Aryl hydrocarbon receptor is activated in patients and mice with chronic kidney disease

clinical investigation

www.kidney-international.org

Aryl hydrocarbon receptor is activated in patients and mice with chronic kidney disease Laetitia Dou1, Ste´phane Poitevin1, Marion Salle´e1,2, Tawfik Addi1, Bertrand Gondouin2, Nathalie McKay1, Michael S. Denison3, Noe´mie Jourde-Chiche1,2, Ariane Duval-Sabatier2,4, Claire Cerini1, Philippe Brunet1,2, Franc¸oise Dignat-George1 and Ste´phane Burtey1,2 1 Aix-Marseille University, INSERM, UMR-S 1076, VRCM, Marseille, France; 2Centre de Néphrologie et Transplantation Rénale, AP-HM, Marseille, France; 3Department of Environmental Toxicology, University of California, Davis, California, USA; and 4Association des dialysés Provence-Corse, Marseille, France

Patients with chronic kidney disease (CKD) are exposed to uremic toxins and have an increased risk of cardiovascular disease. Some uremic toxins, like indoxyl sulfate, are agonists of the transcription factor aryl hydrocarbon receptor (AHR). These toxins induce a vascular procoagulant phenotype. Here we investigated AHR activation in patients with CKD and in a murine model of CKD. We performed a prospective study in 116 patients with CKD stage 3 to 5D and measured the AHR-Activating Potential of serum by bioassay. Compared to sera from healthy controls, sera from CKD patients displayed a strong AHR-Activating Potential; strongly correlated with eGFR and with the indoxyl sulfate concentration. The expression of the AHR target genes Cyp1A1 and AHRR was up-regulated in whole blood from patients with CKD. Survival analyses revealed that cardiovascular events were more frequent in CKD patients with an AHR-Activating Potential above the median. In mice with 5/6 nephrectomy, there was an increased serum AHR-Activating Potential, and an induction of Cyp1a1 mRNA in the aorta and heart, absent in AhR–/– CKD mice. After serial indoxyl sulfate injections, we observed an increase in serum AHR-AP and in expression of Cyp1a1 mRNA in aorta and heart in WT mice, but not in AhR–/– mice. Thus, the AHR pathway is activated both in patients and mice with CKD. Hence, AHR activation could be a key mechanism involved in the deleterious cardiovascular effects observed in CKD. Kidney International (2018) j.kint.2017.11.010

-, -–-;

https://doi.org/10.1016/

KEYWORDS: aryl hydrocarbon receptor; cardiovascular disease; chronic kidney disease; indoxyl sulfate; mouse model Copyright ª 2017, International Society of Nephrology. Published by Elsevier Inc. All rights reserved.

Correspondence: Pr Stéphane Burtey, Team «Dysfonction endothéliale et Insuffisance rénale chronique» VRCM, UMR_S1076 Aix-Marseille Université, Faculté de pharmacie, 27 bd Jean Moulin 13005 Marseille, France. E-mail: [email protected] Received 13 November 2016; revised 18 October 2017; accepted 9 November 2017 Kidney International (2018) -, -–-

C

hronic kidney disease (CKD) is an emerging epidemic. CKD increased by 73% from 1990 to 2013 as a cause of deaths worldwide.1 CKD is associated with an increased risk of death, especially from cardiovascular disease (CVD).2,3 Classical CVD risk factors do not explain the increased rate of cardiovascular events in CKD.4 Uremic toxins that accumulate during CKD could be the missing link between reduced ability of the kidney to eliminate waste and CVD.5 They are divided into 3 groups: small soluble compounds, middle molecules, and protein-bound molecules.6 Among this last group, increased levels of the indolic toxins indoxyl sulfate (IS) and indole-3 acetic acid (IAA) are associated with increased risk of death and cardiovascular events.7,8 In addition, numerous studies have demonstrated the deleterious effect of indolic toxins on renal and vascular cells.9 The cellular receptor of indolic solutes was identified as aryl hydrocarbon receptor (AHR).10,11 AHR is an intracellular receptor for xenobiotics such as 2,3,7,8-Tetrachlorodibenzo-p-dioxin (TCDD), the dioxin-like 3,30 ,4,40 ,5-pentachlorobiphenyl (PCB 126), and benzo[a] pyrene, a chemical found in tobacco smoke.12 AHR also binds endogenous ligands, including metabolites of arachidonic acid and tryptophan, such as kynurenine.13,14 AHR resides in the cytoplasm of mammalian cells in a multiprotein complex that includes HSP90 and AHR-interacting protein.12,15,16 AHR-ligand complex translocates to the nucleus, where it forms a heterodimeric complex with the aryl hydrocarbon nuclear translocator.12,15 The complex binds to a DNA consensus sequence12,15 present in the promoters of a wide variety of genes, including those coding for enzymes involved in xenobiotic detoxification (CYP1A1, CYP1A2, and CYP1B1) and the AHR repressor, AHRR.12,15 A genomic, aryl hydrocarbon nuclear translocator–independent pathway for AHRmediated gene expression has recently been reported,17 as well as a role of AHR as a component of numerous signaling pathways, independent of its ability to bind to DNA.17 In humans exposed to AHR agonists such as TCDD or PCB 126, the risk for cardiovascular events is increased.18 This association was recognized by the US government after exposure of veterans to Agent Orange.19 In ApoE–/– mice, AHR activation by TCDD is associated with acceleration of atherosclerosis.20 In rats, PCB 126 exposure increases CVD risk factors: serum cholesterol, blood pressure, and heart weight.21 1

clinical investigation

L Dou et al.: Aryl hydrocarbon receptor and CKD

The activation of AHR by IS and IAA has been demonstrated to contribute to vascular dysfunction.8,22 In endothelial cells and in vascular smooth muscle cells, AHR activation increases the expression and activity of tissue factor, leading to a procoagulant state.22,23 It also induces an increased expression and activity of the pro-inflammatory enzyme cyclooxygenase-2 in endothelial cells.8 The vascular dysfunction induced by AHR activation could lead to atherothrombosis and plays a role in the increased risk of myocardial infarction, peripheral artery disease, and stroke observed in CKD.24 Despite strong evidence of AHR activation by indolic uremic toxins in vitro, studies of AHR activation in CKD are scarce. The present study aimed to demonstrate that AHR is activated in patients with various stages of CKD. In CKD mice, we also studied AHR activation in the vascular wall and the role of IS as a representative uremic AHR agonist. RESULTS Uremic serum induced AHR activation

We studied a cohort of 116 patients with CKD (51 with stage 3–5 CKD and 65 with stage 5D CKD) (Table 1) and compared

these patients with 52 healthy controls. We analyzed the activation of AHR by serum samples using the AHR-responsive chemically activated luciferase expression cell bioassay, a method commonly used for the screening of samples for the presence of TCDD, dioxin-like compounds, and AHR agonists and/or antagonists.25,26 The serum AHR-activating potential (AHR-AP) was significantly higher in patients with stage 3 to 5 CKD (P < 0.05) and stage 5D CKD (P < 0.0001) than in controls (Figure 1a). Mean  SD values of AHR-AP were 22  9 arbitrary units (AU) (range: 5–44 AU), 37  24 AU (range: 6–121 AU), and 79  56 AU (range: 8–259 AU), respectively, in controls, in patients with stage 3 to 5 CKD, and in patients with stage 5D CKD (Figure 1a). We then examined whether AHR-AP of uremic serum could be counteracted by an AHR antagonist. The addition of the AHR antagonist CH223191 reduced by 46% the AHR-AP of serum from patients with stage 5D CKD (P < 0.01) (Figure 1b). CH223191 alone had no effect, and induced a slight, not significant decrease in the AHR-AP of control serum (Figure 1b). At baseline, AHR-AP from all patients with stage 3 to 5D CKD negatively correlated with hemoglobin (r ¼ –0.31,

Table 1 | Baseline characteristics of the CKD population Characteristics Age (yr) Gender ratio (W:M) Body mass index (kg/m2) Dialyzed patients (%) eGFRa (ml/min per 1.73 m2) Kidney disease Glomerulonephritis ADPKD Vascular Interstitial Other hereditary Unknown Hypertension Systolic blood pressure (mm Hg) Diastolic blood pressure (mm Hg) Current smokers History of cardiovascular diseases Phosphate binders Antihypertensive drugs Statins Antiplatelet drugs Anticoagulant drugs Erythropoietin therapy Serum CRP level (mg/l) Hemoglobin (g/dl) Serum bicarbonate level (mmol/l) Serum albumin level (g/l) Serum calcium level (mmol/l) Serum phosphate level (mmol/l) Serum cholesterol level (mmol/l) Serum LDL cholesterol level (mmol/l) Serum triglyceride level (mmol/l) Serum b2 microglobulin level (mg/l) Serum Indole-3 acetic acid level (mM) Serum indoxyl sulfate level (mM)

All patients (n [ 116)

AHR-AP < 44AU

AHR-AP ‡ 44AU

P value

68 (23; 89) 40:76 24.6 (15.8; 47) 65 (56%) 25 (8; 59)

63 (31; 89) 19:36 24.9 (16.8; 37.9) 19 (34%) 30 (11-59)

72 (23; 89) 21:40 24.5 (15.8; 47) 46 (75%) 14 (8; 53)

<0.05 1 0.7 <0.0001 <0.01

22 (19%) 11 (9%) 32 (28%) 23 (20%) 7 (6%) 21 (18%) 101 (87%) 141 24 77  14 47 (41%) 41 (35%) 60 (52%) 86 (74%) 37 (32%) 47 (40%) 26 (22%) 58 (50%) 4 (0; 78) 12.0 (8.8; 16.3) 22.7  3.0 36 (26; 44) 2.34  0.12 1.33 (0.65; 3.17) 4.6 (1.9; 9.1) 2.9  1.1 1.5 (0.4; 5.9) 22.6 (2.8; 66.9) 2.9 (0.6; 19.1) 43.7 (0.2; 256.2)

8 (15%) 6 (11%) 13 (24%) 15 (27%) 5 (9%) 8 (14%) 47 (85%) 143  28 79  15 20 (36%) 17 (31%) 19 (34%) 43 (78%) 13 (24%) 17 (31%) 11 (20%) 17 (31%) 4.6 (0.1; 54) 12.4 (9.5; 16.3) 23.7  3.1 36 (26; 44) 2.34  0.10 1.23 (0.65; 3.17) 5.5 (2.7; 9.1) 3.2  1.1 1.4 (0.4; 5.9) 7.9 (2.8; 66.9) 2.2 (0.6; 16.3) 13.4 (0.2; 157.8)

14 (23%) 5 (8%) 19 (31%) 8 (13%) 2 (4%) 13 (21%) 54 (89%) 139  19 74  12 27 (44%) 24 (39%) 41 (67%) 43 (70%) 24 (39%) 30 (49%) 15 (24%) 41 (67%) 4 (0; 78) 11.3 (8.8; 14.4) 21.8  2.6 36 (28; 43) 2.34  0.14 1.5 (0.7; 2.9) 4.3 (1.9; 7.3) 2.6  0.9 1.6 (0.5; 3.6) 27.6 (3.8; 56.4) 3.5 (0.7; 19.1) 78.9 (1.2; 256.2)

0.3 0.7 0.4 0.06 0.2 0.4 0.8 0.5 <0.05 0.4 0.4 <0.001 0.4 0.07 0.058 0.6 <0.001 0.5 <0.001 <0.01 0.7 0.9 <0.05 <0.001 <0.01 0.7 <0.0001 <0.05 <0.0001

AHR, aryl hydrocarbon receptor; AP, activating potential; ADPKD, autosomal dominant polycystic kidney disease; AU, arbitrary units; CKD, chronic kidney disease; CRP, Creactive protein; eGFR, estimated glomerular filtration rate; LDL, low-density lipoprotein; M, men; W, women. Results are given as mean  SD if distribution is Gaussian, or in median (min; max) if not. a Calculated by Modification of Diet in Renal Disease (MDRD) study formula for nondialyzed CKD patients (n ¼ 51).

2

Kidney International (2018) -, -–-

clinical investigation

L Dou et al.: Aryl hydrocarbon receptor and CKD

a

b

P < 0.0001 P < 0.05

200 P < 0.01

200

100

P < 0.01

150

AHR-AP (AU)

Serum AHR-AP (AU)

300

P < 0.0001

100

50

0

0 Controls (n = 52)

CKD stage 3–5 (n = 51)

Control serum HD serum CH223191

CKD stage 5D (n = 65)

c

+ -

+ +

+ -

+ +

-

+

AHR-AP of serum from CKD stage 5D patients (AU)

AHR-AP of serum from CKD stage 3–5 patients (AU)

d 150

r = –0.56, P < 0.0001

100

50

0 0

20

40

60

80

eGFR (ml/min per 1.73 m2)

e

300 r = 0.51, P < 0.0001 200

100

0 0

200

100

300

IS (µM)

150 Serum AHR-AP (AU)

** 100

50

0 Before HD

After HD

Figure 1 | Serum of patients with chronic kidney disease (CKD) contains aryl hydrocarbon receptor (AHR) agonists. (a) In vitro AHRactivating potential (AHR-AP) of sera from patients with stage 3 to 5 and stage 5D CKD is higher than AHR-AP of control sera. Data are expressed in arbitrary units (AU). (b) The AHR inhibitor CH223191 reduces AHR-AP. Data are expressed in AU and represent the mean SEM of experiments performed with 8 different uremic sera. (c) AHR-AP of sera from patients with stage 3 to 5 CKD is correlated with estimated glomerular filtration rate (eGFR) calculated according the Modification of Diet in Renal Disease (MDRD) study formula for the 51 nondialyzed CKD patients. (d) Serum AHR-AP is correlated with indoxyl sulfate (IS) serum level in the 65 patients with stage 5D CKD. (e) Hemodialysis (HD) induces a decrease in AHR-AP, reflecting the elimination of AHR agonists. Data, expressed in AU, represent the values obtained with 11 different sera. **P < 0.01.

P < 0.01), serum bicarbonate (r ¼ –0.39, P < 0.001), serum cholesterol (r ¼ –0.37, P < 0.0001), low-density lipoprotein cholesterol (r ¼ –0.37, P < 0.0001), and diastolic blood pressure (r ¼ –0.27, P < 0.01) (Table 2). AHR-AP positively correlated with age (r ¼ 0.26, P < 0.01), CKD stage (r ¼ 0.52, P < 0.0001), phosphate binder therapy (r ¼ 0.37, P < 0.0001), erythropoietin therapy (r ¼ 0.39, P < 0.0001), serum phosphate (r ¼ 0.25, P < 0.01), and the uremic toxins urea (r ¼ 0.33, P < 0.001), creatinine (r ¼ 0.46, P < 0.0001), b2microglobulin (r ¼ 0.41, P < 0.0001), IS (r ¼ 0.62, Kidney International (2018) -, -–-

P < 0.0001), and IAA (r ¼ 0.26, P < 0.01) (Table 2). In multivariate linear regression analysis in stage 3 to 5D CKD patients, age (estimate ¼ 0.68, P < 0.05), serum bicarbonate (estimate ¼ –4.12, P < 0.05), and serum IS (estimate ¼ 0.62, P < 0.001) were significantly associated with AHR-AP (Table 3). We analyzed the correlations of baseline characteristics and serum AHR-AP in the stage 3 to 5 CKD subcohort, and in the hemodialyzed stage 5D CKD subcohort. In the stage 3 to 5 CKD subcohort (Table 4), AHR-AP negatively correlated with 3

clinical investigation

L Dou et al.: Aryl hydrocarbon receptor and CKD

Table 2 | Spearman correlations of baseline characteristics with AHR-AP of serum in the stage 3 to 5D CKD cohort (n [ 116)

Table 4 | Spearman correlations of baseline characteristics with AHR-AP of serum in the CKD stage 3 to 5 subcohort (n [ 51)

AHR-AP Variables Age Gender Body mass index Systolic blood pressure Diastolic blood pressure Current smoking eGFRa CKD stage Kidney disease Phosphate binders Antihypertensive drugs Statins Antiplatelet drugs Anticoagulant drugs Erythropoietin therapy Serum CRP level Hemoglobin Serum bicarbonate level Serum albumin level Serum calcium level Serum phosphate level Serum cholesterol level Serum LDL cholesterol level Serum triglyceride level Serum urea level Serum creatinine level Serum b2 microglobulin level Serum Indole-3 acetic acid level Serum indoxyl sulfate level

AHR-AP

r

P value

Variables

0.26 –0.03 0.02 –0.13 –0.27 –0.01 –0.56 0.52 0.04 0.37 –0.15 0.14 0.18 0.03 0.39 0.07 –0.31 –0.39 –0.05 0.07 0.25 –0.37 –0.37 0.08 0.33 0.46 0.41 0.26 0.62

<0.01 0.7 0.8 0.2 <0.01 0.9 <0.0001 <0.0001 0.6 <0.0001 0.1 0.1 0.06 0.7 <0.0001 0.4 <0.01 <0.001 0.6 0.4 <0.01 <0.0001 <0.0001 0.3 <0.001 <0.0001 <0.0001 <0.01 <0.0001

Age Gender Body mass index Systolic blood pressure Diastolic blood pressure Current smoking eGFRa CKD stage Kidney disease Phosphate binders Antihypertensive drugs Statins Antiplatelet drugs Anticoagulant drugs Erythropoietin therapy Serum CRP level Hemoglobin Serum bicarbonate level Serum albumin level Serum calcium level Serum phosphate level Serum cholesterol level Serum LDL cholesterol level Serum triglyceride level Serum urea level Serum creatinine level Serum b2 microglobulin level Serum Indole-3 acetic acid level Serum indoxyl sulfate level

AHR, aryl hydrocarbon receptor; AP, activating potential; CKD, chronic kidney disease; CRP, C-reactive protein; eGFR, estimated glomerular filtration rate; LDL, low-density lipoprotein. a Calculated by Modification of Diet in Renal Disease (MDRD) study formula for nondialyzed CKD patients (n ¼ 51).

Table 3 | Multivariate linear regression analysis for evaluating the relation between independent variables and AHR-AP in stage 3 to 5D CKD patients (n [ 116) AHR-AP Variables Age Diastolic blood pressure CKD stage Stage 4 Stage 5 Stage 5D Phosphate binders Erythropoietin therapy Hemoglobin Serum bicarbonate level Serum phosphate level Serum cholesterol level Serum urea level Serum creatinine level Serum b2 microglobulin level Serum Indole-3 acetic acid level Serum indoxyl sulfate level

Estimate 0.68 0.12 –6.15 23.7 22.03 3.56 12.68 5.06 –4.12 –10.96 –0.67 –0.15 –0.06 0.15 –0.29 0.62

95% CI

P value

[0.04 to 1.33] [0.49 to 0.73]

<0.05 0.7

[–38.59 to 26.29] [–24.27 to 71.68] [–30.87 to 74.92] [19.44 to 26.57] [6.64 to 32] [1.57 to 11.69] [7.27 to -0.98] [32.29 to 10.37] [7.64 to 6.3] [1.94 to 1.64] [0.12 to 0.01] [1.18 to 1.49] [2.22 to 1.63] [0.41 to 0.83]

0.7 0.3 0.4 0.7 0.2 0.13 <0.05 0.3 0.8 0.8 0.11 0.8 0.7 <0.001

AHR, aryl hydrocarbon receptor; AP, activating potential; CI, confidence interval; CKD, chronic kidney disease.

4

r

P value

0.12 0.04 –0.19 –0.14 –0.19 0.25 –0.56 0.56 0.10 0.46 –0.16 0.24 0.13 –0.01 0.29 0.07 –0.41 –0.39 –0.18 –0.46 0.35 –0.37 –0.27 0.06 0.53 0.64 0.69 0.24 0.60

0.4 0.7 0.2 0.3 0.2 0.08 <0.0001 <0.0001 0.5 <0.01 0.2 0.08 0.4 0.9 <0.05 0.6 <0.001 <0.01 0.2 <0.001 <0.01 <0.01 0.06 0.7 <0.0001 <0.0001 <0.0001 0.1 <0.0001

AHR, aryl hydrocarbon receptor; AP, activating potential; CKD, chronic kidney disease; CRP, C-reactive protein; eGFR, estimated glomerular filtration rate; LDL, lowdensity lipoprotein. a Calculated by Modification of Diet in Renal Disease (MDRD) study formula.

estimated glomerular filtration rate (eGFR; r ¼ –0.56, P < 0.0001) estimated by the MDRD simplified formula (Table 4 and Figure 1c), hemoglobin (r ¼ –0.41, P < 0.001), serum bicarbonate (r ¼ –0.39, P < 0.01), serum calcium (r ¼ –0.46, P < 0.001), and serum cholesterol (r ¼ –0.37, P < 0.01) (Table 4). AHR-AP positively correlated with CKD stage (r ¼ 0.56, P < 0.0001), phosphate binder therapy (r ¼ 0.46, P < 0.01), erythropoietin therapy (r ¼ 0.29, P < 0.05), serum phosphate (r ¼ 0.35, P < 0.01), urea (r ¼ 0.53, P < 0.0001), creatinine (r ¼ 0.64, P < 0.0001), b2-microglobulin (r ¼ 0.69, P < 0.0001), and IS (r ¼ 0.60, P < 0.0001) (Table 4). In multivariate linear regression analysis in stage 3 to 5 CKD patients, serum b2-microglobulin (estimate ¼ 3.7, P < 0.05), and serum IS (estimate ¼ 0.81, P < 0.05) were significantly associated with AHR-AP (Table 5). In the hemodialyzed stage 5D CKD subcohort (Table 6), AHR-AP negatively correlated with serum bicarbonate (r ¼ –0.33, P < 0.01) and positively correlated with IS (r ¼ 0.51, P < 0.0001) (Table 6 and Figure 1d). In multivariate linear regression analysis in hemodialyzed stage 5D CKD patients, serum IS (estimate ¼ 0.62, P < 0.001) was significantly associated with AHR-AP (Table 7). Kidney International (2018) -, -–-

clinical investigation

L Dou et al.: Aryl hydrocarbon receptor and CKD

Table 5 | Multivariate linear regression analysis for evaluating the relation between independent variables and AHR-AP in CKD stage 3 to 5 subcohort (n [ 51)

Table 7 | Multivariate linear regression analysis for evaluating the relation between independent variables and AHR-AP in hemodialyzed CKD stage 5D subcohort (n [ 65)

AHR-AP Variables Age Diastolic blood pressure CKD stage Stage 4 Stage 5 Phosphate binders Erythropoietin therapy Hemoglobin Serum bicarbonate level Serum calcium level Serum phosphate level Serum cholesterol level Serum urea level Serum creatinine level Serum b2 microglobulin level Serum Indole-3 acetic acid level Serum indoxyl sulfate level

95% CI

P value

[–0.49 to 0.68] [–0.4 to 0.94]

0.7 0.4

[–23.67 to 14.61] [–17.4 to 58.13] [25.31 to 30.66] [6.31 to 31.71] [5.78 to 4.82] [3.39 to 1.24] [160.74 to 15] [–61.51 to 5.28] [9.98 to 2.25] [–1.08 to 2.75] [0.44 to 0.03] [0.07 to 7.32] [3.63 to 1.03] [0.21 to 1.41]

0.6 0.3 0.8 0.2 0.9 0.3 0.1 0.1 0.2 0.4 0.1 <0.05 0.3 <0.05

Estimate 0.09 0.27 –4.53 20.37 2.67 12.7 –0.48 –1.07 –72.87 –28.11 –3.86 0.83 –0.2 3.7 –1.3 0.81

AHR-AP

AHR, aryl hydrocarbon receptor; AP, activating potential; CI, confidence interval; CKD, chronic kidney disease.

We then analyzed whether the level of AHR agonists in serum could be decreased after a dialysis session. As shown in Figure 1e, the AHR-AP of sera drawn after a dialysis session was decreased compared with AHR-AP of sera drawn before Table 6 | Spearman correlations of baseline characteristics with AHR-AP of serum in the hemodialyzed CKD stage 5D subcohort (n [ 65) AHR-AP Variables Age Gender Body mass index Systolic blood pressure Diastolic blood pressure Current smoking Kidney disease Phosphate binders Antihypertensive drugs Statins Antiplatelet drugs Anticoagulant drugs Erythropoietin therapy Serum CRP level Hemoglobin Serum bicarbonate level Serum albumin level Serum calcium level Serum phosphate level Serum cholesterol level Serum LDL cholesterol level Serum triglyceride level Serum urea level Serum creatinine level Serum b2 microglobulin level Serum Indole-3 acetic acid level Serum indoxyl sulfate level

r

P value

0.18 –0.02 0.1 –0.09 –0.04 –0.05 –0.12 –0.06 –0.04 0.02 –0.12 –0.08 0.19 –0.14 –0.04 –0.33 –0.2 0.2 –0.09 –0.01 –0.07 0.15 0.04 0.00 –0.19 –0.17 0.51

0.1 0.9 0.5 0.5 0.8 0.7 0.3 0.6 0.7 0.9 0.3 0.5 0.1 0.3 0.8 <0.01 0.1 0.1 0.5 0.9 0.6 0.3 0.7 1 0.1 0.2 <0.0001

AHR, aryl hydrocarbon receptor; AP, activating potential; CKD, chronic kidney disease; CRP, C-reactive protein; LDL, low-density lipoprotein. Kidney International (2018) -, -–-

Variables

Estimate

95% CI

P value

0.71 0.28 1.02 15.66 12.11 –6.71 –2.54 2.54 –0.99 –0.07 0.23 –0.49 0.62

[–0.42 to 1.83] [–0.66 to 1.23] [33.84 to 35.88] [13.86 to 45.18] [0.41 to 24.63] [13.72 to 0.3] [–32.64 to 27.56] [9.81 to 14.88] [–4.06 to 2.08] [0.17 to 0.03] [–1.6 to 2.05] [3.32 to 2.34] [0.33 to 0.9]

0.2 0.5 0.9 0.3 0.06 0.06 0.9 0.7 0.5 0.1 0.8 0.7 <0.001

Age Diastolic blood pressure Phosphate binders Erythropoietin therapy Hemoglobin Serum bicarbonate level Serum phosphate level Serum cholesterol level Serum urea level Serum creatinine level Serum b2 microglobulin level Serum Indole-3 acetic acid level Serum indoxyl sulfate level

AHR, aryl hydrocarbon receptor; AP, activating potential; CKD, chronic kidney disease; CI, confidence interval.

dialysis (P < 0.01). This decrease corresponded to an AHRAP reduction ratio of 27% (Supplementary Table S1). However, the values of AHR-AP after hemodialysis sessions remained higher than the values of AHR-AP of control serum. We measured the reduction of AHR agonists, IS and IAA, after a dialysis session. We observed an IS reduction ratio of 62% and an IAA reduction ratio of 60% (Supplementary Figure S1A and S1B, and Supplementary Table S1). The reduction of serum AHR-AP after dialysis was related to the reduction of serum IS (r ¼ 0.47, P < 0.05), but not to the reduction of IAA (Table 8), small soluble uremic toxin urea, or middle molecule b2microglobulin (Table 8). We finally tested the AHR-AP of IS in vitro. Dose-response experiments showed a significant increase in AHR-AP induced by IS at concentrations found in the serum of CKD patients (50, 100, and 200 mM) (Figure 2a). We performed a kinetic study with the highest uremic concentration of IS tested, 200mM. The AHR-AP of IS was detectable after 4 hours of incubation, reached a plateau between 4 and 8 hours, and decreased at 24 hours (Figure 2b). Even in the presence of a serum control, the addition of IS at a uremic concentration of 200 mM strongly increased the AHR-AP (Figure 2c). In CKD, the protein binding of IS is about 90%, and free IS represents 10% of total IS. So we tested the AHR-AP of free IS by exposing cells to concentrations of free Table 8 | Spearman correlations between the decrease in serum AHR-AP and the decrease in uremic toxin levels during hemodialysis session AHR-AP reduction ratio r

P value

0.47 –0.11 –0.04 –0.11

<0.05 0.3 0.4 0.4

Variables Indoxyl sulfate reduction ratio Indole-3 acetic acid reduction ratio Urea reduction ratio b2 microglobulin reduction ratio

AHR, aryl hydrocarbon receptor; AP, activating potential.

5

clinical investigation

a

L Dou et al.: Aryl hydrocarbon receptor and CKD

150

100 **

**

b AHR-AP (AU)

AHR-AP (AU)

**

50

150

100

50

0

0 Control

IS 50 µM

IS 100 µM

c

IS 200 µM

1h IS 200 µM - +

4h - +

8h + -

24 h - +

*

120

AHR-AP (AU)

100 80 60 40 20 0 Control serum

-

+

+

IS 200 µM

-

-

+

Figure 2 | The aryl hydrocarbon receptor (AHR) agonist indoxyl sulfate (IS) induces AHR activation in vitro. (a) Dose effect of AHR activation in HG2L7.5c1 cells induced by IS was studied after 4 hours of incubation in complete medium. Values are expressed as mean  SEM of 4 independent experiments. **P < 0.01 versus potassium chloride control. (b) Kinetic of AHR activation in HG2L7.5c1 cells induced by IS at the uremic concentration of 200 mM. Values are expressed as mean  SEM of 3 independent experiments. (c) Effect of IS (200 mM) supplementation of control serum on AHR activating potential (AP). Data, expressed in AU, represent the values obtained with 6 different sera. *P < 0.05 versus control serum without IS.

IS found in uremic serum (1–20 mM),27 in medium without protein to rule out any IS protein-binding. All uremic concentrations of free IS significantly increased AHR-AP (Supplementary Figure S2), without dose-effect. AHR pathway is activated in blood cells from patients with CKD

The increase in AHR agonists in serum from CKD patients led us to study AHR activation in blood cells that are chronically exposed to these agonists. To demonstrate AHR pathway activation in cells from CKD patients, we compared the expression of AHR target genes CYP1A1 and AHRR in whole blood from a subgroup of 20 patients with CKD stage 5D and 17 control subjects. In blood, CYP1A1 is known to be expressed in lymphocytes, monocytes, and platelets; AHRR is expressed in lymphocytes and monocytes; both are up-regulated by AHR agonists.28–30 mRNA levels of AHR target genes CYP1A1 (Figure 3a) and AHRR (Figure 3b) were significantly increased in CKD patients compared with control subjects. Indeed, the relative levels of CYP1A1 mRNA were 7.7  4.0 in patients with stage 5D CKD versus 1.9  2.8 in control subjects (P < 0.05). In addition, the AHRR mRNA levels were significantly higher in patients with CKD than in control subjects (respectively, 18.1  6.4 vs. 4.8  2.6; P < 0.01). No correlation between blood cell count and CYP1A1 or AHRR mRNA level was found. 6

AHR-AP is associated with cardiovascular events in patients with CKD

We performed a crude analysis to study the relation between AHR-AP of serum, mortality, and cardiovascular events in the 116 patients with stage 3 to 5D CKD. After a mean follow-up of 871  314 days, 22 patients (18.9%) died. Among the 22 patients who died, 20 were on dialysis. Eight patients who died had AHR-AP values below the median value (44 AU) versus 14 patients with AHR-AP values above the median. Thirty-two patients (27%) displayed a cardiovascular event, which included a new nonfatal cardiovascular event (myocardial infarction n ¼ 10, peripheral vascular event with amputation or need for angioplasty n ¼ 3, and stroke n ¼ 5) or death from cardiovascular cause (n ¼ 14). Among the patients, 27 were undergoing dialysis. Eleven patients with a cardiovascular event had AHR-AP values below the median and 21 above it. During the follow-up, 24 patients were excluded at some point, 10 because of kidney transplantation and 14 because of loss to follow-up. The Kaplan-Meier survival analyses revealed that all-cause mortality (Figure 4a) was not significantly different in CKD patients with AHR-AP above the median (44 AU) than in those with AHR-AP below the median (Gehan-Breslow-Wilcoxon comparison of the curves: P ¼ 0.21). However, survival without cardiovascular events (Figure 4b) was lower in CKD patients with AHR-AP above the median than in those with AHR-AP below Kidney International (2018) -, -–-

clinical investigation

L Dou et al.: Aryl hydrocarbon receptor and CKD

a

*

b

**

25

10 AHRR mRNA (normalized expression)

CYP1A1 mRNA (normalized expression)

12

8 6 4 2 0 Controls (n = 17)

CKD stage 5D (n = 20)

20 15 10 5 0

Controls (n = 17)

CKD stage 5D (n = 20)

Figure 3 | The aryl hydrocarbon receptor (AHR) pathway is activated in blood cells from patients with chronic kidney disease (CKD). The AHR target genes (a) CYP1A1 and (b) AHRR are overexpressed in whole blood of patients with stage 5D CKD (n ¼ 20) compared with controls (n ¼ 17). *P < 0.05, **P < 0.01.

the median (Gehan-Breslow-Wilcoxon comparison of the curves: P ¼ 0.05). Because age was significantly different in CKD patients with AHR-AP above and below the median, we performed Cox analyses with serum AHR-AP and age entered as continuous variables. In these analyses, age was a significant predictor of cardiovascular events, whereas serum AHRAP failed to reach significance (Supplementary Table S2).

a

AHR-AP <44 AU

Cumulative survival (%)

100 80

AHR-AP ≥44 AU

60 P = 0.19

40 20 0 0

400

200

600

800

1000

Time (days)

Fraction alive and free of cardiovascular events (%)

b 100

AHR-AP <44 AU

80 60 P = 0.05

AHR-AP ≥44 AU

40 20 0 0

200

400

600

800

1000

Time (days)

Figure 4 | Kaplan-Meier estimates of (a) cumulative survival and (b) major cardiovascular events of all patients according to aryl hydrocarbon receptor–activating potential (AHR-AP) values above and below the median of 44 arbitrary units (AU). (a) P ¼ 0.19 in a Gehan-Breslow-Wilcoxon comparison of the curves. (b) P ¼ 0.05 in a Gehan-Breslow-Wilcoxon comparison of the curves. Kidney International (2018) -, -–-

The AHR pathway is activated in mice with 5/6 nephrectomy

In order to explore the AHR activation during uremia, CKD was induced in mice by performing 5/6 nephrectomy (5/6 Nx). Within 6 weeks after the 5/6 Nx, serum urea concentration increased approximately 3-fold (Figure 5a), and serum creatinine concentration was significantly elevated (Figure 5b) compared with sham-operated mice. The concentration of IS in serum was significantly higher in 5/6 Nx mice than in sham-operated mice (respectively, 123  33 mM vs. 31  8 mM; mean  SEM, n ¼ 6; Figure 4c). We then determined the AHR-AP of mouse serum. In 5/6 Nx mice, the AHR-AP was significantly increased compared with sham-operated mice (Figure 5d). To demonstrate AHR activation in CKD mice, we examined the expression of the AHR target gene Cyp1a1 in 5/6 Nx wild-type (WT) mice and in 5/6 Nx Ahr-invalidated mice (Ahr–/–). Six weeks after surgery, a significant increase in plasma urea and creatinine was observed in WT and Ahr–/– mice compared with sham-operated mice (Supplementary Table S3). In 5/6 Nx WT mice, we demonstrated that Cyp1a1 mRNA was induced in the aorta and the heart (Figure 5e and f), whereas no induction was observed in 5/6 Nx Ahr–/– mice. Indoxyl sulfate activates AHR in mice

To determine the specific effect of IS, we injected IS (400 mg/kg) in WT and Ahr–/– mice. Time course of IS levels was first determined in WT mice. As shown in Supplementary Figure S3A, IS concentration reached 1222  118 mM (mean  SEM, n ¼ 3) 30 minutes after IS injection, and decreased strongly at 1 hour and 2 hours (respectively, 304  77 mM and 182  35 mM, mean  SEM, n ¼ 4). At 4 hours, IS concentration in serum was 147  81 mM (mean  SEM, n ¼ 4) and significantly higher than IS level before IS injection (16  2 mM, mean  SEM, n ¼ 5). The IS concentration at 4 hours (147  81 mM) was comparable to the IS mean value found in nephrectomized mice (123  33 mM). WT and Ahr–/– mice were then injected with 400 mg/kg of IS for 5 consecutive days. AHR-AP was determined 4 hours after the last IS injection to achieve an IS level in the uremic 7

clinical investigation

a

30

L Dou et al.: Aryl hydrocarbon receptor and CKD

b

**

**

80

c

d

200

100

10

0

20

100

50

Cyp1A1 mRNA in the heart (normalized expression)

0

WT WT sham 5/6 Nx

**

P = 0.07

4

5/6 Nx

0

2

**

8

40

WT WT sham 5/6 Nx

f

12

60

20

0 WT WT sham 5/6 Nx

*

Cyp1A1 mRNA in the aorta (normalized expression)

40

80 150

0 WT WT sham 5/6 Nx

e

60

AHR-AP (AU)

20

Serum Indoxyl Sulfate (µmol/l)

Serum creatinine (µmol/l)

Serum urea (mmol/l)

** **

1

0 -

+ WT

-

+ Ahr

5/6 Nx

-

-/-

+ WT

-

+ Ahr

-/-

Figure 5 | AHR is activated in mice with chronic kidney disease. (a–c) Concentrations of urea, creatinine, and indoxyl sulfate, measured in wild-type (WT) serum obtained from sham-operated or 5/6 nephrectomized (Nx) mice. (d) Aryl hydrocarbon receptor–activating potential (AHR-AP) from sham-operated or 5/6 Nx mice. Expression of Cyp1a1 mRNA in the (e) aorta and (f) heart isolated from sham-operated or 5/6 Nx mice (WT or Ah–/–). The Mann-Whitney U test was used for statistical analysis. *P < 0.05, **P < 0.01. Values are means  SEM (a–d: n ¼ 6/group; e and f: n ¼ 5/group).

range. IS serum concentrations increased approximately 10-fold 4 hours after the last IS injection, and were similar in WT and Ahr–/– mice (Supplementary Figure S3B). In IS-injected WT and Ahr–/– mice, AHR-AP values increased 2-fold compared with control mice injected with potassium chloride (KCl) vehicle (Figure 6a). We then studied IS-induced AHR activation in liver by analyzing the up-regulation of different AHR target genes (Figure 6b–f).31 Hepatic mRNA expression of Cyp1a1, cytochromes P450 1A2 (Cyp1a2) and 2S1 (Cyp2s1), NAD(P)H quinone oxidoreductase 1, and UDP-glucuronosyltransferase 1a1 were increased in IS-injected WT mice (Figure 6b–f), but not in IS-injected Ahr–/– mice (Figure 6b–f). We finally studied IS-induced AHR activation in the cardiovascular system. We observed a strong induction of Cyp1a1 mRNA in the aorta and in the heart of IS-injected WT mice (Figure 7a 8

and b), but not in IS-injected Ahr–/– mice. In Ahr–/– mice, the levels of Cyp1a1 mRNA were identical in both KCl-injected and IS-injected animals. DISCUSSION

Our results demonstrate an activation of AHR in humans and mice with CKD. We first studied the activation of the AHR pathway by uremic serum ex vivo using the AHR-AP, which evaluates the overall load of AHR agonists in serum. An increase in serum AHR-AP was recently demonstrated in 20 patients undergoing chronic hemodialysis using vascular smooth muscle cells.23 In a larger cohort, we confirmed here the increased AHR-AP of serum from patients with CKD stage 5D. In addition, we showed that patients with stages 3 to 5 CKD also display a significant elevation of serum AHR-AP. So the increase in serum AHR-AP appears to be a general Kidney International (2018) -, -–-

clinical investigation

L Dou et al.: Aryl hydrocarbon receptor and CKD

a

b *

**

*

10000 Liver mRNA (normalized expression)

AHR-AP (AU)

150

100

50

0 IS (400 mg/kg)

-

+

-

c

+ Ahr

WT

1 IS (400 mg/kg)

*

20000

10000

-

+

-

+

P = 0.06

-

*

2

1

0 IS (400 mg/Kg)

-

+

f

P = 0.06

-

Liver mRNA (normalized expression)

80 60 40 20

+ Ahr -/-

ugt1a1 5000

100

+ Ahr -/-

WT

Nqo1 120

+

Cyp2s1

Ahr -/-

WT

e

-

d Liver mRNA (normalized expression)

Liver mRNA (normalized expression)

10

WT

30000

Liver mRNA (normalized expression)

100

3

*

* *

4000 3000 2000 1000 0

0 IS (400 mg/Kg)

1000

*

40000

**

**

-/-

Cyp 1a2

0 IS (400 mg/Kg)

Cyp1a1

-

+ WT

-

+ Ahr -/-

IS (400 mg/Kg)

-

+ WT

-

+ Ahr -/-

Figure 6 | Injection of indoxyl sulfate (IS) in mice increased serum aryl hydrocarbon receptor (AHR)–activating potential (AP) and activates AHR in the liver. (a) AHR-AP of serums from mice (wild-type [WT] or Ahr–/–) injected with potassium chloride (KCl; control, –) or IS (þ). (b–f) Expression of Cyp1a1, Cyp 1a2, Cyp2s1, NAD(P)H quinone oxidoreductase 1 (Nqo1), and UDP-glucuronosyltransferase 1a1 (ugt1a1) in the liver isolated from mice (WT or Ahr –/–) injected with KCl (control, –) or IS (þ). The Mann-Whitney U test was used for statistical analysis. *P < 0.05, **P < 0.01. Values are means  SEM (n ¼ 4–5/group).

phenomenon in CKD, found in patients undergoing hemodialysis and in pre-dialysis. The increase in serum AHR-AP we observed in CKD mice is consistent with this view. Higher levels of serum AHR agonists have previously been reported in smokers,32 as well as in patients with diabetes,33 especially with severe diabetic nephropathy.34 The authors suggested increased level of serum AHR agonists as a risk Kidney International (2018) -, -–-

factor for the progression of kidney disease. Indeed, the activation of AHR could lead to the impairment of renal function by podocyte aggression.35 Here, we found a negative correlation between the level of AHR agonists in serum and eGFR, confirming the association between CKD and AHR activation. Furthermore, we observed a decrease in AHR-AP after a dialysis session. This suggests that factors responsible 9

clinical investigation

a

L Dou et al.: Aryl hydrocarbon receptor and CKD

b **

100

10

1 IS (400 mg/kg)

-

+ WT

1000

*

**

** Cyp1A1 mRNA in the heart (normalized expression)

Cyp1A1 mRNA in the aorta (normalized expression)

1000

-

+ Ahr

-/-

100

10

1 IS (400 mg/kg)

-

+ WT

-

+ Ahr

-/-

Figure 7 | Injection of indoxyl sulfate (IS) in mice activates aryl hydrocarbon receptor (AHR) in cardiovascular cells. Expression of Cyp1a1 mRNA in the (a) aorta and (b) heart isolated from mice (wild-type [WT] or Ahr –/–) injected with potassium chloride (control, –) or IS (þ). The Mann-Whitney U test was used for statistical analysis. *P < 0.05, **P < 0.01. Values are means  SEM (n ¼ 4–5/group).

for increased AHR-AP of serum from CKD patients could be uremic toxins. In patients with CKD, AHR-AP could reflect the accumulation of uremic toxins that are AHR agonists. Here, we observe that AHR-AP is correlated with the serum concentration of the uremic toxin IS in CKD patients in simple and multivariate analysis. Furthermore, the addition of IS to serum from healthy controls increased the serum AHRAP, and the AHR-AP of serum is increased after IS injection in mice. Because about 90% of IS circulates in protein-bound form and w10% in free form, we tested in vitro the AHRAP of both forms of IS at uremic concentrations. Highly protein-bound IS, represented by IS added to serum from healthy controls, strongly increased the serum AHR-AP. In addition, free IS at uremic concentrations significantly increased AHR-AP in vitro, but without dose effect. One can suppose that free IS enters into the cell more easily than protein-bound IS, leading AHR-AP of free IS to rapidly reach a plateau in our bioassay. We analyzed the reduction ratio of AHR-AP after a dialysis session. It was rather low (27%), and much lower than the reduction ratio of IS and of other measured uremic toxins. Note that the reduction ratio of IS observed in our study was in the highest range described in the literature.36 The reduction ratio of AHR-AP by dialysis was correlated with that of IS. However, the greater reduction in IS than that in AHR-AP suggests that something other than IS contributes to AHR-AP of uremic serum. Therefore, our data suggest that IS could participate in AHR activation, but also support a role of other uremic factors such as kynurenine10 or other natural or xenobiotic agonists in AHR activation in CKD.13,15 If the uremic serum contains AHR agonists, these agonists could activate the expression of AHR target genes. Here, we demonstrate the up-regulation of mRNA levels of the AHR targets genes CYP1A1 and AHRR in the blood of CKD patients. In CKD mice, the AHR-target gene Cyp1a1 is overexpressed in the heart and vessels, confirming the activation of the AHR pathway in cells of the cardiovascular system. The lack of increased expression of AHR target genes in Ahr–/– 10

CKD mice shows the role of AHR in vascular response to uremia during CKD. We studied the effect of IS in mice with normal renal function after its injection at a concentration of 400 mg/kg. We showed that IS injection increases the expression of AHR target genes in liver, and also in heart and vessels. AHR activation in kidney after injection of IS at 800 mg/kg was previously shown by Ichii et al. in mouse with normal renal function.35 Interestingly, they observed that the main renal lesion was ischemic, certainly related to vascular toxicity of IS besides podocyte aggression. In our conditions, IS levels reached 1000 mM at 30 minutes, and decreased quickly until 4 hours to reach levels observed in dialysis patients or in 5/6 Nx mice. Because IS exhibits a w500-fold lower potency in terms of transcriptional activation of the mouse AHR relative to the human AHR,11 one can suppose that higher concentrations of IS should be necessary to achieve, in mice, a level of AHR activation equivalent to that in humans. AHR activation by IS in mice could be blunt, reducing the impact of IS in a mouse model of CKD. This could explain the better induction of cyp1A1 in a short-term IS-exposed mouse in comparison with the 5/6 Nx mouse. Regardless, the lack of increased expression of AHR target genes in Ahr–/– IS-injected mice confirmed the role of AHR in response to IS. Previous works from Barreto and coworkers and from our group have shown that the serum levels of 2 AHR agonists, IS and IAA, were correlated to mortality and cardiovascular mortality in patients with CKD.7,8 Recently, the role of another uremic toxin and an AHR agonist, kynurenine, was demonstrated in vivo in a mouse model of stroke,37 reinforcing the role of AHR activation in cardiovascular complications of CKD. Here, we observe that an increase in AHR agonists, globally evaluated in the uremic serum by AHR-AP, is associated with an increased risk of cardiovascular events in patients with CKD. However, we failed to show that serum AHR-AP is a significant predictor of cardiovascular events, probably because of the small size of our population and because of the difference in age between patients with high and low AHR-AP. Kidney International (2018) -, -–-

L Dou et al.: Aryl hydrocarbon receptor and CKD

AHR activation in CKD could be a mechanism to explain the harmful effect of uremic AHR agonists on vascular cells. Shivanna et al.23 have demonstrated that tissue factor procoagulant activity increases in vascular smooth muscle cells following exposure to sera from hemodialyzed patients, and is correlated to the ability of uremic serum to activate AHR in vascular smooth muscle cells. In addition, in these cells, IS increases the expression and activation of tissue factor in an AHR-dependent manner, which could explain the increased risk of stent thrombosis in coronary arteries of patients with CKD.38 We and others have previously shown that AHR activation induced by indolic uremic toxins mediates a pro-inflammatory and procoagulant phenotype in vascular cells in vitro8,22,39 and in vivo.40 The uremic toxin agonist of AHR, kynurenine, has a similar effect to TNF, also suggesting a pro-inflammatory effect of uremic AHR agonists.41 The activation of the AHR pathway in the blood and in the cardiovascular tissues may have important physiopathological implications. Because AHR is a critical receptor for various toxins with the ability to promote prothrombotic and pro-inflammatory pathways, it provides a good candidate to link the accumulation of uremic toxins with cardiovascular complications of CKD.24 Indeed, in nonuremic conditions, AHR activation is associated with an increased cardiovascular risk.18 Exposure to TCDD contained in Agent Orange used during the Vietnam War has been associated with various cardiovascular complications from coronary heart disease to stroke.19 Workers exposed to AHR agonists have an increased cardiovascular risk.42 Our results suggest the same effects in CKD. In patients with coronary heart disease, serum concentration of the AHR agonist kynurenic acid is related to cardiovascular events, suggesting that AHR activation could be a more general phenomenon, not restricted to uremic patients but also affecting other patients with increased production or exposure to AHR agonists.43 Our demonstration of AHR activation during CKD provides a rationale to target it and try to limit the harmful effects of uremic toxins on vascular outcome.44 Few therapies are available to reduce the effects of uremic toxins that are poorly dialyzable as indolic toxins. Here, we showed that the AHR-AP of serum could be decreased by an AHR antagonist. The inhibition of AHR could be an interesting way to restrict or reduce the toxicity of uremic AHR agonists.44 Despite its strengths, our study has some limitations. If we observed an association between AHR-AP and cardiovascular events, our population is too small for determining that AHRAP is predictive of cardiovascular events independently of other factors. A well-powered prospective study should be performed to determine the threshold of AHR-AP indicative of an increased risk of cardiovascular events. In addition, the lack of measurement of free IS in serum represents a limitation of the study. Furthermore, we did not measure AHR agonists other than IS and IAA, notably uremic toxins from the kynurenine pathway, nor can we exclude AHR-activating drugs in serum. Kidney International (2018) -, -–-

clinical investigation

In conclusion, the present study in patients and mice with CKD demonstrates a strong increase in serum levels of AHR agonists, leading to AHR activation in cells from the cardiovascular system. IS could participate in AHR activation, but other CKD-related factors are probably involved. MATERIALS AND METHODS Patients We performed a single-center prospective study in 116 CKD patients with CKD stage 3 to 5D, enrolled between April 2008 and July 2013, selected according to the following criteria: age >18 years; no cardiovascular event, infection, or surgical intervention (except for vascular-access angioplasty) in the last 3 months; no pregnancy; no recent history of malignancy; and no intensive treatment. CKD stages 3, 4, and 5 were defined according to eGFR calculated by the Modification of Diet in Renal Disease (MDRD) study formula. CKD stage 3 (n ¼ 20) was defined as 30 ml/min per 1.73 m2 < eGFR < 60 ml/min per 1.73 m2, stage 4 (n ¼ 18) as 15 ml/min per 1.73 m2 < eGFR < 30 ml/min per 1.73 m2, and stage 5 (n ¼ 13) as eGFR < 15 ml/min per 1.73 m2. Stage 5D (n ¼ 65) was defined as patients undergoing hemodialysis. The 65 CKD patients undergoing hemodialysis had been dialyzed at least 3 times a week for a minimum of 6 months. Among them, 12 patients had a residual renal function. Blood samples were drawn in patients during the midweek hemodialysis session. Standard laboratory procedures were used for blood chemistry evaluations. Total IS and IAA were measured by high-performance liquid chromatography, according to methods described by Calaf et al.45 Patients were compared with 52 control subjects with normal renal function and without diabetes. They were recruited by the Centre d’Investigation Clinique of the Assistance Publique Hôpitaux de Marseille. Their mean age was 62 years (range: 40–85 years), and the ratio of women to men was 28:24. During the study period, clinical events including overall mortality and cardiovascular events were recorded. The causes of death were categorized as cardiovascular, infectious, or other. Death occurring outside the hospital with no identified cause was regarded as sudden death and was included as cardiovascular death. Deaths were classified by the treating physician and reviewed by a member of the medical monitoring committee independently of the end point analysis. Cardiovascular events included death from cardiac cause, myocardial infarction, stroke, and peripheral vascular disease. Myocardial infarction was defined as the presence of at least 2 of the following criteria: chest pain of typical duration and intensity, increased cardiac enzyme concentrations, electrocardiographical changes, and endovascular procedure (angioplasty or coronary stenting). Stroke was defined on the basis of neurological symptoms associated with radiologic diagnosis. Peripheral vascular disease was defined as the need for amputation attributable to ischemic disease or symptomatic peripheral vascular disease with need for revascularization. We studied the effect of the hemodialysis session on AHRAP in a subgroup of 11 hemodialyzed patients. Hemodialysis sessions were performed during 240 minutes with synthetic membranes XeniumþH21 (Baxter, Deerfield, IL), TS-2.1SL (Toray Group, Tokyo, Japan), and Elisio-21H (Nipro Corporation, Osaka, Japan). Blood flow rate was 300 ml/min and dialysate flow rate was 500 ml/min; ultrapure dialysate was used. Heparinization during the dialysis session was performed with continuous infusion of heparin. Predialysis blood was drawn at onset. Postdialysis blood was drawn from the arterial needle at the end of the dialysis session after 11

clinical investigation

decreasing blood flow rate to 50 ml/min during 15 seconds to avoid recirculation.46 Reduction ratio (RR) was defined as a function of predialysis (Cpre) and corrected postdialysis (CpostCorr) concentration (RR ¼ [1 – (CpostCorr/Cpre)]  100). Concentration at the dialysis end (Cpost) was corrected for hemoconcentration as follows: CpostCorr ¼ Cpost(1 þ Dweight/0,2weight end).47 Urea reduction ratio (URR) was 68  10% and eKT/V urea was 1.20  0.36. Mice Ahr-KO (Ahr/) mice (B6.129-Ahrtm1Bra/J)48 were purchased from Jackson Laboratories and maintained as a breeding colony in the animal care facility at the Faculty of Medicine of Marseille. C57BL/6J WT mice were used as experimental controls. Genotypes were confirmed by polymerase chain reaction analysis of DNA from tail clippings. Female and male mice were used in this study. Comparisons were made between sex- and age-matched groups. Mice experimental procedures are described in the Supplementary Material. Cell culture and chemically activated luciferase expression bioassay We evaluated the presence of AHR agonists in the serum of CKD patients, control subjects, 5/6 Nx mice, and IS-injected mice using the AHR-responsive chemically activated luciferase expression cell bioassay.25,26 This assay also allowed us to examine the ability of IS to simulate the AHR pathway. A recombinant human hepatoma cell line (HG2L7.5c1) stably transfected with an AHR-responsive dioxinresponsive elements–driven firefly luciferase reporter plasmid was used, as previously described.26 Cells were plated at 750,000 cells/ml into white, clear-bottomed 96-well plates and cultured in aMEM medium containing Glutamax (Life Technologies, Saint-Aubin, France) supplemented with 10% fetal calf serum (Dominique Dutscher, Brumath, France), and 400 mg/ml Geneticin (Gibco, Life Technologies). After 48 hours (37 C, 5% CO2), cells were incubated with 50% serum in aMEM medium with or without the AHR antagonist CH223191 at 10 mM during 24 hours. Cells were also incubated during 24 hours with control sera supplemented with IS at 200 mM. Dose-response and kinetic studies of total IS (compared with its vehicle control KCl at the same concentration) were also performed in complete medium. To study the effect of uremic concentrations of free IS, cells were incubated during 4 hours in aMEM medium (without protein) supplemented with free IS concentrations found in CKD patients (1–20 mM).27 Quality-control solutions of the potent AHR agonist 6-formylindolo [3,2-b]carbazole (FICZ; 1 pM, 1 nM, and 10 nM), FICZ vehicle control DMSO, a medium blank, and the same normal control serum were added in triplicate to every 96-well plate, as internal controls, and incubated for 24 hours. For IS kinetic study, FICZ was incubated during the same time as IS. Afterwards, the medium was removed, and the cells were washed with phosphate-buffered saline (Life Technologies), incubated at room temperature with 40 ml lysis reagent (Promega, Charbonnieres, France) with shaking for 20 minutes, and the plate was stored at –80 C until analysis. After thawing, the plate was placed in an Infinite 200 microplate luminometer (Tecan, Lyon, France), and 100 ml luciferin reagent (Promega) was automatically injected. The light output reflecting luciferase activity was measured in relative light units (RLU) after a read integration time of 5 seconds. The luciferase activity was normalized to the protein concentration, which was determined by the bicinchoninic acid assay method. Values were converted to FICZ equivalents by dividing the normalized RLU of the sample by the normalized RLU obtained with 10 nM FICZ, multiplied by 100, and 12

L Dou et al.: Aryl hydrocarbon receptor and CKD

expressed in AUs. The value obtained was designated as the AHR-AP and reflects the levels of AHR agonists in serum. The dose-response curve obtained with FICZ standards (concentration range of 1–10 nM) was like that obtained with TCDD standards at the same concentration range (not shown), and as such, we therefore used FICZ because of its lesser toxicity when compared with TCDD. RNA extraction and quantitative reverse transcription polymerase chain reaction analysis of mRNA expression Provided in the Supplementary Materials and Methods. Statistical analysis Data are expressed as mean  SD for values with normal distribution or median (min; max) for non-normally distributed values. All numerical variables were tested for normality by the Shapiro-Wilk test. Statistical analyses were performed with Prism software (GraphPad Software Inc, La Jolla, CA). Differences were considered significant when the P value was less than 0.05. Baseline variables were compared by Fisher exact test for categorical variables, t-tests for continuous variables with Gaussian distributions, and MannWhitney U tests for continuous variables with non-Gaussian distributions. Correlations between serum AHR-AP and continuous variables were obtained using Spearman correlation coefficients. To identify factors independently associated with serum AHR-AP, multiple linear regression analyses were performed with R software. The Kaplan-Meier method was used to estimate the cumulative event-free rate for the time to overall mortality and the first cardiovascular event in patients with AHR-AP above and below the median (44 AU). The Gehan-Breslow-Wilcoxon test was used to compare the Kaplan-Meier curves. Cox regression analyses of risk factors for cardiovascular events were performed with serum AHRAP and age entered as continuous variables using R software. For in vitro and animal studies, significant differences were revealed by the Wilcoxon signed-rank test or by the Mann-Whitney U test. Data are expressed as mean SEM. A P value lower than 0.05 was considered significant. Study approval All participants gave their written informed consent. The study was approved by the local ethics committee and conforms to the principles outlined in the Declaration of Helsinki. The animal experiments conformed to the Directive 2010/63/EU of the European Parliament and were approved by the local ethics committee (Comité d’Ethique en Expérimentation Animale de Marseille, C2EA-14). DISCLOSURE

All the authors declared no competing interests. ACKNOWLEDGMENTS

We thank C. Scagliarini and S. Hutter for technical assistance. This work was supported by funding from the Aix-Marseille Université, the Institut National de la santé et de la Recherche médicale (INSERM), the European Uremic Toxins (EUTox) Work Group, and a grant from Fondation du Rein/FNAIR. SUPPLEMENTARY MATERIAL Supplementary Materials and Methods. Table S1. Levels of aryl hydrocarbon receptor activating potential (AHR-AP) and uremic toxins before and after dialysis, and reduction ratios after dialysis session. Kidney International (2018) -, -–-

L Dou et al.: Aryl hydrocarbon receptor and CKD

Table S2. Cox regression analysis of risk factors for cardiovascular events with serum aryl hydrocarbon receptor activating potential (AHR-AP) and age entered as variables. Table S3. Plasma concentration of urea and creatinine in shamoperated and 5/6 nephrectomy (Nx) mice. Figure S1. (A) Indoxyl sulfate (IS) and (B) indole-3 acetic acid (IAA) serum levels decrease after a dialysis session. Data, expressed in mM, represent the values obtained with 11 different sera. **P < 0.01, ***P < 0.001. Figure S2. Free indoxyl sulfate (IS) concentrations corresponding to serum level values in chronic kidney disease (CKD) patients increase aryl hydrocarbon receptor activating potential (AHR-AP). Cells were incubated with different IS concentrations in medium without protein to rule out any IS protein binding. All uremic concentrations of free IS significantly increased AHR-AP. Data represent the mean  SEM of 6 different experiments ***P < 0.001 versus control. Figure S3. Serum indoxyl sulfate (IS) concentrations in mice. (A) Time course of IS levels in serum of wild-type (WT) mice after IS injection (400 mg/kg). IS serum levels were determined by high-performance liquid chromatography. (B) IS serum levels in potassium chloride (KCl) (control, –) or indoxyl sulfate (IS, þ) injected mice (WT or Ahr–/–). Values are means  SEM (n ¼ 3–5/group). *P < 0.05 versus T0. Supplementary material is linked to the online version of the paper at www.kidney-international.org.

REFERENCES 1. GBD 2013 Mortality and Causes of Death Collaborators. Global, regional, and national age-sex specific all-cause and cause-specific mortality for 240 causes of death, 1990-2013: a systematic analysis for the Global Burden of Disease Study 2013. Lancet. 2015;385: 117–171. 2. Gansevoort RT, Correa-Rotter R, Hemmelgarn BR, et al. Chronic kidney disease and cardiovascular risk: epidemiology, mechanisms, and prevention. Lancet. 2013;382:339–352. 3. Tonelli M, Muntner P, Lloyd A, et al. Risk of coronary events in people with chronic kidney disease compared with those with diabetes: a population-level cohort study. Lancet. 2012;380:807–814. 4. Drüeke TB, Massy ZA. Atherosclerosis in CKD: differences from the general population. Nat Rev Nephrol. 2010;6:723–735. 5. Vanholder R, Massy Z, Argiles A, et al. Chronic kidney disease as cause of cardiovascular morbidity and mortality. Nephrol Dial Transplant. 2005;20: 1048–1056. 6. Duranton F, Cohen G, De Smet R, et al. Normal and pathologic concentrations of uremic toxins. J Am Soc Nephrol. 2012;23:1258–1270. 7. Barreto FC, Barreto DV, Liabeuf S, et al. Serum indoxyl sulfate is associated with vascular disease and mortality in chronic kidney disease patients. Clin J Am Soc Nephrol. 2009;4:1551–1558. 8. Dou L, Sallée M, Cerini C, et al. The cardiovascular effect of the uremic solute indole-3 acetic Acid. J Am Soc Nephrol. 2015;26:876–887. 9. Vanholder R, Schepers E, Pletinck A, et al. The uremic toxicity of indoxyl sulfate and p-cresyl sulfate: a systematic review. J Am Soc Nephrol. 2014;25:1897–1907. 10. Heath-Pagliuso S, Rogers WJ, Tullis K, et al. Activation of the Ah receptor by tryptophan and tryptophan metabolites. Biochemistry. 1998;37: 11508–11515. 11. Schroeder JC, Dinatale BC, Murray IA, et al. The uremic toxin 3-indoxyl sulfate is a potent endogenous agonist for the human aryl hydrocarbon receptor. Biochemistry. 2010;49:393–400. 12. Hankinson O. The aryl hydrocarbon receptor complex. Annu Rev Pharmacol Toxicol. 1995;35:307–340. 13. Denison MS, Nagy SR. Activation of the aryl hydrocarbon receptor by structurally diverse exogenous and endogenous chemicals. Annu Rev Pharmacol Toxicol. 2003;43:309–334. 14. Stockinger B, Hirota K, Duarte J, et al. External influences on the immune system via activation of the aryl hydrocarbon receptor. Semin Immunol. 2011;23:99–105. 15. Denison MS, Soshilov AA, He G, et al. Exactly the same but different: promiscuity and diversity in the molecular mechanisms of action of the aryl hydrocarbon (dioxin) receptor. Toxicol Sci. 2011;124:1–22. Kidney International (2018) -, -–-

clinical investigation

16. Ma Q, Whitlock JP. A novel cytoplasmic protein that interacts with the Ah receptor, contains tetratricopeptide repeat motifs, and augments the transcriptional response to 2,3,7,8-tetrachlorodibenzo-p-dioxin. J Biol Chem. 1997;272:8878–8884. 17. Murray IA, Patterson AD, Perdew GH. Aryl hydrocarbon receptor ligands in cancer: friend and foe. Nat Rev Cancer. 2014;14:801–814. 18. Sallée M, Dou L, Cerini C, et al. The aryl hydrocarbon receptor-activating effect of uremic toxins from tryptophan metabolism: a new concept to understand cardiovascular complications of chronic kidney disease. Toxins. 2014;6:934–949. 19. Committee to Review the Health Effects in Vietnam Veterans of Exposure to Herbicides (Tenth Biennial Update), Board on the Health of Select Populations, Institute of Medicine et al. Veterans and Agent Orange: Update 2014. Washington, DC: National Academies Press; 2016. 20. Shan Q, Wang J, Huang F, et al. Augmented atherogenesis in ApoE-null mice co-exposed to polychlorinated biphenyls and 2,3,7, 8-tetrachlorodibenzo-p-dioxin. Toxicol Appl Pharmacol. 2014;276:136–146. 21. Lind PM, Orberg J, Edlund U-B, et al. The dioxin-like pollutant PCB 126 (3,3’,4,4’,5-pentachlorobiphenyl) affects risk factors for cardiovascular disease in female rats. Toxicol Lett. 2004;150:293–299. 22. Gondouin B, Cerini C, Dou L, et al. Indolic uremic solutes increase tissue factor production in endothelial cells by the aryl hydrocarbon receptor pathway. Kidney Int. 2013;84:733–744. 23. Shivanna S, Kolandaivelu K, Shashar M, et al. The aryl hydrocarbon receptor is a critical regulator of tissue factor stability and an antithrombotic target in uremia. J Am Soc Nephrol. 2016;27:189–201. 24. Vanholder R, Fouque D, Glorieux G, et al. Clinical management of the uraemic syndrome in chronic kidney disease. Lancet Diabetes Endocrinol. 2016;4:360–373. 25. Denison MS, Zhao B, Baston DS, et al. Recombinant cell bioassay systems for the detection and relative quantitation of halogenated dioxins and related chemicals. Talanta. 2004;63:1123–1133. 26. Brennan JC, He G, Tsutsumi T, et al. Development of species-specific Ah receptor-responsive third generation CALUX cell lines with enhanced responsiveness and improved detection limits. Environ Sci Technol. 2015;49:11903–11912. 27. Deltombe O, Van Biesen W, Glorieux G, et al. Exploring protein binding of uremic toxins in patients with different stages of chronic kidney disease and during hemodialysis. Toxins. 2015;7:3933–3946. 28. Bahari A, Mehrzad J, Mahmoudi M, et al. Cytochrome P450 isoforms are differently up-regulated in aflatoxin B1-exposed human lymphocytes and monocytes. Immunopharmacol Immunotoxicol. 2014;36:1–10. 29. Jarrar YB, Cho S-A, Oh K-S, et al. Identification of cytochrome P450s involved in the metabolism of arachidonic acid in human platelets. Prostaglandins Leukot Essent Fatty Acids. 2013;89:227–234. 30. Yamamoto J, Ihara K, Nakayama H, et al. Characteristic expression of aryl hydrocarbon receptor repressor gene in human tissues: organ-specific distribution and variable induction patterns in mononuclear cells. Life Sci. 2004;74:1039–1049. 31. Buckley DB, Klaassen CD. Induction of mouse UDPglucuronosyltransferase mRNA expression in liver and intestine by activators of aryl-hydrocarbon receptor, constitutive androstane receptor, pregnane X receptor, peroxisome proliferator-activated receptor alpha, and nuclear factor erythroid 2-related factor 2. Drug Metab Dispos. 2009;37:847–856. 32. Wiest EF, Warneke A, Walsh MT, et al. Association of serum aryl hydrocarbon receptor activity and RBC omega-3 polyunsaturated fatty acids with flow-mediated dilation in healthy, young Hispanic cigarette smokers. Toxicol Lett. 2015;232:422–428. 33. Roh E, Kwak SH, Jung HS, et al. Serum aryl hydrocarbon receptor ligand activity is associated with insulin resistance and resulting type 2 diabetes. Acta Diabetol. 2015;52:489–495. 34. Kim JT, Kim SS, Jun DW, et al. Serum arylhydrocarbon receptor transactivating activity is elevated in type 2 diabetic patients with diabetic nephropathy. J Diabetes Investig. 2013;4:483–491. 35. Ichii O, Otsuka-Kanazawa S, Nakamura T, et al. Podocyte injury caused by indoxyl sulfate, a uremic toxin and aryl-hydrocarbon receptor ligand. PloS One. 2014;9:e108448. 36. Basile C, Libutti P, Di Turo AL, et al. Removal of uraemic retention solutes in standard bicarbonate haemodialysis and long-hour slow-flow bicarbonate haemodialysis. Nephrol Dial Transplant. 2011;26:1296–1303. 37. Cuartero MI, Ballesteros I, de la Parra J, et al. L-kynurenine/aryl hydrocarbon receptor pathway mediates brain damage after experimental stroke. Circulation. 2014;130:2040–2051.

13

clinical investigation

38. Miao Y, Yu-Jie Z, Zhi-Jian W, et al. Chronic kidney disease and the risk of stent thrombosis after percutaneous coronary intervention with drugeluting stents. Catheter Cardiovasc Interv. 2012;80:361–367. 39. Watanabe I, Tatebe J, Namba S, et al. Activation of aryl hydrocarbon receptor mediates indoxyl sulfate-induced monocyte chemoattractant protein-1 expression in human umbilical vein endothelial cells. Circ J. 2013;77:224–230. 40. Ito S, Osaka M, Edamatsu T, et al. Crucial role of the aryl hydrocarbon receptor (AhR) in indoxyl sulfate-induced vascular inflammation. J Atheroscler Thromb. 2016;23:960–975. 41. Loughman JA, Yarbrough ML, Tiemann KM, et al. Local generation of kynurenines mediates inhibition of neutrophil chemotaxis by uropathogenic Escherichia coli. Infect Immun. 2016;84:1176–1183. 42. Humblet O, Birnbaum L, Rimm E, et al. Dioxins and cardiovascular disease mortality. Environ Health Perspect. 2008;116:1443–1448. 43. Metghalchi S, Ponnuswamy P, Simon T, et al. Indoleamine 2, 3-dioxygenase fine-tunes immune homeostasis in atherosclerosis and

14

L Dou et al.: Aryl hydrocarbon receptor and CKD

44.

45.

46. 47.

48.

colitis through repression of interleukin-10 production. Cell Metab. 2015;22:460–471. Pernomian L, da Silva CHTP. Current basis for discovery and development of aryl hydrocarbon receptor antagonists for experimental and therapeutic use in atherosclerosis. Eur J Pharmacol. 2015;764: 118–123. Calaf R, Cerini C, Génovésio C, et al. Determination of uremic solutes in biological fluids of chronic kidney disease patients by HPLC assay. J Chromatogr B Analyt Technol Biomed Life Sci. 2011;879:2281–2286. I. NKF-K/DOQI Clinical Practice Guidelines for Hemodialysis Adequacy: update 2000. Am J Kidney Dis. 2001;37:S7–S64. Krieter DH, Hackl A, Rodriguez A, et al. Protein-bound uraemic toxin removal in haemodialysis and post-dilution haemodiafiltration. Nephrol Dial Transplant. 2010;25:212–218. Schmidt JV, Su GH, Reddy JK, et al. Characterization of a murine Ahr null allele: involvement of the Ah receptor in hepatic growth and development. Proc Natl Acad Sci. 1996;93:6731–6736.

Kidney International (2018) -, -–-