed, ensuring full maintenance of blinding throughout the trial. All study procedures were conducted in accordance with the ethical principles of the Declaration of Helsinki, the International Conference on Harmonization Good Clinical Practice (ICH‐GCP) guidelines, the Malaysian Good Clinical Practice Guidelines, and the Council for International Organizations of Medical Sciences (CIOMS) International Ethical Guidelines. The study protocol and all associated documents were reviewed and approved by the International Islamic University Malaysia Research Ethics Committee (approval ID: IIUM/IREC 2021‐318). The clinical trial was prospectively registered at ClinicalTrials.gov under the identifier NCT05302687.

2.4. Biological Sample Collection and Storage

2.4.1. Vaginal Swab Collection

Vaginal swab samples were collected at week 0 and week 12 via self‐swabs. Each swab had a mark indicating the depth of insertion into the vagina to ensure consistency between all subjects. Upon insertion, swabs were rotated 10 times inside the vagina. Swabs were smeared onto slides for the determination of Nugent's score. Swabs were also placed into a swab collection tube containing RNAlater solution for microbiota analysis. All tubes were tightly capped and stored at ‐80°C until further analyses.

2.4.2. Fecal Sample Collection

Fecal samples were collected at week 0 and week 12 in fecal collection tubes containing RNAlater solution and glass beads by the subjects for microbiota analysis. All tubes were tightly capped and stored at −80°C until further analyses.

2.5. Nugent Scoring for Vaginal Microbiota Classification

Gram staining was performed to determine the Nugent's score from vaginal swab samples. In brief, scoring is based on the number of different morphotypes of bacteria, such as Lactobacillus‐like (large uniform Gram‐positive bacilli; a decrease scored as 0 to 4), Gardnerella vaginalis‐like (small pleomorphic Gram‐variable bacilli) or Prevotella/Bacteroides‐like (small Gram‐negative bacilli) (an increase scored as 0 to 4), and Mobiluncus‐like (curved Gram‐variable bacilli; an increase scored as 0 to 2). A Nugent score of ≥7 is interpreted as consistent with vaginosis as per inclusion criteria [9].

2.6. 16S RRNA Gene Sequencing and Bioinformatics Workflow

2.6.1. DNA Extraction and PCR Amplification

Total genomic DNA was extracted and purified from vaginal swabs and fecal samples following previously described protocols [10]. The concentration and purity of the extracted DNA were assessed using a NanoDrop 2000 UV–vis spectrophotometer (Thermo Scientific, Wilmington, NC, USA). The V3–V4 hypervariable regions of the bacterial 16S rRNA gene were amplified using primers 341F (5′‐CCTAYGGGRBGCASCAG‐3′) and 806R (5′‐GGACTACHVGGGTWTCTAAT‐3′) on a thermocycler PCR system (GeneAmp 9700, Applied Biosystems, San Diego, CA, USA). Each PCR reaction (20 µL) was performed in triplicate and contained 4 µL of 5× FastPfu Buffer, 2 µL of 2.5 mM dNTPs, 0.8 µL of each primer (5 µM), 0.4 µL of FastPfu Polymerase, and 10 ng of template DNA. The thermal cycling protocol included initial denaturation (95°C, 3 min), 27 cycles of denaturation (95°C, 30 s), annealing (55°C, 30 s), and extension (72°C, 45 s), followed by a final extension at 72°C for 10 min. Amplicons were visualized on a 2 % agarose gel, excised, and purified using the AxyPrep DNA Gel Extraction Kit (Axygen Biosciences, Union City, CA, USA). Quantification of purified DNA was performed using the QuantiFluor‐ST fluorometer (Promega, Madison, WI, USA) according to the manufacturer's protocol. Purified amplicons were pooled in equimolar concentrations before paired‐end sequencing (2 × 300 bp) on an Illumina MiSeq platform (Illumina, San Diego, CA, USA).

2.6.2. Bioinformatics and Microbiota Profiling

Raw paired‐end reads in FASTQ format were processed using QIIME V1.9.1 and USEARCH v10.0. Quality control involved trimming low‐quality bases with Trimmomatic, removing barcodes and primers, merging paired reads with USEARCH, and discarding chimeric or low‐quality sequences. Each sample generated at least 50 000 high‐quality merged reads. Operational taxonomic units (OTUs) were clustered at 97% sequence similarity using UPARSE. Taxonomic classification of representative 16S rRNA sequences was performed using the RDP Classifier against the SILVA 132 rRNA database with an 80 % confidence threshold. Microbiota data were analyzed using the MicrobiomeAnalyst phyloseq R package (V3.6.1) to compute alpha‐ and beta‐diversity indices, relative taxonomic abundances, and compositional differences across taxonomic levels.

2.7. Statistical Analysis and Data Interpretation

Statistical analyses were conducted using SPSS V24.0 (SPSS Inc., Chicago, IL, USA). The primary analysis was performed using an intention‐to‐treat (ITT) approach, including all randomized participants. For outcomes requiring paired biological samples (e.g., Nugent score and microbiota analyses), a per‐protocol (PP) or available‐case analysis was applied, including only participants with complete data at both time points.

The primary endpoint was the change in Nugent score from baseline to week 12 between the probiotic and placebo groups, reflecting differences in vaginal health status. Secondary endpoints included changes in vaginal and fecal microbiota composition, alpha‐ and beta‐diversity indices, and relative taxonomic abundances. Because microbial and Nugent data are non‐parametric, inter‐group comparisons were evaluated using the Mann–Whitney U test, while categorical variables were analyzed with the Chi‐square test. All tests were two‐tailed with p < 0.05 considered significant.

Results are presented as mean ± standard error (SE). Effect size measures and confidence intervals were not pre‐specified in the study design and were therefore not included in the analysis.

3. Results and Discussion

3.1. Baseline Characteristics and Study Compliance

A total of 112 participants who met the inclusion and exclusion criteria, were randomized in the intention‐to‐treat (ITT) population (Figure 1). While all participants completed the study follow‐up, not all provided complete biological samples at both time points. Specifically, 108 participants provided fecal samples and 104 provided vaginal swab samples suitable for paired analyses. Missing samples were due to logistical and physiological factors (e.g., menstruation or inability to provide samples at the scheduled time point), rather than study withdrawal. No adverse effects or complications were reported throughout the study period. Baseline demographic and clinical characteristics, including age, body weight, height, and socioeconomic variables, were comparable between groups (Table 1), supporting the validity of between‐group comparisons.

FIGURE 1.

Consort flowchart detailing patients’ recruitment, randomization and allocation.

TABLE 1.

Demographic characteristics of women (n = 112) randomly assigned to a double‐blind administration with either placebo (n = 54) or probiotic Lactiplantibacillus plantarum Probio87 (n = 58).

Characteristics	Placebo	Probiotic	p‐value*
Sample size (n)	54	58
Nugent score Age (year)	4.09 ± 0.184 38.91 ± 0.98	4.03 ± 0.17 38.38 ± 0.85	0.636 0.866
Body weight (kg)	67.73 ± 1.69	63.88 ± 1.53	0.143
Height (cm)	155.61 ± 0.72	157.07 ± 0.75	0.147
Defecation frequency (per day)	0.65 ± 0.07	0.66 ± 0.06	0.938
	% (n)	% (n)	p‐value**
Family income status:
a) Low (< RM2,999)	9.26 (5)	20.69 (12)	0.196
b) Middle (RM3,000‐6,000)	57.41 (31)	55.17 (32)
c) High (> RM6,000)	33.33 (18)	24.14 (14)
Number of people living together in a household:
a) Less than or equal to 4 people	44.44 (24)	44.83 (26)	0.967
b) More than 4 people	55.56 (30)	55.17 (32)
House location:
a) City	64.82 (35)	58.62 (34)	0.763
b) Sub‐urban	22.22 (12)	29.31 (17)
c) Industrial Area	1.85 (1)	3.45 (2)
d) Town	11.11 (6)	8.62 (5)
Type of Residence:
a) Terrace/ Storey/ Bungalow	88.89 (48)	82.76 (48)	0.354
b) Apartment	11.11 (6)	17.24 (10)
Occupation
a) Wage worker b) Self‐employed	92.59 (50)	91.38 (53)	0.813
	7.41 (4)	8.62 (5)

*p‐value obtained via Mann–Whitney U‐test; ** p‐value obtained via Chi‐square test.

3.2. Clinical Outcomes

3.2.1. Primary Outcome: Nugent Score and Vaginal Health Status

Nugent score was <7.0 for both groups as per inclusion criteria for recruitment, which also indicated that all women were healthy and free from vaginosis (Table 2). After 12‐weeks, the placebo group showed a higher Nugent score value compared to the probiotic group (Table 2; p = 0.008). The changes of each group over 12‐weeks also showed a significant difference between groups, where the probiotic group showed a higher reduction than the placebo group (p = 0.043) (Figure 2). The Nugent score is a reliable, standardized diagnostic tool used to assess vaginal health by evaluating the presence of bacterial vaginosis and monitor changes in vaginal microbiota. Scores range from 0 to 10, with value of 0–3 indicating a Lactobacillus‐dominat healthy vaginal flora, 4–6 suggesting an intermediate state, and 7–10 indicating bacterial vaginosis.

TABLE 2.

Nugent score and changes in score values over time of women (n = 104) randomly assigned to a double‐blind administration with either placebo (n = 51) or probiotic Lactiplantibacillus plantarum Probio87 (n = 53).

Nugent score	Week 0		p‐value	Week 12		p‐value
	Placebo	Probiotic		Placebo	Probiotic
Actual score	4.10 ± 0.18	4.08 ± 0.18	0.839	3.27 ± 0.14	2.72 ± 0.12	0.008
Changes over 12‐weeks	−0.82 ± 0.19	−1.36 ± 0.22	0.043

FIGURE 2.

Representatives of Nugent images from the probiotic group at week 0 (A) and week 12 (B), and from the placebo group at week 0 (C), week 12 (D).

Despite evaluating this in a generally healthy women population, our present data illustrated the effects of Probio87 in reducing Nugent score, suggesting its potential role in supporting the maintenance of a balanced vaginal microenvironment. Within the context of the study population, all participants were vaginally healthy at baseline (Nugent score <7), indicating that while the reduction in Nugent score was statistically significant, it may not reflect a clinical shift in disease status. Rather, these findings suggest that Probio87 may contribute to the maintenance or stabilization of a Lactobacillus‐dominant vaginal environment, potentially reducing the risk of dysbiosis over time. This supports a preventive, rather than therapeutic, role of the probiotic in healthy women.

3.3. Vaginal Microbiota Dynamics

3.3.1. Alpha Diversity

Alpha diversity measures the differences within samples using several indices. The Chao1 index estimates the “richness” of OTUs, accounting for both frequent and rare OTUs. Higher Chao1 values indicate greater species richness. Similarly, the ACE index also measures richness but places greater emphasis on low‐abundance OTUs, extrapolating the potential number of undiscovered OTUs within a sample. Higher ACE values therefore reflect higher diversity influenced by both richness and evenness. At baseline, both groups exhibited comparable Chao1 (p = 0.899) and ACE (p = 0.706) indices at the phylum level (Figure S1A,C). However, after 12 weeks, the placebo group demonstrated significantly higher Chao1 and ACE indices at the phylum level compared to the probiotic group (p = 0.011 and p = 0.021, respectively). Likewise, at baseline, both groups showed similar Chao1 (p = 0.912) and ACE (p = 0.543) indices at the genus level (Figure S1B,D). By week 12, the placebo group again exhibited significantly higher Chao1 and ACE indices at the genus level than the probiotic group (p = 0.008 and p = 0.003, respectively).

The observed index measures richness by representing the total number of OTUs detected in a sample, without considering their relative abundances. Conversely, the Shannon and Fisher indices incorporate both richness and evenness, reflecting the frequency distribution of OTUs. Higher Shannon or Fisher values denote greater overall diversity, influenced by both the number and relative abundance of OTUs. At baseline, both groups exhibited comparable Observed indices at the phylum (p = 0.862), family (p = 0.677), and genus (p = 0.644) levels (Figure S2A–C). However, at week 12, the placebo group displayed higher observed indices compared to the probiotic group across phylum, family, and genus levels (p = 0.030, marginally p = 0.084, and marginally p = 0.068, respectively). Similarly, both groups had comparable Shannon (p = 0.526) and Fisher (p = 0.416) indices at the genus level at baseline (Figure S2D,E). After 12 weeks, the placebo group showed marginally higher Shannon and Fisher indices at the genus level than the probiotic group (p = 0.075 and p = 0.075, respectively).

Overall, our findings indicate that Probio87 supplementation limited the increase in within‐group ecological diversity across various taxonomic levels and relative abundances, which was markedly elevated in the placebo group after 12 weeks. This suggests the stabilization of specific dominant OTUs, which are described in greater detail in the subsequent analyses. Although a diverse and balanced vaginal microbiota is generally associated with optimal vaginal health, excessive diversity can indicate microbial imbalance, often linked to infections and disease states. The composition of the vaginal microbiota is critical in maintaining an acidic pH and suppressing the overgrowth of pathogenic microorganisms. Women with BV, for instance, exhibit more diverse and heterogeneous microbial communities with greater species richness and bacterial diversity compared to those without BV [11].

3.3.2. Beta Diversity

Beta diversity assesses differences in microbial community composition between samples. The Bray–Curtis index considers both the shared presence (co‐occurrence) and relative abundance (differential abundance) of OTUs. Beta diversity is visualized using principal coordinate analysis (PCoA) and statistically evaluated through PERMANOVA or ANOSIM, where PERMANOVA assesses differences based on multivariate dispersion and ANOSIM compares average ranks of dissimilarities between groups.

In this study, both PERMANOVA and ANOSIM suggested differences in microbial community structure between groups at week 12, despite no significant differences at baseline. These distinctions were observed across multiple taxonomic levels, including phylum (marginally p = 0.058), class (p = 0.036), order (marginally p = 0.075), family (p = 0.025), and genus (p = 0.027) for PERMANOVA (Figure S3A–E), and phylum (marginally p = 0.079), class (marginally p = 0.055), order (marginally p = 0.080), family (p = 0.036), and genus (p = 0.023) for ANOSIM (Figure S4A–E). However, several comparisons were only marginally significant, indicating that the observed effects may be modest and should be interpreted with caution.

Given the inherent variability and high inter‐individual heterogeneity characteristic of microbiome datasets, these findings are more appropriately considered as indicative of trends rather than definitive shifts in community composition. Overall, the results suggest that Probio87 supplementation may contribute to stabilizing the vaginal microbial community structure, potentially limiting temporal compositional changes observed in the placebo group. Further studies with larger sample sizes and greater statistical power are warranted to confirm these observations.

3.3.3. Core Microbiome and Pathogen Suppression

The core microbiome of vaginal swab samples was identified based on cut‐off values of 20% sample prevalence and 0.01% relative abundance. A temporal shift in the core microbiome was observed in both groups, where certain taxa present at baseline were absent after 12 weeks, and vice versa (Figure 3). In the placebo group, Lactobacillales was part of the core microbiome at baseline but was no longer detected after 12 weeks. Conversely, Gardnerella, Cupriavidus, and Collinsella aerofaciens emerged as core members at week 12 but were absent at baseline. While in the probiotic group, Gardnerella was part of the core microbiome at baseline but was no longer observed after 12 weeks.

FIGURE 3.

Core microbiome of vaginal swab samples as determined based on cut‐off values for sample prevalence of 20% and relative abundance of 0.01%. (A) Placebo at baseline week 0; (B) Placebo at week 12; (C) Probiotic Lactiplantibacillus plantarum Probio87 at baseline week 0; (D) Probiotic at week 12. n = 104 (Probio87; n = 53 and placebo n = 51).

BV is typically characterized by a reduction in beneficial Lactobacillus species (order Lactobacillales), accompanied by an increase in pathogenic Gardnerella populations. Moreover, Cupriavidus has been reported to occur more frequently in the vaginal microbiota of women with BV than in healthy controls [12]. Although Collinsella aerofaciens is not directly associated with BV, its presence in the vaginal environment has been linked to incidents of recurrent abortion [13].

Altogether, our findings indicate that Probio87 supplementation was associated with a reduced presence of pathogenic taxa and a lower relative abundance of opportunistic microorganisms, while preserving beneficial taxa. This suggests that Probio87 may contribute to maintaining a healthier and more protective vaginal microbiome composition over time.

3.3.4. Compositional Changes and Clinical Implications

Alpha and beta diversity analyses, together with core microbiome evaluations, revealed clear between‐group differences in the vaginal microbiota of healthy women. Our findings showed that temporal changes occurred primarily in the placebo group, characterized by increased bacterial abundances across multiple taxonomic levels, particularly among taxa commonly associated with vaginal infections. Many women may remain unaware of such microbial imbalances—marked by increased growth of vaginal pathogens and/or reduced growth of beneficial Lactobacillus—especially in the absence of clinical symptoms. Consistent with this, the reduction in Nugent score observed in the probiotic group compared with the placebo group after 12 weeks suggests that Probio87 supplementation attenuated the proliferation of unhealthy bacterial populations, even within a vaginally healthy cohort. To further clarify these changes, microbiota profiling was conducted to identify specific compositional shifts in the vaginal bacterial community.

The placebo group exhibited a significant increase in the family Bacteroidaceae, while the probiotic group showed a reduction over 12 weeks (p = 0.010; Table 3). A similar pattern was observed at the genus level for Bacteroides (p = 0.010) and other genera, including Paraprevotella (p = 0.008) and Parabacteroides (p = 0.027). Notably, Bacteroides abundance has been linked to HPV infection [14] and urge urinary incontinence [15], while Parabacteroides—though a normal gut and endometrial commensal [16]—has shown dual roles, being associated with both inflammatory and protective outcomes in gastrointestinal disease [17].

TABLE 3.

Changes in abundance of vaginal microbiota represented by number of OTU sequences upon administration of probiotic Lactiplantibacillus plantarum Probio87 (n = 53) or placebo (n = 51) over 12‐weeks at different taxonomic levels.

Class			Order			Family			Genus			Species
Placebo	Probiotic	p‐value	Placebo	Probiotic	p‐value	Placebo	Probiotic	p‐value	Placebo	Probiotic	p‐value	Placebo	Probiotic	p‐value
(Phylum Bacteroidota)
Bacteroidia			Bacteroidales			Bacteroidaceae			Bacteroides
						9.86±9.19	−3.79±2.14	0.01	9.86±9.10	−3.79±2.14	0.010
						Tannerellaceae			Parabacteroides
									0.55±0.39	−0.58±0.35	0.027
						Prevotellaceae			Paraprevotella
									0.27±0.18	−0.17±0.15	0.008
(Phylum Firmicutes)
Bacilli			Lactobacillales			Aerococcaceae			Facklamia			Facklamia hominis CCUG36813
						22.51±18.73	−7.81±19.60	0.011	21.35±20.50	0.09±0.08	0.006	21.29±20.10	0.09±0.08	0.006
Clostridia			Oscillospirales			Hungateiclostridiaceae			Fastidiosipila			Clostridiales bacterium KA00274
									82.03±81.79	50.89±59.80	0.045	82.08±80.21	49.31±58.67	0.049
						Ruminococcaceae			Eubacterium siraeum group
									0.10±0.04	−0.17±0.15	0.008
									Negativibacillus
									0.06±0.05	−0.09±0.06	0.056
									Subdoligranulum
									0.24±0.16	−1.02±0.71	0.082
			Lachnospirales			Lachnospiraceae			Blautia
									3.20±2.45	−1.71±1.13	0.093
									Dorea
									1.29±0.85	−1.00±0.55	0.006
									Lachnoclostridium
									0.65±0.41	−0.64±0.45	0.007
									Marvinbryantia
									0.24±0.13	−0.04±0.03	0.008
									Stomatobaculum
									−1.10±0.85	0.37±0.34	0.096
(Phylum Actinobacteriota)
Actinobacteria			Corynebacteriales			Corynebacteriaceae			Corynebacterium			Corynebacterium simulans
												47.75±33.25	−34.87±27.91	0.047
			Actinomycetales			Actinomycetaceae			Actinomyces
									18.71±17.43	−1.15±2.08	0.014
									Trueperella
									1.06±0.98	0.00±0.00	0.075
			Micrococcales			Brevibacteriaceae			Brevibacterium
						8.96±6.98	−0.02±0.59	0.047	8.96±6.97	−0.02±0.59	0.047
						Micrococcaceae			Micrococcus
									0.06±0.18	−0.68±0.44	0.045
Coriobacteriia			Coriobacteriales			Coriobacteriaceae			Collinsella
						2.96±2.52	−0.57±0.34	0.014	2.96±2.52	−0.57±0.34	0.014
(Phylum Desulfobacterota)
Desulfovibrionia			Desulfovibrionales			Desulfovibrionaceae			Bilophila
0.16±0.07	−0.13±0.12	0.032	0.16±0.07	−0.13±0.12	0.032	0.16±0.08	−0.13±0.12	0.032	0.10±0.06	−0.15±0.12	0.034
(Phylum Proteobacteria)
Gammaproteobacteria			Burkholderiales			Burkholderiaceae			Cupriavidus
									1.10±0.52	−0.25±0.20	0.003
			Enterobacterales			Aeromonadaceae			Aeromonas
						0.04±0.03	−0.06±0.04	0.047	0.04±0.03	−0.06±0.04	0.047
			Xanthomonadales			Xanthomonadaceae			Stenotrophomonas
			0.51±0.38	−0.04±0.04	0.023	0.51±0.38	−0.04±0.04	0.023	0.08±0.06	−0.04±0.04	0.082
Alphaproteobacteria			Rhizobiales			Rhizobiaceae			Ensifer
									0.13±0.09	−0.02±0.02	0.043

Within the phylum Firmicutes, the placebo group showed increased abundance, whereas the probiotic group exhibited reductions over 12 weeks. This included Marvinbryantia (p = 0.008), Dorea (p = 0.006), Lachnoclostridium (p = 0.007), Eubacterium siraeum group (p = 0.008), Fastidiosipila (p = 0.045), and Facklamia hominis CCUG36813 (p = 0.006). Marginal changes were observed for Negativibacillus (p = 0.056), Subdoligranulum (p = 0.082), Blautia (p = 0.093), and Stomatobaculum (p = 0.096). Many of these taxa have been implicated in vaginal dysbiosis or systemic conditions: Dorea correlates with BV [18, 19]; Marvinbryantia is detected in urinary incontinence [15]; Eubacterium siraeum correlates with Group B Streptococcus [20]; Fastidiosipila is linked to high‐risk HPV [21] and BV [22]; and Lachnoclostridium correlates with cervical cancer [23]. In contrast, Blautia and Stomatobaculum are typical commensals of the endometrium and oral cavity, respectively [16, 24], while Subdoligranulum predominates in women with adenomyosis.

Several genera under phylum Actinobacteriota were enriched in the placebo group but reduced in the probiotic group, including Actinomyces (p = 0.014), Brevibacterium (p = 0.047), Micrococcus (p = 0.045), Collinsella (p = 0.014), and Corynebacterium simulans (p = 0.047). Actinomyces has been linked to BV [25], pelvic inflammatory disease in IUD users, and urinary incontinence [15], while Brevibacterium contributes to biofilm formation in BV [26]. Collinsella correlates with Chlamydia trachomatis infection [27]; Micrococcus is associated with endometrial inflammation and cancer [28]; Trueperella (p = 0.075) is enriched in BV [29]; and Corynebacterium simulans, a multidrug‐resistant skin commensal, has emerged as an opportunistic pathogen, particularly in elderly or immunocompromised individuals [30].

Similarly, taxa within phylum Desulfobacterota, including class Desulfovibrionia (p = 0.032), order Desulfovibrionales (p = 0.032), family Desulfovibrionaceae (p = 0.032), and genus Bilophila (p = 0.034), showed increased abundance in the placebo group but decreased in the probiotic group. Bilophila species, though infrequently isolated in clinics, are opportunistic pathogens colonizing the oropharynx, gastrointestinal, and urogenital tracts [31].

Within phylum Proteobacteria, Cupriavidus abundance significantly increased in the placebo group but declined in the probiotic group (p = 0.003). Similar trends were seen for Aeromonas (p = 0.047), Ensifer (p = 0.043), and members of the order Xanthomonadales (p = 0.023), including Stenotrophomonas (p = 0.082). Aeromonas species are known uropathogens [32, 33]; Cupriavidus is more prevalent in BV [11]; Ensifer has been associated with embryo arrest [34]; and Stenotrophomonas, although often a human commensal, can act as an opportunistic pathogen isolated from urine [35].

3.4. Fecal Microbiota Modulation

3.4.1. Alpha Diversity

Over the 12‐week period, the probiotic group exhibited significantly higher richness and evenness than the placebo group, as reflected by the Chao1 (Figure S5), Observed (Figure S6), ACE (Figure S7), and Fisher (Figure S8) indices across multiple taxonomic levels (phylum, class, order, family, and genus). No significant differences between groups at week 0 across all taxonomic levels studied (p > 0.05).

Our findings indicate that Probio87 supplementation preserved within‐group ecological diversity across various taxa and relative abundances—an effect not observed in the placebo group. A healthy gut microbiota is typically characterized by greater diversity, whereas dysbiosis and other gut‐related disorders have been shown to disrupt the gut microenvironment, resulting in reduced microbial diversity [36].

3.4.2. Beta Diversity

PCoA and nonmetric multidimensional scaling (NMDS), assessed using PERMANOVA across different taxonomic levels (phylum, class, order, family, and genus), revealed no significant differences between groups at baseline. However, at week 12, significant compositional differences were observed between the probiotic and placebo groups.

For PCoA, significant differences were detected at the class (p = 0.033) and genus (p = 0.032) levels, with marginal differences at the phylum (p = 0.072), order (p = 0.091), and family (p = 0.088) levels (Figure S9). Similarly, NMDS analysis showed significant differences at week 12 between groups at the phylum (p = 0.045), class (p = 0.028), and genus (p = 0.033) levels, along with marginal effects at the order (p = 0.087) and family (p = 0.085) levels (Figure S10). These findings indicate that Probio87 supplementation altered the overall microbial community structure, contributing to measurable compositional shifts after 12 weeks compared with the placebo group.

3.4.3. Compositional Changes

Alpha and beta diversity evaluations indicated between‐groups differences in bacterial diversity from fecal samples of healthy women across different taxa. Thus, further analyses were performed at each individual taxa to determined changes in composition and profiles.

At phyla level, only phylum Firmicutes showed a difference between group, where the probiotic group showed an increased abundance than the placebo group over 12‐weeks (marginally p = 0.071; Table 4). A similar trend was observed at lower taxonomic levels of classes Clostridia (p = 0.059) and Bacilli (p = 0.096), orders Lachnospirales (p = 0.013), Monoglobales (p = 0.012) and Erysipelotrichales (p = 0.007), families Lachnospiraceae (p = 0.013), and Monoglobaceae (p = 0.012), and several subsequent genera and species. These included genera Dorea (p = 0.001), Eubacterium hallii group (p = 0.031), Eubacterium nodatum group (marginally p = 0.093), Agathobacter (p = 0.003), Lachnospiraceae ND3007 group (p = 0.003), Ruminococcus (p = 0.016), Ruminococcus gauvreauii group (p = 0.002), Ruminococcus torques group (marginally p = 0.081), Oscillibacter (p = 0.029), Monoglobus (p = 0.012), Lactococcus (marginally p = 0.083) and Delma (marginally p = 0.083). The same trend was conveyed through several of subsequent species which included Blautia obeum (p = 0.02), Dorea formicigenerans (p = 0.004), Eubacterium hallii group (p = 0.049), Eubacterium rectale (p = 0.004), Lachnoclostridium sp. YL32 (p = 0.010), Ruminococcus bicirculans (p = 0.011) and Lactobacillus ruminis (p = 0.005). Dorea is a common genus of the human gut microbiota while Dorea formicigenerans are often isolated from human feces and are SCFAs producers typically butyrate [37]. Eubacterium hallii is a butyrate producer and naturally found in feces of healthy human [38]. While E. hallii was shown to improve insulin tolerance and energy metabolism through the production of satiety hormones in obese and diabetic models [39], this species is also able to synthesize cobalamin, which is crucial for the formation of erythrocytes and DNA for the function and development of brain and nerve cells [40]. Meanwhile, Eubacterium nodatum group has a negative association with liver cancer [41].

TABLE 4.

Changes in abundance of fecal microbiota represented by number of OTU sequences upon administration of probiotic Lactiplantibacillus plantarum Probio87 (n = 55) or placebo (n = 53) over 12‐weeks at different taxonomic levels.

Phylum			Class			Order			Family			Genus			Species
Placebo	Probiotic	p‐value	Placebo	Probiotic	p‐value	Placebo	Probiotic	p‐value	Placebo	Probiotic	p‐value	Placebo	Probiotic	p‐value	Placebo	Probiotic	p‐value
Firmicutes			Clostridia			Lachnospirales			Lachnospiraceae			Blautia			Blautia obeum
1352.77±1755.73	6302.85±2130.48	0.071	1832.09±1764.92	6207.15±1909.93	0.059	121.24±1053.02	3168.52±972.59	0.013	122.45±1052.97	3169.22±972.66	0.013				9.00±25.11	24.89±13.32	0.02
												Dorea			Dorea formicigenerans
												−79.08±114.46	198.31±76.08	0.001	−24.69±28.41	30.69±15.71	0.004
												Eubacterium hallii group			Eubacterium hallii group
												83.30±91.37	128.51±50.37	0.031	22.30±25.72	37.49±16.83	0.049
												Eubacterium nodatum group
												−1.06±2.75	0.38±1.48	0.093
												Eubacterium			Eubacterium rectale
															−646.22±390.97	1063.10±397.96	0.004
												Lachnoclostridium			Lachnoclostridium sp YL32
															−5.04±15.94	4029±15.77	0.01
												Agathobacter
												−703.94±412.94	1100.73±416.7	0.0003
												Lachnospiraceae ND3007 group
												−35.49±43.96	29.95±11.59	0.003
						Oscillospirales			Ruminococcaceae			Ruminococcus			Ruminococcus bicirculans
												−109.34±77.92	440.36±145.83	0.016	−7028±52.03	95.82±48.95	0.011
												Ruminococcus gauvreauii group
												−49.66±58.76	30.40±17.25	0.002
												Ruminococcus torques group
												−83.57±114.74	129.42±63.94	0.081
									Oscillospiraceae			Oscillibacter
												2.32±13.21	36.56±10.96	0.029
						Monoglobales			Monoglobaceae			Monoglobus
						−43.83±25.83	28.67±14.87	0.012	−43.83±25.83	28.67±110.25	0.012	−42.83±25.83	28.67±14.87	0.012
			Bacilli			Lactobacillales			Lactobacillaceae			Ligilactobacillus			Lactobacillus ruminis
			−646.04±522.74	399.98±442.36	0.096										−192.70±157.35	45.98±74.02	0.005
									Streptococcaceae			Lactococcus
												−17.66±16.94	66.85±50.17	0.083
						Erysipelotrichales			Erysipelotrichaceae			Dielma
						−259.45±363.92	485.16±301.87	0.07				−0.04±0.72	0.96±0.42	0.083
Actinobacteriota			Coriobacteriia			Coriobacteriales			Coriobacteriaceae			Collinsella			Collinsella aerofaciens
			−735.49±221.61	−443.44±273.09	0.018	−735.49±221.61	−443.55±273.09	0.018	−732.66±211.77	−486.89±1813.47	0.024	−732.36±211.78	−486.80±244.47	0.026	−692.79±206.84	−466.70±240.15	0.028
									Coriobacteriales incertae sedis			Raoultibacter
												−0.62±0.29	0.00±0.54	0.091
Desulfobacterota			Desulfovibrionia			Desulfovibrionales			Desulfovibrionaceae			Desulfovibrio			Desulfovibrio piger
															−3.51±6.91	31.42±16.52	0.009
Bacteroidota			Bacteroidia			Bacteroidales			Marinifilaceae			Odoribacter			Odoribacter splanchnicus
												−0.49±13.86	37.02±16.82	0.029	−11.75±10.96	35.07±16.61	0.008
									Barnesiellaceae			Coprobacter
									−22.00±32.45	71.73±23.71	0.014	3.60±4.61	4.56±2.36	0.073

Eubacterium rectale is a butyrate producer, where a decreased abundance as seen in patients with immunological disorder, showed negative effects involving many organs including the gut [42]. Not much information is available on Dielma, Lachnospiraceae ND3007 group and Lachnoclostridium species, except that they are isolated from human feces and are SCFAs producers typically butyrate [43]. Agathobacter is also a butyrate producer [44] and its reduced abundance has been associated with sleep disorder in children [45]. Ruminococcus are butyrate producers and important gut microbial mutualists via degrading and converting complex polysaccharides into a variety of nutrients for their hosts [46]. Ruminococcus bicirculans can utilize a variety of carbohydrates such as soluble starch, hemicelluloses, barley beta‐glucan, (1,4)‐beta‐d‐mannan and xyloglucan to produce acetate, a gut protective SCFA against pathogenic toxins [47]. Oscillibacter produces valeric acid, which is crucial in the gut to control Clostridium difficile infections [48]. Specialized in pectin degradation, Monoglobus was shown to be negatively associated with neutrophilic inflammation [49], while a decreased in its abundance was associated with enhanced systemic inflammation [50] and seen in hepatitis‐B virus‐induced liver disease patients [51]. Lactococcus species are typically of low virulence and seldom considered pathogenic, with several species used as probiotics [52], while Lactobacillus ruminis is a commensal motile lactic acid bacterium living in the intestinal tract of humans and animals [53]. Taken altogether, the significant increase in abundance of these genera and species, in the probiotic group over time as compared to the placebo group, showed crucial modulation of the gut microbiota typically associated with beneficial metabolites such as SCFAs.

Lower taxonomic levels of phylum Actinobacteria showed a higher decrease in abundance among the placebo group as compared to the probiotic group. This included class Coriobacteriia (p = 0.018), order Coriobacteriales (p = 0.018), family Coriobacteriaceae (p = 0.024), genera Collinsella (p = 0.026) and Raoultibacter (marginally p = 0.091), and species Collinsella aerofaciens (p = 0.028). Meanwhile, species Desulfovibrio piger under the phylum of Desulfobacterota showed increased in abundance over time in the probiotic group but decreased in the placebo group (p = 0.009). Such a similar trend was also observed for lower taxonomic levels of phylum Bacteroidota, namely genus Odoribacter (p = 0.029) and its species Odoribacter splanchnicus (p = 0.008), and family Barnesiellaceae (p = 0.014) and its genus Coprobacter (marginally p = 0.073). Collinsella aerofaciens is a known common gut species in healthy people [54], and has been shown to be enriched in patients with a positive response toward cancer immunotherapy [55]. Desulfovibrio piger is one of the most common sulfate‐reducing bacteria in among healthy adults. They utilize hydrogen as electron donors for reduction of sulfate or other oxidized sulfur compounds which are commonly found in diets including preservatives, antioxidants and in food additives [56]. Odoribacter is a common SCFA producing member of the human intestinal microbiota. A decreased abundance of Odoribacter has been linked to different microbiota‐associated diseases, such as nonalcoholic fatty liver disease, cystic fibrosis and inflammatory bowel disease [57].

While several taxa identified in this study have been previously associated with disease states in other anatomical sites, these associations should be interpreted with caution. Microbial functions and clinical relevance may differ depending on the host niche, and findings from other body sites may not be directly translatable to the vaginal environment. Therefore, the observed taxonomic shifts should be considered in the context of microbial ecology rather than as direct indicators of disease risk.

3.5. Limitations of Study

This study has several limitations that should be considered when interpreting the findings. First, the study population consisted of vaginally healthy women (Nugent score <7), which inherently limits the clinical interpretation of observed changes, as the scope for therapeutic improvement was minimal. Thus, the findings are more reflective of maintenance or stabilization of vaginal microbial balance rather than treatment effects.

Second, microbiota analyses were based on compositional 16S rRNA sequencing data, which describe relative abundance and do not provide functional or mechanistic insights. As such, the observed changes in microbial taxa represent associations and ecological shifts rather than direct causal relationships.

Third, although all randomized participants were included in the intention‐to‐treat population, not all provided complete biological samples at both time points. Consequently, analyses requiring paired samples were conducted on a per‐protocol or available‐case basis, which may introduce some degree of selection bias.

Fourth, the statistical analysis was primarily focused on hypothesis testing using non‐parametric methods, and effect size measures and confidence intervals were not included. This may limit the interpretation of the magnitude and precision of the observed effects.

Finally, while some taxa identified in this study have been previously associated with disease states in other anatomical niches, such associations should be interpreted with caution. Microbial roles may differ depending on the host environment, and findings from other body sites may not be directly applicable to the vaginal ecosystem. Future studies incorporating functional analyses, larger sample sizes, and more comprehensive statistical approaches are warranted to validate and extend these findings.

4. Concluding Remarks

This study demonstrated that supplementation with L. plantarum Probio87 beneficially modulated both the vaginal and gut microbiota of healthy women. Probio87 preserved vaginal microbial stability by maintaining beneficial Lactobacillales and suppressing potentially pathogenic taxa such as Gardnerella, Cupriavidus, and Collinsella aerofaciens. These changes were reflected in improved Nugent scores, suggesting strengthened vaginal health and reduced risk of dysbiosis. In the gut, Probio87 promoted higher microbial richness and evenness, characterized by increased abundances of SCFA producers that are crucial for maintaining intestinal homeostasis, metabolic regulation, and mucosal integrity. The concurrent modulation across both ecosystems supports the existence of a functional gut–vaginal axis, through which oral probiotics can exert systemic benefits on women's microbiota and overall wellbeing. Collectively, the findings establish L. plantarum Probio87 as a promising, safe, and multifunctional probiotic capable of enhancing microbial resilience and promoting eubiosis across gut and vaginal environments—supporting its potential role in the preventive maintenance of female urogenital and gastrointestinal health.

Author Contributions

M.T.L., A.S.A. and E.H.E.I. conceived and designed the experiments. M.N.K. and S.S. provided medical consultation and performed sampling. A.A.N., K.S.G., J.S.T. and M.T.L. analyzed the data, drafted the work, and wrote the manuscript. A.A.N. and U.M. performed the study. All authors revised critically for intellectual content.

Funding

This work is supported by the USM‐Probionic Grant (grant number 304/PTEKIND/6501096).

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Supporting File: mnfr70535‐sup‐0001‐SuppMat.docx.

Data Availability Statement

Data is available upon request from corresponding authors.