The K+ Channel KCa3.1 as a Novel Target for Idiopathic Pulmonary Fibrosis

Background Idiopathic pulmonary fibrosis (IPF) is a common, progressive and invariably lethal interstitial lung disease with no effective therapy. We hypothesised that KCa3.1 K+ channel-dependent cell processes contribute to IPF pathophysiology. Methods KCa3.1 expression in primary human lung myofibroblasts was examined using RT-PCR, western blot, immunofluorescence and patch-clamp electrophysiology. The role of KCa3.1 channels in myofibroblast proliferation, wound healing, collagen secretion and contraction was examined using two specific and distinct KCa3.1 blockers (TRAM-34 and ICA-17043 [Senicapoc]). Results Both healthy non fibrotic control and IPF-derived human lung myofibroblasts expressed KCa3.1 channel mRNA and protein. KCa3.1 ion currents were elicited more frequently and were larger in IPF-derived myofibroblasts compared to controls. KCa3.1 currents were increased in myofibroblasts by TGFβ1 and basic FGF. KCa3.1 was expressed strongly in IPF tissue. KCa3.1 pharmacological blockade attenuated human myofibroblast proliferation, wound healing, collagen secretion and contractility in vitro, and this was associated with inhibition of TGFβ1-dependent increases in intracellular free Ca2+. Conclusions KCa3.1 activity promotes pro-fibrotic human lung myofibroblast function. Blocking KCa3.1 may offer a novel approach to treating IPF with the potential for rapid translation to the clinic.


Introduction
Idiopathic pulmonary fibrosis (IPF) is a progressive fibrosing interstitial pneumonia of unknown etiology [1]. The term IPF is now restricted to patients with radiographic features consistent with the histological pattern of usual interstitial pneumonia (UIP). It occurs primarily in older adults [1] with an incidence of 16 per 100,000 person-years in the USA [2]. In the UK there are over 4,000 cases diagnosed annually, which is an equivalent disease burden with that of ovarian and kidney cancers [3]. There is no effective treatment [1] and prognosis is poor with a median survival of only 2-3 years from diagnosis [3]. IPF therefore represents an important cause of morbidity and mortality and novel approaches to treatment are required urgently to address this unmet clinical need.
The pathogenic mechanisms involved in IPF initiation and progression are poorly understood [4]. Myofibroblasts play a critical role in tissue repair through cell-cell and cell-matrix interactions [5], maintaining and regulating extracellular matrix, interstitial fluid volume, and the extent of tissue contraction needed for optimum function [6]. However, dysregulated or inappropriate myofibroblast function leads to pathological scarring and tissue fibrosis [7]. The myofibroblast is the principle cell responsible for the synthesis and deposition of the fibrotic matrix in IPF and the associated tissue contraction [6]. Targeting pro-fibrotic myofibroblast activity therefore offers the potential to slow down or halt the progression of IPF.
Ion channels are attractive therapeutic targets in many chronic diseases. The Ca 2+ activated K + channel K Ca 3.1 plays an important role in Ca 2+ signalling through its ability to maintain a negative membrane potential during cell activation [8]. The K Ca 3.1 channel modulates the activity of several structural and inflammatory cells, including lymphocytes [9], mast cells [10], and dedifferentiated smooth muscle cells [11], through the regulation of cell proliferation [11], activation [9], migration [10] and mediator release [12]. Pharmacological inhibition or genetic deletion of K Ca 3.1 prevents surgically induced renal fibrosis in mice by targeting myofibroblasts, leading to reduced collagen deposition and fibroblast proliferation while preserving renal parenchyma [13].
Both TGFβ1 and basicFGF are key growth factors which drive myofibroblast-dependent fibrosis in IPF [5,6]. We hypothesise that TGFβ1-and basicFGF-driven K Ca 3.1dependent cell processes are a common denominator in the pathophysiology of IPF. In this study we have investigated the expression and function of the K Ca 3.1 channel in primary human lung myofibroblasts derived from both non-fibrotic and IPF lungs.

Ethics statement
All patients donating tissue gave written informed consent and the study was approved by the National Research Ethics Service (references 07/MRE08/42 and 10/H0402/12).

Human lung myofibroblasts isolation and culture
Non-fibrotic control (NFC) myofibroblasts were derived from healthy areas of lung from patients undergoing lung resection for carcinoma at Glenfield Hospital. No morphological evidence of disease was found in the tissue samples used for myofibroblast isolation. IPF myofibroblasts were derived from patients undergoing lung biopsy for diagnostic purposes at the University of Pittsburgh Medical Center, and were shown to have UIP on histological examination. Myofibroblasts were grown from explanted lung tissue from both sources under identical conditions, using Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% fetal bovine serum (FBS), antibiotic/antimycotic agents and non-essential amino acids [14,15]. The cells were cultured at 37°C in 5% CO 2 /95% air. Cells were studied at passages 4-5 for functional studies. All NFC patients gave informed written consent and the study was approved by the Leicestershire, Northamptonshire and Rutland Research Ethics Committee 2. Written informed consent was also obtained from all IPF subjects, in accordance with the responsible University of Pittsburgh Institutional Review Board.

Immunohistochemistry on human lung tissue
Sections were cut at 4 µm from paraffin-embedded diagnostic lung parenchymal biopsies or healthy control lung obtained at cancer surgery, and immunostained using mouse anti-αSMA mAb (1A4, 0.7 µg/ml, Dako), rabbit polyclonal anti-K Ca 3.1 (AV35098, 5 µg/ml, Sigma-Aldrich), or appropriate isotype control at the same concentration as the primary antibody.

qRT-PCR
Myofibroblast RNA was isolated using the RNeasy Plus Kit (Qiagen, West Sussex, UK) according to the manufacturer's instructions. K Ca 3.1 gene expression was analyzed with the gene-specific Quantitect Primer Assay (Qiagen, Hilden, Germany), Hs_KCNN4_1_SG. The internal normaliser gene used was β-actin (forward primer, 5'-TTCAACTCCATCATGAAGTGTGACGTG-3', reverse primer, 5'-CTAAGTCATAGTCCGCCTAGAAGCATT-3'). All expression data was normalized to β-actin and corrected using the reference dye ROX. Gene expression was quantified by real-time PCR using the Brilliant SYBR Green QRT-PCR 1- Step Master Mix (Strategene, Amsterdam, The Netherlands). PCR products were run on a 1.5% agarose gel to confirm the product amplified was the correct size: K Ca 3.1 130 base pairs (bp), β-Actin 310 bp.
To study the effects of either TGF-β1 (10 ng/ml, R&D Systems, Abingdon, UK) or bFGF (10 ng/ml, R&D Systems) on K Ca 3.1 mRNA expression, cells were grown to confluence and then serum starved for 24 h prior to growth factor exposure for 24 h.

Western blot
Human myofibroblasts were disrupted in lysis buffer and soluble proteins from equivalent numbers of cells were resolved by 12% SDS-PAGE and then transferred to an immunobilon-P polyvinylidene difluoride membrane. Membranes were blocked and incubated with rabbit polyclonal K Ca 3.1 antibodies (Sigma antibodies P4997, AV35098 and GSK gift antibody M20. Protein bands were identified by horseradish peroxidase-conjugated secondary antibody and enhanced chemiluminescence reagent (Amersham, Little Chalfont, UK). Immunolabelled proteins were visualized using an ECL western blot detection system (GE Healthcare Life Sciences, Buckinghamshire, UK).

Human myofibroblast proliferation
Human lung myofibroblasts (n=4) were seeded into 6 well plates, when 50% confluent cells were serum starved for 24h in serum-free medium. Cells were then stimulated with serumfree medium plus 0.1% DMSO, 10% FBS medium plus 0.1% DMSO, 10% FBS plus TRAM-34 20 nM and 10% FBS plus TRAM-34 200 nM. After 48h cells were mobilized with 0.1% trypsin/0.1% EDTA and counted used a standard haemocytometer. Cell viability was assessed by trypan blue exclusion. Results were counted by 2 blinded observers with excellent agreement (intra-class correlation of 0.969). All conditions were performed in duplicate.

Wound healing assay
Human myofibroblasts were grown to confluence in 6 well plates. After 24 h in serum-free medium, 3 artificial wounds were scratched in each monolayer using a 200 µl pipette creating a linear cell-free area. After washing, TRAM-34 (20 and 200 nM) and ICA-17043 (Senicapoc, a gift from Icagen Inc, Durham, NC, USA) (10 and 100 nM) were added (final DMSO concentration 0.1%), and the cells stimulated with either 10 ng/ml bFGF or 10% FBS alone for 48 h. Control wells contained 0.1% DMSO and serum-free medium alone. Cells migrating and proliferating into the wound were observed and photographs taken over 48 h. Wound healing was analysed by measuring the area of the scraped wound using Cell F software by a blinded observer (Olympus, UK), and quantified as a percentage of the starting area of the wound scraped. Data presented represents the mean of the measurements from 2 different scratches. Results were measured by 2 blinded observers with excellent agreement (intra-class correlation of 0.969).

Collagen Assay
Myofibroblasts were cultured in serum-free medium alone or 0.1% DMSO control, and stimulated with TGFβ1 10 ng/ml in the presence of DMSO control, TRAM-34 (20 and 200 nM) and ICA-17043 (10 and 100 nM) for 16 h. Soluble collagen released by myofibroblasts was quantified using the Sircol collagen assay (Biocolor, County Antrim, UK) according to the manufacturer's instructions [20,21].

Collagen Gel Contraction
Cells were pre-treated for 24 hours with serum-free media, 0.1% DMSO, TRAM-34 200 nM, ICA-17043 100 nM, TRAM-7 200 nM or TRAM-85 200 nM. Cells were detached and collagen gels were set up as described in [22]. TGFβ1 and bFGF were then added to appropriate wells to a final concentration of 10 ng/ml. Photographs were taken at 0 h and 22 h. The surface area was measured at each time point using ImageJ software (http://rsbweb.nih.gov/ij/) by 2 blinded observers with excellent agreement (intraclass correlation of 0.974).

Calcium Imaging
Changes in [Ca 2+ ]i were assessed with the use of a flurometric Ca 2+ probe Fura-2 (Sigma-Aldrich). Myofibroblasts were grown on 35 mm fluorodish cell culture dishes (WPI, Hertfordshire, UK) and when 40% confluent loaded with Fura-2 for 45 min at 37°C in normal physiological saline solution (140 mmol/L NaCl, 5 mmol/L KCl, 2 mmol/L CaCl 2 , 1 mmol/L MgCl 2 , 10 mmol/L HEPES, pH 7.4 with NAOH) containing 5 µmol/L fura-2-acetoxymethyl ester and 2.5 mmol/L probenecid. The cells were then washed and [Ca 2+ ]i recorded using the single cell recording system previously described [23]. Data acquisition occurred at a rate of 1 dual wavelength image every second and are presented as the 340 nm/380 nm ratio.

Statistical analysis
Experiments from an individual donor were performed either in duplicate or triplicate and a mean value was derived for each condition. Data distribution was tested for normality using the Kolmogorov-Smirnov test. For parametric data the 1-way ANOVA or repeated measures ANOVA for across-group comparisons was used followed by the appropriate multiple comparison post hoc test; otherwise an unpaired or paired ttest was used. Where appropriate chi-squared tests were used. For non-parametric data the Kruskal-Wallis test was used for across group comparisons with the Dunn's multiple comparison post hoc test, or the Mann Whitney U test was used where there were two unpaired groups. GraphPad Prism for windows (version 6, GraphPad Software, San Diego California USA) was used for these analyses. A value of P<0.05 was taken to assume statistical significance.

Characterization of human lung myofibroblasts
The clinical characteristics of the myofibroblast donors are listed in Table 1. Cultured human lung myofibroblasts displayed the typical morphology of fibroblast-related cells with a spindle or stellate shape (Figure 1). No differences in THY-1, 1B10 and αSMA expression were found between passages 2 and 4. At passage 4, >98% of cells from all donors expressed the fibroblast marker FSP, >97% expressed the fibroblast marker THY-1, and >99% expressed αSMA, confirming that these cells were almost exclusively myofibroblast in phenotype. Collagen 1 staining was present in 100% of cells (Figure 1). There was no difference between NFC and IPF cells. No cells expressed the macrophage/monocyte marker CD68 or mesenchymal cell marker CD34. All isotype controls were negative (Figure 1).

Myofibroblasts express K Ca 3.1 channel mRNA, which is up-regulated by TGFβ1
Human lung myofibroblasts (n=5 NFC, n=5 IPF) expressed K Ca 3.1 mRNA (Figure 2a). No difference in K Ca 3.1 mRNA expression was found between passages, but all experiments were performed between passages 4 and 5 for consistency  (Figure 2b). K Ca 3.1 mRNA was significantly increased in cells from NFC compared to IPF donors (Figure 2c). TGFβ1 (10 ng/ml) stimulation for 24h upregulated myofibroblast K Ca 3.1 channel mRNA relative to β-actin, which was significantly greater in IPF compared NFC cells (Figure 2d). K Ca 3.1 mRNA was up-regulated at least 8 fold in every IPF donor after TGFβ1 exposure (Figure 2e).

Myofibroblasts express K Ca 3.1 protein
K Ca 3.1 protein expression in human lung myofibroblasts was identified by Western blot (n=6 NFC, n=5 IPF). The predicted weight of K Ca 3.1 is 48kDa, but larger forms of ~53kDa and several shorter splice variants exist [11,[24][25][26]. Using two different anti-K Ca 3.1 antibodies, M20 and P4997, a consistent band of ~48kDa was observed (Figure 3a). M20 and P4997 also stained bands of 53kDa as described in human fibrocytes and airway smooth muscle cells [11,26], and 39kDa consistent with the presence of splice variants [25]. K Ca 3.1 protein expression in myofibroblasts was also evident using immunofluorescent staining (Figure 3b).

Myofibroblasts express K Ca 3.1 channel currents which are increased in IPF
At baseline, myofibroblasts from NFC and IPF donors demonstrated strong outwardly rectifying whole-cell currents, and frequent inwardly rectifying currents with features of the Kir2 family (confirmed by blocking with 10 µM barium, results not shown). Interestingly, 69% of IPF cells had Kir2-like currents at baseline as opposed to 42% of NFC cells (P=0.0007, chi squared). K Ca 3.1 currents in myofibroblasts were evoked using the K Ca 3.1 opener 1-EBIO [27]. To block K Ca 3.1 we used two distinct and selective K Ca 3.1 channel blockers, TRAM-34 (Kd for K Ca 3.1 block 20 nM) [18] and ICA-17043 (Senicapoc)(Kd ~10 nM) [19]. K Ca 3.1 currents were not present at baseline but were frequently elicited following the addition of 1-EBIO (100 µM), and demonstrated the characteristic electrophysiological features of K Ca 3.1 (Figure 3c and d). In addition, the 1-EBIOinduced current was dose-dependently blocked by TRAM-34 (Figure 3c and d) and ICA-17043 (not shown) with complete block at 200 nM and 100 nM respectively. DMSO vehicle (0.1% final concentration) had no effect (not shown).
The addition of 1-EBIO elicited K Ca 3.1 currents in 59% of NFC myofibroblasts (n=7 donors) and 77% of IPF myofibroblasts (n=7 donors)(P=0.0411, Chi squared). Also, the proportion of myofibroblasts per donor responding to 1-EBIO was significantly increased in IPF compared to NFC cells (P=0.0285) (Figure 4a), and the size of the currents induced by 1-EBIO was significantly greater in IPF compared to NFC cells (P=0.0054, Mann Whitney test) (Figure 4b and c). Thus, functional K Ca 3.1 channels were expressed more frequently and the currents were larger in myofibroblasts derived from IPF lung tissue compared to NFC tissue.
Measurements of myofibroblast capacitance were unreliable due to the relatively large cell size, so it was not possible to calculate current density. Primary human lung myofibroblast cultures between passages 4 and 5 were stained with myofibroblast markers. Representative images are shown for: anti-fibroblast surface protein (FSP) and the mouse isotype control IgM; anti-fibroblast antigen which recognises the fibroblast antigen (thy-1/ CD90) and the mouse isotype control IgG1; α-smooth muscle actin, and the isotype control IgG2a; collagen type 1 antibody and rabbit isotype control IgG; CD68 cell staining was negative; and corresponding isotype control IgG3, indicating that there is no contamination of monocytes or macrophage cells; CD34 antibody shows negative staining as does the appropriate isotype control IgG1. Nuclei are stained with DAPI.  Products from quantitative real-time PCR for K Ca 3.1 were visualized on a 1.5% agarose gel to confirm that only one product was amplified and that it was the correct size (130 bp). β-Actin was used as the normalizing control (310bp). b) In preliminary experiments we examined the mRNA expression levels of K Ca 3.1 in the NFC donors at passages 2, 4 and 7 and again found no significant differences. To be sure that passage number had no effect all experiments were performed between passages 4 and 5. Data represent mean±SEM. c) Quantitative real-time PCR showed that K Ca 3.1 mRNA expression was greater in NFC donors (n=5) than IPF donors (n=5), P=0.0262 (unpaired t test). d) K Ca 3.1 mRNA expression increased after TGFβ1 stimulation (All groups; 1-way ANOVA, P<0.0001), NFC donors (n=5), IPF donors (n=5). There was a highly significant increase in IPF myofibroblasts following 24h of TGFβ1 stimulation. P<0.0001 (corrected by Bonferroni's multiple comparison test). e) Quantitative real-time PCR demonstrating the relative fold increase in NFC and IPF myofibroblasts after stimulation with TGFβ1. Following normalization with βactin there was a relative fold increase in K Ca

channels demonstrate greater expression in IPF myofibroblasts compared to NFC myofibroblasts and channel expression is increased by pro-fibrotic growth factors. a)
The mean percentage of IPF myofibroblasts per donor developing a K Ca 3.1 current in response to 1-EBIO was significantly higher than in NFC myofibroblasts (P=0.0285, unpaired t-test). Data presented as mean±SEM. b) The whole-cell current at +40 mV before and after the addition of 1-EBIO in all responding NFC and IPF human lung myofibroblasts. Data presented as median and IQR. c) The subtracted (1-EBIO minus baseline) 1-EBIO-dependent K Ca 3.1 current at +40 mV was significantly larger in IPF cells than in NFC cells (P=0.0054, Mann Whitney test). Data presented as median and IQR. d) The mean percentage of NFC and IPF myofibroblasts expressing K Ca 3.1 currents increased after stimulation with TGFβ1 and bFGF (All groups; 1-way ANOVA, P=0.0013). The proportion of IPF cells responding to 1-EBIO after TGFβ1 stimulation was significantly higher (*P=0.0336, corrected by Bonferroni's multiple comparisons test). Significantly more NFC cells responded to 1-EBIO following bFGF stimulation (**P=0.0035, corrected by Bonferroni's multiple comparisons test). Data presented as mean±SEM.

The proportion of myofibroblasts expressing K Ca 3.1 currents increases following mitogenic stimulation
After 24h of stimulation with 10 ng/ml basic fibroblast growth factor (bFGF), a greater proportion of myofibroblasts from NFC tissue expressed K Ca 3.1 currents in comparison to unstimulated cells (P=0.0035, corrected by Bonferroni's multiple comparisons test), indicating increased functional channel expression. Similar results were seen in IPF myofibroblasts using TGFβ1 (P=0.0336, corrected by Bonferroni's multiple comparisons test) (Figure 4d).

K Ca 3.1 immunoreactivity is expressed in IPF parenchymal lung tissue
There was strong immunostaining for K Ca 3.1 in NFC lung tissue (n=3), particularly in airway and alveolar epithelial cells, but also in cells within the interstitium including vessels and inflammatory cells (Figure 5a). K Ca 3.1 was also expressed strongly in areas of parenchymal fibrosis in IPF and colocalised with areas of αSMA positivity (n=5, IPF) (Figure 5b). As there is no control for parenchymal fibrosis, we did not compare the magnitude of K Ca 3.1 staining between NFC and IPF tissue.

K Ca 3.1 channel inhibition inhibits myofibroblast proliferation
Stimulation with 10% FBS for 48h significantly increased myofibroblast proliferation over control (P=0.0040, corrected by Bonferroni's multiple comparisons test), in both NFC and IPF donors, and to the same extent. However, K Ca 3.1 blockade with TRAM-34 (200 nM) inhibited FBS-induced proliferation (P=0.0076, corrected by Bonferroni's multiple comparisons test) (Figure 6a).

Selective pharmacological blockade of K Ca 3.1 attenuates wound healing
Myofibroblast wound healing was assessed in a 2-D scratch assay (Figure 6b). Myofibroblasts stimulated with 10% FBS showed enhanced wound repair compared to media alone (P<0.0001, paired t-test). There was no difference between NFC-and IPF-derived cells and all data are pooled (Figure 6c). Wound healing in the absence of FBS or bFGF was not affected by K Ca 3.1 blockade (Figure 6d) and no inhibition of wound healing was seen with two structurally related molecules without channel blocking activity, TRAM-85 and TRAM-7 (Figure 6e) [13,28].
FBS-induced wound repair was dose-dependently attenuated by both TRAM-34 and ICA-17043 (repeated measures ANOVA, P<0.0001 and P=0.0095 respectively) (Figure 7a and b) (Figure 7c). At 48 h, total wound healing was reduced by 24.4±6.6% in the presence of ICA-17043 100 nM compared to DMSO control (equivalent to inhibition of the FGF-dependent response by 63.4±13.4%). In summary, mitogen-induced wound healing of human lung myofibroblasts is attenuated by K Ca 3.1 channel block.

TGFβ1-dependent collagen production in human myofibroblasts is attenuated by K Ca 3.1 blockade
Myofibroblasts from both IPF and NFC lung secreted similar amounts of collagen in response to TGFβ1 (10 ng/ml). This was markedly attenuated by both TRAM-34 and ICA-17043 (Figure 8a and b), but was not affected by TRAM-85 or TRAM-7 (Figure 8c).

K Ca 3.1 inhibition attenuates TGF-β1 and bFGF induced myofibroblast contraction
Myofibroblasts cultured in collagen gels from both IPF and NFC donors contracted following TGFβ1 stimulation and bFGF stimulation (Figure 9). After dose response experiments were performed, 10 ng/ml chosen as the optimal concentration for each growth factor (Figure 9a and b). There was no difference between NFC-and IPF-derived cells (Figure 9c,) and all data are therefore pooled. TGFβ1 stimulation increased myofibroblast contraction from (mean±SEM) 30.6±3.6% to 48.6±2.4%, and bFGF stimulation increased myofibroblast contraction from 31.7±3.8% to 51.2±4.8%. Compared to DMSO control, pre-treatment for 24 h in the presence of either TRAM-34 200 nM or ICA-17043 100 nM almost completely inhibited both TGFβ1-and bFGF-stimulated myofibroblast contraction (Figure 10a-d). In contrast, the control molecules TRAM-7 and TRAM-85 were without effect (not shown).
TRAM-34 and ICA-17043 did not inhibit baseline constitutive myofibroblast contraction significantly in the absence of growth factors (Figure 9d and e), indicating that TRAM-34 and ICA-17043 only inhibit TGFβ1-or bFGF-dependent myofibroblast contraction.

K Ca 3.1 regulates myofibroblast pro-fibrotic functions by inhibiting TGFβ1 induced increases in [Ca 2+ ]i
Open K ca 3.1 channels hyperpolarise plasma membranes which in turn promotes Ca 2+ influx [12]. This is believed to account for the ability of K Ca 3.1 inhibition to attenuate many diverse cell responses. To investigate the underlying mechanism behind the attenuation of pro-fibrotic myofibroblast function with K Ca 3.1 blockers, we therefore investigated whether changes in [Ca 2+ ]i occur following stimulation with TGFβ1. TGFβ1 (10 ng/ml) elicited an immediate, rapid rise in [Ca 2+ ]i in 100% of IPF cells and 53% of NFC cells. There was a significant increase in [Ca 2+ ]i for both IPF (n=33, P<0.0001) and NFC (n=28, P<0.0001) (Figure 11a). In the IPF donors only, TRAM-34 was added 5 minutes prior to the addition of TGFβ1 which inhibited the TGFβ1 induced rise in [Ca 2+ ]i (Figure 11b). The change in [Ca 2+ ]i following TGFβ1 stimulation was significantly reduced with 5 minutes pre-treatment of TRAM-34 200 nM compared to controls, P<0.0001 (Figure 11c). Cells were also treated with 0.1% DMSO control which had no effect on subsequent TGFβ1-dependent increase in [Ca 2+ ] I , data no shown.

Discussion
There are three major, novel findings from this study. Firstly we have performed the first detailed electrophysiological  analysis of primary human parenchymal myofibroblasts from both NFC and IPF lung, and show for the first time that these cells express the Ca 2+ -activated K + channel K Ca 3.1 in both health and disease. Secondly we demonstrate that pharmacological blockade of K Ca 3.1 inhibits human lung myofibroblast pro-fibrotic activity such as proliferation wound healing, collagen secretion and growth factor-dependent contraction. Thirdly, TGFβ1-induced myofibroblast pro-fibrotic function is associated with K Ca 3.1-dependent regulation of [Ca 2+ ]i. These findings highlight the importance of the K Ca 3.1 channel and Ca 2+ signaling in the growth factor-dependent profibrotic functions of human lung myofibroblasts.
The widely accepted markers of a myofibroblast are that it co-expresses fibroblast-associated markers and αSMA fibres [6]. The cells studied in these experiments were predominantly of a myofibroblast phenotype, a finding consistent with the parenchymal lung myofibroblast phenotype described by others [29,30]. K Ca 3.1 channel mRNA, protein and functional channels were expressed in myofibroblasts derived from both NFC and IPF donors. Interestingly, K Ca 3.1 currents were present more frequently in IPF lung myofibroblasts and these currents were larger when compared to NFC lung myofibroblasts. This might reflect K Ca 3.1 up-regulation driven by the initiating disease insult in vivo, but the cells studied had been through 4 passages of culture, raising the possibility of an underlying disease-predisposing difference in myofibroblast K Ca 3.1 activity in patients with IPF. Surprisingly, although K Ca 3.1 currents were increased in IPF myofibroblasts, K Ca 3.1 mRNA was decreased. However, chronic exposure of cells to the K Ca 3.1 opener 1-EBIO down-regulates K Ca 3.1 mRNA expression [31,32], suggesting a negative-feedback mechanism which is perhaps operative here. Whether the increased K Ca 3.1 plasma membrane expression in IPF myofibroblasts represents increased trafficking to the cell membrane or decreased turnover requires further study. TGFβ1 is a key mediator of fibrotic diseases, playing an important role in myofibroblast differentiation [33], matrix protein production [34], and possibly epithelial-mesenchymal transition [35]. TGFβ1 stimulation increased myofibroblast K Ca 3.1 mRNA expression, which was greatest in IPF-derived myofibroblasts, and increased the frequency of cells expressing functional membrane channels. bFGF, another key mediator in fibrotic diseases, is a potent chemoattractant and mitogen for myofibroblasts, and regulates extracellular matrix production [34]. bFGF stimulation also increased the frequency of cells expressing K Ca 3.1 currents (mRNA expression was not studied). These observations are in keeping with bFGF-and TGFβ1-dependent K Ca 3.1 upregulation in several cell types [36].
The upregulation of functional K Ca 3.1 channels by both TGFβ1 and bFGF suggests that the biological effects of these growth factors might rely heavily on K Ca 3.1 channel activity. This is supported by the observations that serum-dependent proliferation, serum-and bFGF-dependent myofibroblast wound healing, TGFβ1-dependent collagen secretion, and both bFGF and TGFβ1-dependent myofibroblast contraction were attenuated by two distinct and specific K Ca 3.1 blockers, TRAM-34 and ICA-17043. Importantly, these ion channel blockers inhibited these cell processes at physiologically relevant concentrations. Thus it takes 5-10x the K d to inhibit almost all channels (K d =concentration producing 50% block). The K d for TRAM-34 is 20 nM [11] and for ICA-17043 6-10 nM [10]. At 10x the K d for both blockers, myofibroblast wound , and there were significant differences found between TGFβ1 and ICA-17043 10 nM (**P=0.0067, corrected by Bonferroni's multiple comparisons test) and between TGFβ1and ICA-17043 100 nM (***P=0.0039, corrected by Bonferroni's multiple comparisons test). c) TGFβ1 increased collagen secretion in myofibroblasts (*P=0.0417, paired ttest.), however no inhibition of collagen secretion was evident with TRAM-7 or TRAM-85 (n=3). Data represented as mean±SEM for all the above figures. doi: 10.1371/journal.pone.0085244.g008 Figure 9. Constitutive unstimulated myofibroblast contraction is not altered by K Ca 3.1 blockade, and 10 ng/ml of TGFβ1 and bFGF is optimal concentration. (a and b) dose response was performed before beginning experiments which confirmed 10 ng/ml as optimal concentration to use for both TGFβ1 and bFGF in the myofibroblast contraction assay, n=2.c) demonstrates that at the different time points of the contraction assay there were no significant difference between the response of the NFC and IPF cells to TGFβ1. At 1 hour NFC n=4 and IPF n=3, 4 hours NFC n=3, IPF n=2 and 22 hours NFC n=4 and IPF n=5. d) TRAM-34 does not significantly inhibit myofibroblast contraction in the absence of mitogenic stimulation, n=6. e) ICA-17043 does not significantly inhibit myofibroblast contraction in the absence of mitogenic stimulation, n=5. Data represent mean±SEM for all the above figures. healing, collagen secretion and contraction were significantly attenuated. At these concentrations both drugs are specific and not known to affect other ion channels, receptors or transporters [13]. Furthermore, the effects of TRAM-34 and ICA-17043 on myofibroblast biology were not mimicked by TRAM-7 or TRAM-85, two molecules of similar structure to TRAM-34 which do not block K Ca 3.1 ion channels [13,28].
Upon TGFβ1 stimulation human lung myofibroblast exhibited a rise in [Ca 2+ ]i concentrations which was blocked by TRAM-34. This demonstrates how both membrane hyperpolarization and Ca 2+ influx is critical for myofibroblast function in keeping with observations in other cells types [10,11,26], and provides mechanistic insight for the role of K Ca 3.1 in lung myofibroblast responses.
Although K Ca 3.1 channels were more highly expressed in IPF myofibroblasts than in NFC cells, the functional responses of the cells to growth factors were largely similar irrespective of their source. This is perhaps not surprising because membrane hyperpolarization can be achieved with as few as 12 channels per cell [32], meaning that Ca 2+ dependent cell processes can proceed with very low channel expression. Similar shifts in membrane potential in NFC and IPF myofibroblasts are evident in Figure 3c. However, the presence of significantly more channels in IPF myofibroblasts suggests that it will be more difficult for a physiological process to overcome the effects of the increased channel activity. This however does not apply to channel blockers which will inhibit 10 or 1000 channels equally effectively. Therefore pharmacological K Ca 3.1 blockers are attractive targets as they block all channels and will potentially attenuate pro-fibrotic responses as shown in this study. K Ca 3.1 was expressed in both NFC and IPF lung tissue, supporting the relevance of our in vitro findings. In IPF tissue, . c) Similarly, myofibroblast collagen gel contraction was increased following bFGF stimulation and was also inhibited with 24h pre-treatment with TRAM-34 200 nM (All groups; repeated measures ANOVA P<0.0001, for bFGF compared to control, P<0.0001, and for TRAM-34 compared to bFGF, P<0.0001 (corrected by Bonferroni's multiple comparisons test). d) Similarly, 24h pre-treatment with ICA-17043 100 nM significantly reduced bFGF-dependent myofibroblast collagen gel contraction (P=0.0007, repeated measures ANOVA) (P=0.0005 for bFGF versus control and P=0.0053 for ICA-17043 versus bFGF, corrected by Bonferroni's multiple comparisons test). Data represented as mean±SEM for all the above figures. doi: 10.1371/journal.pone.0085244.g010 K Ca 3.1 was not only expressed in areas of αSMA positivity, but also highly expressed in alveolar epithelial cells, vessels and inflammatory cells. In addition to alveolar epithelial cells, both mast cells and fibrocytes are implicated in IPF progression and express K Ca 3.1 [10,26]. Targeting K Ca 3.1 in IPF may therefore have additional benefits over and above those achieved through the inhibition of myofibroblast function.
Our data indicate that the K Ca 3.1 channel may play a key role in the development of lung fibrosis, in both IPF and other lung disorders. K Ca 3.1 is an attractive pharmacological target as it appears to play a minor role in healthy physiology, but contributes significantly to tissue remodeling and fibrosis [37]. It has been suggested that K Ca 3.1 opening contributes to pulmonary vascular relaxation in rats [38], but this was not the case in human pulmonary vessels [39]. Importantly, K Ca 3.1 knockout mice are viable, of normal appearance, produce normal litter sizes, and exhibit rather mild phenotypes [13]. High doses of TRAM-34 administered to rodents over many weeks are well tolerated [13] and the orally available K Ca 3.1 blocker, ICA-17043, has been administered to humans in phase 2 and 3 trials of sickle cell disease with minor side effects [40]. There is therefore the potential for the rapid investigation of K Ca 3.1 blockade in clinical trials of IPF and other fibrotic lung diseases.