1 Introduction
Diabetes Nephropathy (DN) is widely recognized as the leading factor of Chronic Kidney Disease (CKD) and is considered one of the microvascular complications of Diabetes Mellitus (DM) (Gnudi et al., 2016). Around 30 to 40% of individuals with diabetes eventually develop DN. The prevalence of DN has demonstrated a consistent upward trend, paralleling the evolution of society and economy, alongside shifts in individual lifestyles and dietary patterns (Saeedi et al., 2019). DN is one of the key factors of end-stage renal disease (ESRD), contributing to around 54% of newly diagnosed incidents of ESRD and 30% of individuals requiring maintenance dialysis (Zhang L. et al., 2020). Moreover, DN can result in severe cardiovascular complications (Verma et al., 2020). The pathogenesis of DN is characterized by complex interplays of several factors (Samsu, 2021). At present, the treatment of DN mainly involves glycemic control and the use of pharmacological interventions, including angiotensin II receptor blockers (ARBs) or angiotensin-converting enzyme inhibitors (ACEIs), aimed at the regulation of the conditions (Doshi and Friedman, 2017; Cole and Florez, 2020). However, the risk of developing ESRD remains quite high (Anders et al., 2018; de Boer et al., 2020; Samsu, 2021). Therefore, enhancing our understanding of the pathogenic mechanisms of DN and exploring alternative therapeutic targets is imperative.
The human intestinal microbiome is described as the “second genome” that regulates health (Gilbert et al., 2018). Gut microbiota (GM) has been found to participate in the pathogenesis of various ailments by influencing the permeability of the intestinal barrier, the inflammatory response, and the balance of the immunological microenvironment (Lehto and Groop, 2018). Increasing evidence has suggested that the gut microbiota and kidney diseases can reciprocally influence each other through the induction of metabolic, immunological, and endocrine changes, a phenomenon commonly referred to as the “intestinal-renal axis” (Altamura et al., 2023). Several studies have demonstrated the substantial importance of gut microbial dysbiosis in the initiation and deterioration of DN (Wang Y. et al., 2022; Zhang et al., 2022; Zhao et al., 2023). It has been noted in several studies of GM particular alterations in the intestinal microbial composition of patients with DN, including the increase in the abundance of Clostridium and Aspergillus, and the decrease in the level of Rhodococcus (Salguero et al., 2019; Du et al., 2021; Wang Y. et al., 2022). Additionally, studies have demonstrated that a multitude of factors can contribute to the impairment of intestinal mucosa and an increase in permeability among individuals with DN, which allows the entry of metabolites such as indole and p-cresol into the bloodstream, subsequently instigating renal injury (Lehto and Groop, 2018; Lau et al., 2021). Numerous animal experiments have shown that pharmaceutical interventions induced alterations in the gut microbial composition of DN mice, subsequently leading to an amelioration of the disease’s attributes. This enhancement could potentially be attributed to mechanisms including the reduction of lipopolysaccharide (LPS)-producing microbes and the augmentation of short-chain fatty acids (SCFAs)-producing microbes (Chen et al., 2022; Deng et al., 2022).
The correlation between DN and GM has been well-established, and substantial research has been undertaken to unravel the potential mechanisms and corresponding therapeutic strategies involving GM and the kidney injury induced by their metabolites. However, the majority of these studies have mainly relied on animal models and cross-sectional investigations. Furthermore, the human GM is a complex and extensive ecosystem, which creates difficulties in identifying the causal association between certain GM and DN.
The Mendelian randomization (MR) analysis method has been widely applied in the field of epidemiological causal inference in recent years (Lehto and Groop, 2018). This method of analysis, based on Mendel’s second law, employed single-nucleotide polymorphisms (SNPs) associated with clinical phenotypes as Instrumental Variables (IVs) to build models so as to identify the causal association between exposures and outcomes at the genomic level (Davey Smith and Hemani, 2014). It achieves this without factoring in the influence of confounding variables (Holmes et al., 2017). The reliability of MR analysis has been substantiated by multiple studies. Furthermore, several studies have successfully utilized MR analysis to identify causal relationships between various exposures and outcomes successfully (Yeung and Schooling, 2021; Liu et al., 2022; Ma et al., 2022).
In this study, we aim to assess the causal association between GM and DN by employing MR analysis, thereby enhancing our understanding of the intestinal-renal axis and unveiling novel therapeutic targets for the early non-invasive diagnosis and treatment of DN.
4 Discussion
There has been a rise in the prevalence of both diabetes and DN in recent years. According to the International Diabetes Federation, there will be the largest DM population in China, India, and the United States by the year of 2030, the number of which will hit 140 million, 101 million, and 34 million respectively (Saeedi et al., 2019). Concurrently, along with the proposition of concept of the gut-kidney axis and the subsequent in-depth investigations therein, the substantive role of GM dysbiosis in the intricate pathogenesis of DN has been progressively unveiled. Despite these progressions, the conclusive establishment of a causal association between GM and DN remains an unresolved inquiry. In this study, MR analysis was employed to comprehensively identify the underlying causal association between GM and DN. The results of this study elucidated causal associations between thirteen distinct GM species and DN. Class Verrucomicrobiae, Order Verrucomicrobiales, Family Verrucomicrobiaceae, Genus Akkermansia, Genus Catenibacterium, Genus Coprococcus 1, Genus Eubacterium hallii group and Genus Marvinbryantia were associated with an elevated risk of DN. Conversely, Class Actinobacteria, Group Eubacterium ventriosum group, Group Ruminococcus gauvreauii group, Order Lactobacillales, Phylum Proteobacteria were associated with a reduced risk of DN. Notably, by integrating the results of this study with previous research, we found that that dysbiosis of the gut microbiota and its metabolic byproducts play significant roles in the progression of DN through various mechanisms.
Among patients with DN, the co-occurrence of systemic inflammation and compromised innate immunity has been noted (Han et al., 2023). Research has indicated that the gut microbiota, inhabiting the gastrointestinal tract, can modulate antigen reactivity in lymphoid tissues, thus initiating and gradually maturing the intestinal immune system (Cerf-Bensussan and Eberl, 2012). Specifically, genera such as Bacteroides, Bifidobacterium, Lactobacillus, and Bacillus proteus have been found to notably contribute to immune system enhancement (Wu and Wu, 2012). The dysregulation of the gut microbiota can influence the maturation of macrophages, prompting the release of tumour necrosis factor-alpha (TNF-α) and interleukin-6 (IL-6) upon Toll-like receptor (TLR) stimulation, thereby triggering renal inflammation (Yang et al., 2019). Additionally, this imbalance may activate innate immune cells, amplifying the activity of TLR-2 and TLR-4 pathways, and fostering the production of inflammatory cytokines. Such dysbiosis in patients with DN could potentially lead to immune dysfunction and renal injury (Lin et al., 2012; Mudaliar et al., 2013; Donaldson et al., 2016; Chi et al., 2021). Concurrently, compromised immune function may diminish the body’s defence capabilities and heighten susceptibility to infections (Robinson and Freedman, 2018). Urinary tract infections, a frequent complication in individuals with diabetes, may be linked to systemic inflammation associated with hyperglycaemia and dysbiosis of the gut microbiota, thereby exacerbating renal damage in DN (Micle, 2020; Wang C. et al., 2023).
The GM exerted an impact on the body’s metabolic processes and obesity via insulin resistance (IR) (Hoseini Tavassol et al., 2023). Previous research underscored the positive correlation between the severity of IR and the incidence of DN (Penno et al., 2021). Furthermore, it was established that IR contributed to the pathogenesis of DN independently of hyperglycaemia, as it could elicit an increased salt sensitivity (Karalliedde and Gnudi, 2016). Simultaneously, IR led to a reduction in glucose transport by podocytes, which could precipitate the disruption of the glomerular filtration barrier and the onset of proteinuria (Alqallaf et al., 2022).
SCFAs represent a subset of saturated fatty acids characterized by no more than six carbon atoms, predominantly encompassing acetic, propionic, and butyric acids (Zhang et al., 2021). And SCFAs are generated by GM. These chemicals have multifaceted functions, primarily in the regulation of energy metabolism, maintenance of the intestinal epithelial barrier, facilitation of immune responses, and modulation of inflammatory processes (Du et al., 2022). Prior investigations demonstrated diminished serum concentrations of total SCFAs in individuals diagnosed with DN (Zhong et al., 2021; Cai et al., 2022). SCFAs exerted their influence as signalling moieties, engaging a dedicated G-protein-coupled receptor 43 (GPR43), in which activation of GPR43 signalling instigated a regulation of inflammatory cascades. This effect manifested as a reduction in levels of pro-inflammatory cytokines within the tissues of the colon, thus contributing to the preservation of intestinal homeostasis (Sun et al., 2017; Priyadarshini et al., 2018). Moreover, SCFAs can augment neutrophil chemotaxis and facilitate the differentiation and proliferation of natural killer cells and regulatory T cells, thus activating the immune system (Kim et al., 2014). Concurrently, the augmentation in the abundance of urease-producing bacteria precipitated a surge in intestinal pH, consequently elevating the permeability of the intestinal mucosa (Lehto and Groop, 2018; Thaiss et al., 2018; Lau et al., 2021). This facilitated the entry of specific uremic toxin precursors, including cresol, indole, and trimethylamine into the systemic circulation, thereby facilitating the production of uremic toxins. These entities, in turn, elicited oxidative stress and fostered renal tubulointerstitial fibrosis, resulting in a progressive decline in renal function (Kikuchi et al., 2019; Tan et al., 2021). However, SCFAs can preserve intestinal homeostasis and consequently mitigate the absorption of harmful substances through maintaining the structural integrity of the intestinal mucosa, thus bestowing a protective impact on renal function. Furthermore, it has been revealed that SCFAs, especially butyric acid, can enhance tight junction protein complex, thereby maintaining the perpetuation of the intestinal barrier’s functional integrity (Xia et al., 2020). This mechanistic facet holds potential in averting renal impairments. Furthermore, a hyperglycaemic state typically caused the body to generate an overproduction of reactive oxygen species (ROS), resulting in the development of DN (Miranda-Díaz et al., 2016). It was proposed that SCFAs had the capacity to inhibit the stimulation of glomerular mesangial cells caused by hyperglycaemia and lipopolysaccharides (LPS). Concurrently, SCFAs were found to diminish the generation of ROS and malondialdehyde (MDA), along with inflammatory factors, while elevating the level of superoxide dismutase (SOD). These activities mitigated inflammatory responses, ultimately safeguarding renal function (Zhang et al., 2022). Previous research outlined the role of SCFAs in mitochondrial biosynthesis processes (Andrade-Oliveira et al., 2015). This finding suggested that SCFAs could potentially have a positive impact on alleviating renal epithelial cell hypoxia. Moreover, SCFAs treatment improved renal insufficiency in a murine model of acute kidney injury subsequent to renal ischemia-reperfusion (Antza et al., 2018).
Prior research identified an increased prevalence of the Order Lactobacillales and the Class Actinobacteria in the intestines of patients with DN (Du et al., 2021; Wang Y. et al., 2022). This observation was attributed to the use of hypoglycaemic medications (Forslund et al., 2015; Gu et al., 2017). Bifidobacterium and Lactobacillus are widely acknowledged as significant intestinal probiotics (Bindels et al., 2015; Markowiak-Kopeć and Śliżewska, 2020), generating lactic acid and acetic acid within the intestinal environment, thereby safeguarding the host from invasion by intestinal pathogens (Zhang et al., 2015; Roy and Dhaneshwar, 2023). It was found that an elevation in the abundance of the Class Actinobacteria could lead to heightened production of SCFAs, primarily due to the presence of Bifidobacterium (Binda et al., 2018). Lactobacillus, a notable constituent of the Order Lactobacillales, possesses the capability to convert lactic acid into diverse SCFAs forms. In patients with CKD, a reverse relationship was observed between Lactobacillus levels and markers of kidney impairment, including blood creatinine and urea nitrogen (Ren et al., 2020). These findings imply that Order Lactobacillales and Class Actinobacteria serve as protective factors against DN. This concurs with our own research findings and deepens the understanding of the pivotal role these microbiotas play in DN.
The Phylum Proteobacteria represents one of the most expansive and phenotypically diverse phyla within the field of bacteria (Sharma et al., 2022), and it is considered to have the potential for pathogenicity under specific circumstances (Du et al., 2021). The findings from the present investigation elucidate a correlation between this particular bacterial cohort and a lower risk of DN. Previous research demonstrated an association between the Phylum Proteobacteria and an increased synthesis of SCFA production (Machate et al., 2020). Moreover, an alteration in the abundance of the Phylum Proteobacteria was observed within the intestinal environment of patients with DN, exhibiting a disparity when compared to the composition found in healthy populations (Wang Y. et al., 2022). In alignment with the outcomes of the current study, these findings propose that this genus is linked with and held the potential to function as a protective factor for the progression of DN.
The Genus Eubacterium ventriosum group is a component of the Family Lachnospiraceae. The Genus Ruminococcus gauvreauii group belongs to the Family Ruminococcaceae and is a component of the Phylum Firmicutes (Zhang et al., 2023). The association between these two genera and DN has not been examined by conventional epidemiological studies. Nevertheless, the outcomes derived from the present study suggest that both genera are associated with a reduced risk of DN. This assertion is likely attributable to their membership within the community of microbiota known for producing SCFAs (Kemp et al., 2021; Amamoto et al., 2022; Wang J. et al., 2023). This discovery offers a novel perspective on the involvement of these two genera in DN.
In healthy individuals, the Genus Akkermansia (Class Verrucomicrobiae, Order Verrucomicrobiales, Family Verrucomicrobiaceae) typically constitutes approximately 3 and 5 % of the gastrointestinal microbial community (Hasani et al., 2021). Currently, the Genus Akkermansia represents the only Phylum Verrucomicrobia found in the human intestine. As a result, many 16S rRNA gene sequence analyses considered the Phylum Verrucomicrobia to be a representative of Genus Akkermansia (Cani et al., 2022). It has been demonstrated that there is a correlation between a reduction in the abundance of the Genus Akkermansia, a promising probiotic, and the development of various illnesses, including type 2 diabetes and inflammatory bowel disease (Wang K. et al., 2022). In addition, the level of Genus Akkermansia was negatively correlated with the level of IR (Shin et al., 2014; Watanabe et al., 2023). Moreover, it should be noted that Genus Akkermansia also has the capacity to produce SCFAs (Luo et al., 2022), but its role remains somewhat controversial. In animal models of DN and patients with CKD, the level of Genus Akkermansia was positively correlated with markers of kidney injury, including blood creatinine and urea nitrogen (Forslund et al., 2015; Lan et al., 2023). Additionally, it was also found that the relative abundance of Genus Akkermansia was positively correlated with the severity of the disease in patients with Parkinson’s disease and multiple sclerosis (Jangi et al., 2016; Cekanaviciute et al., 2017; Zhang F. et al., 2020). Based on the results of the current study, there was a positive association between Genus Akkermansia and the risk of developing DN. This may be due to the fact that the relative abundance of Gram-negative bacteria is increased in the gastrointestinal tract of patients with DN, including Genus Akkermansia. Gram-negative bacteria contain LPS as components of the outer membrane of the cell wall. As the intestinal barrier was compromised, LPS entered the circulation and stimulates the body to produce excessive pro-inflammatory factors (e.g., IL-1β, IL-6, and TNFα, etc.), which further exacerbated the systemic inflammatory response in patients with DN (Salguero et al., 2019). Finally, it was demonstrated that LPS could bind to TLR, thereby activating the TLR-4 signaling pathway. This activation subsequently triggered downstream signaling pathways, resulting in tissue damage through mechanisms including oxidative stress and DNA damage (Hung et al., 2017).
To date, no study has reported alterations in the abundance of Genus Catenibacterium, Genus Coprococcus 1, Genus Eubacterium hallii group, and Genus Marvinbryantia in the intestinal tract of DN patients. The Genus Catenibacterium belongs to the Family Erysipelotrichaceae (Ricci et al., 2023). According to the results of this study, the Genus Catenibacterium is a risk factor for DN. This genus was found to be associated with several metabolic diseases (Burakova et al., 2022). Moreover, an increased prevalence of this genus was identified in faecal samples from individuals with ESRD, suggesting a potential correlation between the levels of this genus and the progression of nephropathy (Vaziri et al., 2013). There are very few studies on the Genus Coprococcus 1. However, the results of the current study suggest that this genus is positively associated with the risk of DN, warranting comprehensive elucidation of its underlying mechanisms. Genus Eubacterium hallii group is a member of the Family Lachnospiraceae within the Phylum Firmicutes, and as one of the butyrate-secreting genera, it has the potential to ameliorate DM and other diseases associated with IR (Zhang et al., 2015). Similarly, an elevated abundance of Genus Marvinbryantia was found to align with diminished IR levels, which could also produce butyrate (Chen et al., 2021). This indicates the potential benefits of both Genus Eubacterium hallii group and the Genus Marvinbryantia in ameliorating the condition of DN. Nevertheless, the present study reveals a positive correlation between these microbiotas and the risk of DN, potentially attributable to complex gene-gene and gene-environment interactions.
To the best of our knowledge, this is the first study to employ MR to assess the causal association between GM and DN. The implementation of this analytical methodology serves to reduce potential biases arising from reverse causation and residual confounding. This study provides confirmation of a causal association between GM and DN while delving into the plausible mechanistic pathways of intestinal dysbiosis in DN. These findings underscore the significance for nephrologists to exercise vigilant attention over DN patients exhibiting intestinal dysbiosis in their clinical practice. Furthermore, the identification of distinct GM associated with DN within this study provides novel biomarkers for the prevention, diagnosis and therapeutic intervention of DN, thereby enhancing our comprehension of the gut-renal axis.
However, this study has several limitations. Firstly, the predominant participants within the GWAS cohort were European ancestry, which might potentially affect the applicability of study findings across diverse ethnicities. Secondly, the available data pertaining to GM were solely categorized at a taxonomic level higher than the genus, hence limiting the ability to identify causal connections between GM and DN at more specific taxonomic levels such as species or strain. Additionally, there may be a partial overlap of data on SNPs present in GM across distinct taxonomic tiers, which may impact the reproducibility of the results of MR analyses. Lastly, some of the GM identified in this study have not been previously identified as directly associated with DN, and the findings of this study of certain GM fail to align with the results of prior studies. Hence, further population-based prospective studies and experiments are required to investigate the potential biological mechanisms between these GM and DN.