Skip to main content
NCBI home page
Brain, Behavior, & Immunity - Health logoLink to Brain, Behavior, & Immunity - Health
. 2025 Aug 5;48:101081. doi: 10.1016/j.bbih.2025.101081

Altered gut microbial diversity, composition, and metabolomic potential in patients with major depressive disorder and recent suicide attempt

Emese Prandovszky a,⁎, Hua Liu a, Emily G Severance a, Victor W Splan a, Faith B Dickerson b, Robert H Yolken a
PMCID: PMC12390952  PMID: 40896414

Abstract

This study investigates the role of the gut microbiome in suicidal behavior among individuals with major depressive disorder (MDD). Fecal samples from 50 hospitalized patients with MDD, including 35 with recent suicide attempts (60 % female) and 15 without a history of suicide (73 % female), were analyzed using 16S rRNA and shotgun sequencing to assess microbiome diversity and metabolic potential. Results revealed that suicide attempters exhibited significantly greater microbial richness and distinct beta-diversity patterns. Notably, they had higher levels of Fenollaria timonensis and lower levels of Corynebacterium aurimucosum. Additionally, 25 metabolic pathways differed between groups, with several linked to energy metabolism and amino acid processing—processes previously associated with MDD and suicidal behavior. These findings suggest that microbiome composition may influence suicide risk through gut-brain axis-mediated pathways, although due to the exploratory nature of this study further investigation is needed to validate our findings. Given the microbiome's modifiability, future research should explore microbial-targeted interventions as a potential strategy for suicide prevention in individuals with MDD.

Keywords: Recent suicide attempt, Microbiome, Shotgun sequencing, Metabolomic profiling, Gut-brain axis

HIGHLIGHTS

  • •

    Patients with major depressive disorder are at high risk for suicidal behavior

  • •

    Unexpected behavioral outcomes may result from microbiome/gut-brain axis dysfunction

  • •

    Suicide attempters exhibited distinct gut microbial profiles in fecal samples

  • •

    Suicide attempters showed increased levels of Fenollaria timonensis

  • •

    Enriched microbial pathways are previously linked to suicide and depression

1. Introduction

Suicide is an escalating and under-studied public health concern worldwide. In 2022 alone, the CDC estimates that more than 49,000 individuals in the U.S. died by suicide and 1.6 million individuals made suicide attempts. ((CDC), 2022) Suicide has a complex multifactorial etiology comprising some known, but many more yet-to-be identified psychological and biological factors. Effective methods for the prevention and treatment of suicide behaviors are under development but remain to be validated.

In recent advancements, the discovery of robust interactions between microbes inhabiting the gastrointestinal (GI) tract and biological outcomes in seemingly non-GI endpoints, such as the central nervous system (CNS), has propelled psychiatric research in unexpected directions (Dash et al., 2022). These robust interactions involving immune, endocrine, and stress related bi-directional pathways are characterized as the gut-brain axis (Severance et al., 2016). The sum of genes of the microbes residing in the GI tract, collectively known as the intestinal microbiome, are the major modulators of the gut-brain axis.

Studies of the microbiome in psychiatric and neurological disorders have become quite routine and included study populations of individuals with Alzheimer's Disease, autism, bipolar disorder, first episode psychosis, Parkinson's Disease, schizophrenia, and others (Dash et al., 2022; Zhu et al., 2021). Clinical investigations of the gut-brain axis, however, have only briefly addressed on the topic of suicide, suicidal behaviors, suicide ideation, or suicide attempts. For example, Kim et al. found that psychological pain due to social exclusion was correlated with changes in gut microbiota composition, indicating that environmental and psychological stressors can influence microbiome health, which may, in turn, affect mental health outcomes, including suicidal behavior (Kim et al., 2022).

Persons with major depression disorders (MDD) have a substantially elevated risk of suicide compared to the general population (Bertolote et al., 2004). A recent meta-analysis found that individuals with MDD had a sevenfold higher risk of suicide attempts within a one-year period compared to those without MDD (Cai et al., 2021). Recent studies have suggested that alterations in the gut microbiome are associated with metabolic disturbances and inflammatory processes in MDD (Cruz-Pereira et al., 2020; Refisch et al., 2023). However, the composition along with the metabolomic potential of the gut microbiome in individuals with recent suicide behaviors has not been extensively studied.

We employed amplicon and shotgun DNA sequencing techniques to characterize the gut microbiome of 50 individuals hospitalized with MDD, 35 of whom had undergone a recent suicide attempt within 1 month of hospitalization. In addition to quantifying microbial diversity and species representation, we characterized the metabolic potential of the microbiome with a particular emphasis on pathways associated with suicide behaviors.

2. Materials and methods

2.1. Design and participants

2.1.1. Study participants

Study participants were recruited at Sheppard Pratt, a large non-profit psychiatric health system in central Maryland, USA, between 2020 and 2022. Inclusion criteria were: 1) Receiving treatment on a Sheppard Pratt acute adult inpatient unit or in the day hospital program; 2) Meets criteria for a current major depressive episode; 3) Primary diagnosis of a MDD confirmed by the research team and based on the Structured Clinical Interview for DSM-5 Disorders (First et al., 2015) and available medical records; 4) Age 18–65; 5) Suicide attempt history in either of two categories per assessment on the Columbia Suicide Severity Rating Scale (Posner et al., 2011): a) history of a suicide attempt, an action undertaken with at least some wish to die as a result of the act, within the past 30 days; or b) no lifetime history of a suicide attempt and the absence of active suicidal thoughts in the past 30 days; 6) Ability to communicate in English; 7) Capacity for providing informed consent. Exclusion criteria included: 1) Presence of underlying celiac disease, inflammatory bowel disease, or other systematic conditions known to be associated with persistently elevated levels of markers of GI inflammation; 2) HIV or other conditions associated with compromised immune system; 3) Use of potent systemic anti-inflammatory agents such as corticosteroids within the past 2 weeks; 4) Pregnancy or planning to become pregnant; 5) A diagnosis of substance use disorder (moderate or severe) based on DSM-5 criteria. All participants provided written informed consent after the study was explained.

This study was approved by the Institutional Review Boards of Sheppard Pratt and the Johns Hopkins School of Medicine (IRB Approval:1480364-3).

2.1.2. Demographic and clinical variables

Participants were interviewed by the research staff about demographic variables including gender identity, sexual orientation, maternal education as a proxy for pre-morbid socioeconomic status, and current tobacco smoking. Weight and height measurements were used to calculate Body Mass Index (BMI). Participants were queried about their history of alcohol and drug use, which was classified into one of three groups: no history, past history only, or recent history (within the past 3 months). In addition, participants were also queried about past psychiatric hospitalizations. Current medications were extracted from the medical record, with special attention given to whether each participant was receiving the following types of medication: antidepressants, antipsychotics, or anticonvulsant mood stabilizers, such as lithium. Information about current medical conditions was obtained both from medical records and by self-report, as previously described (Dickerson et al., 2021).

In addition, participants were assessed on the Hamilton Depression Rating Scale (Ham-D)(Cruz-Pereira et al., 2020), a measure of depression severity in individuals with MDD; the Barratt Impulsiveness Scale, Version 11 (BIS)(Patton et al., 1995), a self-report measure for assessing impulsive personality traits; the Stressful Life Events Screening Questionnaire (SLES)(Goodman et al., 1998), a self-report measure of lifetime exposure to traumatic events; the Adverse Childhood Events Questionnaire (ACEQ)(Felitti et al., 1998), an instrument to assess the experience of childhood trauma.

2.2. Sample collection and processing

2.2.1. Microbiome sample collection

Rectal swabs were collected during study visits by self-administration with a sterile swab (Foamtec Medical, Ref: MP1302ST). This sample collection method is well accepted by the patient population and provides a comprehensive representation of the rectal microbiome (Marin et al., 2025). The procedures are detailed in the Supplemental material (Appendix Method A.1). Samples were stored below - 20 °C prior to transfer to the Johns Hopkins Division of Developmental Neurovirology laboratory where they were stored at −80 °C until processed.

2.2.2. DNA extraction

DNA was extracted from rectal swabs manually using the QIAamp® PowerFecal® Pro DNA kit (Qiagen, Hilden, Germany, Ref: 51804) according to the manufacturer's instructions with minor modifications. The collection brush heads from the swabs were excised into the PowerBead Pro tubes containing 800 mL lysis buffer. Bead-beating was performed on a horizontal Vortex Adaptor (Qiagen, Hilden, Germany, Ref: 13000-V1-24) at maximum speed for 25min. DNA was eluted in 30 μL elution buffer.

About 20 samples together with one blank swab and one or two of the different mocks were extracted simultaneously. Mock whole cell mixes employed to monitor DNA extraction were ZymoBIOMICS™ Microbial Community Standards (Ref: D6300) from Zymo Research (Irvine, CA), and Whole Cell Mixes for Gut Microbiome (Ref: MSA-2006) and Mycobiome (Ref: MSA-2010) from ATCC (Manassas, VA).

Each extracted DNA concentration was measured using Qubit® 3.0 Fluorometer (Thermo Fisher Scientific, Waltham, MA) with Qubit® dsDNA HS Assay Kit (Thermo Fisher Scientific, Waltham, MA, Ref: Q32854).

2.3. Amplicon library preparation, sequencing and analysis

2.3.1. Amplicon library preparation

Library preparation was performed according to the standard instructions of the 16S Metagenomic Sequencing Library Preparation guide (Illumina™, Inc., San Diego, CA). Briefly, aliquots of 12.5 ng of the isolated DNA from each sample were used to amplify the V3–V4 region of the bacterial 16S rRNA gene using the 341F/785R primer set (Klindworth et al., 2013) on a Applied Biosystems Veriti™ Thermal Cycler (Thermo Fisher Scientific, Waltham, MA). PCR conditions were as follows: 3min at 95 °C; then 25 cycles of at 30s at 95 °C, 30s at 55 °C and 30s at 72 °C; and a final step of 5min at 72 °C.

The amplified DNA was purified with the AMPure XP reagent (Beckman Coulter Life Sciences, Indianapolis, IN; Ref: A63881). A secondary 8-cycle amplification was performed to attach the Illumina indices and sequencing adapters using the Nextera® XT Index kit v2 (Illumina, San Diego, CA; Ref: FC-131-2001). The final libraries were again purified using the AMPure XP reagent. Each library concentration was determined using Qubit® dsDNA HS Assay Kit (Thermo Fisher Scientific, Waltham, MA, Ref: Q32854) and was assessed using DNA ScreenTape D1000 (Agilent, Santa Clara, CA; Ref: 5067–5582) on a 4200 TapeStation system (Agilent, Santa Clara, CA) to determine library quality and size. Concentration of each prepared library was normalized to 4 nM and then 5 μL of each library was pooled together.

Several mock microbial community genomic DNA standards were used as positive controls of the library preparation: Gut Microbiome Genomic Mix (Ref: MSA-1006), Mycobiome Genomic Mix (Ref: MSA- 1010) from ATCC (Manassas, VA), and ZymoBIOMICS™ Microbial Community DNA Standards (Ref: D6306) from Zymo Research (Irvine, CA). Blank DNA extration and non-template library preparation controls (NTC) were used as negative controls to monitor for environmental contaminants."

2.3.2. Amplicon library sequencing and analysis

Sequencing was performed on the MiSeq Illumina sequencing platform using a 2 × 300 PE sequencing reagent kit (Illumina, San Diego, CA, Ref: MS-102-3003) at the Sequencing Core Facility at Johns Hopkins. Before loading, the library was spiked with a 10 % 20 pM PhiX control.

Raw paired-end reads were first de-multiplexed using 5′ index information, and adapter sequences were trimmed off prior to access. Sequence reads were analyzed using the QIIME 2 (v2023.2) platform (Bolyen et al., 2019). In addition to filtering out any remaining PhiX contaminants and chimeric sequences, the DADA-q2 plugin (Callahan et al., 2016) was utilized to remove low-quality reads. All unique sequences were collected into a feature table and filtered to remove low-abundance features using the feature-table q2 plugin (McDonald et al., 2012). Next, multiple sequence alignment was performed by employing the MAFFT alignment algorithm plugin (Katoh and Standley, 2013), followed by computing a phylogenetic tree using the FastTree program (Price et al., 2010). Prior to the taxonomy assignment, the feature table was filtered for potential contaminants. Features were considered contaminant if presented in at least one blank DNA extraction or non-template library preparation controls with at least a count of 10 or higher. To explore the bacterial composition of the samples, taxonomy was assigned to sequences using a pre-trained naive Bayes classifier (Werner et al., 2012) and q2-feature-classifier plugin (https://github.com/qiime2/q2-feature-classifier). This classifier was trained on Greengenes v13_8 99 % OTUs, where the sequences have been trimmed to only include v3-v4 regions of the 16S rRNA gene bound by Illumina primer pairs.

Analyses of the amplicon sequencing data were focused on alpha and beta diversity analysis computed using the Qiime2 platform (Appendix Fig. A.1). The computed default alpha diversity indices included observed features, Shannon and Faith's phylogenetic diversity indices as well as evenness. The computed beta diversity measures included unweighted Jaccard, Bray-Curtis weighted and unweighted UniFrac indices.

Figures were created using the Qiime2 software (alpha diversity boxplots) or qiime2R (Bisanz, JE (2018) https://github.com/jbisanz/qiime2R.) as an R (v4.0)(Team, 2010) package.

2.4. Shotgun library preparation, sequencing and analysis

2.4.1. Shotgun library preparation

Shotgun sequencing libraries were prepared according to Illumina Nextera XT DNA Library Prep reference guide (Illumina, San Diego, CA) using .5 ng of each isolated DNA from the rectal swabs. The Nextera XT DNA Library Preparation Kit (Illumina, San Diego, CA; Ref: FC-131-1024) was used to simultaneously fragment and tag target DNA with adapter sequences. Next, the tagmented DNA was amplified using a limited-cycle PCR program adding index 1 (i7) and index 2 (i5) in unique combinations and sequences specific for cluster formation using the Illumina® DNA/RNA UD Indexes (Illumina, San Diego, CA; Ref: 20091656). The PCR program was: 3min at 72 °C; 30s at 95 °C; followed by 12 cycles of 10s at 95 °C, 30s at 55 °C, and 30s at 72 °C; and a final extension for 5 min at 72 °C. The library DNA then was purified using the AMPure XP Reagent (Beckman Coulter Life Sciences, Indianapolis, IN; Ref: A63881).

Each library was quantified using Qubit® dsDNA HS Assay Kit (Thermo Fisher Scientific, Waltham, MA, Ref: Q32854) and the quality and average size of each library was assessed using DNA ScreenTape D1000 on a TapeStation system. Concentration of each prepared library was normalized and pooled in equimolar ratios. ZymoBIOMICS™ Microbial Community Standards, ZymoBIOMICS™ Microbial Community DNA Standards, ATCC Gut Microbiome and ATCC Mycobiome Whole Cell Mix and ATCC Gut Microbiome Genomic Mix were used as positive controls. Blank DNA extraction and non-template library preparation controls (NTC) were used as negative controls to monitor for environmental contaminants."

2.4.2. Shotgun sequencing and analysis

Libraries were then sequenced on the Illumina NovaSeq X platform at the sequencing core facility at Johns Hopkins.

Sequence reads (150 bp, single-end reads) with homology to human sequences (T2T), as well as low-quality reads and repeats were removed by KneadData (github.com/biobakery/kneaddata). The metabolic potential of the microbiome was characterized using humanN (v3)(Beghini et al., 2021) as described. Taxonomic profiles of the microbes were characterized using MetaPhlan4 (Blanco-Míguez et al., 2023). Heatmaps were constructed to visualize the individual levels of the most abundant species with the abundance set to 10 by default based on rowMeans.

Differential abundance analysis was performed on the MetaPhlan4 output file applying Analysis of Compositions of Microbiomes with Bias Correction (ANCOM-BC) (Lin and Peddada, 2020). Log fold changes (LFC), p values, adjusted p values (q values) and standard error were computed by ANCOM-BC. ANCOM-BC employs the Benjamini-Hochberg (BH)(Benjamini and Hochberg, 1995) method to control for the false discovery rate. Reference level was set to the group of individuals without a history of a suicide attempt additional covariates were not included in the model due to the small sample size.

Pathway differential abundance analysis was carried out by utilizing MaAslin2 (Mallick et al., 2021) as an R (v4.0)(Team, 2010) package. Unstratified pathway abundance data as an output file of the humanN3 analysis served as an input for MaAslin2. MaAslin2 ran with default parameters. Reference level was set to the group of individuals without a history of suicide attempts, additional covariates were not included in the original model due to the small sample size. The significance value was corrected using BH (Benjamini and Hochberg, 1995).

Spearman rank test (Spearman, 1961) was computed using stats as an R (v4.0)(Team, 2010) core package to study the correlation between differentially abundant pathways and bacterial species among clinical groups. These assays were conducted by comparing the relative abundance of pathways with that of the microbes. Results were expressed as correlation coefficient (rho) and a corrected p value (Appendix Fig. A.1).

Visualization was done by R (v4.0) (Team, 2010) employing the ggplot2 package (Wickham and Sievert, 2009). Flowcharts and summary figures (working model) were created using BioRender.

2.5. Levels of statistical significance

Alpha and beta diversity analyses, using Kruskal-Wallis and PERMANOVA pairwise test respectively, were considered to differ significantly between the groups based on a p value < .05, while p value of > .05 and < .1 was considered to indicate a trend towards significance.

Species that considered significantly different in the differential abundance analysis had an uncorrected p value < .05, due to the large number of identified species (n = 783) and log fold changes with absolute value of ≥ .20.

Pathways that considered significantly different in the differential abundance analysis had a BH adjusted p value < .05 and log fold change with absolute value of ≥ .5.

3. Results

3.1. Altered bacterial diversity in the rectal microbiome of MDD patients with a recent suicide attempt

We evaluated the bacterial component of the intestinal microbiome in 50 hospitalized individuals with a diagnosis of MDD employing self-collected rectal swab samples. Among these individuals, 35 had undergone a suicide attempt within 1 month of hospitalization, while 15 had no history of any suicide attempts. The clinical and demographic characteristics of these two groups are described in Table 1. Suicide attempters had significantly fewer years of education and earlier age of mood disorder onset as compared to non-attempters. The two groups were otherwise largely comparable across demographic and clinical variables, including medication use.

Table 1.

Participant characteristics.

Recent suicide attempt (n = 35)a No recent or past suicide attempt (n = 15)a
Age, years 31.0 ± 12.2 36.8 ± 9.3
Sex Female 21 (60 %) 11 (73.3 %)
Gender Identity:
 Cis man 13 (37.1 %) 4 (23.6 %)
 Cis woman 17 (48.6 %) 11 (73.4 %)
 Trans man 2 (5.7 %) 0
 Trans woman 1 (2.9 %) 0
 Non-binary 2 (5.7 %) 0
Race White 14 (40 %) 7 (46.7 %)
Sexual orientation:
 Heterosexual/Straight 17 (48.6 %) 11 (73.2 %)
 Bisexual 8 (22.8 %) 1 (6.7 %
 Gay 3 (8.6 %) 1 (6.7 %)
 Lesbian 2 (5.7 %) 1 (6.7 %)
 Asexual 3 (8.6 %) 0
 Other 2 (5.7 %) 1 (6.7 %)
Education years b 13.6 ± 2.0 15.6 2.3
Maternal education years 14.6 ± 2.6 13.8 ± 2.1
Current tobacco smoker 5 (14 %) 3 (20 %)
Body Mass Index 30.2 ± 8.6 30.8 ± 7.1
Alcohol or Drug misuse:
 None or past 18 (51 %) 10 (67 %)
 Recent – mild 17 (49 %) 5 (33 %)
Hamilton Depression Scale total 26.9 ± 11.0 27.7 ± 9.1
Adverse Childhood Events (ACE) total scorec 4.1 ± 2.9 3.7 ± 1.8
Social Readjustment Rating Scale (SRRS) total score 467.7 ± 435.3 510 ± 368.7
Stressful Life Events Screening Questionnaire (SLES) total score 4.1 ± 3.0 3.7 ± 2.4
Barratt Impulsiveness Scale, total 70.8 ± 11.4 64.4 ± 7.7
Age mood disorder onset, years d 13.7 ± 8.9 22.5 ± 11.7
Duration mood disorder, years 16.9 ± 12.6 13.9 ± 10.5
Number of psychiatric hospitalizations in the past 2 years 3 (8.5 %) 1 (6.7 %)
Medications:e
 Antidepressant 27 (77 %) 12 (80 %)
 Antipsychotic 16 (46 %) 6 (40 %)
 Anticonvulsant mood stabilizer 10 (29 %) 4 (27 %)
 Lithium 2 (6 %) 0 (0 %)
Open in a new tab
a

Mean ± SD (standard deviation) or number (percentage).

b

Significant difference between groups (F = 9.69, p < .004) One-way analysis of variance.

c

One value missing.

d

Significant difference between groups (F = 10.03, p < .003) One-way analysis of variance.

e

Individuals may be receiving more than one medication.

To assess differences in microbial diversity, we utilized amplicon-based DNA sequencing of the 16S v3v4 region of the bacterial ribosomal RNA gene (Appendix Table B.1). We computed four measures of alpha diversity (species richness) and four beta diversity metrics (community dissimilarity) to evaluate bacterial diversity (Fig. 1C and D). In terms of alpha diversity, individuals with a recent suicide attempt exhibited significantly richer microflora compared to individuals without suicide attempt, as measured by the Faith phylogenetic diversity index (p = .045) (Fig. 1B). No significant differences were found in the other three alpha diversity measures. For beta diversity, the greatest difference between the groups was observed in the unweighted UniFrac metric (p = .020) (Fig. 1A). Significant differences were also measured by the weighted UniFrac (p = .039) and Jaccard metrics (p < .04), while a trend toward significance was noted in the Bray-Curtis metric (p = .053) (Appendix Fig. B.1).

Fig. 1.

Fig. 1

Open in a new tab

Microbial diversity in the rectum of MDD patients with or without recent suicide attempt. A: Principal coordinate analysis was employed to assess the bacterial community dissimilarities in fecal samples obtained via rectal swab, using Unweighted Unifrac metrics (beta diversity index). B: Table displays the results of pairwise PERMANOVA test in all 4 metrics. The test showed significant (p < .05) community differences in rectal swabs between hospitalized MDD patients with recent suicide attempt and with no suicide attempt. C: Alpha diversity analysis was used to assess species richness in fecal samples obtained via rectal swab using Faith phylogenetic diversity metrics. D: Table displays the results of the Kruskal-Wallis test in all 4 metrics. The pairwise Kruskal-Wallis test showed significant (∗, p < .05) species enrichment in hospitalized MDD patients with a recent suicide attempt. Red color represent MDD patients with recent suicide attempt and blue color represents patients with no suicide attempt. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)

3.2. Altered taxonomical composition of the fecal microbiome of MDD patients with a recent suicide attempt

Shotgun sequencing of the rectal swab samples resulted in the identification of 783 bacterial species that were present in at least 2 samples (Appendix Table C.1). We included mock controls and blanks in each run. The blanks were free of bacteria, while the three different types of mock communities consistently contained the intended species. The taxonomy profiling of these samples is provided in the supplement (Appendix Table C.1).

The most differentially abundant species between attempters and non-attempters are depicted in Fig. 2. The term “most differentially abundant” refers to the taxa or features that show the greatest differences in abundance between the two groups through statistical testing. The bar graph highlights the portion of the ANCOM-BC output (Appendix Fig. C.1), including species that met the criteria of p < .05 and an absolute value of log fold change ≥.20 (Fig. 2B). As depicted in Fig. 2A, several species met the cutoff threshold however only the following two taxa had the most favorable values for both LFC and p-values and only these two taxa are highlighted due to the exploratory nature of the study: Fenollaria timonensis (p < .003, LFC = .56), which was increased, and Corynebacterium aurimucosum (p = .034, LFC = .71) which was depleted in individuals with a recent suicide attempt.

Fig. 2.

Fig. 2

Open in a new tab

Altered bacterial composition in hospitalized MDD patients with a recent suicide attempt. A: Heatmap showing the most abundant species that were identified by shotgun sequencing. Heatmaps were constructed to visualize the individual levels of the most abundant species with the abundance set to 10 by default based on rowMeans. B: Differential abundance analysis at species level. Taxa with p<.05 and an absolute value of log fold change ≥ .20 are depicted. Green bars indicate enrichment (positive log fold change) and blue bars indicates depletion (negative log fold change) in patients with a recent suicide attempt (recent attempters). Reference level was set to non-attempters. Log fold Change (LFC), p value, standard error (SE) were computed using ANCOM-BC (Analysis of Compositions of Microbiomes with Bias Correction). Error bars are representing standard error. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)

3.3. Altered microbial metabolic potential in the rectal microbiome of MDD patients with a recent suicide attempt

The microbial sequences obtained by shotgun sequencing were also utilized to identify the metabolic potential of the bacterial microflora in the fecal microbiome of MDD patients with and without recent suicide attempts. We identified 26 metabolic pathways which met the criteria of adjusted p value < .05 and log fold change with absolute value of ≥.5. Among these pathways, the phosphatidylglycerol biosynthesis I (plastidic) pathway (PWY4FS-7) was excluded from downstream analysis, as it is unlikely to be present in non-photosynthetic bacteria within the human gut. (Appendix Table C.1). These 25 pathways can be categorized into four metabolic actions: energy production, lipid metabolism, nitrogen and amino acid metabolism, and nucleotide metabolism (Fig. 3A). Metabolic pathways previously linked to depression or suicide behaviors are highlighted (Fig. 3B–G). These included pathways related to the urea cycle (Fig. 3B), metabolism of pyridoxal (Fig. 3C-D), and inosine (Fig. 3E–G). Adding age and gender to the analyses affected different pathways than those associated with suicidal behavior. However, including age and gender as a covariate decreased the number of pathways that was originally differed significantly between attempters and non-attempters in the initial model. The additional analysis is provided in the supplemental table. Appendix Table C.1.

Fig. 3.

Fig. 3

Open in a new tab

Altered microbial metabolomic profile in the rectum of hospitalized MDD patient with recent suicide attempt. Differential abundance analysis, which was carried out by employing MaAslin2, revealed several pathways that previously has been linked to depression and suicide. A: depicts the pathways, where the absolute log fold change was larger than .5 and BH adjusted p value (q value) was less then .05. The analysis used no suicide attempts as baseline. These 25 pathways can be grouped into four metabolic actions: energy production, lipid metabolism, nitrogen and amino acid metabolism, and nucleotide metabolism. Metabolic pathways previously linked to depression or suicide behaviors are highlighted in Fig. 4B–G. These included pathways related to the urea cycle (B), metabolism of pyridoxal (C-D), and metabolism of inosine (F-G).

We also examined the relationship between these 25 significant pathways and the 2 species, Fenollaria timonsiensis and Corynebacterium aurimucosum, that showed the strongest association with recent suicide attempts (Appendix Table C.1). Spearman correlation analysis unveiled that pyridoxine synthesis (rho = .39; adjusted p = .029), histidine degradation (rho = .5; adjusted p = .003), reductive TCA cycle (rho = .45; adjusted p = .006), and UTP/CTP dephosphorylation (rho = .55; adjusted p < .001) were significantly linked to the abundance of Fenollaria timonsiensis and these associations were positive (Fig. 4A). While we observed significant but negative correlation between the levels of Corynebacterium aurimucosum and pathways related to arginine and polyamine synthesis (rho = −.44; adjusted p = .017), GMP degradation (rho = −.69; adjusted p < .001), and NAD synthesis (rho = −.40; adjusted p = .037) (Fig. 4B).

Fig. 4.

Fig. 4

Open in a new tab

Microbial pathways that are significantly related to the two bacteria with the strongest link to recent suicide behavior. Spearman correlation analysis was carried out to reveal if pathway abundance was correlated with the abundance of Fenollaria timonensis (A) and Corynebacter auimucosum (B). BH adjusted p value, q < .05. On each graph rho is equivalent with the Spearman correlation coefficient. Red dashed line highlights a linear trend. In each graph a particular relative pathway abundance as percentage is plotted against relative abundance of either F. timonensis (A) or C. aurimucosum (B). (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)

4. Discussion

We employed amplicon and shotgun DNA sequencing to characterize the fecal microbiome in hospitalized individuals diagnosed with MDD who had undergone a recent suicide attempt as well as hospitalized individuals with MDD who had no history of suicide attempts. To the best of our knowledge, this is the first study that reports altered bacterial diversity, composition and metabolic potential in MDD patients with recent suicide attempts.

Employing 16S rRNA gene-based amplicon sequencing, we identified significant differences in microbial diversity including community dissimilarity (beta diversity) and species richness (alpha diversity) in MDD patients with a recent suicide attempt, compared to those with no history of suicide attempts. The differences between the two clinical groups were more pronounced in beta diversity measures, suggesting the presence of two distinct microbial communities based on recent suicide attempts. Although Thompson et al. found no association between bacterial diversity and suicide attempt using 16S rRNA gene-based amplicon sequencing, it is likely to be attributed to differences in the selected patient population. Their study included individuals with substance use disorders, had a smaller proportion of patients with suicide attempts, and considered lifetime rather than recent suicide attempts. Moreover, their amplicon sequencing targeted only the V4 region, whereas our 16S amplicons were slightly longer (V3-V4). (Thompson et al., 2021). While decreased microbial diversity is commonly associated with psychiatric symptoms, this is not always the case. For example, a systematic review analyzing gut microbiota in psychiatric disorders found that among 57 studies reporting alpha diversity indices, 12 % showed increased diversity in cases, 18 % found decreased diversity, and 44 % did not detect any significant differences between cases and controls. (Chen et al., 2021)

Additionally, by employing shotgun sequencing, we identified two bacterial species that showed strongest correlation with recent suicide attempts. One species, Fenollaria timonensis, showed increased abundance, while the other, Corynebacterium aurimucosum was decreased in individuals with a recent suicide attempt. Fenollaria timonensis is newly isolated bacteria, that present in the normal gut microflora (Lo et al., 2020). It is an anerobic Gram negative coccobacillus of the order Clostridiales, of the family Lachnospiraceae. Although the biological properties, metabolomic potential and possible pathogenic capacity of this organism have not been studied extensively, it is noteworthy that Fenollaria timonensis has been detected in stool samples obtained from individuals with treatment-resistant depression (Fontana et al., 2020). Interestingly, as depicted in Fig. 5, this organism has a broad sugar metabolism, including pentoses (D-ribose), hexoses, disaccharides, and sugar alcohols, and the ability to utilizes N-acetylglucosamine (GlcNAc) (Lo et al., 2020). GlcNAc could be of interest since the level of N-acetyl-β-glucosaminidase, the enzyme for metabolizing GlcNAc, has been found to be increased in the blood of individuals who have recently made serious suicide attempts (Garvey and Underwood, 1997). Furthermore, mRNA encoding O-GlcNAc transferase, the key enzyme in the post-translational modification of GlcNAc, is elevated in the peripheral blood of unmedicated individuals experiencing their first episode of MDD (Fan et al., 2023). In the same study, Fan et al., demonstrated in a chronic social-defeat stress mouse model, that increased O-GlcNAcylation in astrocytes contributed to depression-like behaviors (Fan et al., 2023). Furthermore, glutamate transporter-1 was identified as an O-GlcNAcylation target, indicating that posttranslational modification of glucose metabolite N-acetylglucosamine contributes to depression behaviors through diminished glutamate clearance from excitatory synapses in the dorsolateral prefrontal cortex (Paton and Menard, 2023). Both glucose metabolic dysfunction and aberrant glutamatergic signaling in the prefrontal cortex has been implicated in MDD.

Fig. 5.

Fig. 5

Open in a new tab

Metabolomic potential of Fenollaria timonensis. Since the metabolomic potential and possible pathogenic capacity of this organism have not been studied extensively, based on literature, the figure highlights some of the prospective metabolomic involvement of Fenollaria timonensis in the rectal microbiome of hospitalized MDD patient with recent suicide. This organism has a broad sugar metabolism, including pentoses (D-ribose), hexoses, disaccharides, and sugar alcohols, and the ability to utilizes N-acetylglucosamine (GlcNAc). Our findings suggest that F. timonensis involved in PLP synthesis, the active form of B6 vitamins, which essential in amino acid metabolism (L-histidine degradation) and may play and indirect role in the synthesis of UDP-GlcNAc. Being closely related to the Lachnospiraceae family, it has the potential to produce short-chain fatty acids; however, no reports have yet confirmed this capability. Created with BioRender.com.

On the other hand, Corynebacterium aurimucosum displayed the most significant depletion in the fecal microbiome of individuals with recent suicide attempts. This anaerobic, Gram-positive rod was discovered in 2002 (Yassin et al., 2002). C. aurimucosum has occasionally been linked to symptomatic infections, such as joint, and urinary tract infection, and should be considered as opportunistic pathogen (Lefèvre et al., 2021). Notably, Corynebacterium aurimucosum has not been previously studied in relation to psychiatric disorders or suicide behaviors.

Previous studies have suggested that alterations in the gut microbiome are associated with metabolic disturbances and inflammatory processes in individuals with mood disorders and suicide behaviors (Cruz-Pereira et al., 2020; Refisch et al., 2023). In line with these studies, we identified 25 predicted metabolic pathways falling into 4 metabolic categories which differed between the two clinical study populations. These pathways included several that have previously been associated with depression or suicide behaviors, such as those involved in the metabolism of glucose (Kennedy et al., 2001), lipids (Zhao et al., 2020; Zheng et al., 2013), nitrogen (Pu et al., 2021) (e.g. urea cycle (Gropman et al., 2007)), amino acids (Zheng et al., 2013), nucleotides (e.g. pyridoxine metabolism (Cellini et al., 2020; Śliwiński and Gawlik-Kotelnicka, 2024), and nucleoside (e.g. inosine (Liu et al., 2021)). We also examined the relationship between the levels of these pathways and the concentrations of the two species that differed the most between the study groups. F. timonensis, which was increased in the microbiome of individuals with recent suicide attempts, showed positive correlations with highlighted pathways linked to recent suicide behaviors (Fig. 5). In contrast, C. aurimucosum, which was decreased in the microbiome of individuals with recent suicide attempts, was associated with lower levels of several suicide-related pathways.

Despite growing evidence that microbial dysbiosis in the GI tract significantly impacts brain functioning and behavior, the role of the gut microbiome in suicidal behavior remains largely unexplored. On the other hand, several previous studies have emphasized the role of systemic and CNS inflammation in depression and suicidal behaviors (Courtet et al., 2016; Dickerson et al., 2017; Katherin and John, 2017). Given the interconnected nature of the microbiome and immune system, our results are consistent with these findings. In addition, GI barrier dysfunction has been implicated in psychiatric disorders and suicide behavior (Borkent et al., 2022; Ohlsson et al., 2019). Our findings support the possibility that persistent bacterial and metabolic dysbiosis, accompanied by barrier dysfunction, can lead to systemic inflammation, which may contribute to subsequent inflammation in the CNS and altered behavior. A model illustrating these potential interactions are depicted(Fig. 6). However, the field has yet to establish a definitive link between altered gut microbiome and suicidal behavior to fully understand the underlying mechanisms.

Fig. 6.

Fig. 6

Open in a new tab

Predicted model of mechanism. Our results indicate in the event of recent suicide persistent bacterial and metabolic dysbiosis may occur, and according to the literature, this can be accompanied by barrier dysfunction. Together, these factors can trigger systemic inflammation, which may then lead to inflammation in the CNS and changes in behavior. Created withBioRender.com.

The strengths of our study include the analysis of a well-matched patient cohort, which was specifically recruited to investigate recent suicide attempts in individuals with the same clinical diagnosis, with the same geographical origin, and within the same intense psychiatric treatment setting. Notably, a recent review authored by many leaders of the field of immunopsychiatry highlighted the needs of subgrouping patients to improve outcome of clinical trials (Miller et al., 2025). We believe, our approach can target this need. Additionally, our study utilized both amplicon-based and shotgun sequencing, allowed species level resolution along with microbial metabolic impact profiling. These combined approaches likely explain why we were able to identify suicide-related differences in the microbiome that have not been observed in other studies (Thompson et al., 2021).

The limitations of our study include a modest sample size and its cross-sectional sample collection as well as lacking the below listed confounders in the analysis and study design. Longitudinal studies to explore the long-term associations between the microbiome and suicidal behaviors are also warranted. Additionally, we assessed the metabolic potential of the microbiome through shotgun sequencing rather than directly measuring metabolites in fecal samples. While recent studies suggest that the algorithm used in this study can predict metabolic pathways from shotgun DNA sequencing data with a relatively high degree of accuracy by linking gene function to metabolic pathways (Beghini et al., 2021), future studies should include direct assessment of metabolites in fecal samples.

Given the limited sample size of our study, our analysis model did not incorporate potential confounding variables. Our study design made it difficult to obtain detailed dietary information. Inclusion of potential confounders such as diet, age, gender, alcohol intake, smoking, and various medications or supplements would require a larger cohort in future studies to address these limitations and validate the findings.

5. Conclusion

In conclusion, our exploratory study demonstrates that the fecal microbiome of individuals with the diagnosis of MDD who were hospitalized after a recent suicide attempt differ from those individuals with MDD who were hospitalized with no history of suicide attempts. These differences encompass both the diversity and composition of the microbiome, as well as the metabolic potential of the microbial organisms. The human microbiome can be modified through various practical interventions, including the administration of prebiotic, probiotic, and postbiotic agents. Thus, the potential to modify suicidal behaviors in high-risk populations through microbiome alterations warrants further exploration in clinical trials. The development of effective strategies for preventing suicidal behaviors would represent a significant advancement in the care of individuals at high risk.

6. Resource avaibility

Request for further information and resources should be directed to and will be fulfilled by the lead contact Emese Prandovszky (eodonne3@jhmi.edu). This study did not generate a new unique reagent or material. Raw amplicon and shotgun seq data of rectal swabs samples have been deposited at NCBI Sequence Read Archive (SRA) with the title " Gut microbiome in patients with depression and recent suicide attempt"and are publicly available as of the date of publication. This paper does not report original code. Any additional information required to reanalyze the data reported in this paper is available from the lead contact upon request.

CRediT authorship contribution statement

Emese Prandovszky: Data curation, Formal analysis, Methodology, Software, Visualization, Writing – original draft, Writing – review & editing. Hua Liu: Investigation, Writing – original draft, Writing – review & editing. Emily G. Severance: Conceptualization, Writing – original draft, Writing – review & editing. Victor W. Splan: Visualization, Writing – original draft, Writing – review & editing. Faith B. Dickerson: Conceptualization, Resources, Writing – original draft, Writing – review & editing. Robert H. Yolken: Conceptualization, Funding acquisition, Resources, Supervision, Writing – original draft, Writing – review & editing.

Declaration of generative AI and AI-assisted technologies

During the preparation of this work, the author(s) used ChatGPT in order to improve readability and language. After using this tool, the author(s) reviewed and edited the content as needed and take(s) full responsibility for the content of the publication.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

8. Acknowledgments

We sincerely acknowledge the support of the Sequencing Core Facility at Johns Hopkins Bayview Campus under the leadership of David Mohr for providing essential resources and technical support with amplicon and shotgun sequencing for this study. We also gratefully acknowledge funding from Stanley Medical Research Institute (SMRI) and American Foundation for Suicide Prevention (SRG-2-039-18) , which made this research possible. Their support has been instrumental in advancing our work. We liked to thank the patients who participated in the study.

Footnotes

Appendix

Supplementary data to this article can be found online at https://doi.org/10.1016/j.bbih.2025.101081.

Appendix. ASupplementary data

The following are the Supplementary data to this article:

Multimedia component 1
mmc1.pdf (49.8KB, pdf)
Multimedia component 2
mmc2.xlsx (233.9KB, xlsx)
Multimedia component 3
mmc3.xlsx (1.3MB, xlsx)
Multimedia component 4
mmc4.pdf (951.7KB, pdf)

Data availability

Raw amplicon and shotgun seq data of rectal swabs samples have been deposited at NCBI Sequence Read Archive (SRA) as [SRA: accession number] and are publicly available as of the date of publication.

References

  1. Beghini F., McIver L.J., Blanco-Míguez A., Dubois L., Asnicar F., Maharjan S., Mailyan A., Manghi P., Scholz M., Thomas A.M. Integrating taxonomic, functional, and strain-level profiling of diverse microbial communities with bioBakery 3. eLife. 2021;10 doi: 10.7554/eLife.65088. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Benjamini Y., Hochberg Y. Controlling the false discovery rate: a practical and powerful approach to multiple testing. J. Roy. Stat. Soc. B. 1995;57:289–300. [Google Scholar]
  3. Bertolote J.M., Fleischmann A., De Leo D., Wasserman D. Psychiatric diagnoses and suicide: revisiting the evidence. Crisis. 2004;25:147–155. doi: 10.1027/0227-5910.25.4.147. [DOI] [PubMed] [Google Scholar]
  4. Blanco-Míguez A., Beghini F., Cumbo F., McIver L.J., Thompson K.N., Zolfo M., Manghi P., Dubois L., Huang K.D., Thomas A.M. Extending and improving metagenomic taxonomic profiling with uncharacterized species using MetaPhlAn 4. Nat. Biotechnol. 2023;41:1633–1644. doi: 10.1038/s41587-023-01688-w. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Bolyen E., Rideout J.R., Dillon M.R., Bokulich N.A., Abnet C.C., Al-Ghalith G.A., Alexander H., Alm E.J., Arumugam M., Asnicar F. Reproducible, interactive, scalable and extensible microbiome data science using QIIME 2. Nat. Biotechnol. 2019;37:852–857. doi: 10.1038/s41587-019-0209-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Borkent J., Ioannou M., Laman J.D., Haarman B.C., Sommer I.E. Role of the gut microbiome in three major psychiatric disorders. Psychol. Med. 2022;52:1222–1242. doi: 10.1017/S0033291722000897. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Cai H., Xie X.-M., Zhang Q., Cui X., Lin J.-X., Sim K., Ungvari G.S., Zhang L., Xiang Y.-T. Prevalence of suicidality in major depressive disorder: a systematic review and meta-analysis of comparative studies. Front. Psychiatr. 2021;12 doi: 10.3389/fpsyt.2021.690130. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Callahan B.J., McMurdie P.J., Rosen M.J., Han A.W., Johnson A.J.A., Holmes S.P. DADA2: high-resolution sample inference from illumina amplicon data. Nat. Methods. 2016;13:581. doi: 10.1038/nmeth.3869. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Cellini B., Zelante T., Dindo M., Bellet M.M., Renga G., Romani L., Costantini C. Pyridoxal 5′-phosphate-dependent enzymes at the crossroads of host–microbe tryptophan metabolism. Int. J. Mol. Sci. 2020;21:5823. doi: 10.3390/ijms21165823. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. (CDC), C.f.D.C.a.P. 2022. Suicide Data and Statistics. [Google Scholar]
  11. Chen L.L., Abbaspour A., Mkoma G.F., Bulik C.M., Rück C., Djurfeldt D. Gut microbiota in psychiatric disorders: a systematic review. Biopsychosoc. Sci. Med. 2021;83:679–692. doi: 10.1097/PSY.0000000000000959. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Courtet P., Giner L., Seneque M., Guillaume S., Olie E., Ducasse D. Neuroinflammation in suicide: toward a comprehensive model. World J. Biol. Psychiatr. 2016;17:564–586. doi: 10.3109/15622975.2015.1054879. [DOI] [PubMed] [Google Scholar]
  13. Cruz-Pereira J.S., Rea K., Nolan Y.M., O'Leary O.F., Dinan T.G., Cryan J.F. Depression's unholy trinity: dysregulated stress, immunity, and the microbiome. Annu. Rev. Psychol. 2020;71:49–78. doi: 10.1146/annurev-psych-122216-011613. [DOI] [PubMed] [Google Scholar]
  14. Dash S., Syed Y.A., Khan M.R. Understanding the role of the gut microbiome in brain development and its association with neurodevelopmental psychiatric disorders. Front. Cell Dev. Biol. 2022;10 doi: 10.3389/fcell.2022.880544. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Dickerson F., Adamos M., Katsafanas E., Khushalani S., Origoni A., Savage C., Schweinfurth L., Stallings C., Sweeney K., Alaedini A. The association between immune markers and recent suicide attempts in patients with serious mental illness: a pilot study. Psychiatry Res. 2017;255:8–12. doi: 10.1016/j.psychres.2017.05.005. [DOI] [PubMed] [Google Scholar]
  16. Dickerson F., Origoni A., Rowe K., Katsafanas E., Newman T., Ziemann R.S., Squire A., Khushalani S., Stallings C., Daumit G., Yolken R. Risk factors for natural cause mortality in a cohort of 1494 persons with serious mental illness. Psychiatry Res. 2021;298 doi: 10.1016/j.psychres.2021.113755. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Fan J., Guo F., Mo R., Chen L.-Y., Mo J.-W., Lu C.-L., Ren J., Zhong Q.-L., Kuang X.-J., Wen Y.-L. O-GlcNAc transferase in astrocytes modulates depression-related stress susceptibility through glutamatergic synaptic transmission. J. Clin. Investig. 2023;133 doi: 10.1172/JCI160016. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Felitti V.J., Anda R.F., Nordenberg D., Williamson D.F., Spitz A.M., Edwards V., Marks J.S. Relationship of childhood abuse and household dysfunction to many of the leading causes of death in adults: the adverse childhood experiences (ACE) study. Am. J. Prev. Med. 1998;14:245–258. doi: 10.1016/s0749-3797(98)00017-8. [DOI] [PubMed] [Google Scholar]
  19. First M.B., Williams J.B., Karg R.S., Spitzer R.L. 2015. Structured Clinical Interview for DSM-5 Disorders. Clinician Version (SCID-5-CV) [Google Scholar]
  20. Fontana A., Manchia M., Panebianco C., Paribello P., Arzedi C., Cossu E., Garzilli M., Montis M.A., Mura A., Pisanu C. Exploring the role of gut microbiota in major depressive disorder and in treatment resistance to antidepressants. Biomedicines. 2020;8:311. doi: 10.3390/biomedicines8090311. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Garvey M.J., Underwood K. Association of N-acetyl-β-glucosaminidase levels with seriousness of suicide attempts. Biol. Psychiatry. 1997;42:286–289. doi: 10.1016/S0006-3223(96)00346-0. [DOI] [PubMed] [Google Scholar]
  22. Goodman L.A., Corcoran C., Turner K., Yuan N., Green B.L. Assessing traumatic event exposure: general issues and preliminary findings for the stressful life events screening questionnaire. J. Trauma Stress: Off. Publ. Int. Soc. Traumatic Stress Studies. 1998;11:521–542. doi: 10.1023/A:1024456713321. [DOI] [PubMed] [Google Scholar]
  23. Gropman A., Summar M., Leonard J. Neurological implications of urea cycle disorders. J. Inherit. Metab. Dis. 2007;30:865–879. doi: 10.1007/s10545-007-0709-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Katherin S., John M.J. Biomarkers of suicide attempt behavior: towards a biological model of risk. Curr. Psychiatry Rep. 2017;19 doi: 10.1007/s11920-017-0781-y. [DOI] [PubMed] [Google Scholar]
  25. Katoh K., Standley D.M. MAFFT multiple sequence alignment software version 7: improvements in performance and usability. Mol. Biol. Evol. 2013;30:772–780. doi: 10.1093/molbev/mst010. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Kennedy S.H., Evans K.R., Krüger S., Mayberg H.S., Meyer J.H., McCann S., Arifuzzman A.I., Houle S., Vaccarino F.J. Changes in regional brain glucose metabolism measured with positron emission tomography after paroxetine treatment of major depression. Am. J. Psychiatr. 2001;158:899–905. doi: 10.1176/appi.ajp.158.6.899. [DOI] [PubMed] [Google Scholar]
  27. Kim C.-S., Shin G.-E., Cheong Y., Shin J.H., Shin D.-M., Chun W.Y. Experiencing social exclusion changes gut microbiota composition. Transl. Psychiatry. 2022;12:254. doi: 10.1038/s41398-022-02023-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Klindworth A., Pruesse E., Schweer T., Peplies J., Quast C., Horn M., Glöckner F.O. Evaluation of general 16S ribosomal RNA gene PCR primers for classical and next-generation sequencing-based diversity studies. Nucleic Acids Res. 2013;41 doi: 10.1093/nar/gks808. e1-e1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Lefèvre C.R., Pelletier R., Le Monnier A., Corvec S., Bille E., Potron A., Fihman V., Farfour E., Amara M., Degand N. Clinical relevance and antimicrobial susceptibility profile of the unknown human pathogen Corynebacterium aurimucosum. J. Med. Microbiol. 2021;70 doi: 10.1099/jmm.0.001334. [DOI] [PubMed] [Google Scholar]
  30. Lin H., Peddada S.D. Analysis of compositions of microbiomes with bias correction. Nat. Commun. 2020;11:3514. doi: 10.1038/s41467-020-17041-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Liu X., Teng T., Li X., Fan L., Xiang Y., Jiang Y., Du K., Zhang Y., Zhou X., Xie P. Impact of inosine on chronic unpredictable mild stress-induced depressive and anxiety-like behaviors with the alteration of gut microbiota. Front. Cell. Infect. Microbiol. 2021;11 doi: 10.3389/fcimb.2021.697640. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Lo C.I., Niang E.H.A., Sarr M., Durand G., Tall M.L., Caputo A., Raoult D., Fournier P.-E., Fenollar F. Fenollaria timonensis sp. Nov., a new bacterium isolated from healthy human fresh stool. Curr. Microbiol. 2020;77:3780–3786. doi: 10.1007/s00284-020-02151-8. [DOI] [PubMed] [Google Scholar]
  33. Mallick H., Rahnavard A., McIver L.J., Ma S., Zhang Y., Nguyen L.H., Tickle T.L., Weingart G., Ren B., Schwager E.H. Multivariable association discovery in population-scale meta-omics studies. PLoS Comput. Biol. 2021;17 doi: 10.1371/journal.pcbi.1009442. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Marin J., Bertoye P.A., Birgy A., Dziri S., Lescat M. Validation of rectal swabbing for total and aerobic gut microbiota study. Microbiol. Spectr. 2025;13:4. doi: 10.1128/spectrum.01823-24. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. McDonald D., Clemente J.C., Kuczynski J., Rideout J.R., Stombaugh J., Wendel D., Wilke A., Huse S., Hufnagle J., Meyer F., Knight R., Caporaso J.G. The biological observation matrix (BIOM) format or: how I learned to stop worrying and love the ome-ome. GigaScience. 2012;1:7. doi: 10.1186/2047-217X-1-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Miller A.H., Berk M., Bloch G., Briquet-Laugier V., Brouillon C., Cuthbert B.N., Dantzer R., Davis M.C., De Picker L.J., Drevets W.C. Advancing precision psychiatry and targeted treatments: insights from immunopsychiatry. Brain Behav. Immun. 2025 doi: 10.1016/j.bbi.2025.01.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Ohlsson L., Gustafsson A., Lavant E., Suneson K., Brundin L., Westrin Å., Ljunggren L., Lindqvist D. Leaky gut biomarkers in depression and suicidal behavior. Acta Psychiatr. Scand. 2019;139:185–193. doi: 10.1111/acps.12978. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Paton S.E., Menard C. Glutamate shall not pass: a mechanistic role for astrocytic O-GlcNAc transferase in stress and depression. J. Clin. Investig. 2023;133 doi: 10.1172/JCI168662. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Patton J.H., Stanford M.S., Barratt E.S. Factor structure of the barratt impulsiveness scale. J. Clin. Psychol. 1995;51:768–774. doi: 10.1002/1097-4679(199511)51:6<768::aid-jclp2270510607>3.0.co;2-1. [DOI] [PubMed] [Google Scholar]
  40. Posner K., Brown G.K., Stanley B., Brent D.A., Yershova K.V., Oquendo M.A., Currier G.W., Melvin G.A., Greenhill L., Shen S. The columbia–suicide severity rating scale: initial validity and internal consistency findings from three multisite studies with adolescents and adults. Am. J. Psychiatr. 2011;168:1266–1277. doi: 10.1176/appi.ajp.2011.10111704. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Price M.N., Dehal P.S., Arkin A.P. FastTree 2–approximately maximum-likelihood trees for large alignments. PLoS One. 2010;5 doi: 10.1371/journal.pone.0009490. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Pu J., Liu Y., Zhang H., Tian L., Gui S., Yu Y., Chen X., Chen Y., Yang L., Ran Y. An integrated meta-analysis of peripheral blood metabolites and biological functions in major depressive disorder. Mol. Psychiatr. 2021;26:4265–4276. doi: 10.1038/s41380-020-0645-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Refisch A., Sen Z.D., Klassert T.E., Busch A., Besteher B., Danyeli L.V., Helbing D., Schulze-Späte U., Stallmach A., Bauer M. Microbiome and immuno-metabolic dysregulation in patients with major depressive disorder with atypical clinical presentation. Neuropharmacology. 2023;235 doi: 10.1016/j.neuropharm.2023.109568. [DOI] [PubMed] [Google Scholar]
  44. Severance E.G., Yolken R.H., Eaton W.W. Autoimmune diseases, gastrointestinal disorders and the microbiome in schizophrenia: more than a gut feeling. Schizophr. Res. 2016;176:23–35. doi: 10.1016/j.schres.2014.06.027. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Śliwiński W., Gawlik-Kotelnicka O. Circulating B vitamins metabolites in depressive disorders-connections with the microbiota-gut-brain axis. Behav. Brain Res. 2024 doi: 10.1016/j.bbr.2024.115145. [DOI] [PubMed] [Google Scholar]
  46. Spearman C. 1961. The Proof and Measurement of Association Between Two Things. [DOI] [PubMed] [Google Scholar]
  47. Team R.D.C. 2010. R: a Language and Environment for Statistical Computing. (No Title) [Google Scholar]
  48. Thompson D.S., Fowler J.C., Bradshaw M.R., Frueh B.C., Weinstein B.L., Petrosino J., Hadden J.K., Madan A. Is the gut microbiota associated with suicidality? non-Significant finding among a large cohort of psychiatrically hospitalized individuals with serious mental illness. J. Affect. Disord. Rep. 2021;6 [Google Scholar]
  49. Werner J.J., Koren O., Hugenholtz P., DeSantis T.Z., Walters W.A., Caporaso J.G., Angenent L.T., Knight R., Ley R.E. Impact of training sets on classification of high-throughput bacterial 16s rRNA gene surveys. ISME J. 2012;6:94–103. doi: 10.1038/ismej.2011.82. [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Wickham H., Sievert C. springer; New York: 2009. ggplot2: Elegant Graphics for Data Analysis. [Google Scholar]
  51. Yassin A., Steiner U., Ludwig W. Corynebacterium aurimucosum sp. Nov. and emended description of Corynebacterium minutissimum collins and jones (1983) Int. J. Syst. Evol. Microbiol. 2002;52:1001–1005. doi: 10.1099/00207713-52-3-1001. [DOI] [PubMed] [Google Scholar]
  52. Zhao K., Zhou S., Shi X., Chen J., Zhang Y., Fan K., Zhang X., Wang W., Tang W. Potential metabolic monitoring indicators of suicide attempts in first episode and drug naive young patients with major depressive disorder: a cross-sectional study. BMC Psychiatry. 2020;20:1–8. doi: 10.1186/s12888-020-02791-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Zheng P., Gao H.-C., Qi Z.-G., Jia J.-M., Li F.F., Chen J.J., Wang Y., Guo J., Melgiri N., Xie P. Peripheral metabolic abnormalities of lipids and amino acids implicated in increased risk of suicidal behavior in major depressive disorder. Metabolomics. 2013;9:688–696. [Google Scholar]
  54. Zhu X., Li B., Lou P., Dai T., Chen Y., Zhuge A., Yuan Y., Li L. The relationship between the gut microbiome and neurodegenerative diseases. Neurosci. Bull. 2021;37:1510–1522. doi: 10.1007/s12264-021-00730-8. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Multimedia component 1
mmc1.pdf (49.8KB, pdf)
Multimedia component 2
mmc2.xlsx (233.9KB, xlsx)
Multimedia component 3
mmc3.xlsx (1.3MB, xlsx)
Multimedia component 4
mmc4.pdf (951.7KB, pdf)

Data Availability Statement

Raw amplicon and shotgun seq data of rectal swabs samples have been deposited at NCBI Sequence Read Archive (SRA) as [SRA: accession number] and are publicly available as of the date of publication.

Articles from Brain, Behavior, & Immunity - Health are provided here courtesy of Elsevier

RESOURCES