Genetic Insights into Associations Between Bowel Diseases and Chronic Kidney Disease by Two-Sample Mendelian Randomization

doi:10.21203/rs.3.rs-4813305/v1

Download PDF

Article

Genetic Insights into Associations Between Bowel Diseases and Chronic Kidney Disease by Two-Sample Mendelian Randomization

https://doi.org/10.21203/rs.3.rs-4813305/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Objective

This study aims to investigate the genetic causal relationships between gastrointestinal diseases—specifically celiac disease (CeD), and inflammatory bowel disease (IBD)—and chronic kidney disease (CKD).

Methods

We conducted a two-sample Mendelian randomization (MR) analysis using publicly available genome-wide association studies (GWAS) data. Two sets of single-nucleotide polymorphisms (SNPs) were chosed as instrumental variables(IVs), 32 SNPs related with CeD and 68 SNPs associated IBD. The primary analysis utilized the inverse variance weighted (IVW) method, supplemented by MR-Egger and weighted median approaches.

Results

Our findings indicate significant genetic causal effects of CeD and IBD on the risk of developing CKD. The IVW method showed a positive association between CeD and CKD (OR = 1.021, 95% CI = 1.002–1.041, P = 0.032), with stronger effects observed for IBD (OR = 1.051, 95% CI = 1.014–1.089, P = 0.006). Reverse MR results of CKD on CeD (P = 0.435; OR = 0.939) and CKD on IBD (P = 0.166; OR = 1.120) were not statistically significant.

Conclusions

The study provides genetic evidence linking gastrointestinal diseases to an increased risk of CKD. These findings highlight the importance of considering genetic predispositions when assessing CKD risk in patients with CeD and IBD.

Mendelian Randomization

Celiac Disease

Inflammatory Bowel Disease

Chronic Kidney Disease

Inflammatory bowel disease (IBD), which includes conditions such as Crohn’s disease (CD)¹ and ulcerative colitis (UC)², alongside celiac disease (CeD), are increasingly recognized for their global health impact. IBD is understood as a complex systemic disorder with extraintestinal effects that extend to cardiovascular, dermatological, and hepatobiliary systems³. CeD, a chronic autoimmune enteropathy known for gastrointestinal distress and malabsorption⁴, similarly presents various non-intestinal symptoms that affect the kidneys and skin. Previously thought to be a mere food sensitivity, CeD is now classified alongside autoimmune conditions like rheumatoid arthritis and type 1 diabetes, owing to its strong genetic ties to specific HLA allotypes crucial for antigen presentation.

Recent research has broadened our insight into how these diseases may predispose patients to renal complications, beyond the direct nephrotoxic effects of treatment. Epidemiological data, for instance, have underscored a linkage between IBD and an elevated risk of chronic kidney disease (CKD), as evidenced by a significant correlation in a retrospective cohort study of over 80,000 subjects⁵. Additionally, studies from Sweden have shown a persistent association between IBD and IgA nephropathy, which can lead to end-stage renal disease (ESRD)⁶. In the context of CeD, dietary adjustments that exclude gluten have been noted to alleviate IgA nephropathy in some cases⁷, suggesting that dietary management could be beneficial for renal health in these patients.

Observational studies, however, often struggle with confounding factors that obscure causal inference. To overcome these challenges, this study utilizes Mendelian randomization (MR), leveraging genetic variants as instrumental variables to establish causality, thus minimizing potential biases from confounding and reverse causation⁸. Through a two-sample MR approach using data from extensive genome-wide association studies (GWAS), this analysis aims to decisively examine the causal relationships between IBD, CeD, and CKD.

Study design

We performed a two-sample MR test to explore the causality of CeD and IBD on CKD. An overview of the study design is shown in Fig. 1A. In the MR analysis of the two samples, there are three main assumptions: (1) single nucleotide polymorphisms (SNPs) must be closely related to exposure. (2) SNPs must not be associated with confounding that may affect exposure to the results. (3) SNPs can only be associated with results through exposure, but not in other ways⁸. The MR study was conducted conforming to the “STROBE-MR” guidelines⁹.

Data sources description

The datasets employed in this study from genome-wide association studies (GWASs; https://gwas.mrcieu.ac.uk/) are openly accessible as listed in Supplementary Table S1.

Study exposure

CeD data were from a large GWAS dataset that included 12,041 individuals of European population, including men and women (97,422 SNPs, 11,812 cases and 229 controls). And for IBD, 9,619,016 SNPs, 25,042 cases and 34,915 controls were considered.

Study outcomes

GWAS summary statistics were chosen as outcome data for CKD including 117,165 (12,385 cases and 104,780 controls, 2,179,497 SNPs) participants. We selected studies only included individuals of European ancestry to avoid population bias. Since all identified data sets involved in this study are publicly available, no additional ethical approval or informed consent was required.

Selection of genetic instrumental variables

The SNPs were employed as instrumental variables (IVs) at the genome-wide significance threshold (P < 5×10^–8). To achieve independent loci, these SNPs were clumped with a linkage disequilibrium (LD) threshold r² = 0.001 and 10000kb as clumping window. Then we harmonized the exposure and outcome SNPs and deleted the potential palindromic SNPs. The SNPs were searched from the outcome data to eliminate the effect of confounding factors of CKD. The F-statistics and R²-value were calculated to estimate the strength of IVs. These values can identify whether IVs are strong enough to lessen the weak-tool bias. After all, we carried out MR-Egger intercepts and MR-PRESSO (MR pleiotropy residual sum and outlier), to identify and exclude potential outliers.

MR Analysis

The causal relationship between exposures and outcome was investigated using several MR approaches, namely the inverse variance weighted (IVW) method, MR-Egger, weighted median, Simple mode, and Weighted mode. IVW was chosen as the primary analytical method for its strong detection power of causal relationships¹⁰. Furthermore, the casual effects of CD and UC on CKD were estimated by subgroup analysis.

All statistical analyses were performed using R packages, including Two-Sample MR (version 0.6.6) and MR-PRESSO (version 1.0) on R platform (version 4.2.2). P < 0.05 was considered to be statistically significant.

Sensitivity analyses

We implied Cochran’Q test to estimate the heterogeneity. The MR-Egger intercepts and MR-PRESSO were utilized to determine whether directional or horizontal pleiotropy is existed¹¹. A leave-one-out analysis was performed to evaluate whether the MR estimate was biased by a single SNP. In order to minimize the bias inherent in the genetic data, we chose another CKD GWAS data for sensitivity analysis.

Genetic Instruments Selection

Our study utilized 32 CeD-specific SNPs and 68 SNPs related to IBD as instrumental variables (IVs). 38 CeD-specific SNPs, and 214 SNPs related to IBD were identified for replication analysis as IVs. Furthermore, there were also 52 SNPs linked to CD and 3 SNPs associated with UC in subgroup analysis. These SNPs were chosen according to strict criteria, ensuring robustness with an F-statistic for each variant exceeding 10, which indicates a low risk of bias¹². The specific details regarding these SNPs can be found in Supplementary Table S2.

Causal Effects between IBD on CKD

The IVW method indicated a statistically significant positive association between CeD and increased risk of CKD (Odds Ratio [OR] = 1.021; 95% Confidence Interval [CI] = 1.002–1.041; P = 0.032; Fig. 1B; Table 1). Similarly, positive genetic causal effect was observed for overall IBD on CKD risk. For IBD, the OR was 1.051 with a CI of 1.014–1.089 (P = 0.006; Fig. 1B; Table 1).

Table 1

MR analysis of the causality of CeD and IBD on CKD.
Outcome	Exposure	Method	nSNPs	β	Pval	OR	or_lci95	or_uci95
Chronic kidney disease \|\| id:ebi-a-GCST003374	Celiac disease \|\| id:ebi-a-GCST005523	MR Egger	32	0.015	0.328	1.015	0.986	1.045
		Weighted median	32	0.018	0.166	1.018	0.993	1.044
		Inverse variance weighted	32	0.021	0.032	1.021	1.002	1.041
		Simple mode	32	0.013	0.614	1.013	0.964	1.064
		Weighted mode	32	0.020	0.091	1.020	0.998	1.044
	Inflammatory bowel disease \|\| id:ebi-a-GCST004131	MR Egger	68	0.052	0.431	1.053	0.926	1.198
		Weighted median	68	0.042	0.110	1.043	0.990	1.098
		Inverse variance weighted	68	0.050	0.006	1.051	1.014	1.089
		Simple mode	68	0.040	0.501	1.041	0.927	1.168
		Weighted mode	68	0.052	0.373	1.053	0.940	1.180
nSNPS: number of SNPs used in MR; OR: odds radio; CI: confidence interval.

For sensitivity analysis on another CKD database, positive causal effects were also corroborated for CeD and overall IBD. Specifically, for CeD, the OR was 1.032 with a CI of 1.000-1.064 (P = 0.049), and for IBD, the OR was 1.045 with a CI of 1.009–1.083 (P = 0.015). MR results observed CD and UC had positive causal relationship with CKD. For CD, the OR was 1.048 with a CI of 1.008–1.090 (P = 0.020), and for UC, the OR was 1.104 with a CI of 1.003–1.216 (P = 0.044). The detailed MR results are listed in Supplementary Table S3 and Supplementary Figure S1.

Alternative MR methods, including MR-Egger and WM, corroborated the direction of the IVW findings across all diseases (Supplementary Table S3). The scatter plots illustrating these MR findings are displayed in Fig. 1C-D. Reverse two-sample MR results did not observe the casual relationship of CKD on CeD (P = 0.435; OR = 0.939). The reverse MR result of CKD on IBD was also negative (P = 0.166; OR = 1.120). The scatter plots of reverse MR results are shown in Supplementary Figure S2.

Assessment of Potential Pleiotropy and Heterogeneity

The MR-Egger intercepts suggested no presence of horizontal pleiotropy across the conditions studied (For CeD, intercept = 0.003, P = 0.557; For IBD, intercept = -0.0003, P = 0.969; Supplementary Table S4). The MR-PRESSO test further supported the absence of pleiotropic bias affecting the causality estimates (P > 0.05; Supplementary Table S4). Additionally, Cochran’s Q test indicated no significant heterogeneity among the instrumental variables (P > 0.05; Supplementary Table S4), and the funnel plots confirmed this finding (Supplementary Figure S3). Lastly, the leave-one-out analysis did not identify any individual SNP that significantly influenced the overall outcomes (Supplementary Figure S4).

This Mendelian randomization study provides compelling evidence supporting a genetic causal link between gastrointestinal diseases, specifically CeD and IBD (CD, UC) collectively, and CKD. The robust findings from our two-sample MR analysis, employing genetic variants as instrumental variables, emphasize the importance of genetic predispositions in influencing kidney health in patients with these conditions. The use of 32 CeD-specific SNPs and 68 SNPs related to IBD as instrumental variables, along with additional replication analyses and subgroup analyses, ensures the robustness and reliability of these results. The statistically significant positive associations found through the IVW method, along with the corroborative results from alternative MR methods, underscore the potential genetic pathways that link bowel diseases to increased CKD risk. However, it is important to note that the reverse two-sample MR results did not observe a causal relationship of CKD on CeD or on IBD. These insights are crucial for understanding the genetic underpinnings of CKD in patients with gastrointestinal diseases and could inform future research and therapeutic strategies.

The observed associations highlight several critical insights. First, the positive causal effect of CeD on CKD risk, though modest, suggests that systemic autoimmune responses in CeD may contribute to renal complications. This aligns with the systemic nature of autoimmune disorders where inflammatory processes potentially extend beyond the primary disease locus¹³. For IBD, including CD and UC, the stronger associations suggest more pronounced renal involvement, possibly due to chronic inflammation or immune dysregulation common in these conditions¹⁴. The findings for CD and IBD were interesting, underscoring the need to monitor kidney function in these patients proactively.

The underlying mechanisms might involve chronic systemic inflammation, immune complex deposition, or direct autoimmune attack, all of which are known to contribute to renal pathology¹⁵. The shared genetic pathways, indicated by HLA associations in CeD and similar immunological markers in IBD, could further elucidate the cross-talk between gastrointestinal and renal systems¹⁶. Future research should explore these pathways in greater detail, potentially guiding the development of targeted therapies that could reduce the risk of CKD in these populations.

One of the strengths of this study is the use of MR to overcome the limitations of observational studies, which are often plagued by confounding factors. However, the study is not without limitations. The genetic instruments used, while robustly associated with CeD, CD, UC, and IBD, may not capture all genetic factors influencing the disease pathways, and the possibility of unmeasured pleiotropy cannot be entirely excluded despite our analytical approaches.Further studies should focus on validating these findings in diverse populations and exploring the impact of specific therapies on the progression of CKD in patients with gastrointestinal diseases. Additionally, integrating these genetic findings with environmental factors could provide a more comprehensive understanding of the disease mechanisms at play.

However, our study contrasted with findings from a MR study¹⁷, which did not find a significant association between IBD and CKD but identified a strong link between IBD and IgA nephropathy. This discrepancy underscores the complexity of the relationship between gastrointestinal diseases and kidney disorders. IgA nephropathy, a specific kidney condition, may have a more direct association with IBD compared to CKD, which encompasses a broader range of kidney pathologies. This highlights the necessity of considering specific kidney outcomes when investigating the impact of gastrointestinal diseases. Furthermore, our study did not find potential significant reverse causation, where CKD could predispose individuals to IBD. A recent study¹⁸ reported that lower estimated glomerular filtration rate (eGFR), a marker of CKD, was associated with a higher risk of subsequently developing IBD. This bidirectional relationship suggests that kidney disease could influence the development of gastrointestinal conditions and vice versa. Including reverse MR analyses in future research could provide a more comprehensive understanding of these complex interactions and help to delineate the directionality of these associations. Additionally, while we addressed pleiotropy and heterogeneity in our analysis, the potential impact of important clinical factors, such as medications and environmental influences, on CKD risk was not thoroughly discussed. Medications commonly used in IBD, like Sulfasalazine¹⁹ and immunosuppressants like Cyclosporine²⁰ or Methotrexate²¹, can have nephrotoxic effects, potentially confounding the observed associations. Similarly, environmental factors such as diet, lifestyle, and comorbidities could play significant roles in the risk of CKD among these patients. Integrating these clinical and environmental aspects into future studies could enhance the validity and applicability of the findings.

Despite the compelling evidence, our study has several limitations. The modest effect sizes and the potential for reverse causation suggest that these findings should be viewed as part of a larger, more complex picture. Additionally, the need to consider broader clinical and environmental factors is essential for a comprehensive understanding. Future research should aim to validate these findings across diverse populations and integrate genetic data with clinical and environmental factors. This integrated approach could inform targeted interventions and improve patient care strategies for individuals with these interconnected conditions.

In conclusion, our study provides potential genetic evidence linking gastrointestinal diseases with CKD risk. The findings employing genetic variants as instrumental variables, highlight the genetic predispositions influencing kidney health in patients with CeD and IBD (CD, UC). These results underscore the significance of genetic factors in the relationship between gastrointestinal diseases and CKD.

Data Availability Statement

The analyses for this study were based on publicly available summary statistics. The datasets generated and analyzed during the current study are available in the IEU open GWAS project (https://gwas.mrcieu.ac.uk/).

Acknowledgments

There are no acknowledgments to declare.

Author Contributions

XC created the study concept and designed the studies. LX conducted data analysis. LX, TH, ZC, LJ, XJ, ZL, MZ, and XC wrote and approved the final version of the manuscript.

Competing Interests Statement

All the authors declared no competing interests.

Fernandez A, Gonzalez L, de la Fuente J. Coeliac disease: Clinical features in adult populations. REV ESP ENFERM DIG. 2010;102(8):466-471. doi:10.4321/S1130-01082010000800002
Hodson R. Inflammatory bowel disease. NATURE. 2016;540(7634):S97-S97. doi:10.1038/540S97a
Rogler G, Singh A, Kavanaugh A, Rubin DT. Extraintestinal Manifestations of Inflammatory Bowel Disease: Current Concepts, Treatment, and Implications for Disease Management. GASTROENTEROLOGY. 2021;161(4):1118-1132. doi:10.1053/j.gastro.2021.07.042
Lebwohl B, Sanders DS, Green PHR. Coeliac disease. LANCET. 2018;391(10115):70-81. doi:10.1016/S0140-6736(17)31796-8
Vajravelu RK, Copelovitch L, Osterman MT, et al. Inflammatory Bowel Diseases Are Associated With an Increased Risk for Chronic Kidney Disease, Which Decreases With Age. CLIN GASTROENTEROL H. 2020;18(10):2262-2268. doi:10.1016/j.cgh.2019.10.043
Rehnberg J, Symreng A, Ludvigsson JF, Emilsson L. Inflammatory bowel disease is more common in patients with IgA nephropathy and predicts progression of ESKD: A swedish population-based cohort study. J AM SOC NEPHROL. 2021;32(2):411-423. doi:10.1681/ASN.2020060848
Cheung CK, Barratt J. Gluten and IgA nephropathy: you are what you eat? KIDNEY INT. 2015;88(2):215-218. doi:10.1038/ki.2015.149
Davies NM, Holmes MV, Davey Smith G. Reading Mendelian randomisation studies: a guide, glossary, and checklist for clinicians. BMJ-BRIT MED J. 2018;362:k601-k601. doi:10.1136/bmj.k601
Skrivankova VW, Richmond RC, Woolf BAR, et al. Strengthening the reporting of observational studies in epidemiology using mendelian randomisation (STROBE-MR): explanation and elaboration. BMJ. 2021;375:n2233-n2233. doi:10.1136/bmj.n2233
Bowden J, Smith GD, Burgess S. Mendelian randomization with invalid instruments: Effect estimation and bias detection through Egger regression. INT J EPIDEMIOL. 2015;44(2):512-525. doi:10.1093/ije/dyv080
Burgess S, Thompson SG. Interpreting findings from Mendelian randomization using the MR-Egger method. Eur J Epidemiol. 2017;32(5):377-392. doi:10.1007/s10654-017-0255-x
Burgess S, Thompson SG. Avoiding bias from weak instruments in Mendelian randomization studies. INT J EPIDEMIOL. 2011;40(3):755-764. doi:10.1093/ije/dyr036
Pohl D, Benseler S. Systemic inflammatory and autoimmune disorders. Handb Clin Neurol. 2013;112:1243-52. doi:10.1016/b978-0-444-52910-7.00047-7
Van Sommeren S, Janse M, Karjalainen J, et al. Extraintestinal manifestations and complications in inflammatory bowel disease: From shared genetics to shared biological pathways. INFLAMM BOWEL DIS. 2014;20(6):987-994. doi:10.1097/MIB.0000000000000032
Oikonomou K, Kapsoritakis A, Eleftheriadis T, Stefanidis I, Potamianos S. Renal manifestations and complications of inflammatory bowel disease. INFLAMM BOWEL DIS. 2011;17(4):1034-1045. doi:10.1002/ibd.21468
Karell K, Louka AS, Moodie SJ, et al. HLA types in celiac disease patients not carrying the DQA1*05-DQB1*02 (DQ2) heterodimer: results from the European Genetics Cluster on Celiac Disease. Hum Immunol. Apr 2003;64(4):469-77. doi:10.1016/s0198-8859(03)00027-2
Lian X, Wang Y, Wang S, et al. Does inflammatory bowel disease promote kidney diseases: a mendelian randomization study with populations of European ancestry. BMC Med Genomics. Sep 26 2023;16(1):225. doi:10.1186/s12920-023-01644-2
Yang Y, Ludvigsson JF, Olén O, Sjölander A, Carrero JJ. Estimated Glomerular Filtration Rate and the Risk of Inflammatory Bowel Disease in Adults: A Swedish Population-Based Study. Inflamm Bowel Dis. May 2 2024;30(5):718-725. doi:10.1093/ibd/izac267
Niknahad H, Heidari R, Mohammadzadeh R, et al. Sulfasalazine induces mitochondrial dysfunction and renal injury. RENAL FAILURE. 2017;39(1):745-753. doi:10.1080/0886022X.2017.1399908
Cattaneo D, Perico N, Gaspari F, Remuzzi G. Nephrotoxic aspects of cyclosporine. TRANSPL P. 2004;36(2):S234-S239. doi:10.1016/j.transproceed.2004.01.011
Campistol JM, Sacks SH. Mechanisms of nephrotoxicity. TRANSPLANTATION. 2000;69(12):SS5-SS10. doi:10.1097/00007890-200006271-00002

No competing interests reported.

SupplementaryMaterial.pdf

Download PDF

Reviews received at journal
15 Oct, 2024
Reviewers agreed at journal
15 Oct, 2024
Reviewers invited by journal
25 Sep, 2024
Editor assigned by journal
25 Sep, 2024
Editor invited by journal
07 Aug, 2024
Submission checks completed at journal
06 Aug, 2024
First submitted to journal
27 Jul, 2024

You are reading this latest preprint version

Genetic Insights into Associations Between Bowel Diseases and Chronic Kidney Disease by Two-Sample Mendelian Randomization

Status:

Version 1

Abstract

Objective

Methods

Results

Conclusions

Figures

Introduction

Methods

Study design

Data sources description

Study exposure

Study outcomes

Selection of genetic instrumental variables

MR Analysis

Sensitivity analyses

Results

Genetic Instruments Selection

Causal Effects between IBD on CKD

Assessment of Potential Pleiotropy and Heterogeneity

Discussion

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1