- Email: [email protected]
A farnesoid X receptor polymorphism predisposes to spontaneous bacterial peritonitis
G Model YDLD-2680; No. of Pages 4
ARTICLE IN PRESS Digestive and Liver Disease xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
Digestive and Liver Disease journal homepage: www.elsevier.com/locate/dld
Short Report
A farnesoid X receptor polymorphism predisposes to spontaneous bacterial peritonitis Philipp Lutz a,d,∗ , Cordula Berger a , Bettina Langhans a , Frank Grünhage c , Beate Appenrodt c , Jacob Nattermann a,d , Frank Lammert c , Achim Hoerauf b,d , Tilman Sauerbruch a , Christian P. Strassburg a,d , Ulrich Spengler a,d , Hans Dieter Nischalke a a
Department of Internal Medicine I, University of Bonn, Bonn, Germany Institute for Medical Microbiology, Immunology and Parasitology, University of Bonn, Bonn, Germany c Department of Medicine II, Saarland University Medical Center, Saarland University, Homburg, Germany d German Center for Infection Research, Germany b
a r t i c l e
i n f o
Article history: Received 27 March 2014 Accepted 4 July 2014 Available online xxx Keywords: Ascites Farnesoid X receptor Liver cirrhosis Spontaneous bacterial peritonitis
a b s t r a c t Background: In mice, the farnesoid X receptor is involved in bacterial translocation, which can result in spontaneous bacterial peritonitis in patients with cirrhosis. We investigated if polymorphisms in the farnesoid X receptor gene influence the risk for spontaneous bacterial peritonitis. Methods: Laboratory and clinical data of 293 cirrhotic patients with ascites and 226 healthy controls were prospectively collected. The rs56163822, rs11110390 and rs12313471 polymorphisms of the farnesoid X receptor were determined. Results: 115 (39%) patients had spontaneous bacterial peritonitis. Distribution of all farnesoid X receptor genotypes matched the Hardy–Weinberg equilibrium. Patients with spontaneous bacterial peritonitis had a higher frequency of the rs56163822 GT genotype (7.0%) than patients without (1.7%, OR = 4.4, p = 0.02). This genotype was confirmed as predictor of spontaneous bacterial peritonitis by binary logistic regression analysis (OR = 6.8, p = 0.018). Conclusion: The farnesoid X receptor rs56163822 GT genotype increases the risk for spontaneous bacterial peritonitis in cirrhotic patients with ascites. © 2014 Editrice Gastroenterologica Italiana S.r.l. Published by Elsevier Ltd. All rights reserved.
1. Introduction The farnesoid X receptor (FXR), also called NR1H4, is a nuclear receptor activated by bile acids and expressed mainly in the liver and the intestine [1]. It has been linked to intestinal bacterial loads and bacterial translocation in the mouse [2]. In humans, FXR is involved in inflammatory bowel disease [3]. Spontaneous bacterial peritonitis is a typical complication of intestinal bacterial translocation in patients with cirrhosis and is associated with a high rate of mortality [4]. Polymorphisms in the Nucleotide-binding oligomerization domain-containing protein 2 (NOD2) and the Toll like receptor 2 (TLR2) gene predispose to SBP [6–8], suggesting that the genetic background of the host
play an essential role for susceptibility to SBP. However, the exact mechanisms leading to SBP are not fully understood. It is assumed that bacterial overgrowth and impaired immune responses facilitate bacterial translocation from the intestine to mesenteric lymph nodes, and ultimately into the ascites [5]. We hypothesized that FXR might be involved in the predisposition of patients with cirrhosis for SBP. Therefore we genotyped a large cohort of patients with liver cirrhosis and ascites for three NR1H4 polymorphisms associated with inflammatory bowel disease (IBD), another disease with impaired intestinal barrier [9], looking for a possible association between different FXR genotypes and the occurrence of SBP. 2. Methods
∗ Corresponding author at: Department of Internal Medicine I, University of Bonn, Sigmund-Freud-Strasse 25, D-53129 Bonn, Germany. Tel.: +49 228 287 15507; fax: +49 228 287 51419. E-mail address: [email protected] (P. Lutz).
2.1. Patients We prospectively collected blood samples and clinical data from all patients with liver cirrhosis and ascites who received
http://dx.doi.org/10.1016/j.dld.2014.07.008 1590-8658/© 2014 Editrice Gastroenterologica Italiana S.r.l. Published by Elsevier Ltd. All rights reserved.
Please cite this article in press as: Lutz P, et al. A farnesoid X receptor polymorphism predisposes to spontaneous bacterial peritonitis. Dig Liver Dis (2014), http://dx.doi.org/10.1016/j.dld.2014.07.008
G Model
ARTICLE IN PRESS
YDLD-2680; No. of Pages 4
P. Lutz et al. / Digestive and Liver Disease xxx (2014) xxx–xxx
2
a diagnostic paracentesis in our department from March 2012 to February 2013 in order to increase our cohort from previous studies on SBP [6,7]. Diagnosis of liver cirrhosis was based on liver biopsy, where available, or on typical complications of portal hypertension in patients with a consistent history, laboratory and ultrasonographic findings. A diagnostic paracentesis was performed whenever indicated by current international guidelines [10]. SBP was diagnosed according to international guidelines as a neutrophil count of at least 250/l in the absence of other causes of peritonitis [10]. In addition, blood from 226 healthy volunteers comprising blood donors (n = 164) and healthy participants of colonic-cancer screening programs (n = 62) was collected to analyze the normal distribution of the genotypes. The study protocol adhered to the ethical guidelines of the Helsinki Declaration and was approved by the local ethics committee. Written informed consent from the patients was obtained prior to inclusion in this study. 2.2. Determination of NR1H4, NOD2 and TLR2 genotypes Genomic DNA was extracted from 200 l EDTA-blood using the QIAamp Blood Mini Kit (Qiagen, Hilden, Germany). Determination of the FXR/NR1H4 (rs12313471, rs11110390, rs56163822 [G-1T]) and the NOD2 polymorphisms was performed by LightCycler real time PCR using LightSNiP (SimpleProbe) assays from TIB-MolBiol (Berlin, Germany). Samples were set up in a final volume of 10 l, containing 1 l of DNA solution, 5 l of Fermentas Maxima Probe qPCR Master Mix (Thermo Fisher Scientific Inc., Waltham, MA), and 0.5 l of LightSNiP reagent mix (Tib MolBiol, Berlin, Germany). The cycling conditions were chosen according to the manufacturer’s protocol. The TLR2 rs4696480 polymorphism was determined by a hybridization probe assay as previously described [7]. 2.3. Statistical analysis Wilcoxon–Mann–Whitney-U test was used for analysis of quantitative data. Fisher’s exact test was applied to qualitative data. Correspondence to the Hardy–Weinberg equilibrium was tested with a web-based software (http://ihg.gsf.de/cgi-bin/hw/). Power analysis on the same webpage yielded a number of 279 patients to achieve a power of 80%. Binary logistic regression analysis was performed on the different risk factors for SBP. A p < 0.05 was considered significant. IBM SPSS Statistics software version 21 (IBM, New York, USA) was used for all analyses. 3. Results 3.1. Patient characteristics In total, 293 patients with liver cirrhosis and ascites were inc luded in the study. Characteristics of the patients and healthy controls are summarized in Table 1. In 31 of the 115 patients with SBP (27%), microbiological culture was positive. The most frequent microorganisms were Escherichia coli (16%), Streptococci (16%), Klebsiella (10%) and Staphylococcus aureus (10%).
Table 1 Clinical characteristics of patients and controls.
Total number Median age years (range) Gender male (%)/female (%) Etiology of cirrhosis n (%) Alcohol Chronic hepatitis B/C Alcohol and chronic hepatitis Cryptogenic liver cirrhosis Primary sclerosing cholangitis Others Median INR (range) Median total serum bilirubin, mg/dL (range) Median serum creatinine mg/dL (range) Median serum albumin g/L (range) Median ascites albumin g/L (range) Median MELD score (range) Child-Pugh score (%) A/B/C SBP (%) Previous SBP/during follow-up (%) Culture positive SBP (% of SBP) Intake of prophylactic quinolones Intake of rifaximin (%) Intake of proton pump inhibitors (%)
Cirrhotic patients with ascites
Healthy controls
293 58.5 (23–87) 209 (71)/84 (29)
226 44 (20–86) 160 (71)/66 (29)
188 (64.2) 32 (10.9) 18 (6.1) 26 (8.9) 9 (3.1) 20 (6.8) 1.3 (0.9–4.7) 2.0 (0.19–38.79) 1.3 (0.38–14.93) 27.3 (10.4–47.8) 5.3 (0.7–29.7) 17 (6–43) 3.4/44.7/51.9 115 (39.2) 41 (14)/74 (25.2) 31 (27) 23 (7.9) 25 (8.5) 238 (81.2)
INR, international normalized ratio; MELD, model of end stage liver disease; SBP, spontaneous bacterial peritonitis.
3.2. NR1H4, NOD2 and TLR2 polymorphisms Distributions of all genotypes were consistent with Hardy–Weinberg equilibrium for each single nucleotide polymorphism both in patients and healthy controls. No patient was homozygous for the NR1H4 G-1T (rs56163822) variant. When assigning all patients with liver cirrhosis and ascites to a single group irrespective of the presence of SBP, no differences were found between patients and controls regarding the distribution of the NR1H4 rs56163822, rs1110390 and rs12313471 genotypes (Table 2). However, when patients with a history of SBP were compared to cirrhotic patients without SBP, the NR1H4 rs56168322 GT genotype was more prevalent in the SBP group (8/115 patients; 7.0%) than in the group without SBP (3/181 patients; 1.7%) (odds ratio OR = 4.36; 95%CI = 1.13–16.80; p = 0.02; Fig. 1). In addition, we found that the rs12313471 SNP was in linkage with the rs56163822 polymorphism, since all carriers of the rs56163822 T allele had the rs12313471 G allele. However, we did not observe an association of the rs12313471 G allele frequency with SBP (3.4% vs 5.2%; p = 0.27) because an increased SBP risk was restricted to the carriers of the rs56163822 risk variant and not found in patients with the rs12313471 G allele who did not carry the rs56163822 risk variant. Concerning the frequency of the rs11110390 C allele, we found a trend towards increased frequencies in patients with SBP (67.1% vs 74.3%, OR = 1.42; 95%CI = 0.98–2.05; p = 0.063).
Table 2 Distribution of genotype frequencies of the three investigated single nucleotide polymorphisms in patients and controls. rs56163822 rs11110390 rs12313471
Patients Healthy controls Patients Healthy controls Patients Healthy controls
(GG/GT/TT) N (%) (GG/GT/TT) N (%) (CC/CT/TT) N (%) (CC/CT/TT) N (%) (AA/AG/GG) N (%) (AA/AG/GG) N (%)
282 (96.2%) 218 (96.5%) 143 (48.8%) 94 (41.6%) 270 (92.2%) 203 (89.8%)
11 (3.8%) 8 (3.5%) 124 (42.3%) 103 (45.6%) 22 (7.5%) 23 (10.2%)
0 (0%) 0 (0%) 26 (8.9%) 29 (12.8%) 1 (0.3%) 0 (0%)
Please cite this article in press as: Lutz P, et al. A farnesoid X receptor polymorphism predisposes to spontaneous bacterial peritonitis. Dig Liver Dis (2014), http://dx.doi.org/10.1016/j.dld.2014.07.008
G Model
ARTICLE IN PRESS
YDLD-2680; No. of Pages 4
P. Lutz et al. / Digestive and Liver Disease xxx (2014) xxx–xxx
3
95%CI = 1.386–33.04; p = 0.018), the presence of at least one of the NOD2 risk alleles (OR = 2.120; 95%CI = 1.174–3.828; p = 0.013) and the TLR2 −16934 TT genotype (OR = 1.795; 95%CI = 1.014–3.177; p = 0.045) remained as independent risk factors for SBP (Table 3B). 56% of our patients with SBP carried at least one of these genetic risk factors. Since only one patient exhibited the NOD2 and only four patients the TLR2 risk variant in combination with the NR1H4 risk variant, the effect of combination of several genetic risk factors could not be assessed. 4. Discussion
Fig. 1. The frequency of the GT genotype of the NR1H4 G-1T polymorphism was equal in healthy controls and patients with liver cirrhosis. When cirrhotic patients with and without a history of spontaneous bacterial peritonitis (SBP) were analyzed separately, a statistically significant increase in the prevalence of the GT genotype was revealed. Statistical analysis by Fisher’s exact test.
The distribution of the Child-Pugh stages and of the etiology of liver cirrhosis did not differ between the FXR genotypes neither in the total cohort nor in the SBP subgroup. In addition, the use of systemic antibiotics and proton pump inhibitors were comparable for patients in these groups. Finally, we compared the relative contributions of the NR1H4 rs56163822 and rs11110390 variants, the TLR2 and the NOD2 genetic variants to SBP susceptibility also taking into account other putative risk factors of SBP (gender, MELD score, serum creatinine, INR, serum bilirubin, use of proton pump inhibitors, use of rifaximin, ascites albumin and ascites total protein). Univariate analysis identified MELD score, rifaximin, rs56163822 GT genotype, carriage of the rs11110390 C allele, TLR2 −16934 TT genotype, and carriage of at least one of the NOD2 risk alleles as potential risk factors for SBP (Table 3A). When these variables were entered into a binary logistic regression model, the NR1H4 rs56163822 GT genotype (OR = 6.766; Table 3 Regression analysis of risk factors for SBP. Parameter
OR
95% CI
p
Lower
Upper
(A) Univariate analysis Albumin (ascites)# Bilirubin >2.5 mg/dL Creatinine Gender (male) INR MELD score NOD2 risk variant NR1H4 rs56163822 T allele NR1H4 rs11110390 C allele Proton pump inhibitors Rifaximin TLR2 −16934 TT genotype Total protein (ascites) <10 g/l
0.985 1.148 1.053 1.075 1.157 1.032 1.836 4.361 0.674 1.063 2.092 1.871 1.373
0.939 0.709 0.881 0.641 0.699 0.999 1.053 1.132 0.421 0.573 0.914 1.095 0.842
1.033 1.860 1.258 1.802 1.914 1.066 3.203 16.80 1.079 1.972 4.785 3.197 2.241
0.535 0.575 0.571 0.785 0.571 0.060 0.031 0.020 0.099 0.847 0.075 0.021 0.203
(B) Multivariate analysis* NR1H4 rs56163822 T allele NOD2 risk allele TLR2 −16934 TT genotype
6.766 2.120 1.795
1.386 1.174 1.014
33.04 3.828 3.177
0.018 0.013 0.045
CI, confidence interval; OR, odds ratio; MELD, model of end-stage liver disease; NOD, nucleotide-binding oligomerization domain containing; SBP, spontaneous bacterial peritonitis; INR, international normalized ratio. * Including all significant (p < 0.1) parameters from the univariate analysis. # Per 1 g/l albumin.
In the present study, we show that the FXR /NR1H4 G-1T rs56163822 polymorphism confers an increased risk for SBP in patients with ascites and liver cirrhosis. In addition, we found a statistical trend for the association of the intronic rs11110390 SNP with an increased SBP risk. Our results suggest a possible role for the bile acid receptor FXR for susceptibility to SBP. Several risk factors for SBP in cirrhotic patients have been identified so far: a total ascites protein of less than 10 g/L [11,12], high serum bilirubin [13], a high score in the model of end-stage liver disease (MELD) [14] and the presence of polymorphisms in the NOD2 and TLR2 genes [6–8]. We included these risk factors in a univariate and a multivariate analysis and found that the FXR G-1T polymorphism was indeed an independent risk factor of SBP. The high recurrence rate of SBP of about 74% within two years [15] may thus be attributed in part to genetic predisposition. The G-1T mutation leads to reduced translation of FXR mRNA [16] and a decreased activity of FXR as transcription factor [17]. In FXR-knock-out mice, the absence of FXR expression leads to increased intestinal permeability which is associated with a decrease in the numbers of tight junctions in the ileal mucosa [2]. FXR−/− mice are also more prone to damage of the intestinal mucosa by non-steroidal anti-inflammatory drugs. In line with these observations, the G-1T polymorphism is associated with the occurrence of inflammatory bowel disease [18] which is characterized by an impaired intestinal barrier [9]. Therefore, we propose that patients with the minor NR1H4 variant are more susceptible to bacterial translocation and subsequent SBP due to an impaired intestinal barrier associated with reduced FXR activity. A major limitation of genetic association studies is that falsepositive results occur. Therefore, large numbers of patients and a second cohort to replicate the results are often requested. Here, we present the largest cohort of cirrhotic patients studied for genetic risk factors for SBP so far. Since our findings are in line with experimental data [2] and the constitution of a replication cohort would require several years, we decided to present our findings without replication cohort. In conclusion, we found in a large cohort of patients with liver cirrhosis and ascites that a polymorphism in the FXR gene is a risk factor for SBP. Since FXR agonists such as obeticholic acid are already tested in clinical studies, it should be also investigated if pharmacological modulation of FXR could provide a novel strategy for SBP prophylaxis, avoiding the use of antibiotics and the subsequent induction of bacterial resistance. Funding This work was supported by the German Center for Infection Research [Clinical leave to P.L.], by the Deutsche Krebshilfe [107865 to H.D.N. and U.S.] by the H.W. & J. Hector Foundation [M42 to J.N.]. Conflict of interest None declared.
Please cite this article in press as: Lutz P, et al. A farnesoid X receptor polymorphism predisposes to spontaneous bacterial peritonitis. Dig Liver Dis (2014), http://dx.doi.org/10.1016/j.dld.2014.07.008
G Model YDLD-2680; No. of Pages 4
ARTICLE IN PRESS P. Lutz et al. / Digestive and Liver Disease xxx (2014) xxx–xxx
4
Acknowledgments We gratefully acknowledge the technical assistance of Jennifer Söhne and Carolin Luda. References [1] Hollman DA, Milona A, van Erpecum KJ, et al. Anti-inflammatory and metabolic actions of FXR: insights into molecular mechanisms. Biochimica et Biophysica Acta 2012;1821:1443–52. [2] Inagaki T, Moschetta A, Lee YK, et al. Regulation of antibacterial defense in the small intestine by the nuclear bile acid receptor. Proceedings of the National Academy of Sciences of the United States of America 2006;103:3920–5. [3] Stojancevic M, Stankov K, Mikov M. The impact of farnesoid X receptor activation on intestinal permeability in inflammatory bowel disease. Canadian Journal of Gastroenterology 2012;26:631–7. [4] Wiest R, Krag A, Gerbes A. Spontaneous bacterial peritonitis: recent guidelines and beyond. Gut 2012;61:297–310. [5] Garcia-Tsao G, Wiest R. Gut microflora in the pathogenesis of the complications of cirrhosis. Best Practice and Research Clinical Gastroenterology 2004;18:353–72. [6] Appenrodt B, Grunhage F, Gentemann MG, et al. Nucleotide-binding oligomerization domain containing 2 (NOD2) variants are genetic risk factors for death and spontaneous bacterial peritonitis in liver cirrhosis. Hepatology 2010;51:1327–33. [7] Nischalke HD, Berger C, Aldenhoff K, et al. Toll-like receptor (TLR) 2 promoter and intron 2 polymorphisms are associated with increased risk for spontaneous bacterial peritonitis in liver cirrhosis. Journal of Hepatology 2011;55: 1010–6.
[8] Bruns T, Peter J, Reuken PA, et al. NOD2 gene variants are a risk factor for culturepositive spontaneous bacterial peritonitis and monomicrobial bacterascites in cirrhosis. Liver International 2012;32:223–30. [9] Sartor RB. Microbial influences in inflammatory bowel diseases. Gastroenterology 2008;134:577–94. [10] EASL clinical practice guidelines on the management of ascites, spontaneous bacterial peritonitis, and hepatorenal syndrome in cirrhosis. Journal of Hepatology 2010;53:397–417. [11] Llach J, Rimola A, Navasa M, et al. Incidence and predictive factors of first episode of spontaneous bacterial peritonitis in cirrhosis with ascites: relevance of ascitic fluid protein concentration. Hepatology 1992;16:724–7. [12] Rimola A, Garcia-Tsao G, Navasa M, et al. Diagnosis, treatment and prophylaxis of spontaneous bacterial peritonitis: a consensus document. International Ascites Club. Journal of Hepatology 2000;32:142–53. [13] Andreu M, Sola R, Sitges-Serra A, et al. Risk factors for spontaneous bacterial peritonitis in cirrhotic patients with ascites. Gastroenterology 1993;104:1133–8. [14] Obstein KL, Campbell MS, Reddy KR, et al. Association between model for endstage liver disease and spontaneous bacterial peritonitis. American Journal of Gastroenterology 2007;102:2732–6. [15] Tito L, Rimola A, Gines P, et al. Recurrence of spontaneous bacterial peritonitis in cirrhosis: frequency and predictive factors. Hepatology 1988;8:27–31. [16] Van Mil SW, Milona A, Dixon PH, et al. Functional variants of the central bile acid sensor FXR identified in intrahepatic cholestasis of pregnancy. Gastroenterology 2007;133:507–16. [17] Marzolini C, Tirona RG, Gervasini G, et al. A common polymorphism in the bile acid receptor farnesoid X receptor is associated with decreased hepatic target gene expression. Molecular Endocrinology 2007;21:1769–80. [18] Attinkara R, Mwinyi J, Truninger K, et al. Association of genetic variation in the NR1H4 gene, encoding the nuclear bile acid receptor FXR, with inflammatory bowel disease. BMC Research Notes 2012;5:461.
Please cite this article in press as: Lutz P, et al. A farnesoid X receptor polymorphism predisposes to spontaneous bacterial peritonitis. Dig Liver Dis (2014), http://dx.doi.org/10.1016/j.dld.2014.07.008
ARTICLE IN PRESS Digestive and Liver Disease xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
Digestive and Liver Disease journal homepage: www.elsevier.com/locate/dld
Short Report
A farnesoid X receptor polymorphism predisposes to spontaneous bacterial peritonitis Philipp Lutz a,d,∗ , Cordula Berger a , Bettina Langhans a , Frank Grünhage c , Beate Appenrodt c , Jacob Nattermann a,d , Frank Lammert c , Achim Hoerauf b,d , Tilman Sauerbruch a , Christian P. Strassburg a,d , Ulrich Spengler a,d , Hans Dieter Nischalke a a
Department of Internal Medicine I, University of Bonn, Bonn, Germany Institute for Medical Microbiology, Immunology and Parasitology, University of Bonn, Bonn, Germany c Department of Medicine II, Saarland University Medical Center, Saarland University, Homburg, Germany d German Center for Infection Research, Germany b
a r t i c l e
i n f o
Article history: Received 27 March 2014 Accepted 4 July 2014 Available online xxx Keywords: Ascites Farnesoid X receptor Liver cirrhosis Spontaneous bacterial peritonitis
a b s t r a c t Background: In mice, the farnesoid X receptor is involved in bacterial translocation, which can result in spontaneous bacterial peritonitis in patients with cirrhosis. We investigated if polymorphisms in the farnesoid X receptor gene influence the risk for spontaneous bacterial peritonitis. Methods: Laboratory and clinical data of 293 cirrhotic patients with ascites and 226 healthy controls were prospectively collected. The rs56163822, rs11110390 and rs12313471 polymorphisms of the farnesoid X receptor were determined. Results: 115 (39%) patients had spontaneous bacterial peritonitis. Distribution of all farnesoid X receptor genotypes matched the Hardy–Weinberg equilibrium. Patients with spontaneous bacterial peritonitis had a higher frequency of the rs56163822 GT genotype (7.0%) than patients without (1.7%, OR = 4.4, p = 0.02). This genotype was confirmed as predictor of spontaneous bacterial peritonitis by binary logistic regression analysis (OR = 6.8, p = 0.018). Conclusion: The farnesoid X receptor rs56163822 GT genotype increases the risk for spontaneous bacterial peritonitis in cirrhotic patients with ascites. © 2014 Editrice Gastroenterologica Italiana S.r.l. Published by Elsevier Ltd. All rights reserved.
1. Introduction The farnesoid X receptor (FXR), also called NR1H4, is a nuclear receptor activated by bile acids and expressed mainly in the liver and the intestine [1]. It has been linked to intestinal bacterial loads and bacterial translocation in the mouse [2]. In humans, FXR is involved in inflammatory bowel disease [3]. Spontaneous bacterial peritonitis is a typical complication of intestinal bacterial translocation in patients with cirrhosis and is associated with a high rate of mortality [4]. Polymorphisms in the Nucleotide-binding oligomerization domain-containing protein 2 (NOD2) and the Toll like receptor 2 (TLR2) gene predispose to SBP [6–8], suggesting that the genetic background of the host
play an essential role for susceptibility to SBP. However, the exact mechanisms leading to SBP are not fully understood. It is assumed that bacterial overgrowth and impaired immune responses facilitate bacterial translocation from the intestine to mesenteric lymph nodes, and ultimately into the ascites [5]. We hypothesized that FXR might be involved in the predisposition of patients with cirrhosis for SBP. Therefore we genotyped a large cohort of patients with liver cirrhosis and ascites for three NR1H4 polymorphisms associated with inflammatory bowel disease (IBD), another disease with impaired intestinal barrier [9], looking for a possible association between different FXR genotypes and the occurrence of SBP. 2. Methods
∗ Corresponding author at: Department of Internal Medicine I, University of Bonn, Sigmund-Freud-Strasse 25, D-53129 Bonn, Germany. Tel.: +49 228 287 15507; fax: +49 228 287 51419. E-mail address: [email protected] (P. Lutz).
2.1. Patients We prospectively collected blood samples and clinical data from all patients with liver cirrhosis and ascites who received
http://dx.doi.org/10.1016/j.dld.2014.07.008 1590-8658/© 2014 Editrice Gastroenterologica Italiana S.r.l. Published by Elsevier Ltd. All rights reserved.
Please cite this article in press as: Lutz P, et al. A farnesoid X receptor polymorphism predisposes to spontaneous bacterial peritonitis. Dig Liver Dis (2014), http://dx.doi.org/10.1016/j.dld.2014.07.008
G Model
ARTICLE IN PRESS
YDLD-2680; No. of Pages 4
P. Lutz et al. / Digestive and Liver Disease xxx (2014) xxx–xxx
2
a diagnostic paracentesis in our department from March 2012 to February 2013 in order to increase our cohort from previous studies on SBP [6,7]. Diagnosis of liver cirrhosis was based on liver biopsy, where available, or on typical complications of portal hypertension in patients with a consistent history, laboratory and ultrasonographic findings. A diagnostic paracentesis was performed whenever indicated by current international guidelines [10]. SBP was diagnosed according to international guidelines as a neutrophil count of at least 250/l in the absence of other causes of peritonitis [10]. In addition, blood from 226 healthy volunteers comprising blood donors (n = 164) and healthy participants of colonic-cancer screening programs (n = 62) was collected to analyze the normal distribution of the genotypes. The study protocol adhered to the ethical guidelines of the Helsinki Declaration and was approved by the local ethics committee. Written informed consent from the patients was obtained prior to inclusion in this study. 2.2. Determination of NR1H4, NOD2 and TLR2 genotypes Genomic DNA was extracted from 200 l EDTA-blood using the QIAamp Blood Mini Kit (Qiagen, Hilden, Germany). Determination of the FXR/NR1H4 (rs12313471, rs11110390, rs56163822 [G-1T]) and the NOD2 polymorphisms was performed by LightCycler real time PCR using LightSNiP (SimpleProbe) assays from TIB-MolBiol (Berlin, Germany). Samples were set up in a final volume of 10 l, containing 1 l of DNA solution, 5 l of Fermentas Maxima Probe qPCR Master Mix (Thermo Fisher Scientific Inc., Waltham, MA), and 0.5 l of LightSNiP reagent mix (Tib MolBiol, Berlin, Germany). The cycling conditions were chosen according to the manufacturer’s protocol. The TLR2 rs4696480 polymorphism was determined by a hybridization probe assay as previously described [7]. 2.3. Statistical analysis Wilcoxon–Mann–Whitney-U test was used for analysis of quantitative data. Fisher’s exact test was applied to qualitative data. Correspondence to the Hardy–Weinberg equilibrium was tested with a web-based software (http://ihg.gsf.de/cgi-bin/hw/). Power analysis on the same webpage yielded a number of 279 patients to achieve a power of 80%. Binary logistic regression analysis was performed on the different risk factors for SBP. A p < 0.05 was considered significant. IBM SPSS Statistics software version 21 (IBM, New York, USA) was used for all analyses. 3. Results 3.1. Patient characteristics In total, 293 patients with liver cirrhosis and ascites were inc luded in the study. Characteristics of the patients and healthy controls are summarized in Table 1. In 31 of the 115 patients with SBP (27%), microbiological culture was positive. The most frequent microorganisms were Escherichia coli (16%), Streptococci (16%), Klebsiella (10%) and Staphylococcus aureus (10%).
Table 1 Clinical characteristics of patients and controls.
Total number Median age years (range) Gender male (%)/female (%) Etiology of cirrhosis n (%) Alcohol Chronic hepatitis B/C Alcohol and chronic hepatitis Cryptogenic liver cirrhosis Primary sclerosing cholangitis Others Median INR (range) Median total serum bilirubin, mg/dL (range) Median serum creatinine mg/dL (range) Median serum albumin g/L (range) Median ascites albumin g/L (range) Median MELD score (range) Child-Pugh score (%) A/B/C SBP (%) Previous SBP/during follow-up (%) Culture positive SBP (% of SBP) Intake of prophylactic quinolones Intake of rifaximin (%) Intake of proton pump inhibitors (%)
Cirrhotic patients with ascites
Healthy controls
293 58.5 (23–87) 209 (71)/84 (29)
226 44 (20–86) 160 (71)/66 (29)
188 (64.2) 32 (10.9) 18 (6.1) 26 (8.9) 9 (3.1) 20 (6.8) 1.3 (0.9–4.7) 2.0 (0.19–38.79) 1.3 (0.38–14.93) 27.3 (10.4–47.8) 5.3 (0.7–29.7) 17 (6–43) 3.4/44.7/51.9 115 (39.2) 41 (14)/74 (25.2) 31 (27) 23 (7.9) 25 (8.5) 238 (81.2)
INR, international normalized ratio; MELD, model of end stage liver disease; SBP, spontaneous bacterial peritonitis.
3.2. NR1H4, NOD2 and TLR2 polymorphisms Distributions of all genotypes were consistent with Hardy–Weinberg equilibrium for each single nucleotide polymorphism both in patients and healthy controls. No patient was homozygous for the NR1H4 G-1T (rs56163822) variant. When assigning all patients with liver cirrhosis and ascites to a single group irrespective of the presence of SBP, no differences were found between patients and controls regarding the distribution of the NR1H4 rs56163822, rs1110390 and rs12313471 genotypes (Table 2). However, when patients with a history of SBP were compared to cirrhotic patients without SBP, the NR1H4 rs56168322 GT genotype was more prevalent in the SBP group (8/115 patients; 7.0%) than in the group without SBP (3/181 patients; 1.7%) (odds ratio OR = 4.36; 95%CI = 1.13–16.80; p = 0.02; Fig. 1). In addition, we found that the rs12313471 SNP was in linkage with the rs56163822 polymorphism, since all carriers of the rs56163822 T allele had the rs12313471 G allele. However, we did not observe an association of the rs12313471 G allele frequency with SBP (3.4% vs 5.2%; p = 0.27) because an increased SBP risk was restricted to the carriers of the rs56163822 risk variant and not found in patients with the rs12313471 G allele who did not carry the rs56163822 risk variant. Concerning the frequency of the rs11110390 C allele, we found a trend towards increased frequencies in patients with SBP (67.1% vs 74.3%, OR = 1.42; 95%CI = 0.98–2.05; p = 0.063).
Table 2 Distribution of genotype frequencies of the three investigated single nucleotide polymorphisms in patients and controls. rs56163822 rs11110390 rs12313471
Patients Healthy controls Patients Healthy controls Patients Healthy controls
(GG/GT/TT) N (%) (GG/GT/TT) N (%) (CC/CT/TT) N (%) (CC/CT/TT) N (%) (AA/AG/GG) N (%) (AA/AG/GG) N (%)
282 (96.2%) 218 (96.5%) 143 (48.8%) 94 (41.6%) 270 (92.2%) 203 (89.8%)
11 (3.8%) 8 (3.5%) 124 (42.3%) 103 (45.6%) 22 (7.5%) 23 (10.2%)
0 (0%) 0 (0%) 26 (8.9%) 29 (12.8%) 1 (0.3%) 0 (0%)
Please cite this article in press as: Lutz P, et al. A farnesoid X receptor polymorphism predisposes to spontaneous bacterial peritonitis. Dig Liver Dis (2014), http://dx.doi.org/10.1016/j.dld.2014.07.008
G Model
ARTICLE IN PRESS
YDLD-2680; No. of Pages 4
P. Lutz et al. / Digestive and Liver Disease xxx (2014) xxx–xxx
3
95%CI = 1.386–33.04; p = 0.018), the presence of at least one of the NOD2 risk alleles (OR = 2.120; 95%CI = 1.174–3.828; p = 0.013) and the TLR2 −16934 TT genotype (OR = 1.795; 95%CI = 1.014–3.177; p = 0.045) remained as independent risk factors for SBP (Table 3B). 56% of our patients with SBP carried at least one of these genetic risk factors. Since only one patient exhibited the NOD2 and only four patients the TLR2 risk variant in combination with the NR1H4 risk variant, the effect of combination of several genetic risk factors could not be assessed. 4. Discussion
Fig. 1. The frequency of the GT genotype of the NR1H4 G-1T polymorphism was equal in healthy controls and patients with liver cirrhosis. When cirrhotic patients with and without a history of spontaneous bacterial peritonitis (SBP) were analyzed separately, a statistically significant increase in the prevalence of the GT genotype was revealed. Statistical analysis by Fisher’s exact test.
The distribution of the Child-Pugh stages and of the etiology of liver cirrhosis did not differ between the FXR genotypes neither in the total cohort nor in the SBP subgroup. In addition, the use of systemic antibiotics and proton pump inhibitors were comparable for patients in these groups. Finally, we compared the relative contributions of the NR1H4 rs56163822 and rs11110390 variants, the TLR2 and the NOD2 genetic variants to SBP susceptibility also taking into account other putative risk factors of SBP (gender, MELD score, serum creatinine, INR, serum bilirubin, use of proton pump inhibitors, use of rifaximin, ascites albumin and ascites total protein). Univariate analysis identified MELD score, rifaximin, rs56163822 GT genotype, carriage of the rs11110390 C allele, TLR2 −16934 TT genotype, and carriage of at least one of the NOD2 risk alleles as potential risk factors for SBP (Table 3A). When these variables were entered into a binary logistic regression model, the NR1H4 rs56163822 GT genotype (OR = 6.766; Table 3 Regression analysis of risk factors for SBP. Parameter
OR
95% CI
p
Lower
Upper
(A) Univariate analysis Albumin (ascites)# Bilirubin >2.5 mg/dL Creatinine Gender (male) INR MELD score NOD2 risk variant NR1H4 rs56163822 T allele NR1H4 rs11110390 C allele Proton pump inhibitors Rifaximin TLR2 −16934 TT genotype Total protein (ascites) <10 g/l
0.985 1.148 1.053 1.075 1.157 1.032 1.836 4.361 0.674 1.063 2.092 1.871 1.373
0.939 0.709 0.881 0.641 0.699 0.999 1.053 1.132 0.421 0.573 0.914 1.095 0.842
1.033 1.860 1.258 1.802 1.914 1.066 3.203 16.80 1.079 1.972 4.785 3.197 2.241
0.535 0.575 0.571 0.785 0.571 0.060 0.031 0.020 0.099 0.847 0.075 0.021 0.203
(B) Multivariate analysis* NR1H4 rs56163822 T allele NOD2 risk allele TLR2 −16934 TT genotype
6.766 2.120 1.795
1.386 1.174 1.014
33.04 3.828 3.177
0.018 0.013 0.045
CI, confidence interval; OR, odds ratio; MELD, model of end-stage liver disease; NOD, nucleotide-binding oligomerization domain containing; SBP, spontaneous bacterial peritonitis; INR, international normalized ratio. * Including all significant (p < 0.1) parameters from the univariate analysis. # Per 1 g/l albumin.
In the present study, we show that the FXR /NR1H4 G-1T rs56163822 polymorphism confers an increased risk for SBP in patients with ascites and liver cirrhosis. In addition, we found a statistical trend for the association of the intronic rs11110390 SNP with an increased SBP risk. Our results suggest a possible role for the bile acid receptor FXR for susceptibility to SBP. Several risk factors for SBP in cirrhotic patients have been identified so far: a total ascites protein of less than 10 g/L [11,12], high serum bilirubin [13], a high score in the model of end-stage liver disease (MELD) [14] and the presence of polymorphisms in the NOD2 and TLR2 genes [6–8]. We included these risk factors in a univariate and a multivariate analysis and found that the FXR G-1T polymorphism was indeed an independent risk factor of SBP. The high recurrence rate of SBP of about 74% within two years [15] may thus be attributed in part to genetic predisposition. The G-1T mutation leads to reduced translation of FXR mRNA [16] and a decreased activity of FXR as transcription factor [17]. In FXR-knock-out mice, the absence of FXR expression leads to increased intestinal permeability which is associated with a decrease in the numbers of tight junctions in the ileal mucosa [2]. FXR−/− mice are also more prone to damage of the intestinal mucosa by non-steroidal anti-inflammatory drugs. In line with these observations, the G-1T polymorphism is associated with the occurrence of inflammatory bowel disease [18] which is characterized by an impaired intestinal barrier [9]. Therefore, we propose that patients with the minor NR1H4 variant are more susceptible to bacterial translocation and subsequent SBP due to an impaired intestinal barrier associated with reduced FXR activity. A major limitation of genetic association studies is that falsepositive results occur. Therefore, large numbers of patients and a second cohort to replicate the results are often requested. Here, we present the largest cohort of cirrhotic patients studied for genetic risk factors for SBP so far. Since our findings are in line with experimental data [2] and the constitution of a replication cohort would require several years, we decided to present our findings without replication cohort. In conclusion, we found in a large cohort of patients with liver cirrhosis and ascites that a polymorphism in the FXR gene is a risk factor for SBP. Since FXR agonists such as obeticholic acid are already tested in clinical studies, it should be also investigated if pharmacological modulation of FXR could provide a novel strategy for SBP prophylaxis, avoiding the use of antibiotics and the subsequent induction of bacterial resistance. Funding This work was supported by the German Center for Infection Research [Clinical leave to P.L.], by the Deutsche Krebshilfe [107865 to H.D.N. and U.S.] by the H.W. & J. Hector Foundation [M42 to J.N.]. Conflict of interest None declared.
Please cite this article in press as: Lutz P, et al. A farnesoid X receptor polymorphism predisposes to spontaneous bacterial peritonitis. Dig Liver Dis (2014), http://dx.doi.org/10.1016/j.dld.2014.07.008
G Model YDLD-2680; No. of Pages 4
ARTICLE IN PRESS P. Lutz et al. / Digestive and Liver Disease xxx (2014) xxx–xxx
4
Acknowledgments We gratefully acknowledge the technical assistance of Jennifer Söhne and Carolin Luda. References [1] Hollman DA, Milona A, van Erpecum KJ, et al. Anti-inflammatory and metabolic actions of FXR: insights into molecular mechanisms. Biochimica et Biophysica Acta 2012;1821:1443–52. [2] Inagaki T, Moschetta A, Lee YK, et al. Regulation of antibacterial defense in the small intestine by the nuclear bile acid receptor. Proceedings of the National Academy of Sciences of the United States of America 2006;103:3920–5. [3] Stojancevic M, Stankov K, Mikov M. The impact of farnesoid X receptor activation on intestinal permeability in inflammatory bowel disease. Canadian Journal of Gastroenterology 2012;26:631–7. [4] Wiest R, Krag A, Gerbes A. Spontaneous bacterial peritonitis: recent guidelines and beyond. Gut 2012;61:297–310. [5] Garcia-Tsao G, Wiest R. Gut microflora in the pathogenesis of the complications of cirrhosis. Best Practice and Research Clinical Gastroenterology 2004;18:353–72. [6] Appenrodt B, Grunhage F, Gentemann MG, et al. Nucleotide-binding oligomerization domain containing 2 (NOD2) variants are genetic risk factors for death and spontaneous bacterial peritonitis in liver cirrhosis. Hepatology 2010;51:1327–33. [7] Nischalke HD, Berger C, Aldenhoff K, et al. Toll-like receptor (TLR) 2 promoter and intron 2 polymorphisms are associated with increased risk for spontaneous bacterial peritonitis in liver cirrhosis. Journal of Hepatology 2011;55: 1010–6.
[8] Bruns T, Peter J, Reuken PA, et al. NOD2 gene variants are a risk factor for culturepositive spontaneous bacterial peritonitis and monomicrobial bacterascites in cirrhosis. Liver International 2012;32:223–30. [9] Sartor RB. Microbial influences in inflammatory bowel diseases. Gastroenterology 2008;134:577–94. [10] EASL clinical practice guidelines on the management of ascites, spontaneous bacterial peritonitis, and hepatorenal syndrome in cirrhosis. Journal of Hepatology 2010;53:397–417. [11] Llach J, Rimola A, Navasa M, et al. Incidence and predictive factors of first episode of spontaneous bacterial peritonitis in cirrhosis with ascites: relevance of ascitic fluid protein concentration. Hepatology 1992;16:724–7. [12] Rimola A, Garcia-Tsao G, Navasa M, et al. Diagnosis, treatment and prophylaxis of spontaneous bacterial peritonitis: a consensus document. International Ascites Club. Journal of Hepatology 2000;32:142–53. [13] Andreu M, Sola R, Sitges-Serra A, et al. Risk factors for spontaneous bacterial peritonitis in cirrhotic patients with ascites. Gastroenterology 1993;104:1133–8. [14] Obstein KL, Campbell MS, Reddy KR, et al. Association between model for endstage liver disease and spontaneous bacterial peritonitis. American Journal of Gastroenterology 2007;102:2732–6. [15] Tito L, Rimola A, Gines P, et al. Recurrence of spontaneous bacterial peritonitis in cirrhosis: frequency and predictive factors. Hepatology 1988;8:27–31. [16] Van Mil SW, Milona A, Dixon PH, et al. Functional variants of the central bile acid sensor FXR identified in intrahepatic cholestasis of pregnancy. Gastroenterology 2007;133:507–16. [17] Marzolini C, Tirona RG, Gervasini G, et al. A common polymorphism in the bile acid receptor farnesoid X receptor is associated with decreased hepatic target gene expression. Molecular Endocrinology 2007;21:1769–80. [18] Attinkara R, Mwinyi J, Truninger K, et al. Association of genetic variation in the NR1H4 gene, encoding the nuclear bile acid receptor FXR, with inflammatory bowel disease. BMC Research Notes 2012;5:461.
Please cite this article in press as: Lutz P, et al. A farnesoid X receptor polymorphism predisposes to spontaneous bacterial peritonitis. Dig Liver Dis (2014), http://dx.doi.org/10.1016/j.dld.2014.07.008