NaCl cotransporter abundance in urinary vesicles is increased by calcineurin inhibitors and predicts thiazide sensitivity

Abstract

Animal studies have shown that the calcineurin inhibitors (CNIs) cyclosporine and tacrolimus can activate the thiazide-sensitive NaCl cotransporter (NCC). A common side effect of CNIs is hypertension. Renal salt transporters such as NCC are excreted in urinary extracellular vesicles (uEVs) after internalization into multivesicular bodies. Human studies indicate that CNIs also increase NCC abundance in uEVs, but results are conflicting and no relationship with NCC function has been shown. Therefore, we investigated the effects of CsA and Tac on the abundance of both total NCC (tNCC) and phosphorylated NCC at Thr60 phosphorylation site (pNCC) in uEVs, and assessed whether NCC abundance in uEVs predicts the blood pressure response to thiazide diuretics. Our results show that in kidney transplant recipients treated with cyclosporine (n = 9) or tacrolimus (n = 23), the abundance of both tNCC and pNCC in uEVs is 4-5 fold higher than in CNI-free kidney transplant recipients (n = 13) or healthy volunteers (n = 6). In hypertensive kidney transplant recipients, higher abundances of tNCC and pNCC prior to treatment with thiazides predicted the blood pressure response to thiazides. During thiazide treatment, the abundance of pNCC in uEVs increased in responders (n = 10), but markedly decreased in non-responders (n = 8). Thus, our results show that CNIs increase the abundance of both tNCC and pNCC in uEVs, and these increases correlate with the blood pressure response to thiazides. This implies that assessment of NCC in uEVs could represent an alternate method to guide anti-hypertensive therapy in kidney transplant recipients.

Fig 1. Representative immunoblots of tNCC and pNCC abundance in uEVs of kidney transplant recipients treated with CsA, Tac or CNI-free immunosuppressive regimens and healthy volunteers.

Panels A and B show the immunoreactive bands in uEVs of patients treated with CsA (n = 4), Tac (n = 5), CNI-free immunosuppressive regimens (n = 3), and healthy volunteers (n = 5). tNCC (A) and pNCC (B) immunoreactive bands in uEVs of both CsA- and Tac-treated kidney transplant recipients were more abundant compared to kidney transplant recipients treated with CNI-free immunosuppressive regimens and healthy volunteers.

Fig 2. Densitometry of tNCC and pNCC immunoreactive bands in uEVs of all kidney transplant recipients treated with CsA (n = 9), Tac (n = 23) or CNI-free immunosuppressive regimens (n = 13) and healthy volunteers (n = 6).

Both tNCC (A) and pNCC (C) abundance in both CsA- and Tac-treated kidney transplant recipients was significantly higher in comparison to kidney transplant recipients treated with CNI-free immunosuppressive regimens and healthy volunteers. Densitometry analysis of CD9 expression of the immunoblots for tNCC (B) and pNCC (D) showed no significant differences between the four groups. The ratio of pNCC to tNCC abundance in uEVs of CsA- and Tac-treated group was not significantly higher in comparison to kidney transplant recipients treated with CNI-free immunosuppressive regimens and healthy volunteers (E). The original immunoblots are shown in Fig 1 and S3 and S4 Figs. Values are mean ± SEM normalized to kidney transplant recipients treated with CNI-free immunosuppressive regimens (one-way ANOVA, *P<0.05, n = 51).

Fig 3. Pre-treatment tNCC and pNCC abundances in uEVs isolated from hypertensive kidney transplant recipients who did or did not respond to chlorthalidone.

Shown are immunoblots of tNCC (panel A) and pNCC (panel B) together with CD9 in uEVs of hypertensive kidney transplant recipients using Tac. ‘Responders’ (n = 10) refer to patients who subsequently had a significant anti-hypertensive response (≥10 mmHg reduction in systolic blood pressure) to 8-week treatment with chlorthalidone. Non-responders (n = 8) did not have an anti-hypertensive response to chlorthalidone (no change or increase in systolic blood pressure). uEVs were isolated before the treatment with chlorthalidone. Both tNCC and pNCC abundance were significantly higher in responders compared to non-responders (non-parametric t-test, *P<0.05, n = 18). Both pNCC and tNCC abundance in uEVs correlated with the blood pressure response (panel C, R² = 0.27 and panel D, R² = 0.30 using log-transformed densitometry data because of non-normal distribution, P<0.05 for both). Abbreviations: SBP, ambulatory systolic blood pressure.

Fig 4. pNCC and tNCC abundances in uEVs before and after treatment with chlorthalidone.

Panel A shows pNCC and tNCC abundances in uEVs before (B) and after (A) the 8-week treatment period with chlorthalidone in both responders (n = 10) and non-responders (n = 8). The fold-changes in the before-after abundances of pNCC and tNCC in uEVs (as measured by densitometry) of both responders and non-responders are shown in panel B (Fold-change of 1 means no change, *P<0.05). The scatter plots represent the fold change in tNCC, pNCC or their ratio after treatment with chlorthalidone (densitometry values before treatment with chlorthalidone were set to 1). S5 and S6 Figs, show the original immunoblots from which the individual panels in Fig 4A were derived.

Fig 5. tNCC and pNCC abundance in mouse cortical tubule suspension exposed to CsA.

Panel A shows representative immunoblots of tNCC and pNCC abundance in mouse cortical tubule suspension exposed to CsA. Immunoblots of protein homogenates of mouse cortical tubule suspensions were incubated in basic (B) buffer for 30 and 90 minutes, in the absence (-) or presence of CsA at final concentrations of 5, 10, or 20 μmol/L (A). tNCC remained at baseline levels (B), while pNCC (C) and ratio of pNCC to tNCC (D) were significantly increased in the mouse cortical tubule suspension after 30 minutes of exposure to 10 and 20 μmol/L of CsA and 90 minutes of exposure to 5, 10, and 20 μmol/L of CsA. Similarly, pNCC and the ratio of pNCC to tNCC were significantly increased in hypotonic low chloride (H) buffer (C and D). The original immunoblots for 30 and 90 minutes, in the absence (-) or presence of CsA at final concentrations of 5, 10, or 20 μmol/L are shown in Fig 5 and S10 Fig. Values are mean ± SEM normalized to basic (B) condition (one-way ANOVA, *P<0.05, 30 minutes exposure n = 3, 90 minutes exposure n = 4).