Lysophosphatidic acid-activated Cl⁻ current activity in human systemic sclerosis skin fibroblasts

Abstract

Objectives. SSc (scleroderma) is an often fatal disease characterized by widespread tissue fibrosis. Fibroblasts play a key role in SSc-associated fibrosis. This study was designed to determine: (i) whether fibroblasts isolated from skin of patients with SSc have increased lysophosphatidic acid-activated Cl⁻ current (ICl_LPA) activity vs healthy controls; (ii) whether myofibroblast differentiation is involved in SSc skin fibrosis; and (iii) whether SSc fibroblasts have different proliferation rates vs controls.

Methods. Skin biopsies were taken from involved and uninvolved skin of SSc patients and controls. Whole-cell perforated patch-clamping was used to measure ICl_LPA activity in fibroblasts isolated and cultured from these biopsies. Western blotting was used to measure α-smooth muscle actin (α-SMA). Proliferation was measured using a colorimetric assay.

Results. Fibroblasts cultured from SSc skin show significantly increased ICl_LPA activity following LPA exposure compared with control skin fibroblasts. α-SMA protein was significantly increased in cultured SSc skin fibroblasts vs controls. No significant differences in proliferation rates were found.

Conclusions. Elevated ICl_LPA activity is a hallmark of SSc skin fibroblasts. Blocking ICl_LPA activation may be a new therapeutic approach for treating SSc-associated fibrosis.

Introduction

SSc (scleroderma) is an often fatal disease characterized by progressive, widespread tissue fibrosis. SSc affects not only the skin but also internal organs, such as the lungs and kidneys, leading to organ dysfunction and failure [1]. Early diagnosis of SSc is difficult, and the only treatments available are for symptomatic and organ-based pathologies, with no treatments available that are disease modifying [2]. SSc has no known cause, although it does have an autoimmune component. SSc-associated fibrosis is similar to that seen during wound healing, but the process is not self-limiting as in wound healing [3].

It is widely accepted that activated fibroblasts (myofibroblasts), which overproduce collagen and other components of the extracellular matrix, play a key role in the pathogenesis of SSc [4, 5]. Unfortunately, the stimuli leading to fibroblast activation in SSc remain unclear. Myofibroblasts, which are commonly identified by expression of α-smooth muscle actin (α-SMA) and by features that are intermediate between those of smooth muscle cells and fibroblasts, are typically derived from fibroblasts or epithelial cells and can also be derived from circulating blood-derived mononuclear cells [6, 7]. Myofibroblast differentiation and activity can be regulated by many factors, including TGF-β, connective tissue growth factor, IL-1, IL-4 and IL-6 [8, 9]. Lysophosphatidic acid (LPA) is a bioactive lysophospholipid synthesized during blood coagulation by enzymes released from activated platelets and other cell types [10]. To date, eight LPA-specific mammalian G-protein-coupled receptors (GCPRs), LPA_1–8, have been identified [11] (LPA_6–8 are also referred to as GPR87, p2y5 and p2y10, respectively), and LPA has recently been associated with fibrotic processes and diseases [12]. Importantly, we have demonstrated that LPA can stimulate fibroblast to myofibroblast differentiation [13]. LPA has also been shown to increase keratinocyte proliferation and migration, skin fibroblast proliferation and can activate TGF-β through Smad-3 in keratinocytes [14, 15].

We have previously described a Cl⁻ current (ICl_LPA) activated by LPA and sphingosine-1-phosphate (S1P) that has a much higher current density in myofibroblasts than in fibroblasts isolated from the same tissue. We have recorded this current in myofibroblasts acutely isolated from wounded corneas [16] as well as in cultured cornea, lung [13, 17], skin, heart and liver (Yin and Watsky 2007, data not published) myofibroblasts. We have demonstrated that ICl_LPA activity plays a critical role in the differentiation of human lung fibroblasts to myofibroblasts [13, 17]. The present study was designed to determine whether fibroblasts isolated and cultured from involved or uninvolved skin biopsies of SSc patients have higher ICl_LPA activity as compared with skin fibroblasts from healthy controls, whether fibroblast to myofibroblast differentiation is involved in SSc skin fibrosis and whether SSc skin fibroblasts have different proliferation rates as compared with control fibroblasts.

Patients and methods

Study subjects

Patients with a history of dcSSc were recruited from the Rheumatology Clinic of the University of Tennessee Health Science Center. All the patients met the 1980 ARA criteria for clinical diagnosis of SSc [18]. Healthy controls were recruited from among the University of Tennessee faculty and staff. Those with a history of an organ or stem cell transplant, those who had used prednisone, CYC, d-penicillamine, CSA, MTX, AZA or other immune modulator therapies within 1 month of the start of the study, and those who were <18 years old were excluded from the study. Written consent for participation in the study was obtained from all participants in accordance with the Helsinki II Declaration, and the study was approved by the University of Tennessee Health Science Center Institutional Review Board. Detailed information regarding the SSc patients and controls is presented in Table 1.

Table 1.

Clinical characteristics of subjects included in the study

Strain	Gender	Age, years	Duration, years
SSc1 involved	F	51	7
SSc1 uninvolved
SSc2 involved	F	48	3
SSc2 uninvolved
SSc3 uninvolved	F	49	3
SSc4 uninvolved	F	35	13
SSc5 involved	M	55	8
SSc6 involved	F	48	10
Control 1	F	51	NA
Control 2	F	48	NA
Control 3	M	37	NA
Control 4	M	48	NA
Control 5	F	51	NA
Control 6	F	57	NA

NA: not applicable; F: female; M: male.

Skin biopsy procedure and fibroblast culture

The forearm was chosen as the biopsy site for involved SSc skin and controls, and the abdomen was chosen as the clinically uninvolved SSc biopsy site. The selected area was draped with a sterile waterproof material and prepped with 70% ethanol and a topical microbiocide swab; then an ethanol swab was used to remove the microbiocide, and a sterile 4 × 4 inch gauze was used to dry the area. The biopsy site was anaesthetized with 2% lidocaine without epinephrine. After 5 min, a sterile disposable 4-mm punch biopsy was used to obtain a full thickness dermal biopsy from the anaesthetized site. Using a sterile technique, the skin biopsy site was sutured with one to two sutures and covered with adhesive gauze for 48 h. Patients returned to the clinic 1 week later for suture removal.

To prepare cells for culture, the biopsied skin was separated from s.c. fat by cutting it free with sterile scissors. The biopsies were placed in 100-mm tissue culture dishes and cut with a number 10 sterile surgical blade into small fragments and placed under sterile coverslips that were secured to the tissue culture dish with sterile siliconized grease. Eagle’s minimum essential medium (Gibco and Invitrogen, Carlsbad, CA, USA) containing 10% fetal bovine serum (FBS; HyClone, Logan, UT, USA), 1× non-essential amino acids (Irvine Scientific, Santa Ana, CA, USA), 1 mM sodium pyruvate (Sigma, St Louis, MO, USA), 2 mM l-glutamine (Sigma), 100 µg/ml streptomycin–penicillin (MP Biomedical, Solon, OH, USA) and 1 mg/ml fungizone (Gibco, Invitrogen) was added to each tissue culture dish and dishes were placed in a tissue culture incubator maintained at 37°C with a humidified atmosphere containing 5% CO₂. Medium was changed every 5 days. Two to three weeks after fibroblasts grew out of the explants they were harvested by trypsinization and transferred to new dishes. Fibroblasts were used between the 5th and 10th passage.

Cells for western blot analyses were prepared in two ways. To detect the level of α-SMA expression in SSc and control fibroblast, cells were cultured in regular 10% FBS medium. To determine the level of α-SMA expression after exposure of the cells to TGF-β or the Cl⁻ channel blocker tamoxifen, cells were cultured in regular 10% FBS medium until 80% confluence, and then serum starved for 3 days. On the last day of serum starving, 12.5 µM tamoxifen, 2 ng/ml TGF-β or both were added to the culture medium. We chose to use tamoxifen because it is an US Food and Drug Administration-approved drug that blocks members of the stretch-activated Cl channel family, of which ICl_LPA is a member [16].

Electrophysiology

The amphotericin B whole-cell perforated patch technique was used to patch clamp fibroblasts [19]. A patch-clamp amplifier (model 200A, Axon Instruments, Burlingame, CA, USA) and pClamp 8.2 software (Axon Instruments) were utilized to record and analyse data. Fibroblasts were clamped at a holding voltage of 0 mV and stepped to depolarized voltages, from −80 to 100 mV in 15-mV steps. Records were capacity compensated by the amplifier circuitry, sampled at 2 kHz, and filtered at 1 kHz. Current density, equal to the peak Cl⁻ current divided by the fibroblast capacitance, was calculated for all fibroblasts. ICl_LPA activation was recorded after adding 10 µM LPA to the NaCl Ringer’s solution bathing the cells in the patch-clamp chamber. The pipette solution contained (mM): 2.5 NaCl, 2.5 CaCl₂, 145 KOH, 120 methanesulphonic acid (MeS), 5 HEPES and 240 mg/ml amphotericin B (Sigma). The bathing solution contained (mM): 145 NaCl, 2.5 CaCl₂, 5 KCl, 5 HEPES and 5 glucose.

Measurement of α-SMA expression

Western blotting was performed to examine α-SMA expression as previously described [13]. For this assay, we chose to examine cells cultured from the involved and uninvolved skin of the two patients from whom we were able to obtain both involved and uninvolved biopsies (SSc1 and SSc2), comparing these cells with those of the two controls (control 2 and control 6). Briefly, fibroblasts were gently scraped off culture plates using a cell lifter and suspended in cold PBS with protease inhibitors (Sigma). This mixture was centrifuged at 151 g. and then divided and put into two Eppendorf tubes that were centrifuged for an additional 5 min at 9660 g, 4°C. Fibroblasts from the first tube were lysed in RIPA lysis buffer (25 mM Tris–HCl, pH 7.6, 150 mM NaCl, 1% NP-40, 1% sodium deoxycholate and 0.1% SDS) plus protease inhibitors; and the bicinchoninic acid protein assay reagent (Pierce, Rockford, IL, USA) was used to determine protein concentration. Loading buffer (2×) was added to the second tube of cells, which was heated to 95°C for 10 min. This tube was used for the western blot analysis. Equal amounts of protein were loaded on each lane and separated by 10% SDS–polyacrylamide gel and transferred to nitrocellulose membranes. Membranes were blocked in 5% non-fat dry milk, and immunoblotting was performed using a monoclonal anti-α-SMA antibody (Sigma) and an horseradish peroxidase-conjugated secondary antibody (Sigma). Enhanced chemiluminescence (Perkin Elmer, Boston, MA, USA) was used for detection.

For loading controls, membranes were stripped and re-probed with β-actin antibody (CP01; Calbiochem, Gibbstown, NJ, USA). All western blots were digitally photographed, and blot density was determined using NIH Image J software (National Institutes of Health, Bethesda, MD, USA).

Proliferation assay

The CellTiter 96® AQ_UEOUS One Solution Cell Proliferation Assay (Promega, Madison, WI, USA) was used to measure cell proliferation. Fibroblasts from each patient were seeded in three individual wells in 24-well plates with regular growth medium at a density of 3 × 10⁴ cells/well. One day after seeding, cells from one plate were harvested using trypsin, spun down, re-suspended with 100 µl of serum-free medium plus 20 µl of CellTiter 96® AQ_UEOUS One Solution, and transferred to 96-well plates. The 96-well plates were incubated for 1–4 h at 37°C, 5% CO₂ and absorbance was read at 490 nm using a Bio-TEK synergy HT 96-well plate reader (Bio-Tek Instruments, Inc., Winooski, VT, USA). This procedure was repeated daily for another 3 days.

Statistical analyses

Differences in current density by patch clamp were analysed using ANOVA (analysis of variance) with P < 0.05 considered to be statistically significant. For the proliferation assays, mixed models (Proc Mixed) were performed for repeated measurements over time with subject (within group of SSc compared with control) as the random effect; the unstructured covariance model was chosen as the best fit. All statistical analyses were performed using the SAS System for Windows (version 9.1; SAS Institute, Cary, NC, USA).

Results

Cl⁻ currents induced by LPA

Figure 1A shows currents recorded from a representative fibroblast isolated and cultured from involved SSc skin. LPA (oleoyl; 10 µM) evoked an ICl_LPA current within 2–3 min after application, reaching a peak value 10–15 min after application. As we have previously shown for the ICl_LPA current [13], 12.5 µM tamoxifen blocked the current. Figure 1B shows similar results from a representative fibroblast isolated from uninvolved SSc skin. Figure 1C shows a representative fibroblast isolated from control skin, which rarely showed any ICl_LPA current activation. Current densities (peak current/capacitance) were calculated for controls, SSc-involved fibroblasts and SSc-uninvolved fibroblasts (Table 2). Current density values of unstimulated cells were not significantly different, whereas LPA stimulation resulted in significantly increased current density values in cells from both uninvolved and involved SSc skin as compared with controls.

Fig. 1.

Cl⁻ current in representative fibroblasts isolated from SSc subjects and a control subject. (A) LPA (10 μM) activated a large, tamoxifen-sensitive (12.5 μM) Cl⁻ current (ICl_LPA) in a representative fibroblast isolated and cultured from involved SSc skin. (B) LPA-stimulated ICl_LPA current in a representative SSc uninvolved skin fibroblast. (C) LPA-stimulated minimal ICl_LPA activity in a representative control skin fibroblast.

Table 2.

Scleroderma skin fibroblast patch-clamp results

	n	Before 10-µM LPA	After 10-µM LPA
Control	28	8.02 (0.68)	18.88 (1.82)
SSc (uninvolved)	26	9.61 (1.07)	31.02 (3.72)*
SSc (involved)	31	8.60 (0.88)	32.45 (4.68)*

Values are means (s.e.) of current density (pA/pF). *P < 0.05 compared with control.

α-SMA analysis

Western blotting was performed as a semi-quantitative assay to measure α-SMA expression at the protein level. The bar graph illustrates the ratios of individual band densities divided by the respective loading control densities. Figure 2 demonstrates that fibroblasts isolated from both involved and uninvolved SSc skin of two individual subjects showed elevated α-SMA protein levels as compared with the control fibroblasts. Elevated α-SMA expression in involved vs uninvolved cells was observed in one of these subjects.

Fig. 2.

Western blot results of α-SMA expression in SSc and control skin fibroblasts. In subject SSc1, uninvolved fibroblasts had an ∼25% increase in α-SMA protein levels as compared with involved fibroblasts from the same subject, whereas uninvolved fibroblasts from subject SSc2 had similar α-SMA protein levels compared with involved fibroblasts from that subject. Both involved and uninvolved SSc skin fibroblasts showed more α-SMA protein than control fibroblasts.

Figure 3 demonstrates the effects of TGF-β and tamoxifen on α-SMA protein expression, a marker for myofibroblast differentiation, in cells cultured from uninvolved (Fig. 3A) and involved SSc skin (Fig. 3B). Figure 3A shows that control fibroblasts grown with no TGF-β have minimal α-SMA protein expression, while TGF-β-treated control fibroblasts have a >9-fold increase in α-SMA protein expression. This TGF-β-induced α-SMA expression is almost completely inhibited by including tamoxifen in the culture medium. Interestingly, Fig. 3A also shows >6-fold higher α-SMA protein expression in untreated fibroblasts cultured from uninvolved SSc skin compared with control fibroblasts. TGF-β treatment of the uninvolved fibroblasts resulted in a 40% increase in α-SMA protein expression which was inhibited by 35% in the presence of tamoxifen. This tamoxifen-induced inhibition of α-SMA expression in uninvolved cells was not as large as that seen in control fibroblasts. Figure 3B shows similar results in control and involved SSc skin fibroblasts, although tamoxifen only reduced α-SMA expression by 8% in cells grown with no TGF-β. Similar unstimulated and stimulated α-SMA expression was observed in cells tested from other SSc and control cells (data not shown).

Fig. 3.

Western blot results of α-SMA expression in SSc and control skin fibroblasts treated with TGF-β, tamoxifen or both. Cells were cultured in regular medium grown until 80% confluence, and then serum starved for 3 days. On the last day of serum starvation, 2 ng/ml TGF-β, 12.5 µM tamoxifen or both were added to the culture medium. (A) Control fibroblasts grown with no TGF-β have minimal α-SMA protein expression, whereas TGF-β-treated control fibroblasts have >9-fold increase in α-SMA protein, which was almost completely inhibited by tamoxifen. Untreated uninvolved SSc skin fibroblasts had >6-fold higher α-SMA protein expression compared with untreated control fibroblasts. TGF-β induced a 40% increase in expression, which was reduced by 35% in the presence of tamoxifen. (B) Fibroblasts from involved SSc skin of the same subject in (A) show similar properties to the uninvolved fibroblasts in (A). Dark bars are tamoxifen treated cells and light bars have no tamoxifen. Bars to the left are from control cells and to the right from SSc involved (A) and uninvolved cells (B).

Proliferation assay

We measured cell numbers 24, 48, 72 and 96 h after plating and found no significant differences between fibroblasts isolated from SSc patient skin vs control fibroblasts (data not shown).

Discussion

Our group previously discovered that corneal keratocytes isolated from wounded rabbit corneas or cultured in the presence of serum contain a volume-regulated anion current that is also activated by the lysophospholipid growth factors LPA and SIP [16]. This channel was named ICl_LPA. We also determined that ICl_LPA activity plays a critical role in the differentiation of human lung fibroblasts to myofibroblasts and that this channel is a marker for the myofibroblast phenotype [13].

In the current study, we demonstrated that fibroblasts cultured from both involved and uninvolved skin of SSc patients have higher ICl_LPA current densities than those of healthy controls. In addition, our western blot results show that fibroblasts isolated from both involved and uninvolved SSc skin have higher unstimulated (no TGF-β) α-SMA protein levels than control fibroblasts.

Taken together, our patch-clamp data and α-SMA western data demonstrate that fibroblasts from both uninvolved and involved skin of SSc patients have myofibroblast characteristics, even though the uninvolved skin appears normal on physical exam. In agreement with our findings, Whitfield and his colleagues [20] reported that tissue from SSc patients’ skin that looks normal still has the distinctive genetic fingerprint of the disease. In contrast, a study from Mayes’ group found that skin cells isolated from the involved skin of SSc subjects had increased α-SMA immunostaining as compared with the controls and one group of cells taken from the uninvolved skin of an SSc subject [21]. While Fig. 2 in our study only shows α-SMA western data from uninvolved skin of two SSc subjects (paired with involved skin from the same subjects), we also observed elevated α-SMA levels by western blotting in a third uninvolved SSc skin culture as compared with controls (thus elevated α-SMA in three of three tested). Differences between our results and those of Mayes’ group could be due to the fact that they only looked at cells from a single uninvolved SSc source, or could be due to the higher sensitivity of western blotting vs immunostaining. It is also not clear whether involved and uninvolved skin fibroblasts from the same subject were examined for comparison in Mayes’ study.

LPA was recently found to be associated with renal fibrosis [22, 23] as well as liver [12, 24, 25] and pulmonary fibrosis [26, 27]. Patients with chronic renal failure have been shown to have elevated LPA concentrations in their plasma as compared with controls [28, 29]. Rho-associated kinases (ROCK), which are activated by LPA, have been found to stimulate myofibroblast differentiation from cultured human skin SSc fibroblasts [30]. Watanabe et al. [25] found that LPA plasma levels rise following CCl4-induced liver fibrosis in rats, and they also found elevated LPA levels in hepatitis C virus-induced liver fibrosis in humans [31]. Recently, Tager et al. [26] found significantly elevated LPA levels in bronchoalveolar lavage fluid (BAL) of bleomycin-treated mice. This group also determined that genetic ablation (LPA₁ receptor knockout mice) or pharmacological knockdown (Ki16425) of the LPA₁ receptor reduced bleomycin-induced pulmonary fibrosis. Furthermore, they found that BAL obtained from idiopathic pulmonary fibrosis patients induced chemotaxis of human fetal lung fibroblasts that was blocked by the LPA₁ receptor antagonist Ki16425 [26]. Further support for a major role for LPA in the pathogenesis of SSc comes from the realization that many of the derangements in immunity, inflammation and cardiovascular function in SSc patients could be attributable to excessive production of LPA from ongoing platelet activation and increased density and number of activated mast cells in lesional tissue in SSc [32–34]. In addition to LPA, S1P, a related member of the bioactive lysophospholipid family, has been associated with liver, eye, lung and cardiac fibrosis [35–38]. Importantly, we previously discovered that S1P also activates ICl_LPA in fibroblasts before their differentiation into myofibroblasts [13] and that S1P levels, as well as 20 : 4 LPA, are elevated in the serum of SSc patients [39]. Given our current findings, along with these previous studies and additional findings from our lab demonstrating ICl_LPA activity in myofibroblasts from lung [13, 17], cornea [17], skin, heart and liver (Yin and Watsky 2007, data not published) myofibroblasts, we conclude that elevated ICl_LPA activity is a hallmark of SSc skin fibroblasts and that LPA or S1P may be involved in SSc skin fibrosis.

In our study, tamoxifen inhibited the ICl_LPA current and partially inhibited α-SMA expression, with the maximum inhibition being in the TGF-β-stimulated control cells. We believe that this is because the majority of SSc cells were already in the myofibroblast phenotype, and tamoxifen inhibition of ICl_LPA activity will mainly affect α-SMA expression in cells undergoing fibroblast to myofibroblast differentiation and not those already in the myofibroblast phenotype.

It has been reported that some SSc patients have relatively low wound healing progression [40, 41]. Wound healing can be affected by many factors, including cell proliferation [42]. While we did see extremely low proliferation numbers in involved skin fibroblasts from one SSc subject, our results did not indicate any overall significant differences in proliferation between any of the groups we examined.

In conclusion, our results support the hypothesis that elevated ICl_LPA activity is a hallmark of SSc skin fibroblasts and that fibroblast to myofibroblast differentiation is involved in SSc skin fibrosis. We also conclude that fibroblasts isolated and cultured from uninvolved skin have many of the same characteristics as fibroblasts from involved skin. These current results, taken together with our previous work demonstrating the requirement of ICl_LPA activity for fibroblast to myofibroblast differentiation, allow us to conclude that ICl_LPA activity precludes fibroblast to myofibroblast differentiation in SSc skin, and suggests that blocking ICl_LPA activation may have potential importance as a treatment for SSc-associated fibrosis.