Multi-omics analysis reveals the host-microbe interactions on the dysbiosis of tissue microbiota in male genital lichen sclerosus- induced urethral strictures

doi:10.21203/rs.3.rs-5274378/v1

Download PDF

Research Article

Multi-omics analysis reveals the host-microbe interactions on the dysbiosis of tissue microbiota in male genital lichen sclerosus- induced urethral strictures

https://doi.org/10.21203/rs.3.rs-5274378/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Background

Male genital lichen sclerosus-induced urethral strictures (MGLSc-US) is a chronic inflammatory disease with significant microbiota dysbiosis. The impact of this dysbiosis on disease progression and gene expression in MGLSc lesions still has a knowledge gap. This study investigated the influence of microbiome-host interactions on microbial dysbiosis and differential gene expression in MGLSc by analyzing the microbiome and transcriptome of foreskin tissues.

Results

Microbiome and transcriptome sequencing were conducted using prepuce samples from MGLSc patients (n = 27) and controls (n = 17). In MGLSc patients, unclassified Muribaculaceae and Escherichia coli. were enriched, while Finegoldia magna, Prevotella timonensis, Bacillus pumilus, and Peptoniphilus harei etc., were reduced. No differences in alpha diversity were observed, but beta diversities were significantly different (p < 0.05) compared to controls. The microbial community exhibited a decrease in Gram-positive bacteria (p < 0.05). The top 15 GO pathways of differentially expressed genes (DEGs) were associated with immune activation, inflammatory response, and innate immunity and pathogen response. Single sample gene set enrichment analysis revealed MGLSc lesions enriched immune cells, including activated CD4 T cells (p < 0.0001), effector memory CD4 T cells (p < 0.0001), activated B cells (p < 0.001), and Type 2 Helper T cells (p < 0.001). DEGs related to pathogen recognition, such as TLR1, TLR2, TLR6, and HLA-DOB, were upregulated (p < 0.05). Clinical correlation analysis indicated that Escherichia coli negatively correlated with age (p < 0.01). The unclassified Muribaculaceae was positively correlated with total cholesterol levels (p < 0.001). The abundance of Peptoniphilus sp. S470 was positively correlated with body mass index (p < 0.05). The abundance of Bacillus pumilus was positively correlated with triglycerides levels (p < 0.05). Enterococcus faecalis (p < 0.05) and Staphylococcus epidermidis (p < 0.01) were negatively correlated with stricture grade.

Conclusions

This study, based on microbiota and transcriptomics, was the first to discover a decrease in Gram-positive bacteria in the lesional prepuce tissue of MGLSc patients. In the MGLSc population, dysbiosis was associated with pathogen-response immune pathways. Additionally, patient clinical characteristics were correlated with the abundance of differential microbe.

Male genital lichen sclerosus

Urethral stricture

Multi-omics

Transcriptome

Intratissue microbiome

Male genital lichen sclerosus (MGLSc) is a chronic inflammatory scarring skin disease(1). Notably, approximately 30% of patients with MGLSc will develop urethral stricture(2). It has also become the major cause of pan-urethral strictures. MGLSc-US (MGLSc-induced urethral stricture, MGLSc-US) is more invasive than iatrogenic and traumatic urethral strictures. Its recurrence rate after urethroplasty can be as high as 70%(2). Until now, the exact molecular mechanism of the pathogenesis and involvement of urethral stricture in MGLSc is still unclear. Although not entirely known, a dysregulated inflammatory response is presumed to be one of the underlying mechanisms in MGLSc and MGLSc-US.

Increasing evidence suggests that the local microbiota plays a crucial role in the host immune response. Previous studies, including results from our laboratory, have reported that patients with MGLSc and MGLSc-US exhibit a distinct microbial composition on the skin surface and in the urine compared to healthy individuals and patients with non-MGLSc urethral strictures.(3–6). However, previous studies employed balanopreputial sac swabs and urine sampling for microbiota characterization, which precluded analysis of local interactions and their immediate impact on host prepuce expression signatures. To dissect the occurrence and progression mechanisms of MGLSc, more attention needs to be paid to local effects, such as host interactions between microbes within the tissue. Modulation of host gene expression by tissue microbiota or effects of gene expression on microbial fitness may expose mechanisms that contribute to MGLSc pathogenesis—a knowledge that could be utilized to explore novel therapeutic targets.

Here, we are first to use transcriptome and 16S rRNA sequencing approaches to identify the core microbiota in MGLSc prepuce lesions and investigate prepuce tissue host-microbe interactions, which could broaden our understanding of the local inflammatory responses and potentially lead to the development of microbiome-directed therapies.

Patient selection and sample collection

Ethical approval for this cohort study was obtained through the ethics committee of Shanghai Sixth People’s Hospital (2021-KY-018(K), SH6THHOSP-KY-2022-201) and all study participants signed the informed consent prior to sample collection. Patients recruited for this study included adult males clinically diagnosed with MGLSc-induced urethral stricture (MGLSc group) and adult males with redundant prepuce (control group) from 2021 to 2023. Exclusion criteria were: I. acute urinary tract infection; II. acute genital skin infection; III. history of sexually transmitted diseases (gonorrhea, syphilis, HIV); IV. congenital genital anomalies not caused by MGLSc; V. genital pigmentary disorders not caused by MGLSc. Patients with phimosis were excluded from the control group. (Detailed clinical information can be found in Table 1 and Table S1).

The obtained foreskin tissues were divided using sterile scalpel blades, with each tissue piece approximately 2mm × 3mm in size. A portion of the excised foreskin tissue was retained for pathological examination. The divided foreskin tissues were placed in sterile, enzyme-free cryogenic vials, properly labeled, and transported in an ice box at 4 ℃. All tissue samples were stored in liquid nitrogen tanks until they were used for DNA extraction and sequencing. Clinical information was collected from patients through medical history and questionnaires.

2.2 Bacteria marker stain, and location in prepuce tissue

All MGLSc patients had their prepuce tissue collected for haematoxylin-eosin (H&E). The degree of inflammatory cell infiltration in MGLSc prepuce tissue sections will be evaluated and graded by a professional pathologist. Based on the degree and distribution patterns of inflammatory cell infiltration in tissue sections from MGLSc patients, the sections were classified into four types: "scattered", "patchy or nodular", "band-like", and "lichenoid". Scattered density showed lymphocytes sparsely infiltrating within the sclerosus dermis. Patchy and nodular density showed lymphocytes patchily infiltrating and nodular infiltrating within sclerosus dermis. Band-like density showed lymphocytes exhibiting the contiguous band infiltration beneath the dermis sclerosus. Lichenoid density showed lymphocytes present in the basal layer of the epidermis and scattered within the epithelium(7, 8). We conducted immunohistochemistry (IHC) using antibodies against bacterial lipopolysaccharides (LPS) (HM6011-20UG, 1:100) and lipoteichoic acid (LTA) (HM2048-200UG, 1:100) to detect Gram-negative and Gram-positive bacteria in prepuce tissues, respectively.

2.3 16s rRNA extraction, sequence and microbiome diversity analysis

Total DNA was extracted from frozen prepuce samples. The specific primers employed were as follows: Forward primer 27F (AGRGTTTGATYNTGGCTCAG) and Reverse primer 1492R (TASGGHTACCTTGTTASGACTT). Following polymerase chain reaction (PCR) amplification, we conducted a quality assessment of the sequencing library. The classification of microbes was obtained based on specific features. Characterization of the alpha and beta diversity of microbial communities and their composition at the species level was conducted. Linear Discriminant Analysis (LDA) Effect Size (LEfSe) analysis was utilized to identify statistically significant biomarkers between the groups, with a threshold LDA score greater than 4 set for biomarker selection. Operational Taxonomic Units (OTU) sequences were compared against the Greengenes database to perform Bugbase phenotype prediction analysis(9).

2.4 Transcriptome sequencing, immune infiltration and data analysis

Total RNA was extracted from frozen prepuce tissues using TRIzol Reagent (Life Technologies, California, USA). The mRNA libraries were prepared by poly-A selection, cDNA synthesis, fragmentation, adapter ligation, and PCR amplification through the Hieff NGS Ultima Dual-mode mRNA Library Prep Kit (Yeasen Biotechnology, Shanghai, China), then were sequenced on the Illumina NovaSeq 6000 platform. Raw reads underwent quality control with FastQC and trimming with Trimmomatic. HISAT2 aligned reads to the GRCh38 reference genome, and FeatureCounts quantified gene expression through the bioinformatics pipeline tool BMKCloud. DESeq2 identified differentially expressed genes (adjusted p-value < 0.05, |log2 fold change| >1). Differential gene functional annotation and enrichment analyses were performed using the Metascape online analysis tool. The top 15 enriched GO pathways were categorized into different functional modules based on their GO functional descriptions. Based on gene expression results, we conducted deconvolution analyses of immune cell infiltration through the single sample gene set enrichment analysis (ssGSEA) to calculate the enrichment score of 28 distinct immune cell types to compare cellular heterogeneity between the MGLSc and control groups. Furthermore, to reveal the particular patterns of two groups, the immune cells infiltration score in two groups were evaluated using the ggpubr package of R software, and statistical significance was assessed using the Wilcoxon test.

2.5 Correlation Analysis between microbiome and host gene expression

To explore the association between microbial dysbiosis and the occurrence of MGLSc, a correlation analysis was performed using differentially expressed gene expression and microbial abundance data from MGLSc and control groups(10). Patients who underwent both microbiome and transcriptome sequencing were included in the analysis cohort. All calculations were done in R 4.4.1 using “dplyr” for data cleaning, “reshape2” for data formatting, and “Hmisc” for Spearman correlation and p-values.

Then, we analyzed the correlation between microbial abundance and immune cell abundance within each group to observe whether microbe-host interactions vary across different populations(10). We also utilized the “cor.test” function to identify genes significantly correlated with differential microbes in MGLSc and controls, respectively (p < 0.05). Genes belonging to the “immune activation,” “inflammatory response,” and “innate immune and pathogen response” modules were then subjected to correlation analyses with differential microbes in both MGLSc and controls.

2.6 Correlation Analysis between Microbiome and Clinical Data of MGLSc cohort

MGLSc is an inflammatory condition, and systemic conditions commonly associated with chronic inflammatory diseases are expected to occur at higher rates in the MGLSc cohort(11). Correlation analysis was performed between microbes and risk factors related to MGLSc, including age, body mass index (BMI), blood pressure, diabetes, hyperlipidemia, lymphocyte distribution patterns, stenosis grade, and stenosis score in the MGLSc urethral stricture group. The evaluation of the stenosis grade and stenosis score of MGLSc was conducted based on the TURNS classification system for urethral strictures (Supplemental materials) (12). Based on the types of clinical data, the association between binary variables and microbes was represented by calculating the point-biserial correlation coefficient. For continuous clinical variables, the association with microbes was represented by calculating the Spearman correlation coefficient. Each clinical data point was analyzed separately for its association with the microbes. All analyses were based on data collected from the enrolled participants through medical history and the questionnaires.

Composition and diversity of MGLSc prepuce microbiota and normal prepuce microbiota

According to the relative abundance analysis of taxa at species level in samples, MGLSc and normal prepuce had unique tissue microbiome compositions. The top nine most abundant microbes in each group's microbiota were entirely different. These microbes formed the main part of each group's microbiota, with the top nine microbes accounting for 37.77% of the microbiota in the control group and 41.99% in the MGLSc group. Microbes with high abundance in the control group showed extremely low abundance in the MGLSc group, while those with high abundance in the MGLSc group were extremely low in abundance in the control group. (Fig. 1A-B, Fig. S1). In addition, alpha diversity indices including Simpson (p = 0.079), Chao1 (p = 0.14), Shannon (p = 0.2), and ACE (p = 0.26), did not show significant differences between the MGLSc and control groups (Fig. 2A). Notably, beta diversity analysis using Bray-Curtis (R = 0.244, p = 0.001) and Weighted UniFrac (R = 0.128, p = 0.008) based PCoA revealed significant differences between the MGLSc and normal prepuce samples (Fig. 2B). These results indicated that the tissue microbiome dysbiosis significantly existed in MGLSc prepuce lesions compared to normal prepuce.

Furthermore, the LEfSe analysis identified species-level dominant and differential microbes in both normal prepuce and MGLSc prepuce. In the MGLSc group, the abundance of unclassified Muribaculaceae and Escherichia coli significantly increased (p < 0.05, ∣LDA∣ > 4) (Fig. S2). In contrast, the abundance of Finegoldia magna, Prevotella timonensis, Campylobacter ureolyticus, Bacillus pumilus, Enterococcus faecalis, Staphylococcus epidermidis, Bacteroides coagulans, Peptostreptococcaceae bacterium oral taxon 929, Peptoniphilus harei, and Peptoniphilus sp. S470 significantly decreased in the MGLSc group (p < 0.05, ∣LDA∣ > 4) (Fig. 2C, Fig. S2).

Bugbase phenotype prediction analysis and microbiota distribution patterns of Microbes in MGLSc prepuce and normal prepuce

In the microbial functional prediction analysis, the microbial communities colonising the prepuce tissue of MGLSc patients showed a significantly lower proportion of Gram-positive bacteria and exhibited a greater potential pathogenic capability. The microbes in the MGLSc group exhibit greater biofilm formation ability, pathogenicity, and environmental tolerance (Fig. 3A). In the microbiomes of the two groups, among the top 30 most abundant microbes on average, Gram-positive bacteria accounted for approximately 39.4% of the microbiome in the control group and about 15.4% in the MGLSc group. In contrast, Gram-negative bacteria accounted for about 13.6% in the control group and approximately 30.1% in the MGLSc group (Fig. 3B).

Based on the results of microbial functional prediction analysis, to further explore the spatial distribution pattern of microbes in MGLSc and normal prepuce, we performed MGLSc prepuce histological examination and immunohistochemical staining of Gram-positive bacteria and Gram-negative bacteria in the two groups. Among the 27 MGLSc patients, 26 underwent an examination of inflammatory cell distribution. Based on the distribution patterns of inflammatory cells in the lesional prepuce, the samples were classified into four categories: scattered (n = 4, 15.38%), patchy or nodular (n = 11, 42.31%), band-like (n = 5, 19.23%), and lichenoid (n = 5, 19.23%) (Fig. 4A, Table 1). Patchy or nodular was the main distribution pattern in MGLSc prepuce.

In addition, although all enrolled participants were in the non-acute infection phase, we still could find Gram-negative and Gram-positive bacteria were distributed in the epidermal and dermal layer of both groups’ prepuce tissue. In normal prepuce, Gram-negative and Gram-positive bacteria were primarily distributed in the epidermal layer (Fig. 4B). Notably, in dermis, we found most of the microbial positive areas were concentrated in the specific inflammatory cell infiltration zones of MGLSc prepuce and normal prepuce did not have inflammatory cell infiltration zones (Fig. 4B).

Transcriptome and immune infiltration analysis in MGLSc prepuce and normal prepuce

Among 23 MGLSc patients and 11 control patients, a total of 1243 differentially expressed genes were identified, with 1139 upregulated genes and 104 downregulated genes (Fig. S3). Functional annotation and enrichment of DEGs in MGLSc patients revealed that the top 20 pathways primarily focus on immune activation and regulation. The top 15 enriched GO pathways were categorized into three functional modules: "immune activation" (‘GO:0001775’, ‘GO: 0050778’, etc.), "inflammatory response" (‘GO:0006954’, ‘GO:0002683’, etc.), and "innate immune and pathogen response" (‘GO:0009617’, ‘GO:0045087’, etc.) (Fig. 5A). In the MGLSc population, the expression levels of TLR1, TLR2, and TLR6, which are associated with pathogen recognition, were significantly elevated (Fig. 5B). Additionally, the expression levels of genes from the matrix metalloproteinase (MMP) family, which are involved in extracellular matrix degradation and inflammation, also showed a significant up-regulation, including MMP7, MMP9, MMP11, MMP12 and MMP25 (Fig. S4). The upregulated expression of these genes may lead to skin barrier damage that could induce pathogen immune response. The characteristics of immune dysregulation in MGLSc were explored through ssGSEA. The immune cell infiltration analysis showed a significant increase in innate (e.g., macrophages, dendritic cells) and adaptive immune cells (e.g., activated CD4, CD8, type 1 T cells, type 2 T cells) in the MGLSc group compared to the control group, indicating elevated immune activity across various immune processes (Fig. 5B-C).

Transcriptome and microbiome integration analysis in MGLSc prepuce and normal prepuce

To explore the potential relationship between the formation of specific microbial composition and the occurrence and development of MGLSc inflammation, we performed the correlation analysis by using immune cell infiltration score and microbial abundance data from MGLSc and control groups(10). We found dominant bacteria in normal prepuce, such as Finegoldia magna, Prevotella timonensis, etc., which showed significant negative correlation with the immune cell infiltration score. The MGLSc prepuce dominant bacteria, such as unclassified Muribaculaceae and Escherichia coli, showed positive correlations with immune cells. Notably, Escherichia coli was significantly positively correlated with Neutrophil, Mast cell, immature dendritic cells, and activated dendritic cells. Among all bacteria, Finegoldia magna showed the strongest correlation with immune cell activated in the MGLSc prepuce, except for neutrophils (Fig. 5D).

Then we analyzed the correlation between microbial abundance and immune cell abundance in each group. In the control group, dominant bacteria in control cohort (like Finegoldia magna) were positively correlated with plasmacytoid dendritic cells (pDCs) and Th2 cell infiltration. These microbes were negatively correlated with monocytes, natural killer (NK) cells and natural killer T (NKT) cells infiltration. In addition, these bacteria significantly positivly correlated with pathways involved in “immune activation”, “inflammatory response”, “innate immune and pathogen response”. Notably, three functional module genes were positively correlated with these microbial abundance. These microbes positive correlated with GBP5 and CCL8, TLR1, TLR2, and TLR6 in the innate immune and pathogen response module (Fig. 6A).

Conversely, these microbes had low abundance in MGLSc prepuce and were not only negatively correlated with monocytes, NK cells and NKT cells infiltration but also negatively correlated with Th1 cells, activated CD8 T cells, central memory CD8 T cells, effector memory CD8 T cells and so forth. Moreover, these microbes exhibited different trends in their correlation with three pathway modules and genes compared to those in normal prepuce, which suggested that most microbes showed a decrease in abundance as the expression of these genes increased (Fig. 6. A-B).

Differential microbe abundance and clinical phenotypes correlation analysis in MGLSc-US patietns

To further explore the potential relationship of differential microbes in the MGLSc prepuce with clinical phenotypes in MGLS-US patients, in the high abundance of distinct bacteria in the MGLSc prepuce, we found that Escherichia coli was negatively correlated with the patient's age (R = -0.541, p = 0.004) and the lymphocyte infiltration pattern (R = -0.441, p = 0.024). The higher the abundance of Escherichia coli in the prepuce tissue of MGLSc patients, the sparser the distribution of lymphocytes in the tissue. The unclassified Muribaculaceae was positively correlated with total cholesterol (CHOL) levels (R = 0.793, p = 0.0007). In the low abundance of distinct bacteria in the MGLSc prepuce, the abundance of Peptoniphilus sp. S470 was positively correlated with BMI (R = 0.433, p = 0.024). The abundance of Bacillus pumilus was positively correlated with triglycerides levels (R = 0.652, p = 0.012). Enterococcus faecalis and Staphylococcus epidermidis were negatively correlated with Stenosis Grading (Fig. 7). The fewer Enterococcus faecalis and Staphylococcus epidermidis present, the longer the stricture segments.

The inflammatory response associated with MGLSc was one of the critical mechanisms underlying its pathogenesis, leading to fibrosis and structural remodeling of the genitalia, such as phimosis, fusion of the preputial corona, and urethral strictures. However, the exact mechanisms remained unclear. Researchers suggested that, compared to vulvar lichen sclerosus, the etiology of MGLSc may be more closely related to the long-term stimulation of urine within the preputial cavity and the resulting microbial dysbiosis(1). Microbial communities play a crucial role in modulating host inflammatory responses. Previous studies have demonstrated microbial dysbiosis on urine and the prepuce swab in MGLSc patients. However, the nature of the interactions between these microbes and the host remains poorly understood. In this study, we first described the unique tissue microbiome composition and the host-microbe interactions within the MGLSc prepuce tissue.

Through microbiome analysis, MGLSc prepuce showed unique microbial community composition compared to normal skin microbiome with a lower proportion of Gram-positive bacteria (Fig. 3A-B). In microbiome diversity, we found alpha diversities did not show a significant difference, but beta diversities showed significant difference (Fig. 2A-B). These results were similar to the previous research in urine and balanopreputial sac swab(3). Additionally, in MGLSc prepuce lesions, we found the abundance of Finegoldia magna, Prevotella timonensis, and Staphylococcus epidermidis, etc that normal prepuce dominant microbes, were decreased (Fig. 2C). Similarly, in the balanopreputial swab, Finegoldia magna was also found to have a higher abundance in the control group compared to the MGLSc group (25). Finegoldia magna was reported as a common microbe on skin, and could cause infections when the skin was damaged or the host's immunity was compromised(13). Prevotella timonensis, though commonly detected in healthy skin, could cause bacterial vaginosis under certain conditions(14). Staphylococcus epidermidis was considered related to skin barrier homeostasis. These microbes were typically skin surface colonizers. Notably, the high-abundance distinct microbes were unclassified Muribaculaceae and Escherichia coli. Escherichia coli was a common gut microbe and commonly associated with UTI (urinary tract infection, UTI) (15). Muribaculaceae was also a common gut microbe and involved in producing short-chain fatty acids (SCFAs) to maintain gut barrier homeostasis(16). Interestingly, these microbes were closely associated with the urinary tract(17, 18) and our cohort of MGLSc patients had a history of urethral stricture and genital structural remodeling, increasing the likelihood of these microbe colonising prepuce.

At the host transcriptomic level, the upregulated differentially expressed genes, after GO functional annotation, were primarily associated with biological functions related to immune activation and regulation. The top 15 dysregulated GO pathways could be categorized into three main functional modules: immune activation, inflammatory response, and innate immune and pathogen response (Fig. 5A). Immune infiltration analysis results further indicated that excessive immune activation existed in MGLSc prepuce lesions (Fig. 5C). Notably, although the results of ssGSEA showed Th1 cells (p < 0.05) and CD8 T cells (p < 0.01) were present in high scores, the most significant different cells between the two groups were activated B cells (p < 0.0001), activated CD4 T cells (p < 0.0001), effector memory CD4 T cells (p < 0.0001), and Th2 cells (p < 0.001). The significant increase in activated B cells and Th2 cells in the MGLSc group reflects active humoral immunity, which typically occurs in the early stages of microbial infection and is crucial for clearing exogenous microbes. The notable difference in effector memory CD4 T cells between the two groups, generally seen in chronic or recurrent infectious diseases, indicates persistent antigen exposure in the disease group, leading to long-term retention and immune surveillance of memory T cells(19).

Corresponding to these results, in the MGLSc group, genes in the Toll-like receptor (TLR) family that primarily recognised Gram-positive bacterial markers, such as TLR1, TLR2, and TLR6(20), were highly expressed in the lesional tissues (Fig. 5B), while the expression of TLR4, which primarily recognized Gram-negative bacterial markers(21), showed no significant difference between the two groups (Fig. 5B). These results corresponded to the discovery in the microbial sequencing data, where the MGLSc microbial phenotype showed a significant decrease in Gram-negative bacteria. This gene expression of the MGLSc prepuce could be related to the formation of its specific microbial community.

In addition, we also found that MMP genes, including MMP3, MMP7, MMP9, and MMP25, were highly expressed in the lesional tissues of the MGLSc group (Fig.S4). These MMPs are key members of the MMP family and can degrade the extracellular matrix (ECM), suggesting that the skin barrier may be compromised (22). The high expression of these genes could cause compromised skin barrier function through damaged basement membrane. Meanwhile, as the ECM was degraded by MMPs, it released a large number of cytokines, further contributing to the inflammatory response in MGLSc tissues and forming the inflammatory loop. Based on these findings, the unique microbiome composition in MGLSc prepuce lesions might be caused by persistent inflammation and the resulting disruption of the skin barrier. These factors created conditions that facilitated common skin microbial clearance and the colonization of new dominant microbiota in MGLSc prepuce lesions. Related studies suggested that micro-incontinence and the residual urine on the skin were the potential causes of MGLSc. The residual urine could damage the skin barrier and introduce new microbes(23). A damaged skin barrier could provide a gateway for microbes such as Muribaculaceae and Escherichia coli, which typically do not colonize the external genital skin. After the clearance of common prepuce-colonizing bacteria such as Finegoldia magna, Prevotella timonensis, and Staphylococcus epidermidis, non-conventional prepuce-colonizing bacteria like unclassified Muribaculaceae and Escherichia coli may rapidly colonise and become the dominant species in the new microbial community, thriving in the absence of competition from common prepuce-colonising microorganisms. The anatomical proximity between the external genitalia and the anus may explain why the dominant microorganisms ultimately colonizing the prepuce tissue in MGLSc are common gut microbes.

Furthermore, through host-microbiota interactions analysis between the abundance of distinct microbes and immune cell infiltration score of all samples, we found the low-abundance dominant bacteria of MGLSc prepuce were common colonized microbes and exhibited a significant negative correlation with immune cell infiltration score. Meanwhile, the investigation of host-microbiota interactions within each group revealed that the opposite trend between microbes and immune cells and immune response-related genes (Fig. 6). In the control group, an increase in the abundance of microbes, especially Finegoldia magna, often coincided with the activation of host immune response pathways and innate immune cells (DC cells, macrophage). Notably, we found GBP5 and CCL8 were significantly positively correlated with the abundance of Finegoldia magna (Fig. 5B and Fig. 6A-B). GBP5 and CCL8 were associated with macrophage polarization and the activation of the NF-κB signaling pathway(24, 25). This finding might suggest that when the host skin barrier was intact, Finegoldia magna helped maintain homeostasis by enhancing host cytokine regulation to prevent pathogen invasion. In contrast, the dominant microbes of the control group in MGLSc patients' prepuce tissues exhibited a negative correlation with innate immune response module genes (GBP5 and CCL8, TLR1, TLR2, and TLR6) and immune cells, which implied that as immune response gene expression increased and immune cell infiltrated, the abundance of these bacteria significantly decreased. Based on these findings, we further validated our hypothesis that when the skin barrier was damaged, the invasion of mostly dominate microbes of normal prepuce could activate inflammatory responses through activating innate immune and pathogen response pathways and genes (TLR signaling pathway, NFKB signaling pathway).

During this process, the original dominant microbes no longer thrived in inflammation environment, the MGLSc microbiome composition was unstable and those with greater environmental stress tolerance like unclassfied Muribaculaceae and Escherichia coli, become the new dominant microbes (Fig. S1, Fig. 2C and Fig. 3A). Notably, because of the inflammation loop, after being cleared from inflammatory sites, they could continuously migrate from healthier skin areas to the lesions, acting as a trigger for chronic inflammation and resulting in prolonged inflammatory responses in MGLSc patients. The inflammatory loop could further compromise the integrity of the epithelial barrier to enlarge the impaired range through the increased degree of dermis sclerosus (26). Within this process, normal prepuce microbes acted as a persistent antigen to prompt effector memory T cells, maintaining the inflammatory loop of MGLSc and promoting the development of MGLSc.

Additionally, we further explored the potential interaction of microbes with more clinical phenotypes including demographic, systematic disease, histology and urethral stricture characteristics. Previous researches had reported some risk factors such as hypertension, hyperglycemia, and hyperlipidemia, which made patients more susceptible to developing MGLSc. We found that high abundance of Escherichia coli was primarily concentrated in younger MGLSc patients with shorter history of urethral stricture and correlated with less immune cell infiltration, like scatter sparse in sclerosus dermis. Escherichia coli is the most common microbe in urinary tract infections. The residual urine or high-pressure urination caused by phimosis and urethral stricture could enable it to migrate and colonise easier in prepuce with the development of MGLSc. Another microbe that significantly increased in the MGLSc community, Muribaculaceae was significantly positively correlated with CHOL levels. This microbe might involve in the lipid metabolism of MGLSc patients. Unclassified Muribaculaceae was reported to produce SCFA and the excessive SCFA on the skin could promote the expression of the HIF-1a gene, thereby contributing to skin inflammation(27). Interestingly, the decerased abundance of Enterococcus faecalis and Staphylococcus epidermidis was correlated with the longer urethral stricture segment. Staphylococcus epidermidis, as a common skin colonizer, is generally considered to play a crucial role in maintaining skin barrier function. The reduction in this microorganism might reflect a greater degree of local skin barrier damage along with an active inflammatory response. Due to the similar pathological features of MGLSc in urethral and prepuce lesions, the inflammatory response, associated with skin barrier dysfunction, might have contributed to a more severe urethral stricture grade in the patients.This phenomenon needs cautious interpretation and further explore.

Although we preliminary proposed the new insight of the MGLSc prepuce tissue microbiome features and the interaction with host, the sample size needs to be expanded to improve the reliability of our findings. Additionally, the lack of appropriate in vitro and in vivo models limited our ability to further explore and validate the host-microbe interactions in MGLSc. Future research should focus on validating the mechanisms of host-microbe interactions.

This study was the first to demonstrate that the prepuce in MGLSc had a lower proportion of Gram-positive bacteria. Our results suggested the microbial composition in MGLSc patients might result from the selective interactions between the host and the microbiota. Microbe-host interaction might through activation of pathogen immune response-related pathways including TLR and NF-κB pathways, and immune cell infiltration.

MGLSc: Male Genital Lichen Sclerosis;

LSP: Lichen Sclerosis prepuce;

nLSP: non-Lichen Sclerosis prepuce;

DEGs: differentially expressed genes;

IHC: immunohistochemistry;

LPS: lipopolysaccharides;

LTA: lipoteichoic acid;

LEfSe: Linear Discriminant Analysis Effect Size;

LDA: Linear Discriminant Analysis;

OTU, Operational Taxonomic Units;

PCR, polymerase chain reaction;

ssGSEA: single sample gene set enrichment analysis;

MMP: Matrix metalloproteinase;

UTI: urinary tract infection;

TLR: Toll-like receptor;

ECM: extracellular matrix.

Supplementary Information

Additional file 1: Table S1. Detailed clinical information for Control cohort.

Additional file 2: Supplementary Materials

Immunohistochemistry assays

Evaluation methods of stricture grade and stricture score.

Fig. S1. (A) Microbial species abundance bar chart in the prepuce of the MGLSc group. (B) Microbial abundance bar chart in the prepuce of the control group. Fig. S2. LDA scores of differential microbes between the MGLSc group and the control group. Fig. S3. Volcano plot of up-regulated and down-regulated DEGs. Fig. S4. Differential expression of MMP family genes in the MGLSc and control cohorts.

Acknowledgements

We thank the patients who participated in this study.

Author contributions

L.S., X.X. and Z.Y. conceived the study. Z.Y. and X.X. designed the computational analysis. Z.Y., Z.W. and X.X. performed 16srRNA-seq and bulk RNA-seq analysis. Z.Y. and X.X. wrote the manuscript. Z.W., L.S., X.X. revised manuscripts. J.T. is in charge of the pathology examination of the tissue sections. Z.Y., J.T., R.Z. and X.X. are responsible for immunohistochemistry. Z.Y., X.X., G.M., J.L., and R.Z. collected and analyzed clinical data for all patient samples. Z.Y., G.M. and R.Z. cared for patients analyzed in the study and handled the patient samples.

Funding

This work was supported by the National Natural Science Foundation of China (No. 81974085), the Discipline Leader of Shanghai Municipal Health (No.2022XD015) and the Summit Plateau Program, Research Physician Program, Shanghai Jiao Tong University School of Medicine (20240817).

Data availability

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Ethics approval and consent to participate:

This study was approved by the Shanghai Sixth People’s Hospital Ethics Committee (2021-KY-018(K), SH6THHOSP-KY-2022-201).

Consent for publication:

The MGLSc prepuce tissues and unpaired control redundant prepuce samples were obtained from Shanghai Sixth People’s Hospital. All patients signed an informed consent to participate in the study. The study was reviewed and approved by the Shanghai Sixth People’s Hospital ethics committee (2021-KY-018(K), SH6THHOSP-KY-2022-201).

Conflict of interest statement

The authors declare that they have no competing interests.

C. B. Bunker, T. N. Shim, Male genital lichen sclerosus. Indian J Dermatol 60, 111-117 (2015).
A. Levy et al., Insights into the Pathophysiology of Urethral Stricture Disease due to Lichen Sclerosus: Comparison of Pathological Markers in Lichen Sclerosus Induced Strictures vs Nonlichen Sclerosus Induced Strictures. J Urol 201, 1158-1163 (2019).
R. E. Watchorn et al., Balanopreputial sac and urine microbiota in patients with male genital lichen sclerosus. Int J Dermatol 60, 201-207 (2021).
A. J. Cohen et al., Synchronous genitourinary lichen sclerosus signals a distinct urinary microbiome profile in men with urethral stricture disease. World journal of urology 39, 605-611 (2021).
M. L. Jamil et al., Urinary microbiome differences between lichen sclerosus induced and non-lichen sclerosus induced urethral stricture disease. World J Urol 41, 2495-2501 (2023).
Z. Wang et al., MP03-08 STUDY ON THE INFLUENCE OF URINE AND SKIN MICROBIOTA IN THE PATHOGENESIS OF MALE GENITAL LICHEN SCLEROSUS. Journal of Urology 206, e24 (2021).
A. Piris et al., Topographical Evaluation of Penile Lichen Sclerosus Reveals a Lymphocytic Depleted Variant, Preferentially Associated With Neoplasia: A Report of 200 Cases. Int J Surg Pathol 28, 468-476 (2020).
H. J. Pyle, J. C. Evans, T. W. Vandergriff, M. M. Mauskar, Vulvar Lichen Sclerosus Clinical Severity Scales and Histopathologic Correlation: A Case Series. Am J Dermatopathol 45, 588-592 (2023).
T. Ward et al., BugBase predicts organism-level microbiome phenotypes. BioRxiv, 133462 (2017).
S. Hu et al., Mucosal host-microbe interactions associate with clinical phenotypes in inflammatory bowel disease. Nat Commun 15, 1470 (2024).
B. A. Erickson et al., Understanding the Relationship between Chronic Systemic Disease and Lichen Sclerosus Urethral Strictures. J Urol 195, 363-368 (2016).
B. A. Erickson et al., Development and Validation of A Male Anterior Urethral Stricture Classification System. Urology 143, 241-247 (2020).
F. Walser et al., Antimicrobial susceptibility testing is crucial when treating Finegoldia magna infections. Eur J Clin Microbiol Infect Dis, (2022).
N. H. van Teijlingen et al., Vaginal dysbiosis associated-bacteria Megasphaera elsdenii and Prevotella timonensis induce immune activation via dendritic cells. J Reprod Immunol 138, 103085 (2020).
P. Katongole, F. Nalubega, N. C. Florence, B. Asiimwe, I. Andia, Biofilm formation, antimicrobial susceptibility and virulence genes of Uropathogenic Escherichia coli isolated from clinical isolates in Uganda. BMC Infect Dis 20, 453 (2020).
I. Lagkouvardos et al., Sequence and cultivation study of Muribaculaceae reveals novel species, host preference, and functional potential of this yet undescribed family. Microbiome 7, 28 (2019).
M. F. M. Gonçalves et al., Microbiota of Urine, Glans and Prostate Biopsies in Patients with Prostate Cancer Reveals a Dysbiosis in the Genitourinary System. Cancers (Basel) 15, (2023).
Y. Wang et al., Increased abundance of bacteria of the family Muribaculaceae achieved by fecal microbiome transplantation correlates with the inhibition of kidney calcium oxalate stone deposition in experimental rats. Front Cell Infect Microbiol 13, 1145196 (2023).
L. Sun, Y. Su, A. Jiao, X. Wang, B. Zhang, T cells in health and disease. Signal Transduct Target Ther 8, 235 (2023).
N. J. Bitto et al., Staphylococcus aureus membrane vesicles contain immunostimulatory DNA, RNA and peptidoglycan that activate innate immune receptors and induce autophagy. J Extracell Vesicles 10, e12080 (2021).
A. Ciesielska, M. Matyjek, K. Kwiatkowska, TLR4 and CD14 trafficking and its influence on LPS-induced pro-inflammatory signaling. Cell Mol Life Sci 78, 1233-1261 (2021).
G. A. Cabral-Pacheco et al., The Roles of Matrix Metalloproteinases and Their Inhibitors in Human Diseases. Int J Mol Sci 21, (2020).
G. Kravvas et al., Male genital lichen sclerosus, microincontinence and occlusion: mapping the disease across the prepuce. Clin Exp Dermatol 47, 1124-1130 (2022).
T. Yamane et al., Iron accelerates Fusobacterium nucleatum-induced CCL8 expression in macrophages and is associated with colorectal cancer progression. JCI Insight 7, (2022).
L. Zhou et al., GBP5 exacerbates rosacea-like skin inflammation by skewing macrophage polarization towards M1 phenotype through the NF-κB signalling pathway. J Eur Acad Dermatol Venereol 37, 796-809 (2023).
S. Saha, D. Müller, A. G. Clark, Mechanosensory feedback loops during chronic inflammation. Front Cell Dev Biol 11, 1225677 (2023).
M. R. Shakespear et al., Histone deacetylase 7 promotes Toll-like receptor 4-dependent proinflammatory gene expression in macrophages. J Biol Chem 288, 25362-25374 (2013).

Table 1 is available in the Supplementary Files section.

No competing interests reported.

Download PDF

Editorial decision: Revision requested
24 Oct, 2024
Editor assigned by journal
24 Oct, 2024
Submission checks completed at journal
16 Oct, 2024
First submitted to journal
16 Oct, 2024

You are reading this latest preprint version

Multi-omics analysis reveals the host-microbe interactions on the dysbiosis of tissue microbiota in male genital lichen sclerosus- induced urethral strictures

Status:

Version 1

Abstract

Background

Results

Conclusions

Figures

Background

Methods and materials

Patient selection and sample collection

2.2 Bacteria marker stain, and location in prepuce tissue

2.3 16s rRNA extraction, sequence and microbiome diversity analysis

2.4 Transcriptome sequencing, immune infiltration and data analysis

2.5 Correlation Analysis between microbiome and host gene expression

2.6 Correlation Analysis between Microbiome and Clinical Data of MGLSc cohort

Results

Composition and diversity of MGLSc prepuce microbiota and normal prepuce microbiota

Transcriptome and immune infiltration analysis in MGLSc prepuce and normal prepuce

Transcriptome and microbiome integration analysis in MGLSc prepuce and normal prepuce

Differential microbe abundance and clinical phenotypes correlation analysis in MGLSc-US patietns

Discussion

Conclusion

Abbreviations

Declarations

References

Table 1

Additional Declarations

Supplementary Files

Status:

Version 1