Renal fibrosis is attenuated by targeted disruption of K_Ca3.1 potassium channels

Abstract

Proliferation of interstitial fibroblasts is a hallmark of progressive renal fibrosis commonly resulting in chronic kidney failure. The intermediate-conductance Ca²⁺-activated K⁺ channel (K_Ca3.1) has been proposed to promote mitogenesis in several cell types and contribute to disease states characterized by excessive proliferation. Here, we hypothesized that K_Ca3.1 activity is pivotal for renal fibroblast proliferation and that deficiency or pharmacological blockade of K_Ca3.1 suppresses development of renal fibrosis. We found that mitogenic stimulation up-regulated K_Ca3.1 in murine renal fibroblasts via a MEK-dependent mechanism and that selective blockade of K_Ca3.1 functions potently inhibited fibroblast proliferation by G₀/G₁ arrest. Renal fibrosis induced by unilateral ureteral obstruction (UUO) in mice was paralleled by a robust up-regulation of K_Ca3.1 in affected kidneys. Mice lacking K_Ca3.1 (K_Ca3.1^−/−) showed a significant reduction in fibrotic marker expression, chronic tubulointerstitial damage, collagen deposition and αSMA⁺ cells in kidneys after UUO, whereas functional renal parenchyma was better preserved. Pharmacological treatment with the selective K_Ca3.1 blocker TRAM-34 similarly attenuated progression of UUO-induced renal fibrosis in wild-type mice and rats. In conclusion, our data demonstrate that K_Ca3.1 is involved in renal fibroblast proliferation and fibrogenesis and suggest that K_Ca3.1 may represent a therapeutic target for the treatment of fibrotic kidney disease.

Progressive renal fibrosis is the final common manifestation of various chronic kidney diseases (CKD) resulting in irreversible loss of kidney parenchyma and end-stage renal failure. Up to now, fibrotic kidney disease has remained a major unresolved problem in clinical medicine because of incomplete pathophysiological understanding and the lack of effective therapeutic strategies.

The characteristics of renal fibrosis comprise tubulointerstitial fibroblast proliferation, elevated matrix production and mixed leukocytic cell infiltration resulting in tubular cell apoptosis and necrosis (1, 2). Controversy persists over the origin of activated, matrix-producing fibroblasts and myofibroblasts. For instance, there have been reports that transformed tubular epithelial and endothelial cells as well as bone marrow-derived cells may potentially constitute a part of this cell population in the diseased kidney (3, 4). However, ample evidence suggests that resident interstitial fibroblasts represent a significant, if not major, source of (myo)fibroblast recruitment (5–8). Hence, suppression of fibroblast proliferation in the injured kidney may diminish the number of scar tissue generating cells and could thus yield an option to halt progression of renal fibrosis.

In recent years, great efforts have been made to gain further insight into the mechanisms of fibrogenesis and several molecules with antifibrotic properties, such as bone morphogenic protein 7 (9), hepatocyte growth factor (10), and pirfenidone (11), have been proposed. According to current understanding, resident fibroblast activation and proliferation in the kidney is triggered by locally secreted fibrogenic chemokines including TGFβ₁, PDGF, CTGF, and bFGF. TGFβ₁ potently promotes proliferation of renal fibroblasts via a downstream mechanism that is largely mediated by bFGF (7, 12).

Increasing evidence suggests that ion channels also play a role in cell proliferation by enhancing intracellular Ca²⁺ signaling and affecting cell cycle progression (13, 14). In this regard, Ca²⁺-permeable cation channels and Ca²⁺-activated K⁺ channels (K_Ca) seem to be of paramount importance. Whereas the first directly mediate Ca²⁺ inflow, the latter regulate membrane potential and thus provide the driving force for Ca²⁺ entry (13). In particular, the intermediate-conductance K_Ca (K_Ca3.1) has been shown to play an important role in promoting mitogenesis in several tissues such as the endothelium (15, 16), vascular smooth muscle (17–19), and T lymphocytes (20) as well as in several cell lines including A7r5 neonatal aortic smooth muscle cells (21), 10T1/2-MRF4 myogenic fibroblasts (22, 23), HMEC-1 dermal endothelial cells (15), and some cancer cell lines (24, 25).

Interestingly, mitogens such as bFGF, PDGF, or VEGF have been found to distinctly up-regulate K_Ca3.1 in several cell types (15, 21). Conversely, we and others have recently demonstrated in models of experimental angiogenesis (15), post-interventional arterial restenosis (17, 19), atherosclerosis (26), and endometrial cancer (27), that selective pharmacological inhibition or knockdown of K_Ca3.1 suppresses mitogen-driven cell proliferation and ameliorates disease progression. Thus, K_Ca3.1 could have a pivotal role in disease states characterized by excessive cell proliferation and may emerge as a promising pharmacotherapeutic target for antiproliferative treatment. However, the potential involvement of K_Ca channels in renal fibrogenesis has so far not been investigated. In the present study, we hypothesized that K_Ca3.1 channels also promote renal fibroblast proliferation and the development of tubulointerstitial fibrosis in the kidney. Our data provide strong evidence that targeted disruption of K_Ca3.1 inhibits proliferation of renal fibroblasts and attenuates renal fibrosis.

Results

Mitogenic Stimulation Induces K_Ca3.1 Expression in Renal Fibroblasts.

We first determined the basal gene expression pattern of K_Ca channel subtypes in murine fibroblasts of renal origin (TFB). We found appreciable mRNA expression of K_Ca subtype K_Ca3.1, which was distinctively exceeding transcript levels of the related channel subtypes K_Ca1.1 (large-conductance K_Ca) by ≈1,000× and K_Ca2.3 (small-conductance K_Ca) by ≈100×. However, in electrophysiological patch-clamp studies we registered only very small K_Ca-like currents (<1 pA/pF) indicating weak functional expression of K_Ca in cell membranes of quiescent, nonproliferating renal fibroblasts.

Next, we studied the regulation and functional role of K_Ca in renal fibroblast proliferation. Upon mitogenic stimulation with bFGF (10 ng/mL) we observed a sharp 4.4-fold increase in K_Ca3.1 mRNA expression levels as compared with unstimulated controls (Fig. 1A). In contrast, mRNA expression levels of K_Ca1.1 and K_Ca2.3 were not up-regulated in these fibroblasts. Consistently, Western blot analyses also revealed a substantial up-regulation of K_Ca3.1 protein levels in mitogenically stimulated renal fibroblasts (Fig. 1B). Moreover, patch-clamp analyses showed a time-dependent increase of K_Ca3.1-like K⁺ currents (voltage-independent activation and characteristic inward rectification at membrane potentials >+40 mV) in these cells, reaching a maximum at 24 h of bFGF stimulation (Fig. 1C Left and Center). The identity of K_Ca3.1 was corroborated by means of specific pharmacological channel blockers allowing dissection of individual K_Ca current components. K_Ca currents were found to be insensitive to the K_Ca1.1 blocker iberiotoxin or the K_Ca2.3 blocker apamin. However, TRAM-34, a highly selective K_Ca3.1 inhibitor, virtually abolished K_Ca currents in these fibroblasts (Fig. 1C Right), proving that K_Ca3.1 was solely responsible for the increase in K_Ca currents after mitogenic stimulation. In line, determination of the IC₅₀ of TRAM-34 yielded a concentration of 14 nM (Fig. S1), which is comparable with values reported for the cloned K_Ca3.1 channel (28, 29). Of note, pretreatment of renal fibroblasts with the MEK inhibitor PD98059 or the tyrosine kinase inhibitor genistein, but not the p38 MAP kinase inhibitor SB203580, largely prevented bFGF-induced up-regulation of K_Ca3.1 mRNA levels and currents (Fig. 1 D and E).

Fig. 1.

Mitogenic stimulation induces up-regulation of K_Ca3.1 in murine renal fibroblasts (TFB) involving tyrosine kinase and MEK activity. (A) Quantitative RT-PCR analysis of K_Ca mRNA expression levels after mitogenic stimulation of quiescent fibroblasts with bFGF (10 ng/mL) for 4 h. Charts: Left ordinate: ΔΔC_t values (ΔΔC_t = ΔC_tw/o-ΔC_tx) represent change in expression over control (unstimulated TFB). Right ordinate: Expression relative to control (2^ΔΔC_t × 100%). Data are given as mean ± SEM, n = 9–19 for each data point. **, P < 0.001. (B) Representative Western blot analysis showing rise of K_Ca3.1 protein levels in membrane fractions of TFB upon bFGF (10 ng/mL) or 5% FCS stimulation (24 h) as compared with unstimulated quiescent controls (Ctrl). (C) Time-dependent increase of fibroblast K_Ca3.1 currents in response to bFGF stimulation. (Left) Representative whole-cell recordings of K_Ca3.1 currents before (Ctrl) and after stimulation with bFGF (10 ng/mL). K_Ca3.1 channels were maximally activated by standardized cell dialysis with a 3 μM-free Ca²⁺ containing pipette solution in addition to bath application of DC-EBIO (10 μM). Currents were normalized to cell capacitance (I/C). (Center) Mean normalized K_Ca3.1 currents in unstimulated (Ctrl) and bFGF-stimulated (6, 12, 24 h) TFB. Data represent mean ± SEM, n = 4–7 for each data point. *, P < 0.01; **, P < 0.001. (Right) Whole-cell recordings illustrating the effect of K_Ca3.1 blocker TRAM-34 (100 nM) on mitogen-induced K_Ca currents (Ctrl) in renal fibroblasts. (D) Quantitative RT-PCR analysis of K_Ca3.1 mRNA expression after 4 h bFGF stimulation (10 ng/mL) with or without pretreatment with signal transduction inhibitors. MEK inhibitor PD98059 (20 μM) and tyrosine kinase inhibitor genistein (50 μM), but not p38 MAP kinase inhibitor SB203580 (10 μM), prevented bFGF-induced up-regulation of K_Ca3.1. *, P < 0.01; **, P < 0.001. (E Left) Representative whole-cell recordings of K_Ca3.1 currents in pretreated and 24 h bFGF-stimulated fibroblasts. bFGF-driven induction of K_Ca3.1 functions was largely obviated by PD98059 and genistein but not by SB203580. (Right) Mean K_Ca3.1 currents normalized to cell capacitance. Values are given as mean ± SEM, n = 4–7 for each data point. *, P < 0.005 vs. bFGF without pretreatment.

Taken together, these data demonstrate that mitogenic/profibrotic stimulation by bFGF leads to up-regulation of K_Ca3.1 expression in renal fibroblasts, which is mediated by receptor tyrosine kinase activity and the Ras/Raf/MEK/ERK-signaling cascade.

Selective Pharmacological Disruption of K_Ca3.1 Functions Suppresses Proliferation of Renal Fibroblasts.

We tested whether disruption of K_Ca3.1 functions had any effect on renal fibroblast proliferation. In in vitro proliferation studies we found that pharmacological blockade of K_Ca3.1 with TRAM-34 dose-dependently decreased bFGF-induced incorporation of BrdU, whereas treatment with TRAM-85, a compound closely related to TRAM-34 in structure but without specific K_Ca3.1-inhibiting properties, had no effect (Fig. 2A). Consistently, application of TRAM-34 resulted in a significant 65% growth reduction of renal fibroblast counts stimulated by bFGF (Fig. 2B).

Fig. 2.

K_Ca3.1 inhibition suppresses proliferation of murine renal fibroblasts via G₀/G₁ arrest. (A) Dose-dependent suppression of bFGF-stimulated BrdU incorporation (24 h) in renal fibroblasts by TRAM-34 (●, n = 8–14 for each data point) but not by the structurally related yet inactive compound TRAM-85 (Δ, 1 μM, n = 15). (B) TRAM-34 (1 μM) inhibited bFGF-induced fibroblast proliferation (cell count at 24 h, Ctrl, 0% FCS). (C) Analysis of cell cycle progression in renal fibroblasts. Treatment with TRAM-34 inhibited bFGF-induced (10 ng/mL) cell cycle progression via arrest in phase G₀/G₁. Data points represent mean ± SEM, n = 11–12 for each data point. *, P < 0.01; **, P < 0.001.

We next studied the impact of K_Ca3.1 inhibition on cell cycle progression in renal fibroblasts. To ensure G₀/G₁ phase synchronization, cells were switched to quiescent medium 48 h before mitogen exposure. Stimulation with bFGF resulted in a distinct cell shift from phase G₀/G₁ (35.9 ± 1.2%) to phases S (19.9 ± 0.7%) and G₂/M (45.4 ± 1.1%) as compared with unstimulated cells (G₀/G₁: 49.5 ± 2.0%; S: 14.5 ± 0.9%; G₂/M: 36.9 ± 1.4%) (Fig. 2C). In the presence of TRAM-34, this bFGF-induced shift was largely prevented (G₀/G₁: 45.0 ± 1.5%; S: 17.2 ± 0.4%; G₂/M: 38.9 ± 1.7%), indicating that K_Ca3.1 inhibition caused a cell cycle arrest in phase G₀/G₁.

To rule out potential proapoptotic effects of TRAM-34 which could affect cell counts, we quantified apoptosis by annexin-V labeling and flow cytometric analysis. Incubation with TRAM-34 at concentrations of up to 10 μM produced no increase in annexin-V positive fibroblasts suggesting that selective K_Ca3.1 blockade did not promote apoptosis in these cells (Fig. S2). In line, cell morphology appeared well maintained in the presence of TRAM-34 as visualized by phase contrast microscopy.

K_Ca3.1-Deficiency Attenuates Progression of Kidney Fibrosis After Unilateral Ureteral Obstruction (UUO).

Results from our in vitro experiments suggest that K_Ca3.1 may have a significant function in promoting proliferation of renal fibroblasts. Because enhanced fibroblast proliferation is a hallmark of active renal fibrosis, we hypothesized that K_Ca3.1 functions may play a role in promoting fibrogenesis in the kidney in vivo. To test this hypothesis we used murine UUO, a well-established model of renal fibrosis. We first examined whether K_Ca3.1 expression in total renal tissue was altered between UUO kidneys and sham-operated control kidneys. We observed a more than 20-fold increase of K_Ca3.1 expression in fibrotic UUO kidneys paralleled by a rise in FSP-1, collagen I, collagen III and TGFβ expression levels (Fig. 3A). In contrast, subtypes K_Ca1.1 and K_Ca2.3 were not up-regulated in UUO kidneys, bearing analogy to the K_Ca expression pattern in proliferating renal fibroblasts. In line, Western blot analysis revealed considerable staining for K_Ca3.1 channel protein in total renal tissue of UUO kidneys, whereas in sham control kidneys the respective signals were very faint or not detectable (Fig. 3B). Taken together, these findings provide evidence for substantial K_Ca3.1 up-regulation in fibrotic kidneys indicating a potential pathophysiological involvement of K_Ca3.1 channels in renal fibrogenesis.

Fig. 3.

In vivo up-regulation of K_Ca3.1 expression in fibrotic kidneys after UUO. (A) Comparative quantitative RT-PCR analysis of mRNA expression levels of K_Ca channel subtypes and markers of fibrosis in total renal tissue of fibrotic kidneys (14 d UUO) and sham-operated control kidneys (Ctrl) of wild-type mice. K_Ca3.1, but not related K_Ca subtypes K_Ca1.1 and K_Ca2.3, was distinctly up-regulated in UUO kidneys along with FSP1, collagen I, collagen III, and TGFβ. Values of the ordinate show target mRNA levels given as percent of related reference gene (GAPDH) expression levels (100/2^{ΔCt(x−GAPDH)}; n = 4–10 for each data point.) *, P < 0.001. Note logarithmic scaling. (Inset) Representative ethidium bromide-stained gel depicting respective RT-PCR amplicons. (B) Representative Western blot analysis of total renal tissue indicating K_Ca3.1 protein induction in fibrotic kidneys (14 d UUO) as compared with sham-operated control kidneys (Ctrl).

Next, we studied whether K_Ca3.1 disruption could have an impact on the development and progression of renal fibrosis. For that purpose, we chose to apply a classical knockout approach using K_Ca3.1 deficient mice that were recently generated in our laboratory (30). Age- and sex-matched knockout (K_Ca3.1^−/−) and wild-type (K_Ca3.1^+/+) mice were subjected to UUO maneuver and killed after 14 days. The caudal half of each kidney was immediately shock frozen and processed for RT-PCR studies whereas the cranial half was used for histopathological evaluation. Strikingly, histological examination and morphometric analysis of UUO kidneys revealed a significant attenuation of chronic tubulointerstitial damage, reduced collagen deposition, fewer αSMA-positive cells, and a better preservation of differentiated proximal tubules and total renal parenchyma in mice deficient for K_Ca3.1. In detail, chronic tubulointerstitial damage was reduced by ≈−48% (score 44.0 ± 6.0 vs. 84.4 ± 12.9; P = 0.009), interstitial collagen I/III deposition by ≈−30% (score 132.5 ± 6.8 vs. 187.7 ± 8.9; P = 0.0006), and cells staining positive for (myo)fibroblast marker αSMA by ≈−45% (cells per HPF 18.25 ± 1.8 vs. 33.1 ± 4.1; P = 0.008) in UUO kidneys from K_Ca3.1^−/− as compared with K_Ca3.1^+/+ littermates (Fig. 4 A and B and Fig. S3b). Moreover, analysis of differentiated proximal tubules revealed a significantly higher percentage of preserved proximal tubules in UUO kidneys from K_Ca3.1^−/− compared with the wild-type (17.1 ± 1.2% vs. 10.6 ± 1.0%; P = 0.00095) (Fig. S3a Left). Likewise, the width of functional kidney parenchyma after UUO was maintained to a higher extent in mice lacking K_Ca3.1 (245.3 ± 12.7 μm vs. 199.9 ± 11.8 μm in wild-type; P = 0.02) suggesting better conservation and less scarring of functional renal tissue in these animals (Fig. S3a Right). In line with the histopathological findings, renal gene expression levels of established fibrotic markers (FSP-1, αSMA, collagen I/III) were considerably lower in UUO kidneys from K_Ca3.1^−/− compared with K_Ca3.1^+/+ mice (Fig. S3c). Of note, no appreciable difference in mononuclear cell infiltration was observed between genotypes (Fig. S4 a and b).

Fig. 4.

Genetic deficiency of K_Ca3.1 attenuates progression of renal fibrosis after UUO in mice. (A) PAS, collagen I/III, and αSMA staining. Pictures display representative areas from diseased UUO kidneys of wild-type (K_Ca3.1^+/+) and K_Ca3.1 knock out (K_Ca3.1^−/−) mice. Large areas of atrophic tubuli with high density of condensated nuclei within the tubules and interstitium of a wild-type; in comparison, kidney parenchyma of a K_Ca3.1^−/− presenting reduced chronic tubulointerstitial damage and tubular segments with well preserved architecture (PAS). Note also reduced collagen deposition and number of αSMA+ cells in kidneys from K_Ca3.1^−/− animals. (B) Histological evaluation of UUO kidney cross sections in K_Ca3.1^+/+ and K_Ca3.1^−/−. Each data point (▴,○) represents one individual UUO kidney (n = 9–10 for PAS, n = 6 for collagen, and n = 6 for αSMA); bars represent mean values. **, P < 0.01; ***, P < 0.001.

Collectively, these in vivo data provide evidence that progression of UUO-induced renal fibrosis is significantly attenuated by the absence of K_Ca3.1 channel functions and suggest the possibility that pharmacological K_Ca3.1 inhibition might constitute a therapeutic approach for fibrotic kidney disease.

Pharmacological Blockade of K_Ca3.1 Channels Reduces UUO-Induced Renal Fibrosis.

To test whether pharmacological K_Ca3.1 inhibition indeed attenuates renal fibrosis in the wild-type, we conducted a comparative interventional study using again the UUO model. Sex- and age-matched wild-type mice were subjected to UUO and subsequently assigned to either the treatment group (TRAM-34) or the vehicle control group (peanut oil). The small molecule TRAM-34 was used in the treatment group for its highly selective K_Ca3.1-blocking properties in the submicromolar range and its good tolerance in vivo (17, 26). Determination of pharmacokinetic parameters revealed an average plasma compound concentration of ≈200 nM 1 day after a single administration (i.p.) of TRAM-34 at 120 mg/kg (Fig. S5a), thus providing sufficient K_Ca3.1 inhibition in vivo. In the treatment group, mice were treated with daily TRAM-34 i.p.-injections over a period of 3 weeks starting on the day of UUO maneuver. Control animals received the same volumes of vehicle alone. Remarkably, histopathological analysis of renal cross-sections demonstrated an ≈−38% reduction of chronic tubulointerstitial damage in UUO kidneys of K_Ca3.1 blocker-treated animals as compared with vehicle-treated controls (score 78.0 ± 9.1 vs. 125.5 ± 8.6; P = 0.004). Consistent with this finding, we also observed a decrease in collagen I/III deposition by ≈−27% (score 137.6 ± 5.4 vs. 189.4 ± 6.8; P = 0.0006) and a significant reduction in interstitial αSMA+ cells by ≈−49% (cells per HPF 15.8 ± 1.5 vs. 31.1 ± 1.8; P = 0.0004) (Fig. 5 A and B and Fig. S5b), whereas the extent of mononuclear cell infiltration was not different (Fig. S5 c and d). We went on to test the effect of K_Ca3.1 inhibition in another animal model and obtained similar results in a rat model of renal fibrosis (Fig. S6 a–e). Of note, animals treated with TRAM-34 showed no overt signs of toxicity such as behavioral abnormalities or visible macroscopic organ changes.

Fig. 5.

Administration K_Ca3.1 blocker TRAM-34 attenuates development of UUO-induced renal fibrosis in wild-type mice. (A) PAS-, collagen I/III-, and αSMA-staining. Pictures display representative areas from diseased UUO kidneys of wild-type mice receiving treatment with TRAM-34 (120 mg/kg, once daily, i.p.) or vehicle (peanut oil) alone. Areas of renal parenchyma are dominated by chronically damaged tubular segments and moderate interstitial mononuclear cell infiltration in a vehicle-treated animal; in comparison, fewer atrophic tubuli and large areas of well differentiated tubular segments can be observed in the kidney of a TRAM-34-treated mouse (PAS). Note also reduced collagen deposition and number of αSMA+ cells in animals with TRAM-34-treatment. (B) Histological evaluation of UUO kidney cross sections from TRAM-34- and vehicle-treated wild-type mice. Each data point (▾, ◇) represents one individual UUO kidney (n = 7 for PAS, n = 5 for collagen, and n = 5 for αSMA); bars represent mean values. *, P < 0.01; **, P < 0.001.

Taken together, these data demonstrate that selective pharmacological K_Ca3.1 inhibition can effectively reduce the number of (myo)fibroblasts in diseased UUO kidneys and mitigate the progression of renal fibrosis.

Discussion

Excessive proliferation of activated renal fibroblasts is a hallmark of progressive renal fibrosis (2, 31–33). In this report, we investigated the role of calcium-activated potassium channels K_Ca in renal fibroblast proliferation in vitro and in the development of renal fibrosis in vivo. We found that mitogenic stimulation with bFGF induced a robust and rather selective up-regulation of the intermediate-conductance K_Ca subtype K_Ca3.1 in renal fibroblasts. The principle signal transduction pathway involved in this mitogen-induced K_Ca3.1 up-regulation appeared to rely on the Ras/Raf/MEK/ERK-signaling cascade, which is in line with observations in other cell types (15, 21, 23, 34). In contrast, expression levels of the related K_Ca channels K_Ca1.1 and K_Ca2.3 were unchanged in response to bFGF, underscoring a certain specificity of K_Ca3.1 up-regulation and suggesting a pivotal role for K_Ca3.1 in the promotion of renal fibroblast proliferation.

In support of these observations, we show that selective pharmacological inhibition of K_Ca3.1 is capable of suppressing mitogen-driven proliferation of renal fibroblasts via cell cycle arrest in G₀/G₁. The importance of K_Ca3.1 in renal fibroblast mitogenesis may be explained by its ability to enhance the driving force for Ca²⁺ influx via membrane hyperpolarization and thus sustain a high intracellular Ca²⁺ concentration needed for gene transcription, as has been reported in VSMC, T lymphocytes and cancer cells (17, 20, 24, 27, 28, 35, 36). Membrane hyperpolarization mediated by potassium channels is known to promote Ca²⁺ inflow during G₁ of the cell cycle allowing transition through the G₁/S checkpoint (37). Ouadid-Ahidouch et al. have recently reported in a cell culture study, that in MCF-7 breast cancer cells this mechanism is critically dependent on the activity of K_Ca3.1. Conversely, blockade of these channels diminished [Ca²⁺]_i, halted cell cycle progression and suppressed breast cancer cell proliferation (25). Here, we demonstrate that selective K_Ca3.1 inhibition with TRAM-34 is able to suppress the proliferation of renal fibroblasts in a dose-dependent manner. Cytotoxic or proapoptotic actions potentially arising from K_Ca3.1 blockade by TRAM-34 were not observed. Although our data provide evidence for a cell cycle arrest in G₀/G₁, we cannot exclude additional effects of K_Ca3.1 inhibition at other points of the cell cycle, particularly those involving precise cell volume regulation (38). In any case, induction of K_Ca3.1 might constitute an important step for the initiation of renal fibroblast proliferation in response to mitogenic and profibrotic factors.

However, whether K_Ca3.1 functions play a role in the pathogenesis of renal fibrosis in vivo, had so far not been investigated. To address this question we conducted a series of experiments applying the mouse model of UUO to induce experimental renal fibrosis. First, we observed a strong correlation between the increase of mRNA expression levels of (myo)fibroblast markers such as αSMA and FSP1 and the concomitant up-regulation of K_Ca3.1 in the fibrotic kidney. In contrast, the related subtypes K_Ca1.1 and K_Ca2.3 were not altered in expression, which bears analogy to the in vitro expression pattern of proliferating renal fibroblasts and thus further corroborates that K_Ca3.1 may specifically contribute to renal fibrogenesis in vivo.

To study the impact of K_Ca3.1 channels on renal fibrosis in more detail, we decided to use a definitive approach and generated a K_Ca3.1 knockout mouse. Our findings demonstrate that K_Ca3.1 deficiency significantly attenuated the development of renal fibrosis after UUO, which implies an important role for K_Ca3.1 in the pathogenesis of renal fibrotic disease. Most importantly, beside a substantial reduction in extracellular matrix including collagen deposition, we also found significantly lower numbers of interstitial (myo)fibroblasts in diseased kidneys from K_Ca3.1^−/− mice when compared with the wild-type. This indicates that loss of K_Ca3.1 likely constrains proliferation of renal (myo)fibroblasts in a profibrotic setting, thus resulting in fewer matrix-producing cells. As shown previously, fibroblasts from fibrotic kidneys are characterized by an enhanced proliferative activity and matrix synthesis capacity (7, 8). However, it should be noted that activated resident interstitial fibroblasts may not be the only cells participating in extracellular matrix production. There have been reports that bone marrow-derived cells (39, 40), transdifferentiated tubular epithelial and endothelial cells [via epithelial-to-mesenchymal (EMT) (39, 41) and endothelial-to-mesenchymal (EndMT) transition (4), respectively], and periadventitial cells (42, 43) could also contribute to the population of interstitial matrix-producing cells in renal fibrotic disease.

Our data promote the idea that the beneficial effect of K_Ca3.1 deficiency on the progression of renal fibrosis may be due to reduced proliferative activity of interstitial fibroblasts giving rise to fewer matrix-producing cells with lower net matrix synthesis. Evidence in support of this notion is provided by the well-established correlation between the number of interstitial fibroblasts and the prognosis of both human and experimental renal disease progression (44–46). Whether the lack of K_Ca3.1 functions exerts protection by additional mechanisms, as for example interference with EMT/EndMT, alteration of TGFβ signaling or modulation of inflammatory responses, warrants further investigation.

In addition to the genetic knockout of K_Ca3.1, we also treated wild-type mice subjected to UUO with the small molecule K_Ca3.1 blocker TRAM-34 and showed that pharmacological inhibition of K_Ca3.1 was similarly effective in reducing tubulointerstitial fibrosis. Reminiscent of genetic K_Ca3.1 loss, this pharmacological disruption of K_Ca3.1 functions was again associated with a substantial reduction of interstitial (myo)fibroblasts in UUO kidneys. These findings could be reproduced in a rat model of renal fibrosis, lending strength to the notion that this protective effect is not restricted to one particular species but might rather imply a more general applicability. Thus, the results from both studies, using either a genetic or a pharmacological approach, corroborate the proposed concept that K_Ca3.1 channels play an important role in the pathogenesis of renal fibrosis. Of note, in animal models of acute vascular injury (17, 19), atherosclerosis (26), angiogenesis (15), and endometrial cancer (27), administration of selective K_Ca3.1 blockers has been shown to also prevent excessive cell proliferation in vivo and ameliorate the course of disease. Importantly, long-term treatment with TRAM-34 at therapeutic concentrations caused no discernible toxicity and did not compromise immune responses in mice and rats (17, 26), which is in line with observations of the present study. Moreover, results from safety studies on the TRAM-34 analog ICA-17043, which was in Phase-3 clinical trials for sickle cell anemia (47), further support these findings. It is therefore intriguing to speculate that K_Ca3.1 may emerge as a promising target for modulating the progression of complex fibroproliferative disorders such as renal fibrosis.

In summary, our findings demonstrate that K_Ca3.1 channels play an important role in the proliferation of renal fibroblasts and contribute to renal fibrosis. Targeted disruption of K_Ca3.1 may provide a therapeutic rationale for combating fibrotic diseases such as progressive renal fibrosis.

Materials and Methods

Transgenic Mice.

Mice lacking the K_Ca3.1 channel protein were generated as described in ref. 30. See also SI Text.

Unilateral Ureteral Obstruction.

Renal fibrosis was induced in K_Ca3.1^−/− mice and wild-type littermates by UUO as described in ref. 48. Mice weighing 22–26 g were anesthetized by i.p. injection of ketamine (80 mg/kg) and xylazine (10 mg/kg). Through a flank incision, the left ureter was exposed and completely ligated using fine suture material (4–0 silk). Sham animals underwent the same surgical intervention except for ureter ligation. Mice were allowed to recover from anesthesia and were housed in standard rodent cages with ad libitum access to water and food until sacrificed. In another set of experiments, wild-type mice were subjected to UUO maneuver and treated with daily i.p.-injections of K_Ca3.1 blocker TRAM-34 at 120 mg/kg or vehicle (peanut oil) alone for 3 weeks. All experimental procedures were performed in accordance to the German animal care and ethics legislation and were approved by the local governmental authorities.

Cell Culture.

A cell line of tubulointerstitial fibroblasts (TFB) derived from murine kidney (49) was maintained in growth media (DMEM with 5% FCS, 2% penicillin/streptomycin) and switched to serum-free, quiescent medium 48 h before stimulation to induce growth arrest.

Real Time RT-PCR.

RNA from TFB cells and homogenized sections of total kidney tissue was extracted using a high-purity RNA isolation kit (Roche) according to standard procedures. Reverse transcription and PCR (RT-PCR) were performed as described in detail elsewhere (21). Control experiments in the absence of RT were routinely conducted, and all PCR products were validated for correct amplification. Real time RT-PCR was performed using Platinum SYBR Green qPCR SuperMix-UDG (Invitrogen) and the real-time PCR system ABI 7500 (Applied Biosystems). For primer sequences see SI Text.

Western Blot Analysis.

Immunoblotting was performed as described in ref. 21 with some modifications. See also SI Text.

Patch Clamp.

Electrophysiological experiments were conducted in the whole-cell configuration of the patch-clamp technique as reported before (15, 21). See also SI Text.

Proliferation Assays.

Cell proliferation was examined by use of bromodeoxyuridine (BrdU) incorporation assays and cell counts. To induce growth arrest, TFB were switched to quiescent medium 48 h before mitogenic stimulation with bFGF (10 ng/mL). In a first series of experiments, renal fibroblasts were placed in 96-well microtiter plates at 4 × 10³ cells per well and proliferation was determined at 24 h by nonradioactive BrdU incorporation assays (BrdU Labeling and Detection Kit III, Roche Applied Science) as specified earlier (12) and based on the method by Gratzner et al. (50). Optical densities were measured by photometer (Multiskan Ex, Thermo Fisher Scientific) at 405 nm. Cell count experiments were performed as described in ref. 15. Briefly, at 10–20% confluence, photo-micrographs of cells were taken in fixed fields before and 24 h after stimulation and the percent increase was calculated for each experiment. For blocker studies, cells were treated with TRAM-34, the structurally related compound TRAM-85, or vehicle (DMSO) alone.

FACS Analysis.

To determine the cell distribution in the different phases of the cell cycle, TFB cells were loaded with propidiumiodide and counted by a FACSCalibur Flow Cytometer (BD Biosciences), thus allowing allocation of cells to the G₀/G₁, S or G₂/M phase. Annexin-V-labeling (Annexin-V-Fluos, Roche Applied Science) was used to assess potential proapoptotic effects of applied substances. Here, the 35–36 kDa protein annexin-V binds with high affinity to the membrane phospholipid phosphatidylserine, which is externalized from the inner to the outer plasma membrane leaflet during apoptotic processes. Experiments were conducted according to the manufacturer's recommendations and cells were analyzed by flow cytometry for fluorescence. Sucrose-treated cells exposed to hypertonic shock [900 mOsm (Ringer plus 600 mM sucrose)] served as positive controls.

Histological Analysis.

Cranial kidney halves were fixed at room temperature in 4% neutral buffered formalin for 24 h and then embedded in paraffin. Light microscopy was performed on 3-μm sections stained by periodic acid-Schiff. Tubulointerstitial damage was evaluated in whole kidney sections, including cortex and outer medulla as described in refs. 51 and 52. For more information including immunohistochemical staining see SI Text.

Statistics.

Data are given as mean ± SEM. Statistical analysis was performed using the unpaired Student's t test. P values of <0.05 were considered statistically significant.