Calcineurin inhibitor effects on kidney electrolyte handling and blood pressure: tacrolimus versus voclosporin

Abstract

Background: Calcineurin inhibitors (CNIs) affect kidney electrolyte handling and blood pressure (BP) through an effect on the distal tubule. The second-generation CNI voclosporin causes hypomagnesaemia and hypercalciuria less often than tacrolimus. This suggests different effects on the distal tubule, but this has not yet been investigated experimentally.

Methods: Rats were treated with voclosporin, tacrolimus or vehicle for 28 days. Dosing was based on a pilot experiment to achieve clinically therapeutic concentrations. Drug effects were assessed by electrolyte handling at day 18 and 28, thiazide testing at day 20, telemetric BP recordings and analysis of messenger RNA (mRNA) and protein levels of distal tubular transporters at day 28.

Results: Compared with vehicle, tacrolimus but not voclosporin significantly increased the fractional excretions of calcium (>4-fold), magnesium and chloride (both 1.5-fold) and caused hypomagnesaemia. Tacrolimus but not voclosporin significantly reduced distal tubular transporters at the mRNA and/or protein level, including the sodium-chloride cotransporter, transient receptor melastatin 6, transient receptor potential vanilloid 5, cyclin M2, sodium-calcium exchanger and calbindin-D28K. Tacrolimus but not voclosporin reduced the mRNA level and urinary excretion of epidermal growth factor. The saluretic response to hydrochlorothiazide at day 20 was similar in the voclosporin and vehicle groups, whereas it was lower in the tacrolimus group. The phosphorylated form of the sodium-chloride cotransporter was significantly higher at day 28 in rats treated with voclosporin than in those treated with tacrolimus. Tacrolimus transiently increased BP, whereas voclosporin caused a gradual but persistent increase in BP that was further characterized by high renin, normal aldosterone and low endothelin-1.

Conclusions: In contrast to tacrolimus, voclosporin does not cause hypercalciuria and hypomagnesaemia, but similarly causes hypertension. Our data reveal differences between the distal tubular effects of tacrolimus and voclosporin and provide a pathophysiological basis for the clinically observed differences between the two CNIs.

Keywords: aldosterone; calcineurin inhibitors; calcium; hypertension; hypomagnesaemia; mineral metabolism.

Graphical Abstract

Figure 1:

Overview of the study design. Following a 10-day recovery period, we collected 24-hour urine in metabolic cages, blood samples from the tail vein and 2-day radiotelemetry measurements to assess baseline physiological parameters (Supplementary Table S1). The treatment with vehicle, tacrolimus and voclosporin started on day 0 and continued until day 28. On day 18, blood and 24-hour urine were collected to measure trough levels of the drugs and to analyse fractional excretions of electrolytes. A thiazide test was performed on days 19–20. On day 28, blood samples were collected at 8 hours and 24 hours after the last dose and the rats were sacrificed. i.p.: intraperitoneal.

Figure 2:

The single-dose pharmacokinetics of tacrolimus and voclosporin in rats. Mean whole blood concentration–time profiles of (A) tacrolimus (n = 3 rats; 0.5 mg/kg by intraperitoneal injection) and (B) voclosporin (n = 3 rats; 0.4 mg/kg and 1.0 mg/kg by intraperitoneal injection) following a single-dose administration in rats. Data shown are mean ± SEM. C_max: maximum concentration.

Figure 3:

Effects of tacrolimus and voclosporin on fractional excretions of calcium, magnesium, sodium and chloride and urinary EGF excretion after 18 days of treatment. (A) Fractional Ca²⁺ excretion. (B) Fractional Mg²⁺ excretion. (C) Fractional Na⁺ excretion. (D) Fractional Cl⁻ excretion. (E) Urinary EGF excretion. Calculated values are normalized to the mean values of the vehicle group defined as 1.0. Shown are data points with mean ± SD (n = 8–9 rats/group). Fractional excretions of individual electrolytes were calculated by the equation: 100 × (U_electrolyte × P_creatinine)/(P_electrolyte × U_creatinine), where U_electrolyte is urinary excretion of electrolyte (mmol/l), P_creatinine is plasma creatinine (μmol/l), P_electrolyte is plasma electrolyte concentration (mmol/L) and U_creatinine is urinary excretion of creatinine (mmol/l). For the fractional magnesium excretion, plasma Mg²⁺ is multiplied by 0.7, because only 70% of the circulating Mg²⁺ is filterable. *P < 0.05; **P < 0.01; ***P < 0.001; N.S.: not significant; one-way ANOVA with Dunnett's T3 post hoc testing. Ca²⁺: calcium; Cl⁻: chloride; Mg²⁺: magnesium; Na⁺: sodium.

Figure 4:

Effects of tacrolimus and voclosporin on mRNA levels of transport proteins in the distal convoluted tubule and connecting tubule after 28 days of treatment. Results of quantitative reverse transcription PCR of total RNA isolated from the kidney cortex. Gene expression was normalized to Actb and to mean values of the vehicle group defined as 1.0. Shown are data points with mean ± SD (n = 8 rats/group; *P < 0.05; **P < 0.01; ***P < 0.001; N.S.: not significant; one-way ANOVA with Dunnett's T3 post hoc testing, except for Cnnm2 and Trpv5 by Kruskal–Wallis test. Calb1: calbindin 1; Cnnm2: cyclin M2; Egf: epidermal growth factor; Slc8a1: solute carrier family 8 member A1; Slc12a3: solute carrier family 12 member A3; Trpm6: transient receptor potential melastatin 6; Trpv5: transient receptor potential vanilloid 5.

Figure 5:

Effects of tacrolimus and voclosporin on distal tubular transporters and thiazide sensitivity. (A) Immunoblotting of whole kidney homogenates from vehicle-, tacrolimus- and voclosporin-treated rats. (B) Densitometry of immunoblots. Band intensities are normalized to GAPDH and to the mean intensity of the vehicle group defined as 1.0. Values displayed are mean ± SD (n = 8 rats/group, one-way ANOVA with Dunnett's T3 post hoc test). (C) The 6-hour response of a single dose of hydrochlorothiazide (25 mg/kg i.p.) on Δ urinary Na⁺ and Cl⁻ excretion. Values displayed are mean ± SD (n = 8–9 rats/group; one-way ANOVA with Dunnett's T3 post hoc test; Δ: thiazide − vehicle; data obtained on day 19 for vehicle and on day 20 for thiazide). *P < 0.05; **P < 0.01; ***P < 0.001; N.S.: not significant. pNCC-T53: phosphorylated NCC at threonine 53; tNCC: total abundance of the sodium–chloride cotransporter; NCX1: sodium–calcium exchanger 1; TRPV5: transient receptor potential vanilloid 5; GAPDH: glyceraldehyde 3-phosphate dehydrogenase; Na⁺: sodium; Cl⁻: chloride.

Figure 6:

Effects of tacrolimus and voclosporin on MAP and heart rate. Time course of (A) MAP, (B) changes in MAP and (C) heart rate. Data are presented as mean ± SEM (n = 6–8 rats/group; two-way ANOVA with mixed effects model analysis). ***P < 0.001 versus vehicle; ^‡P < 0.001 versus indicated group.

Figure 7:

Effects of tacrolimus and voclosporin on renin, aldosterone, COX-2 and endothelin-1 after 28 days of treatment. (A) Plasma renin activity. (B) Renin mRNA. (C) Plasma aldosterone. (D) COX-2 mRNA. Gene expression was normalized to Actb and to mean values of the vehicle group defined as 1.0. Shown are data points with mean ± SD (n = 8–9 rats/group, one-way ANOVA with Dunnett's T3 post hoc testing, except for plasma aldosterone and COX-2 mRNA by Kruskal–Wallis test). *P < 0.05; **P < 0.01; ***P < 0.001; N.S.: not significant.