Skip to main content
Download PDF

Urobiome of patients with diabetic kidney disease in different stages is revealed by 2bRAD-M

Journal of Translational Medicine volume 23, Article number: 414 (2025) Cite this article

Abstract

Background

Knowledge of the urinary microbiome (urobiome) in diabetic kidney disease (DKD) remains limited. The most commonly used 16S rRNA sequencing technique can only provide bacterial identification at the genus level. As a novel technique, 2bRAD sequencing for microbiome (2bRAD-M) can be used to identify the low-biomass microbiome at the species level. In this study, we used 2bRAD-M to compare the urobiome composition of patients with DKD at different stages with healthy individuals and those with type 2 diabetes mellitus (T2DM), with the expectation that we would find discriminative species correlated with DKD.

Method

Healthy controls, patients with DKD with microalbuminuria (DKD1 group) or macroalbuminuria (DKD2 group), and patients with T2DM were recruited (n = 20 for each group). The first-morning urine was collected for 2bRAD-M testing. The albumin-to-creatinine ratio (ACR) was also measured with urine samples. Serum samples were collected for detecting clinical indicators. The microbial diversity and composition based on abundance were calculated. Differential bacteria for different groups were identified. Besides, the correlation between discriminative bacteria and clinical indices was also analyzed.

Results

Urobiome diversity was significantly reduced in the DKD groups. In the DKD1 group, was the dominant genus, followed by Pseudomonas_E, whereas in the DKD2 group, Pseudomonas_E became the dominant genus and Escherichia was notably reduced. Both Bifidobacterium and Streptococcus, which were the top genera in the control group, were substantially decreased in the DKD groups. The discriminative species for DKD1 included Escherichia coli and Acinetobacter johnsonii, while for DKD2, Pseudomonas_E oleovorans, Enterococcus faecalis, and Morganella morganii were identified. Pseudomonas_E, Enterococcus and Morganella showed a strong correlation with renal function indicators and urinary protein levels.

Conclusion

The urobiome diversity and composition in patients with DKD were markedly different from those in healthy individuals and T2DM patients. These findings provide valuable insights into the onset and progression of DKD, driven by changes in the urinary bacterial community structure.

Introduction

Diabetic kidney disease (DKD) is a severe microvascular complication of diabetes and a major cause of end-stage renal disease in both China and developed countries. As approximately 30–40% of individuals with diabetes develop DKD, its prevention and treatment has always been a global public health challenge [1,2,3]. However, due to its relatively complicated pathogenesis, the understanding of DKD remains limited.

Accumulated studies have shown that dysbiosis of the microbiome likely occurs in DKD, especially in the gut microbiota [4,5,6,7]. With the development of high-throughput sequencing technology, researchers have realized that even urine is not sterile in a healthy condition, but rather has a colonized microbiota, termed the urinary microbiome (urobiome) [8,9,10,11]. Using 16S rRNA sequencing, scientists have discovered the presence of various colonizing bacteria in the urinary tract, including Lactobacillus, Corynebacterium, Staphylococcus, Streptococcus, Veillonella, and Prevotella, which appear to be unassociated with urinary tract infections (UTIs) [12]. Researchers have reported that a urobiome shift may play essential roles in several diseases, including kidney stones, bladder cancer, chronic kidney disease (CKD), and type 2 diabetes mellitus (T2DM) [12,13,14,15]. However, knowledge of the urobiome in DKD remains limited [16].

In previous studies, microbiome analysis relied on two high-throughput research technologies: amplicon sequencing (16S rRNA) and shotgun metagenome sequencing (WMS) [17,18,19]. Currently, 16S rRNA sequencing is the most widely adopted technique for urinary microbial diversity research [12, 20,21,22,23]. However, because this technique only sequences the 16S rRNA region, it cannot identify bacteria up to the species level [24]. WMS allows sequencing of the entire genome from samples and is able to classify the bacteria to species level; however, it is much more expensive and requires a large amount of high-quality DNA as the starting material, so it is unsuitable for urine samples, which have much lower biomass [25]. Enhanced quantitative urine culture (EQUC) can also be used to identify a wide variety of urinary bacteria [26,27,28,29]. However, the limitation of culture-based techniques compared with sequencing techniques is that they are insufficient for identifying the complete urine microbiome.

The novel 2bRAD sequencing for microbiome (2bRAD-M) technology is a sequence-based method for studying urobiomes [25]. In this method, IIB-type restriction endonuclease is used to digest double-stranded DNA to produce equal-length fragments (20–33 bp), which are then amplified and sequenced. The tag sequences can be mapped to reference genomes to reconstruct the taxonomic composition. This technology can efficiently process low biomass and degraded samples and accurately generate species-level taxonomic profiles [30].

In this study, 2bRAD-M was applied to evaluate the urobiome of patients with T2DM without albuminuria, patients with DKD at different stages (with micro- or macroalbuminuria), and matched healthy controls. We expected to identify the specific bacteria closely associated with diabetes or DKD to better clarify the pathogenesis of DKD and identify possible species markers. The serum biochemical indicators and urine protein were also included in the analysis.

Method

Recruitment of participants

Sixty patients with T2DM with or without albuminuria and 20 healthy participants from eight medical centers in China were recruited for this study. T2DM was diagnosed based on the American Diabetes Association guidelines [31]. The diagnostic criteria for DKD were determined according to a previous report [32]: albumin-to-creatinine ratio (ACR) > 30 mg/g and/or abnormal estimated glomerular filtration rate (eGFR < 60 mL/min/1.73 m2). Microalbuminuria (DKD1) was defined by an ACR < 300 mg/g, while macroalbuminuria (DKD2) was defined by an ACR ≥ 300 mg/g. The exclusion criteria were as follows: 1) disease duration < 5 years; 2) rapid decline in eGFR; 3) rapid elevation in urinary albumin or nephrotic syndrome; 4) urinary active sediment (red blood cells, white blood cells, or cell casts); 5) resistant hypertension; and 6) combined with other systemic diseases. We also excluded subjects with a lack of clinical data, type I diabetic nephropathy and other special types of diabetic nephropathy, acute and chronic infectious diseases, and antibiotic and probiotic use 3 months prior to sample collection. All healthy volunteers were older than 18 years and had no underlying diseases.

Sample collection

The first morning urine was required for this study. Before collecting, the participants were asked to wash their external genitalia. Then, they were allowed to collect midstream urine during continuous urination into a sterile container. Fasting blood (≥ 8 h) was collected at the same time, and serum was separated within 1 h. All samples were stored at − 80 °C until testing.

This study was approved by the Ethics Committee of China-Japan Friendship Hospital. All participants were clearly informed of the purpose and process of the trial. Informed consent was obtained from the patients for the use of their samples (2022-KY-246).

Serum indices and urinary protein examination

Serum creatinine (CRE), blood urea nitrogen (BUN), Cystatin C (CysC), estimated glomerular filtration rate (eGFR), and urine albumin-to-creatinine ratio (ACR) were measured using an automatic biochemical analyzer (Beckman AU5800).

DNA extraction, library construction, and sequencing

The 2bRAD sequencing for microbiome (2bRAD-M) was performed according to the protocol established by Sun et al. (https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s13059-021-02576-9), including all formulas and methodologies as detailed in their study.

Genomic DNA was extracted from urine samples using a TIANamp Micro DNA Kit (Tiangen). The 2bRAD library was constructed according to previous studies [30, 33]. The brief steps were as follows: First, genomic DNA was digested with 4 U of BcgI restriction enzyme (NEB) at 37 °C for 3 h. The product was then run on a 1% agarose gel to verify digestion; second, the ligation reaction was performed at 4 °C for 16 h. The reaction volume contained 10 μL of digested product, 0.8 μM each of library-specific adaptors (Ada1 and Ada2), 1 mM ATP (NEB), 1 × T4 DNA Ligase Buffer, and 800 U T4 DNA ligase (NEB). BcgI was heat-inactivated at 65 °C for 20 min, and finally, the ligated products in the 20-μL reaction volume consisted of 7 μL ligated DNA, 0.4 μM each primer (primer 1 and primer 2 for Illumina), 0.6 mM dNTP, 1 × Phusion HF buffer, and 0.4 U Phusion high-fidelity DNA polymerase (NEB). The PCR reaction was performed as follows: 98 °C for 5 s, 60 °C for 20 s, 72 °C for 10 s (16–28 cycles), and then a final extension at 72 °C for 10 min. The library products were purified using a purification kit (Qiagen) and subjected to sequencing via Illumina HiSeq X™ Ten platform. Library construction and Illumina sequencing were performed at OE BioTech Co., Ltd. (Qingdao, China). All adaptor and primer sequences were listed in Supplement Table 1.

Identification of species-specific 2bRAD-M markers

First, 404,199 microbial genomes, including bacteria, fungi, and archaea, were downloaded from GTDB and Ensembl. Then, built-in Perl scripts were used to sample restriction fragments from microbial genomes for each of 16 type 2b restriction enzymes, resulting in a large 2bRAD microbial genome database. The set of 2bRAD tags sampled from each genome was assigned under the GCF number, as well as the taxonomic information of GCF corresponding to the whole genome. Finally, all 2bRAD tags from each GCF that occurred once within the genome were compared with the other tags. The 2bRAD tags specific to a species-level taxon (having no overlap with other species) were developed as species-specific 2bRAD markers, collectively forming a 2bRAD marker database.

Calculation of relative abundance

First, to identify microbial species within each sample, all sequenced 2bRAD tags after quality control were mapped (using a built-in Perl script) against the 2bRAD marker database, which contains all 2bRAD tags theoretically unique to each of the 86,022 microbial species in the database. To control false-positives in species identification, the G score was derived for each species identified within a sample as follows: a harmonious mean of read coverage of 2bRAD markers belonging to a species and the number of all possible 2bRAD markers of this species. The threshold of the G score for false-positive discovery of microbial species was set at 5.

$$G\,score_{species\,i} = \sqrt {S_{i} \times t_{i} }$$

S: Number of reads assigned to all 2bRAD markers belonging to species i within a sample.

t: Number of all 2bRAD markers of species i sequenced within a sample.

Then, we calculated the average read coverage of all 2bRAD markers for each species, which represented the number of individuals belonging to a species present in a sample at a given sequencing depth. The relative abundance of a given species was then calculated as the ratio of the number of microbial individuals belonging to a species to the total number of individuals from known species that could be detected within a sample.

$${\text{Re}} lative{\kern 1pt} abundance_{species\,i} = \frac{{S_{i} /T_{i} }}{{\sum {_{i = 1}^{n} S_{i} /T_{i} } }}$$

S: Number of reads assigned to all 2bRAD markers of species i within a sample.

T: the number of all theoretical 2bRAD markers of species i.

Diversity analysis and identification of differential taxa

The alpha diversity of the microbial composition was measured using the Chao1, Shannon, and Simpson indices and then calculated using the “Vegan” package. The Wilcoxon test was used for group comparisons of alpha diversity and microbial communities. The “Vegan” package was also used to calculate the Bray–Curtis distance, Binary Jaccard distance, and Euclidean distance to measure the beta diversity of the multi-omics data. Differences in dissimilarity were determined using a nonparametric multivariate analysis of variance (Permanova test) and visualized through principal coordinate analysis (2D-PCoA).

Kruskal–Wallis was used to analyze the overall differences among groups. The discriminant bacteria of each group were identified using LEfSe. Indicator analysis was performed to identify the indicative species for each group, and the correlation of different genera was calculated using Spearman correlation analysis based on relative abundance. P-values < 0.05 indicated a statistically significant difference.

Correlation analysis of bacterial abundance and clinical indices

The correlation between different bacteria and various clinical indices (eGFR, CRE, ACR, 24 h-UP, CysC, Age, HbA1c, BMI and BUN) was calculated using Spearman’s correlation analysis.

Results

Clinical characteristics of the participants

A total of 80 participants were recruited for the study, including 20 healthy controls (control group), 20 diabetic patients (T2DM group), 20 diabetic patients with microalbuminuria (DKD1 group), and 20 diabetic patients with macroalbuminuria (DKD2 group). Compared with the control group, the body mass index of the other three groups was significantly higher. The ACR, CRE, and CysC were significantly higher in the DKD groups than in the control and T2DM groups. The BUN was significantly higher and eGFR was significantly lower in the DKD2 group than in the other groups. The clinical characteristics of the participants are listed in Table 1.

Table 1 Clinical characteristics of enrolled patients (n = 60)
Full size table

Biodiversity of the urobiome in different groups

Alpha diversity analysis (Fig. 1A–C) reflects the richness and the evenness of their distribution within that sample. The Chao1 index revealed that the microbial diversity of the DKD groups (DKD1 and DKD2) was significantly lower than that in the control and T2DM groups, with no significant difference observed between the control and T2DM groups or between the DKD1 and DKD2 groups. The Shannon index revealed significant differences between the control and DKD1 groups. The Simpson index revealed no significant difference between the groups. Beta diversity compares microbial differences across multiple groups. Permanova analysis revealed the overall microbial composition was significantly different (P = 0.001) among groups based on the Euclidean distance, Binary Jaccard distance, and Bray–Curtis distance (Fig. 1D–F).

Fig. 1
figure 1

Microbial diversity of the urobiome in different groups. Comparison of the alpha diversity among groups (A Chao1 index, B Shannon index, C Simpson index). Comparison of the beta diversity among groups based on the Euclidean distance (D), Binary Jaccard distance (E), and Bray–Curtis distance (F). *p < 0.05, **p < 0.01, ***p < 0.001

Full size image

Microbial community structures of different groups

The numbers of identified taxa in the four groups are presented in Table 2. At all levels (phylum, genus, species), the numbers of bacterial taxa in the T2DM group (23, 426, 1071) were similar to those in the control group (20, 391, 925). In the DKD1 (20, 281, 645) and DKD2 (10, 267, 650) groups, the numbers of identified taxa were significantly lower than those in the control and T2DM groups. As shown in Fig. 2, some identified taxa were shared among the four groups, but with different relative abundances.

Table 2 Number of identified taxa in four groups at different level
Full size table
Fig. 2
figure 2

Microbial community structures of different groups. Barplots show the top 15 most abundant (A) phyla, (B) genera, and (C) species

Full size image

The top 15 identified taxa at different levels are shown in Fig. 2. In the control group, Actinomycetota was the dominant phylum (29.8%), followed by Bacillota (18.5%), Bacteroidota (18.3%), Pseudomonadota (15.8%), and Bacillota_A (9.2%). In the T2DM group, Actinomycetota was also dominant (32%), followed by Pseudomonadota (24.5%), Bacteroidota (19.6%), and Bacillota_A (5.1%). In the DKD1 and DKD2 group, Pseudomonadota was the most abundant phylum, accounting for 43.6% and 44.6%, respectively (Fig. 2A).

At the genus level, the top two genera in the control group were Bifidobacterium (16.2%) and Streptococcus (10.2%); in the T2DM group, they were Bifidobacterium (15.1%) and Lactobacillus (6.2%); in the DKD1 group, they were Escherichia (17.2%) and Pseudomonas_E (10.7%); and in the DKD2 group, they were Pseudomonas_E (18.6%) and Enterococcus (11.1%) (Fig. 2B).

At the species level, Streptococcus_anginosus was mostly abundant in the control group (6.0%), followed by Bifidobacterium_piotii (4.6%) and Bifidobacterium_leopoldii (4.3%). In the T2DM group, Bifidobacterium vaginale was the most dominant, followed by Escherichia coli (6.0%) and Pseudescherichia sp002298805 (5.4%). Escherichia_coli (17.2%), Pseudomonas_E_oleovorans (10.0%), and Lactobacillus_crispatus (4.9%) were the top three most abundant bacteria in the DKD1 group, whereas in the DKD2 group, the top three most abundant bacteria were Pseudomonas_E_ oleovorans (14.8%), Enterococcus_faecalis (11.2%), and Morganella_Morganii (6.5%) (Fig. 2C).

Top ten abundant differential bacterial taxa

We applied Kruskal–Wallis analysis to compare the top ten differentially expressed bacteria across the groups at three levels (Table 3). Some major differential bacteria are also shown in Fig. 3. Compared with the control and T2DM groups, an increased abundance of Pseudomonadota and a decreased abundance of Actinomycetota were identified in the DKD groups (DKD1 and DKD2) at the phylum level (Fig. 3A and B). In addition, the Bacteroidota abundance was particularly lower in the DKD2 group than in the other three groups (Fig. 3C), whereas the abundance of Bacillota_A (Fig. 3D) was obviously lower in the DKD1 group compared with the control group.

Table 3 The top 10 differential bacterial taxa
Full size table
Fig. 3
figure 3

Comparison of the major differential bacterial taxa among groups. A–D Differential bacteria at the phylum level. E–H Differential bacteria at the genus level. I–L Differential bacteria at the species level

Full size image

In terms of genus, the abundance of Enterococcus was significantly increased, while the abundance of Streptococcus (Fig. 3E, Table 3) was significantly reduced in all three groups compared to the control group. The abundance of Bifidobacterium (Fig. 3F) was notably lower in the DKD groups compared to both the control and T2DM groups.

Furthermore, Escherichia (Fig. 3G) was particularly abundant in the DKD1 group, whereas Pseudomonas_E, Enterococcus and Morganella were especially enriched in the DKD2 group (Fig. 3H, Table 3). Compared to the control group, Pseudescherichia was exclusively enriched in the T2DM group, with no significant enrichment observed in the DKD groups. Lactobacillus was enriched in the T2DM and DKD1 groups compared to the control group, whereas no significant difference was found between the DKD2 group and the control group (Table 3).

At the species level, Escherichia coli showed higher enrichment in the DKD1 group relative to the other groups (Fig. 3I, Table 3). The abundances of Pseudomonas E oleovorans, Enterococcus faecalis, and Morganella morganii were notably higher in the DKD2 group (Fig. 3J–K, Table 3). In contrast, Bifidobacterium leopoldii and Streptococcus anginosus were significantly reduced in all diabetic groups compared to the control group (Fig. 3L, Table 3). Pseudescherichia sp002298805 and Bifidobacterium vaginale were most enriched in the T2DM group compared to the other three groups (Table 3).

Indicator bacteria for different groups

LEfSe analysis was performed to identify the distinct bacterial taxa of each group. As shown in Fig. 4, 27 discriminative taxa were identified. In the DKD1 group, s__Escherichia_coli, g__Escherichia, f__Lactobacillaceae, and s__Lactobacillus_crispatus were identified as discriminative bacteria. Eleven discriminative taxa were identified in the DKD2 group, five of which were species: s__Pseudomonas_E_oleovorans, s__Enterococcus_faecalis, s__Morganella_morganii, s__Citrobacter_portucalensis, and s__Pseudomonas_E_fulva. In the T2DM group, o__Bacteroidales, g__Pseudescherichia, and s__Pseudescherichia sp002298805 were identified as discriminative bacteria. Among the nine discriminative taxa for the control group, three were species: s__Streptococcus_anginosus, s__Bifidobacterium_piotii, and s__Klebsiella_pneumoniae.

Fig. 4
figure 4

Discriminative bacterial taxa in each group. A The histogram of the LDA score showed 27 biomarkers with significant differences among the four groups. B The Cladogram represents the taxonomic hierarchical structure of the identified 27 discriminative biomarkers using LEfSe

Full size image

Indicator analysis was used to identify the most indicative species in each group. As shown in Fig. 5, Escherichia coli and Acinetobacter johnsonii were found to be the most indicative species in the DKD1 group. Although Acinetobacter johnsonii was not among the top 10 in terms of abundance, it exhibited the highest indicator value in this analysis. In the DKD2 group, Morganella morganii was identified as the most indicative species. For the T2DM group, Pseudescherichia_sp002298805 displayed the highest indicator value. In the control group, Streptococcus_anginosus emerged as the most indicative species.

Fig. 5
figure 5

Indicator bacteria map. The vertical axis represents bacterial taxa. The bar chart represents the relative abundance of each bacterium. The horizontal axis of the bubble chart represents the sample groups, and the bubble size represents the indicator value of each bacterium

Full size image

Connections among the top 30 most abundant genera

We next performed Spearman correlation analysis to evaluate correlations among the top 30 most abundant genera. Both positively and negatively correlated genera are shown in Fig. 6. For example, Pseudomonas_E was positively correlated with Ralstonia (R = 0.27, P = 0.01) and was negatively correlated with Bifidobacterium (R = − 0.22, P = 0.04). Streptococcus was positively correlated with Haemophilus_D (R = 0.29, P = 0.009). Cutibacterium was positively correlated with Sphingomonas (R = 0.45, P = 2.66E-05), Ralstonia (R = 0.35, P = 0.001), and Alloprevotella (R = 0.29, P = 0.009).

Fig. 6
figure 6

Spearman correlation analysis heatmap of the top 30 most abundant genera

Full size image

Correlations between clinical indicators and differential bacteria

The Spearman coefficient was calculated to estimate the correlation between various clinical indices and the abundance of the top 10 differential bacteria. The results indicated that the BUN, CRE, ACR, 24-UP and CysC were positively correlated with Pseudomonas_E, Morganella and Enterococcus (all the P values < 0.05, R 0.32–0.70). Conversely, eGFR was negatively correlated with Pseudomonas_E, Morganella and Enterococcus (all the P values < 0.05, R − 0.44 ∽ − 0.62), while positively correlated with Lactobacillus (P values < 0.05, R = 0.52). In addition, ACR was negatively correlated with Pseudescherichia (P < 0.05, R = − 0.64). (Fig. 7).

Fig. 7
figure 7

Correlation Heatmap of Clinical Indicators and the Top 10 Differential Bacteria. Red indicates a positive correlation, while blue indicates a negative correlation

Full size image

Discussion

The pathogenesis of DKD is complex, and knowledge about DKD remains limited. With the development of bioinformatics and high-throughput sequencing technologies, as well as the finding that urine is not sterile, some researchers have begun to pay attention to the role of the urobiome in CKD, including DKD. Kramer et al. reported that the midstream voided urine microbiome of older adults with stage 3–5 non-dialysis-dependent CKD was diverse and that greater microbiome diversity was associated with a higher eGFR [34]. Yang et al. found that patients with DKD had distinct urinary microbiota from healthy individuals, and Acidobacteria was the most prevalent microbiota in DKD [16]. However, researches on the urobiome in DKD are still rare. The urobiome of DKD still needs to be analyzed more accurately. Since 2bRAD-M can efficiently process low biomass and degraded samples, we applied this method to identify the urobiome of patients with DKD (with micro- or macroalbuminuria) and compared it with healthy individuals, also analyzing the urobiome of patients with T2DM without DKD.

Consistent with a previous finding [16],our study observed significantly lower alpha diversity in the DKD groups compared to the control group. This reduction in alpha diversity in DKD may suggest that the disease is associated with a narrowing of the microbial community, which could impact kidney function and albuminuria progression. In contrast, the control and T2DM groups exhibited higher alpha diversity, indicating that a more diverse microbial community may be present in individuals without kidney injury. Beta diversity, which compares microbial composition between different groups, provides further insights into how the urobiome varies across groups. The significant differences observed in beta diversity, as indicated by metrics such as Bray–Curtis, Euclidean distance, and Jaccard distance, highlight that the microbial communities in DKD patients are distinct from those in healthy controls and T2DM patients. This suggests that the progression of DKD is associated with specific shifts in the microbial composition of the urinary microbiome, which may be linked to the disease’s progression.

Actinomycetota, Pseudomonadota, Bacteroidota, and Bacillota_A were the dominant four phyla in all groups, collectively accounting for over 80% of the urobiome, but with substantial differences in their relative abundances across the groups. Actinomycetota and Bacteroidota were the top two phyla in healthy individuals, which is somewhat different from an earlier report that Furmicutes and Bacteroidota were the major two bacteria in healthy people [16]. In patients with T2DM, Actinomycetota was also the dominant phylum, followed by Pseudomonadota. In patients with DKD, the abundance of Actinomycetota was much lower than that in healthy individuals and those with T2DM. By contrast, Pseudomonadota became the dominant phylum in the DKD urobiome, which has not been previously reported.

At the genus level, the dominant bacterial profile also varied greatly. We found that Bifidobacterium was the most abundant in the urine of healthy individuals and patients with T2DM. As an important member of the gut microbiota, Bifidobacterium was previously reported to be depleted in patients with DKD [35, 36], but its role in urine has been rarely reported [37]. Similar to the gut microbiome, in the current study, the abundance of Bifidobacterium was also sharply reduced in patients with DKD, which indicates that Bifidobacterium may also be important in maintaining the microbiota balance in urine. Streptococcus was previously reported to be more abundant in healthy people [12]. Similarly, Streptococcus was also more abundant in healthy individuals, but showed significantly lower abundance in patients with T2DM and DKD. Lactobacillus was found to be enriched in patients with T2DM and DKD with microalbuminuria, but reduced in those with DKD with macroalbuminuria. This change, along with its positive correlation with eGFR and negative correlation with CRE, suggests that Lactobacillus may have a potential protective role in the early stage of DKD. However, to validate this hypothesis, further studies with larger sample sizes or mechanistic investigations are required. The higher abundance of Lactobacillus in T2DM and DKD1, compared to the DKD2 group, could indicate that Lactobacillus is associated with better kidney function or less severe disease, potentially contributing to maintaining the urobiome balance. In DKD with macroalbuminuria (DKD2), the absence or reduced abundance of Lactobacillus may reflect a disrupted microbial balance, potentially signaling a decline in kidney function. Thus, reduction of Lactobacillus may be associated with worsening disease severity. Another noteworthy finding is that Pseudescherichia was almost exclusively enriched in the T2DM group, and its abundance showed a negative correlation with ACR, a relationship not previously reported in the literature. Its presence in T2DM, but not in other groups, may indicate that it plays a specific role in the process of diabetes-related kidney changes, potentially affecting kidney function before the onset of more severe disease stages. However, further studies would be needed to confirm its exact role in this context. Compared with healthy individuals, the abundance of Enterococcus was also increased in patients with T2DM and DKD, especially in those with macroalbuminuria. This could be due to the changes in the urinary tract microenvironment caused by elevated urinary protein levels and kidney dysfunction, which may create favorable conditions for the overgrowth of certain bacteria such as Enterococcus. The bacterial composition also differed significantly between the two stages of DKD. In the microalbuminuria stage, Escherichia was dominant, followed by Pseudomonas_E. However, in the macroalbuminuria stage, Pseudomonas_E became dominant bacteria, followed by Enterococcus, while the abundance of Escherichia was significantly reduced. This shift from Escherichia to Pseudomonas across different stages of DKD has not been reported in previous studies. While the causal relationship between the urobiome composition and DKD remains uncertain, this findings suggest two important insights: first, the dominance of Escherichia and Pseudomonas, along with a decrease in Bifidobacterium, may be associated with DKD; second, the shift in dominance between Escherichia and Pseudomonas may be linked to the progression of DKD.

In DKD groups, distinct urinary species were identified through LEfSe and Indicator analyses. Escherichia coli emerged as the most indicative species for DKD with microalbuminuria. Although Escherichia coli is commonly associated with urinary tract infections (UTIs), previous studies, including one by Garretto et al. [38], suggest that its presence and abundance are weak predictors of UTI status and that UTI may be influenced by the overall urobiome. Other studies have indicated an association between Escherichia coli and urolithiasis [39]. In our study, UTI cases were excluded, and the increased abundance of Escherichia coli appeared to be more closely associated with T2DM and the occurrence of DKD (stage of microalbuminuria). However, once ACR exceeded 300 mg/g, indicating the macroalbuminuria stage, Escherichia coli abundance sharply declined. This decline may be attributed to the significant rise in urinary albumin levels, which could alter the bacterial composition, although this hypothesis requires further investigation.

In DKD with macroalbuminuria Pseudomonas_E oleovorans, Enterococcus faecalis, and Morganella morganii were identified as key indicative species. Among these, Morganella morganii and Enterococcus faecalis were the most distinguishing species distinguished compared to other groups, while Pseudomonas_E oleovorans was the most abundant species in DKD with macroalbuminuria. The abundance of Pseudomonas_E was positively correlated with renal function indicators (BUN, CRE, CysC, 24 h-UP and ACR) and negatively correlated to eGFR. While Pseudomonas is typically considered a pathogen linked to UTI [40], a recent study has also suggested its association with urolithiasis [39]. Our findings indicated that the abundance of Pseudomonas_E was positively correlated with declining renal function and increased urine protein levels, rather than UTI, suggesting that it could be an indicator of DKD progression.

Conclusions

In summary, urobiome diversity was significantly reduced in DKD patients (both micro- or macroalbuminuria stages) compared to healthy individuals and those with T2DM. Bifidobacterium and Streptococcus abundances were notably lower in DKD patients, with a similar reduction in Streptococcus observed in T2DM patients. In DKD, Escherichia was dominant in the microalbuminuria stage (DKD1), followed by Pseudomonas_E, whereas Pseudomonas_E became dominant in the macroalbuminuria stage (DKD2), with a marked reduction in Escherichia. These findings suggest that Escherichia may be linked to the onset of DKD, while Pseudomonas_E may indicate disease progression. Additionally, Pseudomonas_E, Enterococcus, Morganella, and Escherichia correlated with renal function indicators and urinary protein. The most indicative species for each group were identified: Streptococcus anginosus for the healthy individuals, Pseudescherichia sp002298805 for the patients with T2DM, Escherichia coli and Acinetobacter johnsonii for the DKD with microalbuminuria, and Pseudomonas_E oleovorans, Enterococcus faecalis and Morganella morganii for the DKD with macroalbuminuria.

Limitations

While this study provides new insights, certain limitations need to be acknowledged, which may affect the generalizability and robustness of the findings. First, the sample size in this study is relatively small, with only 20 participants in each group. A larger cohort with more participants is needed to validate the results and provide more robust conclusions. Second, sex imbalance in the T2DM group may introduce biases in the microbiome analysis, especially for the sex-specific microbial (Lactobacillus is predominantly found in women, while Corynebacterium or Streptococcus are more common in men). Therefore, the influence of sex imbalance cannot be excluded when analyzing these bacteria. Third, this study does not explore the precise mechanisms by which these indicative bacteria contribute to the progression of DKD. Further mechanistic studies are necessary to better understand and confirm their role in this process.

Availability of data and materials

The data sets used or analyzed in the current study could be obtained from the corresponding authors upon reasonable request.

Abbreviations

Urobiome:

Urinary microbiome

DKD:

Diabetic kidney disease

2bRAD-M:

2BRAD sequencing for microbiome

T2DM:

Type 2 diabetes mellitus

ACR:

Albumin-to-creatinine ratio

eGFR:

Estimated glomerular filtration rate

CysC:

Cystatin C

UTIs:

Urinary tract infections

CKD:

Chronic kidney disease

WMS:

Shotgun metagenome sequencing

EQUC:

Enhanced quantitative urine culture

CRE:

Creatinine

BUN:

Blood urea nitrogen

References

  1. Gupta S, Dominguez M, Golestaneh L. Diabetic kidney disease: an update. Med Clin North Am. 2023;107(4):689–705.

    Article  PubMed  Google Scholar 

  2. Sinha SK, Nicholas SB. Pathomechanisms of diabetic kidney disease. J Clin Med. 2023;12(23):7349.

    Article  PubMed  PubMed Central  Google Scholar 

  3. Vučić Lovrenčić M, Božičević S, Smirčić Duvnjak L. Diagnostic challenges of diabetic kidney disease. Biochem Med. 2023;33(3):030501.

    Google Scholar 

  4. Cheng G, et al. Molecular mechanisms of gut microbiota in diabetic nephropathy. Diabetes Res Clin Pract. 2024;213: 111726.

    Article  PubMed  Google Scholar 

  5. Han YZ, et al. Role of gut microbiota, immune imbalance, and allostatic load in the occurrence and development of diabetic kidney disease. J Diabetes Res. 2023;2023:8871677.

    Article  PubMed  PubMed Central  Google Scholar 

  6. Ma L, et al. The potential mechanism of gut microbiota-microbial metabolites-mitochondrial axis in progression of diabetic kidney disease. Mol Med. 2023;29(1):148.

    Article  PubMed  PubMed Central  Google Scholar 

  7. Das S, Gnanasambandan R. Intestinal microbiome diversity of diabetic and non-diabetic kidney disease: current status and future perspective. Life Sci. 2023;316: 121414.

    Article  PubMed  Google Scholar 

  8. Chorbińska J, et al. Urinary microbiome in bladder diseases-review. Biomedicines. 2023;11(10):2816.

    Article  PubMed  PubMed Central  Google Scholar 

  9. Elsayed NS, Wolfe AJ, Burk RD. Urine microbiome in individuals with an impaired immune system. Front Cell Infect Microbiol. 2023;13:1308665.

    Article  PubMed  Google Scholar 

  10. Jeries LM, et al. Synthesis of current pediatric urinary microbiome research. Front Pediatr. 2024;12:1396408.

    Article  PubMed  PubMed Central  Google Scholar 

  11. Roth RS, Liden M, Huttner A. The urobiome in men and women: a clinical review. Clin Microbiol Infect. 2023;29(10):1242–8.

    Article  PubMed  Google Scholar 

  12. Jayalath S, Magana-Arachchi D. Dysbiosis of the human urinary microbiome and its association to diseases affecting the urinary system. Indian J Microbiol. 2022;62(2):153–66.

    Article  PubMed  Google Scholar 

  13. Razi A, et al. Unraveling the association of bacteria and urinary stones in patients with urolithiasis: an update review article. Front Med (Lausanne). 2024;11:1401808.

    Article  PubMed  Google Scholar 

  14. Kustrimovic N, et al. The urinary microbiome in health and disease: relevance for bladder cancer. Int J Mol Sci. 2024;25(3):1732.

    Article  PubMed  PubMed Central  Google Scholar 

  15. Suarez Arbelaez MC, et al. The emerging role of the urinary microbiome in benign noninfectious urological conditions: an up-to-date systematic review. World J Urol. 2023;41(11):2933–48.

    Article  PubMed  Google Scholar 

  16. Yang Y, et al. Urinary microbiota and serum metabolite analysis in patients with diabetic kidney disease. Heliyon. 2023;9(8): e17040.

    Article  PubMed  PubMed Central  Google Scholar 

  17. Regueira-Iglesias A, et al. Critical review of 16S rRNA gene sequencing workflow in microbiome studies: from primer selection to advanced data analysis. Mol Oral Microbiol. 2023;38(5):347–99.

    Article  PubMed  Google Scholar 

  18. Yang MQ, et al. Research progress on the application of 16S rRNA gene sequencing and machine learning in forensic microbiome individual identification. Front Microbiol. 2024;15:1360457.

    Article  PubMed  PubMed Central  Google Scholar 

  19. Xia Y. Statistical normalization methods in microbiome data with application to microbiome cancer research. Gut Microbes. 2023;15(2):2244139.

    Article  PubMed  PubMed Central  Google Scholar 

  20. Isali I, et al. Current trends and challenges of microbiome research in bladder cancer. Curr Oncol Rep. 2024;26(3):292–8.

    Article  PubMed  PubMed Central  Google Scholar 

  21. Bschleipfer T, Karl I. Bladder microbiome in the context of urological disorders-is there a biomarker potential for interstitial cystitis? Diagnostics (Basel). 2022;12(2):281.

    Article  PubMed  Google Scholar 

  22. Patel SR, et al. The microbiome and urolithiasis: current advancements and future challenges. Curr Urol Rep. 2022;23(3):47–56.

    Article  PubMed  Google Scholar 

  23. Hong SY, et al. The role of microbiome: a novel insight into urolithiasis. Crit Rev Microbiol. 2023;49(2):177–96.

    Article  PubMed  Google Scholar 

  24. Alcazar CG, et al. The association between early-life gut microbiota and childhood respiratory diseases: a systematic review. Lancet Microbe. 2022;3(11):e867–80.

    Article  PubMed  PubMed Central  Google Scholar 

  25. Hong SY, et al. The renal pelvis urobiome in the unilateral kidney stone patients revealed by 2bRAD-M. J Transl Med. 2022;20(1):431.

    Article  PubMed  PubMed Central  Google Scholar 

  26. Gasiorek M, Hsieh MH, Forster CS. Utility of DNA next-generation sequencing and expanded quantitative urine culture in diagnosis and management of chronic or persistent lower urinary tract symptoms. J Clin Microbiol. 2019; 58(1).

  27. Boyanova L, et al. Urinary tract infections: should we think about the anaerobic cocci? Anaerobe. 2022;77: 102509.

    Article  PubMed  Google Scholar 

  28. Deen NS, et al. Clinical relevance of expanded quantitative urine culture in health and disease. Front Cell Infect Microbiol. 2023;13:1210161.

    Article  PubMed  PubMed Central  Google Scholar 

  29. Kim DS, Lee JW. Urinary tract infection and microbiome. Diagnostics (Basel). 2023;13(11):1921.

    Article  PubMed  Google Scholar 

  30. Sun Z, et al. Species-resolved sequencing of low-biomass or degraded microbiomes using 2bRAD-M. Genome Biol. 2022;23(1):36.

    Article  PubMed  PubMed Central  Google Scholar 

  31. ElSayed NA, et al. 2. Classification and diagnosis of diabetes: standards of care in diabetes-2023. Diabetes Care. 2023;46(Suppl 1):S19-s40.

    Article  PubMed  Google Scholar 

  32. Liu T, et al. Serum high mobility group box 1 as a potential biomarker for the progression of kidney disease in patients with type 2 diabetes. Front Immunol. 2024; 15.

  33. Sun Z, et al. Removal of false positives in metagenomics-based taxonomy profiling via targeting Type IIB restriction sites. Nat Commun. 2023; 14(1).

  34. Kramer H, et al. Diversity of the midstream urine microbiome in adults with chronic kidney disease. Int Urol Nephrol. 2018;50(6):1123–30.

    Article  PubMed  PubMed Central  Google Scholar 

  35. Wang Y, et al. The specific alteration of gut microbiota in diabetic kidney diseases—a systematic review and meta-analysis. Front Immunol. 2022;13: 908219.

    Article  PubMed  PubMed Central  Google Scholar 

  36. Hong J, et al. Specific alterations of gut microbiota in diabetic microvascular complications: a systematic review and meta-analysis. Front Endocrinol (Lausanne). 2022;13:1053900.

    Article  PubMed  Google Scholar 

  37. Bi H, et al. Urinary microbiota—a potential biomarker and therapeutic target for bladder cancer. J Med Microbiol. 2019;68(10):1471–8.

    Article  PubMed  Google Scholar 

  38. Garretto A, et al. Genomic survey of E. coli from the bladders of women with and without lower urinary tract symptoms. Front Microbiol. 2020;11:2094.

    Article  PubMed  PubMed Central  Google Scholar 

  39. Jung HD, Cho S, Lee JY. Update on the effect of the urinary microbiome on urolithiasis. Diagnostics (Basel). 2023;13(5):951.

    Article  PubMed  Google Scholar 

  40. Timm MR, Russell SK, Hultgren SJ. Urinary tract infections: pathogenesis, host susceptibility and emerging therapeutics. Nat Rev Microbiol. 2024;23:72.

    Article  PubMed  Google Scholar 

Download references

Funding

The National Natural Science Foundation of China [Grant Number 82104601, 82174144, U23A20504, 82374226], National High Level Hospital Clinical Research Funding [NO. 2023-NHLHCRFDJMS-05], Elite Medical Professionals project of China-Japan Friendship Hospital [NO.ZRJY2024-GG11] supported this study.

Author information

Authors and Affiliations

  1. Beijing Key Laboratory for Immune-Mediated Inflammatory Diseases, Institute of Clinical Medical Sciences, China-Japan Friendship Hospital, Beijing, China

    Nan Li, Ying Wang, Haojun Zhang, Liang Ma, Xinhui Huang, Yuhang Qiao, Han Li, Hailing Zhao & Ping Li

  2. Clinical Laboratory, China-Japan Friendship Hospital, Beijing, China

    Nan Li & Liang Ma

  3. Dongzhimen Hospital, Beijing University of Chinese Medicine, Beijing, China

    Ying Wang

  4. Heilongjiang Academy of Chinese Medicine Sciences, Harbin, China

    Xinhui Huang & Yuhang Qiao

  5. Guang’anmen Hospital, China Academy of Chinese Medical Sciences, Beijing, China

    Han Li

Authors
  1. Nan Li
    View author publications

    You can also search for this author inPubMed Google Scholar

  2. Ying Wang
    View author publications

    You can also search for this author inPubMed Google Scholar

  3. Haojun Zhang
    View author publications

    You can also search for this author inPubMed Google Scholar

  4. Liang Ma
    View author publications

    You can also search for this author inPubMed Google Scholar

  5. Xinhui Huang
    View author publications

    You can also search for this author inPubMed Google Scholar

  6. Yuhang Qiao
    View author publications

    You can also search for this author inPubMed Google Scholar

  7. Han Li
    View author publications

    You can also search for this author inPubMed Google Scholar

  8. Hailing Zhao
    View author publications

    You can also search for this author inPubMed Google Scholar

  9. Ping Li
    View author publications

    You can also search for this author inPubMed Google Scholar

Contributions

PL and HZ designed and implemented the experiments. YW, HX, YQ collected and prepared clinical samples. HZ and HL performed the tests. LM and YW analyzed and interpreted the data. NL and HZ were the major contributors in writing the manuscript. All authors read and approved the final manuscript.

Corresponding authors

Correspondence to Hailing Zhao or Ping Li.

Ethics declarations

Ethics approval and consent to participate

This study was approved by the Ethics Committee of China-Japan Friendship Hospital. All participants were clearly informed of the purpose and process of the trial. Informed consent was obtained from the patients for the use of their samples (2022-KY-246).

Consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Li, N., Wang, Y., Zhang, H. et al. Urobiome of patients with diabetic kidney disease in different stages is revealed by 2bRAD-M. J Transl Med 23, 414 (2025). https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s12967-025-06405-6

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s12967-025-06405-6

Keywords

Journal of Translational Medicine

ISSN: 1479-5876

Contact us