Exploring metabolism in scleroderma reveals opportunities for pharmacological intervention for therapy in fibrosis

Abstract

Background: Recent evidence has indicated that alterations in energy metabolism play a critical role in the pathogenesis of fibrotic diseases. Studies have suggested that 'metabolic reprogramming' involving the glycolysis and oxidative phosphorylation (OXPHOS) in cells lead to an enhanced generation of energy and biosynthesis. The aim of this study was to assess the molecular basis of changes in fibrotic metabolism in systemic sclerosis (Scleroderma; SSc) and highlight the most appropriate targets for anti-fibrotic therapies.

Materials and methods: Dermal fibroblasts were isolated from five SSc patients and five healthy donors. Cells were cultured in medium with/without TGF-β1 and with/without ALK5, pan-PIM or ATM kinase inhibitors. Extracellular flux analyses were performed to evaluate glycolytic and mitochondrial respiratory function. The mitochondrial network in TMRM-stained cells was visualized by confocal laser-scanning microscopy, followed by semi-automatic analysis on the ImageJ platform. Protein expression of ECM and fibroblast components, glycolytic enzymes, subunits of the five OXPHOS complexes, and dynamin-related GTPases and receptors involved in mitochondrial fission/fusion were assessed by western blotting.

Results: Enhanced mitochondrial respiration coupled to ATP production was observed in SSc fibroblasts at the expense of spare respiratory capacity. Although no difference was found in glycolysis when comparing SSc with healthy control fibroblasts, levels of phophofructokinase-1 isoform PFKM were significantly lower in SSc fibroblasts (P<0.05). Our results suggest that the number of respirasomes is decreased in the SSc mitochondria; however, the organelles formed a hyperfused network, which is thought to increase mitochondrial ATP production through complementation. The increased mitochondrial fusion correlated with a change in expression levels of regulators of mitochondrial morphology, including decreased levels of DRP1, increased levels of MIEF2 and changes in OPA1 isoform ratios. TGF-β1 treatment strongly stimulated glycolysis and mitochondrial respiration and induced the expression of fibrotic markers. The pan-PIM kinase inhibitor had no effect, whereas both ALK5 and ATM kinase inhibition abrogated TGF-β1-mediated fibroblast activation, and upregulation of glycolysis and respiration.

Conclusions: Our data provide evidence for a novel mechanism(s) by which SSc fibroblasts exhibit altered metabolic programs and highlight changes in respiration and dysregulated mitochondrial morphology and function, which can be selectively targeted by small molecule kinase inhibitors.

Keywords: fibrosis; glycolysis; kinase inhibitors; mitochondrial morphology; mitochondrial respiration; myofibroblasts; oxidative phosphorylation; systemic sclerosis.

Figure 1

The fibrotic markers α-SMA and COL-1 are overexpressed in SSc fibroblasts. (A) Western blots with samples from five control, and five paired uninvolved and lesional SSc fibroblast cultures probed with antibodies against α-SMA, COL-1 and β-tubulin. Migration of protein standards is indicated. (B, C) Mean α-SMA and COL-1 signals, normalized for the β-tubulin signal and expressed relative to the mean value of the control samples. Data points show relative values of the individual samples. Error bars indicate standard deviations. Asterisks denote statistically significant differences (*P<0.05, **P<0.01).

Figure 2

Mitochondrial OXPHOS is increased in SSc fibroblasts but glycolysis is unaffected. (A) Extracellular acidification rate (ECAR) profiles in representative glycolysis stress tests of a control, an uninvolved SSc and a corresponding lesional SSc fibroblast culture. Glucose, oligomycin A and 2-deoxyglucose were sequentially added to dissect glycolytic function. Error bars indicate standard deviation of technical triplicates. (B) Mean basal glycolysis, glycolytic capacity and glycolytic reserve of five control and five paired uninvolved and lesional SSc fibroblast cultures. Error bars indicate standard deviation. (C) Oxygen consumption rate (OCR) profiles in representative mitostress tests of a control, an uninvolved SSc and a corresponding lesional SSc fibroblast culture. Oligomycin A, FCCP and rotenone + antimycin A were sequentially added to dissect mitochondrial respiratory function. Error bars indicate standard deviation of technical triplicates. (D) Mean basal respiration, maximal respiration, spare respiratory capacity and respiration coupled to ATP production of five control and five paired uninvolved and lesional SSc fibroblast cultures. Error bars indicate standard deviation. Asterisks denote statistically significant differences (*P<0.05, **P<0.01).

Figure 3

Muscle-type phosphofructokinase (PFKM) protein expression is decreased in SSc fibroblasts but the expression of other key enzymes of the glycolytic pathway and lactate secretion are unaffected. (A) Western blots with samples from five control, and five paired uninvolved and lesional SSc fibroblast cultures probed with antibodies against enzymes of the glycolytic pathway and lactate secretion as indicated. Blots were reprobed with an antibody against β-tubulin to confirm even loading. Migration of protein standards is shown. (B−P) Mean signals of the indicated proteins or protein ratios, normalized for the β-tubulin signal and expressed relative to the mean value of the control samples. Data points show relative values of the individual samples. Error bars indicate standard deviations. Asterisks denote statistically significant differences (*P<0.05).

Figure 4

Subunits of OXPHOS enzyme complexes I, III and IV show decreased expression in SSc fibroblasts but subunits of OXPHOS enzyme complexes II and V, and citrate synthase activity are unaffected. (A) Western blots with samples from five control, and five paired uninvolved and lesional SSc fibroblast cultures probed with antibodies against subunits of the OXPHOS complexes as indicated. Blots were reprobed with an antibody against β-tubulin to confirm even loading. Migration of protein standards is shown. (B−F) Mean signals of the indicated proteins, normalized for the β-tubulin signal and expressed relative to the mean value of the control samples. Data points show relative values of the individual samples. (G) Mean citrate synthase activity of five control, and five paired uninvolved and lesional SSc fibroblast cultures. Data points show values of the individual samples. Error bars indicate standard deviations. Asterisks denote statistically significant differences (*P <0.05, **P<0.01).

Figure 5

Cultured SSc fibroblasts have hyperfused mitochondrial networks. (A) Representative fluorescent micrographs of TMRM-stained control, uninvolved SSc and corresponding lesional SSc fibroblasts. (B) Mean length of mitochondrial rods and branches in arbitrary units (a.u.), and (C) mean number of branches per mitochondrial network in fibroblasts of five control and five paired uninvolved and lesional SSc cultures. Data point represent the mean length or the number of branches per network in 17−58 cells of the 15 individual cultures. Error bars indicate standard deviations. Asterisks denote statistically significant differences (*P <0.05, **P<0.01).

Figure 6

Expression of proteins engaged in mitochondrial fission and fusion shift in SSc fibroblasts to promote fusion. (A) Western blots with samples from five control, and five paired uninvolved and lesional SSc fibroblast cultures probed with antibodies against proteins involved in mitochondrial fission and fusion as indicated. Blots were reprobed with an antibody against β-tubulin to confirm even loading. Migration of protein standards is shown. (B−I) Mean signals of the indicated proteins, normalized for the β-tubulin signal and expressed relative to the mean value of the control samples or given as a ratio. Data points show relative values of the individual samples. Error bars indicate standard deviations. Asterisks denote statistically significant differences (*P<0.05, **P<0.01).

Figure 7

The kinase inhibitors SB431542 and KU55933, but not AZD1208, prevent TGF-β1-induced overexpression of α-SMA and ECM proteins in lesional SSc and control fibroblasts. (A−C) Western blots with samples from a lesional SSc and a control fibroblast culture treated vehicle or 2 ng/ml of TGF-β1 and/or 10 μM SB431542, 20 μM AZD1208 or 5 μM KU55933, and probed with antibodies against α-SMA or ECM proteins as indicated. Blots were reprobed with an antibody against β-tubulin to verify even loading.

Figure 8

The kinase inhibitors SB431542 and KU55933, but not AZD1208, prevent TGF-β1-induced increases in glycolysis and mitochondrial respiration in lesional SSc fibroblasts. (A) Extracellular acidification rates (ECAR) in representative glycolysis stress tests of a lesional SSc fibroblast culture treated vehicle or 2 ng/ml of TGF-β1 with or without 10 μM SB431542, 20 μM AZD1208 or 5 μM KU55933. Glucose, oligomycin A and 2-deoxyglucose were sequentially added to delineate glycolytic function. Error bars indicate standard deviation of technical triplicates. (B) Mean basal glycolysis, glycolytic capacity and glycolytic reserve of four lesional SSc fibroblast cultures treated vehicle or 2 ng/ml of TGF-β1 with or without 10 μM SB431542, 20 μM AZD1208 or 5 μM KU55933. Error bars indicate standard deviation. (C) Oxygen consumption rates (OCR) in representative mitostress tests of a lesional SSc fibroblast culture treated vehicle or 2 ng/ml of TGF-β1 with or without 10 μM SB431542, 20 μM AZD1208 or 5 μM KU55933. Oligomycin A, FCCP and rotenone + antimycin A were sequentially added to delineate mitochondrial respiratory function. Error bars indicate standard deviation of technical triplicates. (D) Mean basal respiration, maximal respiration, spare respiratory capacity and respiration coupled to ATP production of four lesional SSc fibroblast cultures treated vehicle or 2 ng/ml of TGF-β1 with or without 10 μM SB431542, 20 μM AZD1208 or 5 μM KU55933. Error bars indicate standard deviation. Asterisks denote statistically significant differences (*P<0.05).