Sex-Specific Dysbiotic Bladder Microbiome in CKD Uncovered via High-Throughput Sequencing and Culture

doi:10.21203/rs.3.rs-3407275/v1

Download PDF

Research Article

Sex-Specific Dysbiotic Bladder Microbiome in CKD Uncovered via High-Throughput Sequencing and Culture

https://doi.org/10.21203/rs.3.rs-3407275/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Background

High-throughput techniques confirm the human bladder microbiome, but its relation to chronic kidney disease (CKD) remains unexplored. Bladder bacteria may migrate to the kidneys, affecting CKD.

Methods

A cross-sectional study recruited 66 females diagnosed with CKD, 66 males diagnosed with CKD, as well as 22 HCfemales and 22 healthy control (HC) males. Transurethral catheterized urine and fecal samples were collected for 16S rRNA gene sequencing and EQUC. Urinary analysis, kidney function, and serum cytokines were evaluated.

Results

Bladder microbiomes of CKD females and males versus HC females and males differed (FDR < 0.05); however, the difference was more obvious in females. In CKD females, sequencing revealed a depletion of 5 genera, including Lactobacillus, and enrichment of 14 genera, including Escherichia/Shigella, Bifidobacterium, and several clostridial genera (FDR < 0.05), while EQUC detected increased Escherichia and decreased Lactobacillus CKDB (P < 0.05). Escherichia-Shigella was positively associated, whereas Lactobacillus was negatively associated, with CKDB-female serum creatinine (r = 0.285, P = 0.020; r=-0.337, P = 0.006, respectively). Lactobacillus was positively associated with eGFR (r = 0.251, P = 0.042). Some CKD-related serum cytokines were negatively associated with clostridial genera. In contrast, the fecal microbiomes of CKD and HC females and males did not significantly differ in bacterial diversity or composition. However, bladder and fecal microbiomes of CKD females resembled each other more than those of controls, as assessed by the Bray-Curtis Dissimilarity Index (FDR < 0.05).

Conclusions

CKD bladder microbiomes were dysbiotic, especially in females, associated with kidney damage and serum cytokine dysregulation. Increased bladder-fecal microbiome similarity in CKD females implies potential “leaky gut”.

bladder microbiome

chronic kidney disease

cytokine

fecal microbiome

sex differences

leaky gut

Chronic kidney disease (CKD) is a prevalent non-communicable condition affecting over 10% of the global population[1], characterized by progressive nephron loss leading to kidney failure. Its underlying pathology remains unclear. Research suggests microbial dysbiosis in the human gut and blood linked to CKD, highlighting the importance of host-microbiome interactions in CKD pathophysiology[1].

The human bladder hosts its unique microbial community[4–6]. Studies often investigate microbiomes in urine samples collected via transurethral catheter (bladder) or midstream void (urogenital) in relation to urinary tract disorders, e.g., urinary stone disease[7], urinary tract infection (UTI)[8], and bladder cancer[9]. A previous study explored the urogenital microbiome in CKD patients, revealing high diversity associated with estimated glomerular filtration rate (eGFR)[10]. Nevertheless, this study lacked controls and examined voided urine samples[10], which may contain microbes from both urinary and genital tracts[11, 12].

The kidneys play a major role in eliminating various cytokines. Dysregulation of proinflammatory cytokines and their inhibitors in CKD is associated with gut dysbiosis[13–16]. Notably, tumor necrosis factor (TNF-α) and interleukin-6 (IL-6) dysregulation is tied to gut dysbiosis in CKD[15, 16].

Emerging evidence suggests a gut-kidney axis[15, 16], typically centered on metabolite impacts on the kidneys by gut bacteria, rather than urinary tract microbes[18]. Given microbes’ sensitivity to their surroundings and interactions within their niches, we propose that (1) CKD patients exhibit a dysbiotic bladder microbiome, (2) the bladder microbiome distinguishes CKD patients more effectively than the fecal microbiome, and (3) the bladder microbiome correlates with serum cytokines.

Study participants

Participants were recruited between July 1st, 2019, and March 30th, 2020, from the Department of Urology and the Department of Urology at the Affiliated Wuxi Second Hospital of Nanjing Medical University. The study received approval from the ethics committee of the Affiliated Wuxi Second Hospital of Nanjing Medical University (Reference: 2018051). Informed consent was provided by all subjects prior to sample collection. Figure 1and Figure S1 displays the study design, including participant recruitment, methods, and objectives. Since our current study represents the first cross-sectional investigation of the urinary microbiome in CKD patients, we lacked predefined parameters for sample size calculation. However, we adopted a sample size of 20 subjects for both the CKD and control groups, based on a previous cross-sectional study that observed differences in bacterial communities between 20 samples from interstitial cystitis Chinese patients and controls¹. Diagnostic criteria of CKD included decreased eGFR [< 60 mL/min/1.73m² or evidence of kidney damage, such as albuminuria [(albumin excretion rate ≥ 30 mg/24 h; urinary albumin creatinine ratio [UACR] ≥ 30 mg/g], urine sediment abnormalities, electrolyte and other abnormalities due to tubular disorders, abnormalities detected by histology, and/or structural abnormalities detected by imaging or history of kidney transplantation[18]. CKD patients undergoing hemodialysis were excluded. Asymptomatic controls (presumably “healthy” controls, HC) were recruited if they had an eGFR ≥ 90 ml/min/1.73m² and were free of current renal damage, a history of liver disease, cancer, diabetes, hypertension, and hyperlipidemia. After recruitment, HC subjects were excluded if they had an indicator of urinary tract infection (UTI) by urinalysis, such as > 5 WBCs/HPF, > 3 RBCs/HPF, positive urine nitrites, positive leucocyte esterase, positive occult, positive protein, and abnormal values of diabetes, hypertension, or hyperlipidemia. Any participant with acute intercurrent disease or infection, diarrhea, kidney transplantation, pregnancy, breastfeeding was excluded. Those who used antibiotics, probiotics, or immunosuppressive drugs within 30 days before enrollment also were excluded.

Urine collection and storage

Bladder urine samples were collected by transurethral catheterization and stored as reported previously. Before catheter insertion, 5% iodophor was applied to disinfect the genital and perineal areas. Fifty mL urine were obtained and separated into three portions (i.e., 30 mL, 5 mL, and 15 mL) that were used for 16S rRNA gene sequencing, Expanded Quantitative Urine Culture (EQUC) and clinical urinary analysis, respectively. All samples were processed in a biosafety cabinet. Samples for 16S rRNA gene sequencing and EQUC were placed in sterile, DNA- and enzyme-free centrifuge tubes. Samples for 16S rRNA gene sequencing were stored at -80°C within 15 mins after collection. Samples for EQUC were placed in an insulated sterile container, immediately cooled on ice, and transferred to the research personnel within 4 hours.

Expanded Quantitative Bladder Urine Culture

Methods for EQUC were modified according to previous studies[4, 20]. 0.1 mL of catheterized bladder urine was inoculated onto BAP, chocolate and colistin, as well as nalidixic acid agars (Kemajia inc., Shanghai, China), streaked for quantification, and incubated in 5% CO₂ at 35°C for 48h. Next, 0.1 mL of urine was inoculated onto two CDC anaerobe 5% sheep blood agar plates and incubated in an anaerobic incubator with a Campy gas mixture (5% O₂, 10% CO₂, 85% N₂) at 35°C for 5 days. 1.0 mL of urine was placed in thioglycolate medium and incubated aerobically at 35°C for 5 days. If growth was visually detected in the thioglycolate medium, the medium was mixed, and 0.01 mL of urine was plated on BAP and CDC anaerobe 5% sheep blood agars for isolation and incubated aerobically and anaerobically at 35°C for 48h. Each morphologically distinct colony type was isolated on a different plate of the same media to prepare a pure culture that was used for identification, as described below.

Bioinformatics

16S rRNA gene sequencing

Bacterial DNA isolation from urine was described previously[4, 20]. The quantity and quality of the extracted DNA was measured using a NanoDrop ND-1000 spectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA) and agarose gel electrophoresis. Using the isolated DNA as the template, the bacterial 16S rRNA V3-V4 region was PCR-amplified for 32 cycles using universal primers 338F and 806R. The resultant PCR amplicons were purified with Agencourt AMPure XP Beads (Beckman Coulter, Indianapolis, IN, USA) and quantified using the PicoGreen dsDNA Assay Kit (Invitrogen, Carlsbad, CA, USA). Amplicon pools were prepared for sequencing, and the size and quantity of the amplicon library were assessed using an Agilent 2100 Bioanalyzer (Agilent, Santa Clara, MA, USA) and Library Quantification Kit for Illumina (Kapa Biosciences, Woburn, MA, USA), respectively. The libraries were sequenced using the NovaSeq 6000 platform. As urine are low-biomass samples, they were sequenced in duplicate for reproducibility. When the Bray-Curtis dissimilarity of the duplicated urinary microbiome samples was above 0.3, they were re-sequenced. If their Bray-Curtis dissimilarity remained above 0.3, they were removed from further analysis[6]. Six negative controls (2 without urine, 2 without bacterial DNA, and 2 without PCR product) also were sequenced.

Bioinformatic analysis of 16S rRNA gene sequencing data

Raw reads of the 16S rRNA gene sequences were trimmed using Cutadapt (cutadapt.readthedocs.io) to remove barcodes and adaptors. The overlapping paired-end reads were merged into a longer tag using FLASH (v1.2.8). Reads were quality trimmed using fdtrim (v0.94) from the 3’ end to remove bases with low-quality scores. Reads shorter than 100bp, with more than 5% Ns, or with an average quality below 20 were discarded. Chimeras were removed using Vsearch (v2.3.4). QIIME2 was used to process the clean reads to generate an ASV table, and taxonomy of microbes was identified using the Silva database (v138). Environmental contaminants of urine samples were removed as previously described[6]. Bacterial ASVs whose counts did not exceed five times the maximum number of counts in the negative controls were considered as contaminates and removed, as described[6]. We also manually removed bacteria that have been reported to be environmental contaminants from soil and water[6].

Identification of bacterial isolates

Identification by 16S rRNA gene sequencing

Bacteria isolated by EQUC as described above were harvested: if the colonies were large enough, a single colony was chosen; if not, then the plate was scraped. To the harvested bacteria, 200µL lysis buffer (Guhe, Hangzhou, China) was added and its DNA extracted using the method described above for 16S rRNA gene sequencing. Identification of the isolates was confirmed by partial sequencing of the 16S rRNA gene. Briefly, the 16S rRNA gene was amplified for 30 cycles by PCR with universal primers 27F and 1492R, which results in an amplicon of approximately 1500 bp. PCR was performed in a DNA thermocycler (Thermo Scientific, USA) with PCR Premix (Ensure biologicals, Shanghai, China) in 50 µL reaction mixtures. The resultant amplicons were sequenced with the ABI Prism™ Bigdye™ terminator cycle sequencing reaction kit (Applied Biosystems, USA) using the 27F and 1492R primer pair. The sequencing product was purified using the Millipore-Montage dye removal kit (Ensure Biologicals, Shanghai, China) and sequenced on an ABI 3730XL capillary DNA sequencer (Applied Biosystems, USA). For each run, negative controls containing water instead of template DNA were run in parallel. All sequencing procedures were performed by BGI (Shanghai, China). For taxonomic identification, a portion of the sequence was compared to sequences in GenBank public database (http://www.ncbi.nlm.nih.gov/BLAST/).

Identification by MALDI-TOF MS

When a bacterial isolate could not be identified by 16S rRNA gene sequencing, we used Matrix-Assisted Laser Desorption Ionization-Time-Of-Flight Mass Spectrometry (MALDI-TOF MS), as described previously. Briefly, a small portion of a single colony was overlaid with 1 µL of a saturated solution of alpha-cyano-4-hydroxycinnamic acid matrix in 50% acetonitrile and 2.5% trifluoroacetic acid (VITEK MSCHCA, bioMérieux), then air dried. Escherichia coli (ATCC 8739) was used for system calibration. Mass spectra were acquired using a VITEK MS Plus (bioMérieux, Marcy l’Etoile) and the bioMérieux VITEK MS (IVD Knowledgebase v.3.2) used to analyze a composite mass spectrum for accurate identification. A probability score between 60% and 99.9% was considered to be a high discrimination value and thus a reliable identification. A probability score < 60% was considered to be a low discrimination identification. No identification resulted when either no match was found for the composite spectra, or the analysis did not obtain enough spectral peaks[21]. Isolates that yielded no identification results were redeposited on the target plate and reanalyzed.

Other sample collection and processing

Blood samples for assessment of renal function and serum cytokines were collected on the day of urinary sample collection. The Bio-Plex™ 200 System (Bio-Rad) and Bio-Plex Pro™ Human Cytokine Screening 48-plex Panel (Bio-Rad, California, USA) were used to detect serum cytokines. An aliquot of bladder urine (described above) was used for urinalysis. The 24-hour urine volume was collected one day prior to the day of bladder microbiome sample collection. Fecal samples were collected on the day of bladder urine sample collection. The participants were told to excrete their feces into a sterile container. A nurse then put 30 mg feces into a sterile tube using sterile technique and stored them at -80°C refrigerator within 15 mins of collection. DNA isolation and sequencing of fecal samples was the same as for the urine samples, except that they were sequenced only once, as we have found that Bray-Curtis values for feces samples are typically < 0.3.

Disease profiles assessment

Participants’ demographics data were collected during a face-to-face interview. Blood pressure and fasting blood glucose were assessed on the day of urine and fecal sample collection.

Statistical analysis

Pearson’s Chi-square or Fisher’s exact tests were used for categorical variables; Student’s t test was used for normalized continuous variables and Wilcoxon rank-sum test was used for non-normal continuous variables. The P-value was adjusted for multiple comparisons using the Benjamini–Hochberg (BH) false discovery rate (FDR). The significance threshold was set at an FDR-corrected value < 0.05. Pearson correlation analysis was used to assess associations between variables, and the significance threshold was set at a P-value < 0.05.

Study participants

Figure S1 summarizes participant recruitment, methods, and study objectives. We collected urine via catheter from 230 CKD patients and 176 controls. Participants with undetectable bacterial DNA in bladder samples and controls with abnormal urinalysis were excluded. To ensure reproducibility, we sequenced bladder urine samples in duplicate, excluding those with dissimilar samples (Bray-Curtis Dissimilarity score > 0.3). From the remaining participants, we selected age/BMI-matched CKD and HC males and females. This resulted in 66 CKD male bladder samples (CKDB-male), 22 HC male bladder samples (HCB-male), 66 CKD female bladder samples (CKDB-female), and 22 HC female urine samples (HCB-female).

Compared to the HCB groups, both male and female CKDB groups had higher rates of hypertension, diabetes, occult blood, and cast in their urine (Fisher’s exact test, P < 0.05; Table 1). While both male and female CKDB groups had higher systolic blood pressure than controls, only female patients exhibited higher diastolic blood pressure (t-test, P < 0.05; Table 1). The CKDB-males and CKDB-females had lower eGFR and higher serum creatinine levels compared to HCB-males and HCB-females, respectively, while CKDB-males had higher blood urea nitrogen levels, and CKDB-females had elevated uric acid levels (t-test, P < 0.05; Table 1). Nutrient intake did not differ between CKD and HC groups (t-test, P < 0.05, Table S1), ruling out its confounding influence on downstream analyses.

Table 1

Demographics of urine donors of CKD patients and controls
Characteristics	CKDB-male (n = 66)	HCB-male (n = 22)	P value	CKDB-female (n = 66)	HCB-female (n = 22)	P value
Demographics
Age-yrs	58.55 ± 17.85	53.32 ± 12.48	0.207	64.45 ± 16.86	57.41 ± 13.19	0.078
Male sex-no.(%)	66 (100.00)	22 (100.00)	1.000	66 (100.00)	22 (100.00)	1.000
Body mass index-kg/m²	25.58 ± 6.01	23.72 ± 3.07	0.618	24.81 ± 4.14	24.56 ± 3.22	0.800
Duration of CKD-years	3.33 ± 3.79	NA	NA	2.68 ± 3.60	NA	NA
Smoking history-no.(%)			0.730			0.570
Never smoker	61 (92.42)	20 (90.91)		63 (95.45)	22 (100.00)
Former smoker	1 (1.52)	0 (0.00)		0 (0.00)	0 (0.00)
Current smoker	4 (6.06)	2 (9.09)		3 (4.55)	0 (0.00)
Drinking history-no. (%)			1.000			1.000
Never drinker	64 (96.97)	21 (95.45)		63 (95.45)	22 (100.00)
Former drinker	0 (0.00)	0 (0.00)		1 (1.52)	0 (0.00)
Current drinker	2 (3.03)	1 (2.33)		2 (3.03)	0 (0.00)
Medical history
Hypertension-no. (%)	46 (69.70)	0 (0.00)	< 0.001	39 (59.09)	0 (0.00)	< 0.001
Diabetes-no. (%)	20 (30.30)	0 (0.00)	< 0.001	16 (24.24)	0 (0.00)	0.004
Coronary artery disease-no. (%)	3 (4.55)	0 (0.00)	0.570	2 (3.03)	0 (0.00)	1.000
HbA1c-%	6.43 ± 1.44	5.93 ± 0.41	0.014	6.04 ± 1.03	5.99 ± 0.38	0.825
Fasting blood glucose-mmol/L	5.48 ± 2.57	4.85 ± 0.62	0.069	5.45 ± 2.54	5.09 ± 1.12	0.513
Systolic blood pressure-mmHg	144.97 ± 21.09	126.45 ± 6.88	< 0.001	140.00 ± 24.40	128.86 ± 14.69	0.013
Diastolic blood pressure-mmHg	82.89 ± 12.44	81.04 ± 7.52	0.408	82.03 ± 13.78	76.23 ± 6.09	0.008
Renal function
eGFR-mL/min/1.73m²	55.06 ± 31.52	114.73 ± 17.26	< 0.001	62.04 ± 35.43	118.44 ± 15.86	0.013
CKD stage-no. (%)			NA			NA
Stage 1	7 (10.61)	NA		12 (18.18)	NA
Stage 2	27 (40.91)	NA		25 (37.88)	NA
Stage 3	12 (18.18)	NA		12 (18.18)	NA
Stage 4	10 (15.15)	NA		8 (12.12)	NA
Stage 5	10 (15.15)	NA		9 (13.64)	NA
24h urinary protein-mg/dL	3099.03 ± 2255.45	NA	NA	3383.98 ± 3107.22	NA	NA
Serum creatinine-µmol/L	269.68 ± 214.47	69.30 ± 10.36	< 0.001	286.87 ± 204.42	52.68 ± 5.17	< 0.001
Blood urea nitrogen-mmol/L	12.26 ± 11.67	6.55 ± 5.30	0.003	9.03 ± 6.24	6.53 ± 6.89	0.116
Serum uric acid-umol/L	394.89 ± 134.11	344.27 ± 105.86	0.111	386.31 ± 139.24	269.16 ± 69.97	< 0.001
Urinary analysis
White blood cells > 5 WBCs/HPF-no.(%)	6 (9.09)	0 (0.00)	0.330	3 (9.09)	0 (0.00)	0.570
Red blood cells > 3 RBCs/HPF-no.(%)	7 (10.61)	0 (0.00)	0.111	3 (9.09)	0 (0.00)	0.570
Nitrites positive-no.(%)	1 (1.52)	0 (0.00)	1.000	1 (1.52)	0 (0.00)	1.000
Leucocyte esterase positive-no. (%)	9 (13.64)	0 (0.00)	0.068	11 (16.67)	0 (0.00)	0.041
Occult blood positive-no. (%)	35 (53.03)	0 (0.00)	< 0.001	34 (51.52)	0 (0.00)	< 0.001
Urine casts positive-no.(%)	35 (53.03)	0 (0.00)	< 0.001	32 (48.48)	0 (0.00)	< 0.001
Urine pH	5.92 ± 0.66	6.20 ± 0.80	0.107	6.03 ± 0.70	5.95 ± 0.58	0.602
24-hour urine volume-mL	1932.74 ± 10.72	NA	NA	1903.10 ± 568.45	NA	NA
Medicine usages
Antihypertensive agent	13 (59.09)	NA	NA	18 (27.27)	NA	NA
Glucocorticoid agent	2 (3.03)	NA	NA	4 (6.06)	NA	NA
Hypoglycemic agent	1 (1.52)	NA	NA	6 (9.09)	NA	NA
Hypolipidemic agent	0 (0.00)	NA	NA	6 (9.09)	NA	NA
Pearson Chi-square or Fisher’s exact test was used with categorical variables; Student’s t test on normalized continuous variables was used.
Abbreviations: CKDB-male: bladder urine samples provided by males with CKD; CKDB-female: bladder urine samples provided by females with CKD; HCB-male: bladder urine samples provided by healthy males; HCB-female: bladder urine samples provided by healthy females.

Bladder microbiomes are altered in CKD males and females

As sex difference has been reported to impact the bladder microbiome in asymptomatic subjects and disease groups, we wondered whether the bladder microbiome composition of females and males differed between our CKDB and HCB groups. To test this, we performed principal-coordinate analysis (PCoA) of Bray-Curtis Dissimilarity indices for all microbial taxa (ASVs) present within and among the microbiomes of each group. As expected, the bladder microbiome of HCB-males and HCB-females was significantly distinct (R² = 0.077, FDR = 0.002; Fig. 1A); the bladder microbiome of CKDB-males and CKDB-females was also distinct but less so (R² = 0.017, FDR = 0.016; Fig. 1A). Since the bladder microbiome of males and females differed, further analyses were performed separately.

When comparing CKDB to HCB, we found that the bladder microbiomes of both males and females differed (PERMANOVA, R² = 0.033, FDR = 0.001 and R² = 0.072, FDR < 0.001, respectively; Fig. 1A). Hypertension, glucocorticoid levels, hypoglycemia, and hypolipidemic agents were not confounding factors, as the bladder microbiome of the medication users and sex-age matched non-users did not differ (PERMANOVA, FDR > 0.05; Figure S2A-D). The bladder microbiome of CKDB-males and CKDB-females was significantly more diverse than those of their respective HCB controls. For females, this was reflected in both the number of ASVs (Fig. 1B) and the Chao 1 index, which estimates bacterial richness (Wilcoxon rank-sum test, FDR < 0.001; Fig. 1C); for males, this increased richness was reflected only in the number of ASVs (Fig. 1B). To assess whether bacterial richness was associated with renal function in CKDB females, we performed a Pearson correlation analysis between Chao1 and estimators of renal function (i.e., eGFR, serum creatinine, blood urea nitrogen, and serum uric acid), but no association was observed (P > 0.05).

Bladder microbiome composition is altered in CKD patients

We initially conducted a comparative analysis of ASV abundances between two groups: CKDB-male vs. HCB-male, and CKDB-female vs. HCB-female. Our analysis revealed that in the CKDB-male samples, 60 ASVs exhibited a decrease compared to the HCB-male samples, whereas 26 ASVs showed an increase (FDR < 0.05; Table S2). Similarly, in the CKDB-female samples, 56 ASVs exhibited a decrease in abundance compared to the HCB-female samples, while 71 ASVs displayed an increase (FDR < 0.05; Table S3). Subsequently, we proceeded to compare differences in bacterial phylum composition between the two groups. Proteobacteria, Firmicutes, Actinobacteriota, Bacteroidota and Fusobacteriota were the 5 most abundant phyla in the bladder microbiomes of all 4 groups (Figure S3A). After correction for multiple testing, we observed no significant difference when we compared CKDB-males and HCB-males, and CKDB-females and HCB-females (Wilcoxon rank-sum test, FDR > 0.05; Table S4 & S5). However, at the genus level, both the CKDB-males and CKDB-females were enriched for Escherichia-Shigella (9.59% and 17.71%, respectively) and Enterococcus (3.34% and 5.63% respectively; Figure S3B). Next, we compared bacterial genera that accounted for > 0.1% of the total relative abundance in males and females. After correction for multiple testing, we observed no significant difference when we compared CKDB-males and HCB-males (Wilcoxon rank-sum test, FDR > 0.05; Table S6). In contrast, 19 genera differed significantly between CKDB-females and HCB-females (Wilcoxon rank-sum test, FDR < 0.05; Table S7). Five genera were significantly less abundant in CKDB-females relative to HCB-females (Wilcoxon rank-sum test, FDR < 0.05). These included one unknown genus in the family Comamonadaceae and the genera Lactobacillus, Sphingobium, Sphingomonas, and Streptomyces (Table S7; Fig. 1D). Fourteen genera were significantly more abundant in CKDB-females, including Escherichia-Shigella. Others included Agathobacter, Anoxybacillus, Bacteroides, Bifidobacterium, Blautia, Comamonas, Diaphorobacter, [Eubacterium] coprostanoligenes group, Faecalibacterium, Sphingobacterium, Subdoligranulum, and CAG-352 in the family Ruminococcaceae (Wilcoxon rank-sum test, FDR < 0.05; Table S7; Fig. 1E). Many of these taxa are members of the class Clostridia. As Escherichia-Shigella is one of most common causes of UTI, and the participants with current UTI were excluded, we compared the abundance of Escherichia-Shigella between the age-matched CKDB-female urine samples with positive urinary leucocyte esterase and negative ones, but there was no significant difference between the positive samples and negative ones (29.82 ± 45.23 vs. 24.73 ± 39.24 (Wilcoxon rank-sum test, P = 0.781, FDR = 0.814). In addition, since Lactobacillus can impede Escherichia growth in the urinary tract, we performed Pearson correlation analysis to examine whether they were connected, but there was no significant correlation (r=-0.221, P = 0.075).

To determine whether the detected sequences came from live bacteria, we performed EQUC (Table S8), an enhanced culture method that vastly outperforms the standard clinical microbiology urine culture method. Microbes (including both bacteria and fungi) were more prevalent in HCB-females than in HCB-males (68.2 vs 13.6%, P < 0.001; Table S9), whereas microbes were only slightly more prevalent in CKDB-females than in CKDB-males (51.1 vs 42.4%, P = 0.295; Table S10). Microbes also were more prevalent in CKDB-males than in HCB-males (42.2 vs. 13.6%, P = 0.014; Table S11). In contrast, microbes were equally prevalent in CKDB-females and HCB-females (51.5 vs 68.2%, P = 0.173; Table S12).

Only two identified genera and one unidentified genus were detected in HCB-males: the fungus Cladosporium was detected in one (4.5%), the bacterium Curtobacterium was detected in another (4.5%), and an unidentified genus was detected in a third (4.5%). In contrast, multiple genera were detected in CKDB-males. The most prevalent genera were Bacillus [18.18% (11/66)], Staphylococcus [9.09% (6/66)], Kocuria [3.03% (2/66)], Micrococcus [3.03% (2/66)], and Paenibacillus [3.03% (2/66)]. The genus Bacillus was significantly more prevalent in the CKDB-males relative to HCB-males (Fisher’s exact test, P = 0.031; Fig. 2A).

While the overall prevalence of microbes did not differ between HCB-females and CKDB-females, many genera were significantly different (Fig. 2B). The most prevalent in HCB-females were Bacillus [22.73% (5/22)], Lactobacillus [22.73% (5/22)], and Herbaspirillum [13.64% (3/22)], whereas the most prevalent in CKDB-females were Bacillus [15.15% (10/66)], Escherichia [7.58% (5/66)], and Rothia spp. [4.55% (3/66)]. The genera Lactobacillus and Herbaspirillum were significantly less prevalent in CKDB-females than in HCB-females (Fisher’s exact test, P < 0.05; Fig. 2B).

Altered genera in CKD was associated with demographics and renal function

Because the bladder microbiome composition of CKDB-females and HCB-females differed, we used Pearson correlation analysis to determine whether any of the significantly different taxa (as determined by 16S rRNA gene sequencing) were associated with any demographic category or renal function. For CKDB-females, multiple associations were observed (Fig. 3A). For example, Escherichia-Shigella exhibited a positive association with CKDB-female serum creatinine (r = 0.285, P = 0.020). Lactobacillus was negatively associated with CKDB-female age and serum creatinine (r=-0.330, P = 0.007 and r=-0.337, P = 0.006; respectively), but positively associated with CKDB-female eGFR (r = 0.251, P = 0.042). Bifidobacterium showed a positive association with CKDB-female disease duration (r = 0.335, P = 0.006). In contrast, significant associations were not observed for HCB-females (r < 0.03, P > 0.05; Fig. 3B), with one exception: Streptomyces was positively associated with blood urea nitrogen (r = 0.790, P < 0.001).

Bacterial community altered in all stages of CKD patients

For each CKD stage, we performed PCoA of Bray-Curtis Dissimilarity indices, comparing their bladder microbiomes to sex- and age-matched controls; these microbiomes differed significantly in stages 2–5 (PERMANOVA, FDR < 0.05), but not stage 1 (PERMANOVA, FDR > 0.05; Figure S4A). These differences were primarily due to changes in composition, as only CKDB microbiomes from stages 3 and 5 were significantly richer than their respective HCB (as measured by Chao 1; Wilcoxon rank-sum test, FDR < 0.05), and the Shannon indices did not differ (Figure S4B).

Altered bladder microbiome associated with altered serum cytokines in CKD patients

To determine if associations exist between the bladder microbiome and serum cytokines, we first measured serum cytokines and then compared them to the bladder microbiome. Of the 48 serum cytokines we assessed, 44 cytokines were detected. Of these, 23 cytokines differed significantly between the 66 CKD females and 22 female controls (Wilcoxon rank-sum test, FDR < 0.001). Some, such as IL-8, IL-18, IL-1β and TNF-α were increased in CKD females, whereas others, such as IL-9 and TNF-β, were decreased in CKD females (Fig. 4A). To determine whether associations existed between serum cytokines and bladder microbial composition, we performed Pearson correlation analysis. Several taxa enriched in HCB-females were negatively associated with increased serum cytokines (r > 0.03, P < 0.05; Fig. 4B). For example, the genera Sphingobium and Streptomyces were negatively associated with IL-18, IL-8, and/or IL-1β. In contrast, several taxa enriched in CKDB-females were positively associated with increased levels of serum cytokines (r > 0.03, P < 0.05; Fig. 4B). For example, Bifidobacterium was positively associated with IL − 2Rα, and MIG (r > 0.03, P < 0.05), and IL-8 was positively associated with the members of Clostridia, including Agathobacter, Blautia, Faecalibacterium, Subdoligranulum (r > 0.03, P < 0.05). Intriguingly, the genus Lactobacillus was enriched in HCB-females, but none of the correlations were statistically significant (Fig. 4B).

Alteration of microbiome not observed in feces

To assess the gut-bladder axis in CKD, we compared the fecal and bladder microbiomes of CKD-males (n = 10) to age-matched HC-males (n = 10) and CKD-females (n = 22) with age-matched HC-females (n = 22) who provided both fecal and catheterized urine samples. We observed no difference in gut microbiome compositions (PERMANOVA, FDR > 0.05; Fig. 5A) or alpha diversity (Wilcoxon rank-sum test, FDR > 0.05; Fig. 5B) between fecal samples from CKD males (CKDF-males) and fecal samples from HC males (HCF-males) or between fecal samples from CKD females (CKDF-females) and fecal samples from HC females (HCF-females). To test the hypothesis that gut-kidney leakage occurs in CKD patients, we first performed PCoA on the bladder and fecal microbiomes in CKD patients and controls. The bladder microbiomes differed significantly from the corresponding fecal microbiomes for both CKD-males and females and their age-matched HC-controls (PERMANOVA, FDR < 0.05; Figs. 5C-F). We then compared the Bray-Curtis dissimilarity between the bladder and gut microbiomes from either CKD males/females or HC males/females. These microbiomes tended to be less dissimilar in CKD-females than in HC-females (Wilcoxon rank-sum test, FDR < 0.001), but not in the males (Wilcoxon rank-sum test, FDR > 0.05; Fig. 5G). In addition, the Bray-Curtis dissimilarity between the bladder and gut microbiomes from CKD females was slightly less than that from CKD males (0.965 ± 0.068 vs. 0.983 ± 0.048, Wilcoxon rank-sum test, FDR = 0.089), while the Bray-Curtis dissimilarity between the bladder and fecal microbiomes from healthy females was almost equal to that from healthy males (0.998 ± 0.005 vs. 0.998 ± 0.004, Wilcoxon rank-sum test, FDR = 0.472).

The role of the bladder microbiome in kidney function is understudied. Our research delves into the bladder microbiome of CKD patients, revealing unique microbial profiles linked to kidney dysfunction and serum cytokine imbalances.

First, we found that both male and female CKD patients had dysbiotic bladder microbiomes, as assessed by 16S rRNA gene sequencing. This dysbiosis may be associated with loss of kidney function, as the bladder microbiomes of CKD patients in stages 2–5 differed from those of controls. A similar phenomenon in the gut was described by Wu and co-authors, who reported that the fecal microbiome could reflect the CKD stage[22].

We observed an increased microbial richness in CKD patients relative to controls and found that this increase was greater in females than in males, as assessed by both sequencing and EQUC. Since population-based studies show that CKD affects more females than males[23], bladder dysbiosis in the context of CKD sex differences is worthy of future exploration.

One striking finding of this study is that CKD females’ bladder microbiome was enriched for the genus Escherichia/Shigella, as assessed by both 16S rRNA gene sequencing and EQUC. This genus was not associated with urinary leucocyte esterase or serum cytokine profile, but its increase in CKD females was positively correlated to the level of serum creatinine, a kidney function estimator. Given that Escherichia/Shigella is the most commonly detected genus in UTI[24, 25], the increased prevalence of potentially uropathogenic microbe Escherichia/Shigella in CKD females warrants attention during treatment.

Another notable finding is that both detection methods (sequencing and EQUC) revealed a depletion of Lactobacillus in CKD females. Decreased Lactobacillus has been reported in patients with UTI[24, 25], and may be a sign of an unhealthy bladder. Even though we did not observe an association between Lactobacillus and serum cytokines like previous urinary microbiome studies on interstitial cystitis and systemic lupus erythematosus suggested[24, 25], Lactobacillus was associated with the kidney function indicators. It was positively associated with eGFR and negatively associated with serum creatinine, indicating that Lactobacillus in the bladder may be a biomarker for kidney function. Although the effects of bladder Lactobacillus on kidney functions have never been reported, Lactobacillus oral supplementation in mouse models has been suggested to attenuate the loss of kidney function, in a process that involved inflammation pathways[6, 27]. Thus, Lactobacillus in the bladder might protect kidney function in females.

While decreased Lactobacillus in the vagina is a well-documented feature of older females, recent studies have shown this decrease to also be true of the bladder. Komesu and colleagues found that the relative abundance of Lactobacillus in catheterized urine had a strong negative relationship with age when study participants with mixed urinary incontinence and controls were assessed together[6, 27]. Another study demonstrated that Lactobacillus in mid-stream voided urine was more common in pre-menopausal females comparing to post-menopausal ones[29]. In contrast, our study found a negative association between bladder Lactobacillus and age only in female CKD patients. This finding suggests that loss of kidney function, a sign of getting older[30], may be attributable to Lactobacillus depletion.

Bifidobacterium is widely advocated as a probiotic due to its potential beneficial effects[31]. However, in our study, bladder Bifidobacterium was enriched in CKD females and positively associated with disease duration and MIG, a chemoattractant cytokine for activated T lymphocytes and tumor infiltrating T-lymphocytes[32], and an indication of an unhealthy kidney, e.g. kidney cancer or stones[32]. Increased Bifidobacterium also has been reported in patients with kidney stones[35]. Bifidobacterium has been reported in bladder infections and other diseases[35–37]; thus, its role as a probiotic in the bladder remains unclear.

The upregulation of IL-1β, IL-8, IL-18 and TNF-α in CKD patients have been reported[38–40]. As expected, all these cytokines were upregulated in the serum of CKD females, and almost all were negatively associated with decreased Sphingobium in CKD female bladders. As Sphingobium has been rarely reported in association with humans[41], the roles of these associations remain to be examined. It is worth noting that half of the bacterial genera enriched in CKD females are members of the class Clostridia, including Agathobacter, Blautia, Faecalibacterium and Subdoligranulum, and all were positively linked to the increase of the level of IL-18 in patients. Some members of Clostridia, including Clostridiales, Lachnospiraceae and Ruminococcaceae, have been reported to be enriched in the voided urine of patients with chronic prostatitis[42]. It is reasonable to conclude that Clostridia plays a pathogenic role in CKD females.

To explore the gut-kidney axis, we examined fecal and bladder microbiomes in age-sex-matched CKD patients and controls. A significant difference was only found in the bladder microbiome. Analyzing the Bray-Curtis Dissimilarity Index, we observed a significant decrease in female CKD patients but not in matched controls, supporting the hypothesis of a “leaky gut” in CKD females.

The differences in the bladder microbiome between CKD patients and healthy controls may stem from two hypothesized scenarios. The first suggests bladder-residing bacteria migrate through the urinary tract, potentially affecting renal physiology and CKD progression via reflux to the kidneys. The second posits gut bacteria translocate into the circulatory system, influencing renal physiology, CKD progression, and eventually reaching the bladder. Our data lean toward supporting the second hypothesis. However, further in-depth investigations are needed to uncover the precise mechanisms behind this complex process.

In summary, our study highlights significant differences in the CKD bladder microbiome, especially in females, marked by greater richness and distinct taxonomic variations compared to controls. Notably, CKD females exhibited a notable increase in Escherichia/Shigella and a decline in Lactobacillus, potentially linked to reduced kidney function. Associations between specific bladder bacterial genera and CKD-related serum cytokines suggest potential pathogenic roles. Furthermore, bladder dysbiosis coincided with “gut-leakage” in CKD females. Future research employing suitable CKD animal models should investigate the impact of the bladder microbiome on kidney function, inflammation, and “leaky gut” to elucidate these relationships.

Limitations of the study

Our study has limitations. Firstly, the unequal 3:1 ratio of CKD patients to age-sex-BMI matched controls may introduce potential confounders within the groups[43]. However, recruiting healthy matched controls was challenging due to the prevalent health issues in the elderly population[30], including kidney-related concerns Secondly, enrolling participants solely from one institution limited the power of genetic profiling[44].

CKD: chronic kidney disease

CKDF-females: fecal samples from CKD females

CKDF-males: fecal samples from CKD males

CKDS: serum samples from CKD patients

CKDB-female: bladder urine samples from CKD females

CKDB-male: bladder urine samples from CKD males

eGFR: estimated glomerular filtration rate

EQUC: Expanded Quantitative Urine Culture

HC: healthy controls

HCF-females: fecal samples from HC females

HCF-males: fecal samples from HC males

HCS: serum samples from HC

HCB-female: bladder urine samples from HC females

HCB-male: bladder urine samples from HC males

PCoA: principal-coordinate analysis

UACR: urinary albumin creatinine ratio

UTI: urinary tract infection

Acknowledgements

The authors thank the study participants and the clinic staff in the Affiliated Wuxi No. 2 Hospital, Nanjing Medical University.

Author’s contributions

Conceptualization: FPL, AJW, AWM and NHF; Data curation: FPL, JJD; Formal analysis: FPL, JJD; Funding acquisition: YFG, NHF; Investigation: FPL, AJW, NHF; Methodology: FPL, JJD; Project administration: FPL, HL, YFG, YFX, FY, YFT, YW, WG, ZYX, JLH, PJ, CQG, JYS, WJH, LH; Resources: NHF, FY; Software: FPL, JJD; Supervision: NHF; Validation: AJW; Visualization: FPL, JJD, AJ. W; Writing-original draft: FPL, JJD, AJW, AWM; Writing-review & editing: AWM and AJW.

Funding

This study received funding from the National Natural Science Foundation of China (Grants 81874142 and 82073041), where Yifeng Gu was the recipient; Zhejiang Provincial Natural Science Foundation of China (Grant LXR22H160001), also secured by Yifeng Gu; and the Wuxi “Taihu Talents Program” Medical and Health High-level Talents Project (THRCJH20200901), with funding acquired by Ninghan Feng.

Availability of data and materials

Raw data from 16S rRNA sequencing are available in the Sequence Read Archive under BioProject ID SRP360378 (https://www.ncbi.nlm.nih.gov/bioproject/PRJNA807874).

Ethics approval and consent to participate

The study received approval from the ethics committee of the Affiliated Wuxi Second Hospital of Nanjing Medical University (Reference: 2018051). Informed consent was provided by all subjects prior to sample collection.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Bikbov B, Purcell CA, Levey AS, Smith M, Abdoli A, Abebe M, et al. Global, regional, and national burden of chronic kidney disease, 1990-2017: a systematic analysis for the Global Burden of Disease Study 2017. Lancet. 2020;395(10225):709-33.
Meijers B, Evenepoel P, Anders HJ. Intestinal microbiome and fitness in kidney disease. Nat Rev Nephrol. 2019;15(9):531-45.
Shah NB, Allegretti AS, Nigwekar SU, Kalim S, Zhao S, Lelouvier B, et al. Blood Microbiome Profile in CKD : A Pilot Study. Clin J Am Soc Nephrol. 2019;14(5):692-701.
Hilt EE, Mckinley K, Pearce MM, Rosenfeld AB, Zilliox MJ, Mueller ER, et al. Urine is not sterile: use of enhanced urine culture techniques to detect resident bacterial flora in the adult female bladder. J Clin Microbiol. 2014;52(3):871-6.
Wolfe AJ, Toh E, Shibata N, Rong R, Kenton K, Fitzgerald M, et al. Evidence of uncultivated bacteria in the adult female bladder. J Clin Microbiol. 2012;50(4):1376-83.
Liu F, Du J, Zhai Q, Hu J, Miller AW, Ren T, et al. The Bladder Microbiome, Metabolome, Cytokines, and Phenotypes in Patients with Systemic Lupus Erythematosus. Microbiol Spectr. 2022:e21222.
Kachroo N, Lange D, Penniston KL, Stern J, Tasian G, Bajic P, et al. Meta-analysis of Clinical Microbiome Studies in Urolithiasis Reveal Age, Stone Composition, and Study Location as the Predominant Factors in Urolithiasis-Associated Microbiome Composition. Mbio. 2021;12(4):e200721.
Vaughan MH, Mao J, Karstens LA, Ma L, Amundsen CL, Schmader KE, et al. The Urinary Microbiome in Postmenopausal Women with Recurrent Urinary Tract Infections. J Urol. 2021;206(5):1222-31.
Wu P, Zhang G, Zhao J, Chen J, Chen Y, Huang W, et al. Profiling the Urinary Microbiota in Male Patients With Bladder Cancer in China. Front Cell Infect Microbiol. 2018;8:167.
Kramer H, Kuffel G, Thomas-White K, Wolfe AJ, Vellanki K, Leehey DJ, et al. Diversity of the midstream urine microbiome in adults with chronic kidney disease. Int Urol Nephrol. 2018;50(6):1123-30.
Frimodt-Møller N. The urine microbiome – Contamination or a novel paradigm? Ebiomedicine. 2019;44:20-1.
Brubaker L, Gourdine JF, Siddiqui NY, Holland A, Halverson T, Limeria R, et al. Forming Consensus To Advance Urobiome Research. Msystems. 2021;6(4):e137120.
Carrero JJ, Yilmaz MI, Lindholm B, Stenvinkel P. Cytokine dysregulation in chronic kidney disease: how can we treat it? Blood Purif. 2008;26(3):291-9.
Amdur RL, Feldman HI, Gupta J, Yang W, Kanetsky P, Shlipak M, et al. Inflammation and Progression of CKD: The CRIC Study. Clin J Am Soc Nephrol. 2016;11(9):1546-56.
Salguero MV, Al-Obaide M, Singh R, Siepmann T, Vasylyeva TL. Dysbiosis of Gram-negative gut microbiota and the associated serum lipopolysaccharide exacerbates inflammation in type 2 diabetic patients with chronic kidney disease. Exp Ther Med. 2019;18(5):3461-9.
Linh HT, Iwata Y, Senda Y, Sakai-Takemori Y, Nakade Y, Oshima M, et al. Intestinal Bacterial Translocation Contributes to Diabetic Kidney Disease. J Am Soc Nephrol. 2022;33(6):1105-19.
Tao S, Li L, Li L, Liu Y, Ren Q, Shi M, et al. Understanding the gut-kidney axis among biopsy-proven diabetic nephropathy, type 2 diabetes mellitus and healthy controls: an analysis of the gut microbiota composition. Acta Diabetol. 2019;56(5):581-92.
Wolfe AJ, Brubaker L. Urobiome updates: advances in urinary microbiome research. Nat Rev Urol. 2019;16(2):73-4.
Perez-Gomez MV, Bartsch LA, Castillo-Rodriguez E, Fernandez-Prado R, Fernandez-Fernandez B, Martin-Cleary C, et al. Clarifying the concept of chronic kidney disease for non-nephrologists. Clin Kidney J. 2019;12(2):258-61.
Price TK, Dune T, Hilt EE, Thomas-White KJ, Kliethermes S, Brincat C, et al. The Clinical Urine Culture: Enhanced Techniques Improve Detection of Clinically Relevant Microorganisms. J Clin Microbiol. 2016;54(5):1216-22.
Mesureur J, Arend S, Celliere B, Courault P, Cotte-Pattat PJ, Totty H, et al. A MALDI-TOF MS database with broad genus coverage for species-level identification of Brucella. PLoS Negl Trop Dis. 2018;12(10):e6874.
Wu IW, Gao SS, Chou HC, Yang HY, Chang LC, Kuo YL, et al. Integrative metagenomic and metabolomic analyses reveal severity-specific signatures of gut microbiota in chronic kidney disease. Theranostics. 2020;10(12):5398-411.
Carrero JJ, Hecking M, Chesnaye NC, Jager KJ. Sex and gender disparities in the epidemiology and outcomes of chronic kidney disease. Nat Rev Nephrol. 2018;14(3):151-64.
Flores-Mireles AL, Walker JN, Caparon M, Hultgren SJ. Urinary tract infections: epidemiology, mechanisms of infection and treatment options. Nat Rev Microbiol. 2015;13(5):269-84.
Huttner A, Hatz C, van den Dobbelsteen G, Abbanat D, Hornacek A, Frolich R, et al. Safety, immunogenicity, and preliminary clinical efficacy of a vaccine against extraintestinal pathogenic Escherichia coli in women with a history of recurrent urinary tract infection: a randomised, single-blind, placebo-controlled phase 1b trial. Lancet Infect Dis. 2017;17(5):528-37.
Kim H, Nam BY, Park J, Song S, Kim WK, Lee K, et al. Lactobacillus acidophilus KBL409 Reduces Kidney Fibrosis via Immune Modulatory Effects in Mice with Chronic Kidney Disease. Mol Nutr Food Res. 2022:e2101105.
Tungsanga S, Katavetin P, Panpetch W, Udompornpitak K, Saisorn W, Praditpornsilpa K, et al. Lactobacillus rhamnosus L34 attenuates chronic kidney disease progression in a 5/6 nephrectomy mouse model through the excretion of anti-inflammatory molecules. Nephrol Dial Transplant. 2022;37(8):1429-42.
Komesu YM, Dinwiddie DL, Richter HE, Lukacz ES, Sung VW, Siddiqui NY, et al. Defining the relationship between vaginal and urinary microbiomes. Am J Obstet Gynecol. 2020;222(2):151-4.
Curtiss N, Balachandran A, Krska L, Peppiatt-Wildman C, Wildman S, Duckett J. Age, menopausal status and the bladder microbiome. Eur J Obstet Gynecol Reprod Biol. 2018;228:126-9.
Ellam T, Twohig H, Khwaja A. Chronic kidney disease in elderly people: disease or disease label? BMJ. 2016;352:h6559.
Ming J, Yu X, Xu X, Wang L, Ding C, Wang Z, et al. Effectiveness and safety of Bifidobacterium and berberine in human hyperglycemia and their regulatory effect on the gut microbiota: a multi-center, double-blind, randomized, parallel- controlled study. Genome Med. 2021;13(1):125.
Whiting D, Hsieh G, Yun JJ, Banerji A, Yao W, Fishbein MC, et al. Chemokine monokine induced by IFN-gamma/CXC chemokine ligand 9 stimulates T lymphocyte proliferation and effector cytokine production. J Immunol. 2004;172(12):7417-24.
Lee MH, Laajala E, Kreutzman A, Jarvinen P, Nisen H, Mirtti T, et al. The tumor and plasma cytokine profiles of renal cell carcinoma patients. Sci Rep. 2022;12(1):13416.
Suen JL, Liu CC, Lin YS, Tsai YF, Juo SH, Chou YH. Urinary chemokines/cytokines are elevated in patients with urolithiasis. Urol Res. 2010;38(2):81-7.
Liu F, Zhang N, Jiang P, Zhai Q, Li C, Yu D, et al. Characteristics of the urinary microbiome in kidney stone patients with hypertension. J Transl Med. 2020;18(1):130.
Pathak P, Trilligan C, Rapose A. Bifidobacterium--friend or foe? A case of urinary tract infection with Bifidobacterium species. BMJ Case Rep. 2014;2014.
Bhaskar MM, Sistla S, Kumaravel S. A case of pyometrocolpos with Bifidobacterium species. Anaerobe. 2017;44:48-50.
Komada T, Muruve DA. The role of inflammasomes in kidney disease. Nat Rev Nephrol. 2019;15(8):501-20.
Vianna HR, Soares CM, Silveira KD, Elmiro GS, Mendes PM, de Sousa TM, et al. Cytokines in chronic kidney disease: potential link of MCP-1 and dyslipidemia in glomerular diseases. Pediatr Nephrol. 2013;28(3):463-9.
Gupta J, Mitra N, Kanetsky PA, Devaney J, Wing MR, Reilly M, et al. Association between albuminuria, kidney function, and inflammatory biomarker profile in CKD in CRIC. Clin J Am Soc Nephrol. 2012;7(12):1938-46.
Burton EN, Cohn LA, Reinero CN, Rindt H, Moore SG, Ericsson AC. Characterization of the urinary microbiome in healthy dogs. Plos One. 2017;12(5):e177783.
Shoskes D, Altemus J, Polackwich A, Tucky B, Wang H, Eng C. The Urinary Microbiome Differs Significantly Between Patients With Chronic Prostatitis/Chronic Pelvic Pain Syndrome and Controls as Well as Between Patients With Different Clinical Phenotypes. Urology. 2016;92:26-32.
Zhu Z, Satten GA, Mitchell C, Hu YJ. Constraining PERMANOVA and LDM to within-set comparisons by projection improves the efficiency of analyses of matched sets of microbiome data. Microbiome. 2021;9(1):133.
Mcwilliams R, Hoover-Fong J, Hamosh A, Beck S, Beaty T, Cutting G. Problematic variation in local institutional review of a multicenter genetic epidemiology study. JAMA. 2003;290(3):360-6.

No competing interests reported.

Supplementaryfigures.rar
Figure S1 Stud design. Participants enrollment and study procedures. Abbreviations: CKDB: urine samples provided by patients with CKD; HCB: urine samples provided by healthy controls; CKDF: fecal samples provided by patients with CKD; HCF: fecal samples provided by healthy controls; CKDS: serum samples provided by CKD patient; HCS: serum samples provided by healthy controls. Figure S2 Medication usages on urinary microbiome. (A) Comparison of urinary microbiome between antihypertensive agent users and non-users; (B) Comparison of urinary microbiome between glucocorticoid agent users and non-users; (C) Comparison of urinary microbiome between hypoglycemic agent users and non-users; (D) Comparison of urinary microbiome between hypolipidemic agent users and non-users. Permutational multivariate analysis of variance (PERMANOVA) was performed for statistical comparisons of samples in the two groups. P value was adjusted by the Benjamini and Hochberg false discovery rate (FDR). Figure S3Microbial profile at the phylum and genus level. Only the top 5 most abundant phylum and the top 15 most abundant genus are shown. Figure S4 Urinary microbiome in various CKD stages. Comparisons of microbial community between patients in CKD stage1-stage5 and controls. Permutational multivariate analysis of variance (PERMANOVA) was performed for statistical comparisons of samples in the two groups. P value was adjusted by the Benjamini and Hochberg false discovery rate (FDR). (B)Comparisons of microbial richness and diversity between patients in CKD stage1-stage5 and controls. Wilcoxon rank-sum test was performed and adjusted by Benjamini and Hochberg false discovery rate (FDR).
Supplementarytables.rar
Table S1 Comparison of nutrient intake between CKD and HC cohort. Student’s ttest on normalized continuous variables, and Wilcoxon rank-sum test on non-normal continuous variables. Table S2Comparative Analysis of ASV Abundances between CKDB-male and HCB-male Groups. Wilcoxon rank-sum test was applied. Abrreviations: CKDB-male: bladder urine samples provided by males with CKD; HCB-male: bladder urine samples provided by healthy males. Table S3Comparative Analysis of ASV Abundances between CKDB-female and HCB-female Groups. Wilcoxon rank-sum test was applied. Abrreviations: CKDB-female: bladder urine samples provided by females with CKD; HCB-female: bladder urine samples provided by healthy females. Table S4Comparison of bacterial phylum abundance between CKDB-male and HCB-male. Wilcoxon rank-sum test was applied. Abbreviations: CKDB-male: urine samples provided by males with CKD; HCB-male: urine samples provided by healthy males. Table S5 Comparison of bacterial phylum abundance between CKDB-female and HCB-female. Wilcoxon rank-sum test was applied. Abbreviations: CKDB-female: urine samples provided by females with CKD; HCB-female: urine samples provided by healthy females. Table S6Comparison of bacterial genus abundance between CKDB-male and HCB-male. Wilcoxon rank-sum test was applied. Abbreviations: CKDB-male: urine samples provided by males with CKD; HCB-male: urine samples provided by healthy males. Table S7Comparison of bacterial genus abundance between CKDB-female and HCB-female. Wilcoxon rank-sum test was applied. Abbreviations: CKDB-female: urine samples provided by females with CKD; HCB-female: urine samples provided by healthy females. Table S8 Bacteria identified using EQUC. Abbreviations: CKDB-male: urine samples provided by males with CKD; CKDB-female: urine samples provided by females with CKD; HCB-male: urine samples provided by healthy males; HCB-female: urine samples provided by healthy females. Table S9 Comparisons of samples with visible cultures between HCB-male and HCB-female. Pearson Chi-square or Fisher’s exact test was used Abbreviations: CKDB-male: urine samples provided by males with CKD; HCB-male: urine samples provided by healthy males. Table S10 Comparisons of samples with visible cultures between HCB-male and HCB-female. Pearson Chi-square or Fisher’s exact test was used. Abbreviations: CKDB-female: urine samples provided by females with CKD; HCB-female: urine samples provided by healthy females. Table S11Comparison of samples with visible cultures between CKDB-male and HCB-male. Pearson Chi-square or Fisher’s exact test was used. Abbreviations: CKDB-male: bladder urine samples provided by males with CKD; HCB-male: bladder urine samples provided by healthy males. Table S12Comparison of samples with visible cultures between CKDB-female and HCB-female. Pearson Chi-square or Fisher’s exact test was used. Abbreviations: CKDB-female: bladder urine samples provided by females with CKD; HCB-female: bladder urine samples provided by healthy females.

Download PDF

Version 1

posted

You are reading this latest preprint version

Sex-Specific Dysbiotic Bladder Microbiome in CKD Uncovered via High-Throughput Sequencing and Culture

Status:

Version 1

Abstract

Background

Methods

Results

Conclusions

Figures

INTRODUCTION

METHODS

Study participants

Urine collection and storage

Expanded Quantitative Bladder Urine Culture

Bioinformatics

16S rRNA gene sequencing

Bioinformatic analysis of 16S rRNA gene sequencing data

Identification of bacterial isolates

Identification by 16S rRNA gene sequencing

Identification by MALDI-TOF MS

Other sample collection and processing

Disease profiles assessment

Statistical analysis

RESULTS

Study participants

Bladder microbiomes are altered in CKD males and females

Bladder microbiome composition is altered in CKD patients

Altered genera in CKD was associated with demographics and renal function

Bacterial community altered in all stages of CKD patients

Altered bladder microbiome associated with altered serum cytokines in CKD patients

Alteration of microbiome not observed in feces

DISCUSSION

Limitations of the study

Abbreviations

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1