AST-120 to Target Protein-Bound Uremic Toxins Improves Cardiac Output and Kidney Oxygenation in Experimental Chronic Kidney Disease

Chronic kidney disease (CKD) is a worldwide health problem affecting approximately 700 million people, with rapidly increasing incidence and prevalence [1, 3]. The leading causes of CKD in the adult population are diabetes and hypertension [4, 5]. Development of kidney tissue hypoxia has been proposed as a unifying pathway to CKD independent of the initial disease [6, 8]. The mechanisms causing intrarenal hypoxia are not fully understood, but oxidative stress, reduced bioavailability of nitric oxide, and mitochondria uncoupling have been shown to contribute [8]. The most common treatment strategy to reduce the progression of CKD in patients is to target angiotensin II signaling, using either angiotensin-converting enzyme inhibitors (ACEI) or antagonists directed toward angiotensin II AT₁-receptors [1]. However, these available strategies do not constitute a final cure for the disease, and additional interventions are needed.

CKD is associated with cardiovascular dysfunction, where malfunctioning of either organ system negatively affects the other, creating a negative cycle where both heart and kidney function deteriorate. During the last 2 decades, this has been referred to as the cardiorenal syndrome. Early on, cardiorenal syndrome was defined as heart failure, causing the kidneys to fail due to low cardiac output. However, today it is widely accepted that the primary insult can be to either the heart or kidneys and that the resulting cycle of negative feedback mechanisms causes further damage to both organs [9].

Loss of kidney function in CKD is also accompanied by accumulation of uremic retention solutes, some of which have deleterious effects per se, i.e., uremic toxins. Uremic toxins are divided into three major groups: (1) small size molecules (<500 D) free water soluble with no known protein binding; (2) protein bound; and (3) middle size molecules (>500 D) [10]. Protein-bound uremic toxins pose an additional problem in CKD patients since protein-bound uremic toxins are poorly cleared by hemodialysis [11].

Indoxyl sulfate is a protein-bound uremic toxin known to accumulate in patients with CKD, and plasma levels correlate with the rate of decline in function [12, 14]. In type 2 diabetic patients, coronary artery disease and reduced glomerular filtration rate (GFR) correlated with increased plasma levels of indoxyl sulfate [15]. Studies by Niwa and colleagues [16] demonstrated a causal relationship between indoxyl sulfate, glomerular sclerosis, and kidney dysfunction in rats. Further, in proximal tubular cells, indoxyl sulfate increases oxidative stress and oxygen utilization, resulting in intrarenal tissue hypoxia [17].

AST-120 is an oral absorbent that binds small molecular weight precursors of common uremic toxins in the intestines, thereby reducing the systemic load and impact of these toxic compounds. AST-120 is an approved treatment strategy to slow progression of disease in CKD in several Asian countries. Although AST-120 is not exclusively specific, it primarily reduces the protein-bound uremic toxin indoxyl sulfate [18]. The plasma levels of indoxyl sulfate depend on gut bacteria production of indole, which is reabsorbed by the intestines and converted to water-soluble indoxyl sulfate by hepatic metabolism. Indoxyl sulfate is cleared from the circulation via passive filtration across the glomerular filtration barrier but mainly via active secretion by the proximal tubule [19, 21]. AST-120 binds tightly to indole in the intestines and thus prevents systemic uptake, thereby reducing plasma levels of indoxyl sulfate [21]. In rats with decreased kidney function, AST-120 decreased pimonidazole staining in kidney cortex, indicating improved oxygenation in kidney tissue [17].

In a rat model of CKD, we investigated the effects on heart and kidney function of chronic AST-120 treatment to reduce plasma levels of protein-bound uremic toxins. This was compared to the effects of chronic ACEI (enalapril), the gold standard treatment for patients with proteinuric CKD. As kidney function declines, the excretory capacity to compensate for excessive salt intake may constitute an increasing challenge. Furthermore, excessive salt intake is a common phenomenon in most developed countries, and we therefore investigated the effects of each treatment in response to a chronic high salt challenge. The hypothesis of the study was that AST-120 administration to reduce protein-bound uremic toxins would improve cardiac output and kidney oxygenation.

Male Sprague-Dawley rats (Taconic, Lille Skensved, Denmark) weighing approximately 300 g at the start of experiments were used. Animal health and well-being were assessed daily in accordance with national guidelines for the care and use of animals in research and were housed under controlled conditions with a 12-h light/dark cycle with free access to food and water. All chemicals were from Sigma-Aldrich (St. Louis, MO, USA) unless otherwise stated.

Surgery was performed in a two-step antiseptic intervention under isoflurane anesthesia (∼2.5% in 100% oxygen, catalog number [Cat No.] 506949). Analgesia (carprofen 1.75 mg/rat, Orion Pharma AB, Sollentuna, Sweden, Cat No. 014920) was given subcutaneously perioperatively and once again the day following surgical procedures. Body temperature was maintained at 37°C using a servo-controlled heating pad and the abdomen opened through a midline incision. In the first step, the vessels and ureter of the right kidney were ligated and the whole kidney was removed. The abdomen was thereafter closed in two layers of continuous stitches using polyethylene thread (SweVet-Piab AB, Sjobo, Sweden, Cat No. S180430). Four days after the removal of the right kidney, the midline incision was reopened and the left kidney was dissected free from surrounding fat and connective tissue. Blood flow was temporarily occluded proximally to the kidney with a ligature around the abdominal aorta. The upper and lower poles of the kidney were removed surgically, leaving 1/3 of the original kidney tissue intact. Bleeding was controlled using absorbable hemostatic gelatin sponge (Spongostan; Ethicon, Cat No. MS0002), and blood flow to the kidney was restored thereafter. The abdomen was closed as described above.

The technique to surgically remove the poles of the left kidney by cutting was used to create a model of salt-sensitive hypertension, which is normotensive at baseline. This is in contrast to the technique in which the artery branches to the poles are ligated to reduce kidney mass. Such procedure results in ischemic injury, renin release, and baseline hypertension [22]. To verify the integrity of our model, we measured continuous blood pressure using implantable wireless telemetry probes (Telemetry Research, Auckland, NZ). These probes were implanted into the abdominal aorta during the procedure of removing the right kidney, as described above. The pressure catheter was inserted into the abdominal aorta below the bifurcation of the renal arteries. The transmitter was attached to the peritoneum with a suture in order to prevent any dislocation of the transmitter that would interfere with the signal transmission. Mean arterial pressure (MAP) was recorded immediately before and after the week of high-salt diet. Data were collected for 5 min every 30 min during the dark cycle. Data were recorded and analyzed using a digital data acquisition system (PowerLab, AD Instruments, Hastings, UK).

Five days postsurgery, rats were allocated to one of 3 treatment groups to receive either no treatment (control; n = 12); the oral absorbent AST-120 (8.7% mixed with powdered chow; Kureha Corp., Tokyo, Japan, lot number 10YM16; n = 12) to bind gut-derived precursors of protein-bound uremic toxins, such as indoxyl sulfate; or the ACEI enalapril (10 mg/kg bw/day in drinking water, Cat No. E6888; n = 12) to reduce angiotensin signaling. Rats were treated for a total of 6 weeks. All animals were allowed free access to water and standard rat chow for the first 5 weeks and the same rat chow containing 4% NaCl during the sixth week. The response to high salt intake was evaluated at the end of the study period by measuring kidney function and cardiac output (see experimental design in Fig. 1).

Prior to the start of high-salt diet, all rats were placed in metabolic cages for 24 h to measure food and water intake. Furthermore, GFR was measured using plasma clearance of fluorescein isothiocyanate (FITC)-labeled inulin (Cat No. F3272) as previously described [23]. FITC-inulin was dissolved in sterile phosphorus-buffered saline (500 mm, pH 7.4). 1.5% FITC-inulin was dialyzed light protected at 4°C overnight using a 1,000 Da cut-off dialysis membrane (Spectra/Por^® 6 Membrane, Spectrum Laboratories Inc., Rancho Dominguez, CA, USA, Cat No. 514-0073) and filtered through a 0.45-μm syringe filter before use to remove any free or crystalline FITC, respectively. The conscious rat was placed in a restrainer and 1 mL of FITC-inulin was rapidly injected into the tail vein. The syringe was weighed before and after the injection to calculate the exact dose. Blood samples were obtained from the tip of the tail at 1, 3, 7, 10, 15, 35, 55, 75, 95, 125, and 155 min after the injection and centrifuged for 3 min at 13,800 g. Plasma samples were diluted in HEPES buffer (pH 7.4) and fluorescence was measured in a black 384-well microplate (Greiner Bio-One GmbH, Kremsmuenster, Austria) using a plate reader (Safire II, Tecan Austria GmbH, Grödig, Austria) with 496 nm excitation and 520 nm emission. Plasma clearance of FITC-inulin (Cl_pl) was calculated using a non-compartmental pharmacokinetic data analysis as described by Sällström and Fridén [24].

At the end of the study period, following 1 week of high-salt diet (6 weeks of total treatment), animals were anesthetized with Inactin (sodium thiobarbitural i.p, 100 mg/kg bw, Cat No. T133) and placed on a servo-rectally controlled heating pad to maintain body temperature at 37°C. An endotracheal tube was inserted in order to facilitate spontaneous respiration, and a catheter was inserted into the left femoral vein for infusion of FITC-inulin (0.3% in saline; 7 mL/kg bw/h) in order to determine GFR. Cardiac output was measured by thermal dilution. A polyethylene catheter containing a T-type implantable probe (AD Instruments, Oxford, UK) was inserted via the left carotid artery to the level of the aortic arch. A thermodilution curve was generated by infusion of room temperature saline, and the area under the curve was calculated as an estimate of cardiac output. An average of the four highest measurements was used for analysis. After cardiac output measurements, the catheter for thermal dilution was replaced by a polyethylene catheter for continuous measurement of blood pressure and blood sampling. MAP and heart rate (HR) were recorded and later used for calculations of total peripheral resistance (TPR) and stroke volume (SV).

The remnant kidney was exposed by a flank incision on the left side and immobilized in a plastic cup. A catheter was placed in the ureter for urine collection and determination of urine flow rate. Surgery was followed by a 40-min recovery period and a 60-min experimental period with measurements of kidney parameters. RBF using an ultrasound probe (Transonic, Ichita, USA) placed on the left renal artery and kidney cortex oxygen tension (PO₂) at 1 mm depth was measured using a Clark-type microelectrode (OXY10; Unisense A/S, Aarhus, Denmark).

At the end of the experimental period, blood gas parameters in samples from the left renal vein and carotid artery were analyzed (iSTAT, Abbott, Princeton, NJ, USA, Cat No. 3P7725) in order to determine total kidney oxygen consumption (QO₂). The left kidney weight and urine volumes were determined. Fluorescence of FITC-inulin in plasma and urine were analyzed as described above to calculate GFR. Tissue samples were placed in liquid nitrogen and stored at −80°C for later analysis. Urinary concentrations of Na⁺ and K⁺ were determined by flame photometry (IL943, Instrumentation Laboratory, Milan, Italy) and urinary protein excretion by DC Protein Assay (BioRad Laboratories, CA, USA) and multiplied by urine flow to determine urinary excretion rates.

Thiobarbituric acid reactive substances (TBARS) in urine were determined fluorometrically, as previously described [25]. In brief, 50 µL of sample was mixed with 62.5 µL of 42 mmol/L thiobarbituric acid and heated to 97°C for 60 min. Malondialdehyde was used to prepare standard samples. After cooling on ice, samples were precipitated with a mixture of methanol and 1 mol/L NaOH (91:9) and centrifuged at 840 g for 5 min. The top layer was transferred to a 384-well plate; fluorescence intensity was analyzed (excitation, 532 nm; emission, 553 nm) and corrected for protein concentration. Protein carbonyl content in kidney cortex was determined spectrophotometrically by protein Carbonyl Colorimetric Assay (Cayman Chemicals, MI, USA, Cat No. 10005020-96-CAY) and normalized for protein concentration.

TPR was calculated as MAP/cardiac output and SV was calculated as cardiac output/HR. GFR was calculated using inulin clearance: GFR = U x V/P, where U and P denote the FITC fluorescence in urine and plasma, respectively, and V denotes the urine flow (mL/min). QO₂ (μmol/min/kidney) was calculated from the arteriovenous difference in oxygen content (O_2ct) with a standard equation (O_2ct = [Hb] × O₂ saturation × 1.34 + PO₂ × 0.003) multiplied by RBF. Transported Na⁺ (T_Na) was calculated as GFR × [Na⁺]_plasma − urinary Na⁺ excretion. Electrolyte transport efficiency was calculated as T_Na/QO₂.

Levels of indoxyl sulfate in serum were measured by a mobile phase, 5% tetrahydrofuran/0.1 M KH₂PO₄ (pH 6.5) at a flow rate of 1 mL/min, and fluorescence detection (excitation 295 nm, emission 390 nm) utilizing HPLC (Shimadzu, Kyoto, Japan) as described previously [26].

All values are presented as mean ± SD. Data were tested for normality using Shapiro-Wilk test (see online suppl. material at www.karger.com/doi/10.1159/000529272). For normally distributed data, statistical comparisons between groups were performed using one-way analysis of variance and followed by Tukey’s post hoc test. Comparisons of data before and after high salt challenge within each group were performed using Student’s t test (paired, two tailed). For data not following normal distribution, nonparametric testing was performed using Kruskal-Wallis test followed by Dunn’s multiple comparisons test. All significant comparisons (p < 0.05) are indicated in figures and tables.

During normal salt intake, all animals were normotensive (101 ± 2 mm Hg; N = 11) and had comparable blood pressure, independently of treatment, verifying the integrity of our model. In response to high-salt diet, untreated controls increased MAP by 19 ± 5 mm Hg (N = 8), whereas rats treated with AST-120 developed a significantly smaller increase in MAP of 10 ± 1 mm Hg (N = 3) compared to untreated nephrectomized controls (two-tailed t test, p = 0.049).

Indoxyl sulfate is the primary uremic toxin affected by AST-120 and was used in this study as indicator of AST-120 treatment effectiveness. To confirm the treatment effect of the AST-120, indoxyl sulfate was measured in plasma of all groups, collected at the end of the study period. AST-120 treatment reduced plasma levels of indoxyl sulfate (1.6 ± 0.4 µg/mL) compared to untreated controls (3.0 ± 0.3 µg/mL; p < 0.05) and animals treated with enalapril (3.4 ± 0.4 µg/mL; p < 0.05).

Food and water intake were measured in metabolic cages before the start of salt diet. There was no difference in food and water intake between groups. Body and kidney weight at the end of the study period were similar in all groups (Table 1).

To evaluate the cardiovascular effects of reduced kidney function, cardiac output was measured by thermal dilution at the end of the study period. Cardiac output was improved by AST-120 (Fig. 2a) and TPR was decreased by both AST-120 and enalapril treatment compared to controls (Fig. 2b). Also, AST-120 increased SV. There was no difference in MAP or HR between the groups under Inactin anesthesia (Table 2).

At the end of the study period, i.e., six weeks after 5/6 nephrectomy, the total GFR in this rat CKD model was 1.1 ± 0.1 mL/min (control group). In comparison to previous published results from healthy Sprague-Dawley rats, this is approximately 45% of a normal GFR [23, 27, 29], which equals CKD stage 3b (i.e., moderate to severe reduction in GFR). GFR was lower during normal salt intake in animals treated with enalapril compared to other groups. However, in response to the high salt intake, this group significantly increased GFR (Fig. 3) compared to baseline and also compared to untreated controls.

Increased oxidative stress has been demonstrated to be an important mechanism that contributes to development of kidney hypoxia in diseases such as diabetes and hypertension. We measured TBARS in urine and protein carbonylation in kidney cortex at the end of the study period to evaluate if treatments were effective in reducing oxidative stress. Urinary excretion of TBARS was reduced in animals treated with AST-120 compared to enalapril treatment (Fig. 4a). Protein carbonylation was reduced in kidney tissue in animals treated with enalapril compared to other groups (Fig. 4b).

To evaluate the effect on kidney oxygen availability, kidney cortex PO₂ was measured using a Clarke-type electrode at the end of the study period. Treatment with both enalapril and AST-120 improved PO₂ compared to untreated controls after the high salt intake (Fig. 5). Proteinuria is a marker of kidney injury in both patients and in experimental models. Both AST-120 and enalapril had a reduced urinary protein excretion compared to untreated controls at the end of the study period (Table 3). The ability to excrete excess sodium to control blood pressure is an important function of the kidney. To evaluate the treatment effects on sodium handling, fractional urinary Na⁺ excretion was measured and increased in the group treated with enalapril compared to AST-120 (Table 3). Other important kidney function parameters such as QO₂, T_Na/QO₂, and RBF were similar in all groups (Table 3).

One animal in the control group developed hematuria due to development of a kidney stone in the ureter and was euthanized before terminal experiments. From the data presented in figures and tables, eight single values were considered outliers (mean ± 2 SD) and removed (control: 3, AST-120: 1, enalapril: 4 ). Remaining missing values were due to technical difficulties preventing specific measurements to be obtained, e.g., intra-abdominal adhesions and fibrosis development or insufficient sample volume.

In this study, we investigated the effects of AST-120 treatment to reduce plasma levels of protein-bound uremic toxins on cardiovascular function, kidney function, and oxygen handling and compared them to the effects of ACEI, the gold standard treatment for patients with CKD. We also included a high salt challenge in each group since loss of kidney function is accompanied by reduced capacity to excrete excessive salt and high salt intake is an increasing health problem worldwide.

The main finding of this study was that administration of AST-120 to reduce protein-bound uremic toxins, such as indoxyl sulfate, improved cardiac output, implying cardio-protective properties of AST-120. Also, decreased TPR together with the blunted blood pressure response to high salt intake measured by telemetry further supports the conclusion that targeting uremic toxins in CKD has beneficial effects on systemic circulation. Further, AST-120 improved oxygen availability in kidney tissue and reduced urinary excretion of TBARS, a marker of oxidative stress. This demonstrates positive effects on both cardiovascular and kidney function and further supports the stipulated beneficial effects of reducing plasma levels of uremic toxins in CKD. The protective cardiovascular effects of reducing protein-bound uremic toxins have also been demonstrated by others. In a canine model of heart failure, AST-120 improved left ventricle function and decreased myocardial apoptosis and fibrosis [30]. In rat models of CKD, AST-120 decreased cardiac oxidative stress and fibrosis [31] and ameliorated endothelial dysfunction in isolated aortic rings by improving relaxation in response to acetylcholine [32]. In patients with nondiabetic CKD, AST-120 treatment decreased arterial stiffness, a risk factor of cardiovascular morbidity and mortality. Interestingly, over the 2-year study period, the curve of creatinine increase was less steep in patients compared to those who did not receive AST-120 treatment [33], further supporting the dual action of AST-120 on both heart and kidney function.

Cardiac output is calculated from HR multiplied by SV. Since HR was similar between groups, treatment with AST-120 improves cardiac output by increasing SV. Either this is via a direct inotropic effect on the cardiac muscle or it is an effect of decreased myocardial stiffness. Decreased stiffness would improve diastolic relaxation and increase preload, thereby increasing contractility and SV. Based on available literature that shows AST-120 inhibits cardiac fibrosis, decreases arterial stiffness, and counteracts endothelial dysfunction, decreased myocardial stiffness seems more likely.

Both treatment with AST-120 and angiotensin II signaling had striking effects on cortical oxygen availability. This is important since tissue hypoxia has been demonstrated to both initiate [34] and accelerate the progression of CKD [35], as originally proposed by Fine and colleagues [7]. Improved kidney oxygenation after reduction of indoxyl sulfate by AST-120 is in line with previous results [17], and we now extend the evidence to show these beneficial effects also during high salt intake.

In contrast to AST-120, ACEI reduced kidney tissue oxidative stress and improved the ability to increase GFR in response to high salt intake. Interestingly, the increase in GFR was not associated with increased urinary protein leakage, indicating protected kidney function. A preserved functional renal reserve and decreased intrarenal oxidative stress demonstrate the renoprotective effects of angiotensin II reduction by ACEI. Furthermore, ACEI alone induced a profound increase in fractional urinary Na⁺ excretion during the high salt intake, which may have additional beneficial effects for long-term salt balance and, together with decreased TPR, arterial blood pressure regulation.

Previous studies have demonstrated that accumulating levels of indoxyl sulfate modify the RAS system, influencing both the ACE/angiotensin II/AT₁R and the ACE2/angiotensin (1–7)/Mas receptor pathway. In mice, treatment with indoxyl sulfate increased kidney AT₁R protein expression and decreased the expression of AT₂R [36], promoting vasoconstriction and increased oxidative stress. Stimulation of the AT₂R and binding of the Mas receptor by angiotensin (1–7) counteracts the effects of angiotensin II by stimulating vasodilation and anti-inflammatory and anti-oxidative pathways. Niwa and colleagues demonstrated an indoxyl sulfate-dependent downregulation of the Mas receptor in nephrectomized rats and a restoration of the expression by AST-120 treatment [37]. In our study, a reduction of indoxyl sulfate may have enabled the normal NO-dependent vasodilation induced by angiotensin 1–7 binding to Mas receptors and restored AT₂R expression, which would provide an explanation for the decreased TPR and improved PO₂ also in the group treated with AST-120.

The effectiveness of AST-120 to slow progression of disease in CKD patients is under debate. Some trials report a positive result in patients’ progression of CKD [38]. However, this was not supported by the multinational, phase 3 EPPIC trial [39]. However, a subanalysis of these data, focusing on the US population, indicated a longer time to reach the primary endpoint in patients treated with AST-120 compared to placebo group [40]. Further post hoc analysis of data from clinical trials reveals that in patients with fast-progressing CKD, treatment with AST-120 slowed rate of progression and decreased the proportion of patients that reached primary endpoint (i.e., initiation of dialysis) [41]. Taken together, this indicates beneficial effects of AST-120 in some patient populations, especially those with fast-progressing CKD. Recently published results from the study by Kula et al. [42] demonstrate that young black and Hispanic participants with CKD had a higher incidence rate of disease progression compared to participants of other ethnicities, possibly identifying a subgroup of patients where treatment aimed to target protein-bound uremic toxins could be useful to slow the progression rate.

Even though indole, and subsequently indoxyl sulfate, is the primary uremic toxin targeted by AST-120, other uremic toxins derived from gut bacteria metabolism, such as phenyl sulfate and p-cresyl sulfate, can also be affected [18]. Therefore, it cannot be ruled out that the effects seen in the AST-120-treated animals are partly an effect of other uremic toxins being reduced.

The broad absorptive qualities of AST-120 can also include binding of other substances in the GI tract, e.g., enalapril. Patients treated with AST-120 (Kremezin^®) are instructed to separate intake of Kremezin and other prescribed drugs to avoid this negative interaction. This prevented the possibility of investigating the effects of combining AST-120 and enalapril treatment in this experimental setting, where these interventions were given in food and drinking water, respectively. The lack of a combined AST-120 + enalapril treatment group constitutes a limitation with our study. Further, the lack of a sham-operated control group poses another limitation of this study.

AST-120 treatment successfully improved cardiac output and reduced urinary oxidative stress but fails to reduce intrarenal oxidative stress. This suggests a more systemic effect of AST-120 than compared to the direct renoprotective effects of ACEI. If the reduced proteinuria in AST-120-treated animals is due to a direct effect on kidney function or a secondary effect of improved cardiac function, it is still unclear and warrants further studies.

In conclusion, AST-120 treatment to decrease circulating levels of protein-bound uremic toxins, e.g., indoxyl sulfate, improves cardiac output, reduces urinary oxidative stress and proteinuria in this animal model of CKD. However, reduced angiotensin II signaling was essential in order to decrease intrarenal oxidative stress, protect renal reserve, and increase urinary Na⁺ excretion during high salt intake. Both treatments had striking effects on cortical oxygen availability, which is of critical importance since tissue hypoxia has been demonstrated to both initiate and accelerate the progression of CKD [34, 35]. To target protein-bound uremic toxins and angiotensin II signaling simultaneously could be an efficient strategy to target both heart and kidney dysfunction in CKD in order to further slow progression of disease.

This study protocol was reviewed and approved by the Swedish Ethical Review Authority (approval number C321/11) and carried out in accordance with the NIH Guide for the Care and Use of Laboratory Animals.

Kureha Corporation, who manufacture and market AST-120 for the treatment of uremia, kindly provided AST-120 for this study. However, Kureha Corporation had no influence on experimental design, data analysis, or interpretation of the results.

This study was supported by the Swedish Medical Research Council, the Swedish Society for Medical Research, the Swedish Heart-Lung Foundation, and the Swedish Diabetes Foundation and in part supported by the Grant-in-Aid for Scientific Research (B) of Japan (15H04835).

Ebba Sivertsson: acquisition of data, analysis and interpretation of data, and drafting and revising manuscript. Sara Ceder: acquisition and interpretation of data and revising manuscript. Masaomi Nangaku: acquisition of data, revising manuscript, and funding. Peter Hansell: conception and design, revising manuscript, and supervision. Lina Nordquist: conception and design, revising manuscript, and funding. Fredrik Palm: conception and design, interpretation of data, revising manuscript, supervision, and funding. All authors read and approved the final manuscript.

The data that support the findings of this study are openly available in Figshare at https://figshare.com/articles/dataset/CKD_uremic_toxins/21104617.