Autotaxin/lysophosphatidic acid signaling mediates obesity‐related cardiomyopathy in mice and human subjects

Abstract Obesity is associated with increased cardiovascular morbidity and mortality, but the direct signals to initiate or exaggerate cardiomyopathy remain largely unknown. Present study aims to explore the pathophysiological role of autotaxin/lysophosphatidic acid (LPA) in the process of cardiomyopathy during obesity. Through utilizing mouse model and clinical samples, present study investigates the therapeutic benefits of autotaxin inhibitor and clinical correlation to obesity‐related cardiomyopathy. The elevated circulating levels of autotaxin are closely associated with cardiac parameters in mice. Administration with autotaxin inhibitor, PF‐8380 effectively attenuates high fat diet‐induced cardiac hypertrophy, dysfunction and inflammatory response. Consistently, autotaxin inhibition also decreases circulating LPA levels in obese mice. In in vitro study, LPA directly initiates cell size enlargement and inflammation in neonatal cardiomyocytes. More importantly, circulating levels of autotaxin are positively correlated with cardiac dysfunction and hypertrophy in 55 patients. In conclusion, present study uncovers the correlation between circulating autotaxin and cardiac parameters in mice and human patient, and provided solid evidence of the therapeutic application of autotaxin inhibitor in combating obesity‐related cardiomyopathy.

cardiomyopathy is frequently observed in obese patients. 6 Clinical studies show increased fatty acid uptake and oxidation damages the cardiac function in patients. 7 The accumulation of specific lipid intermediates affects a set of biological process, including mitochondrial function, metabolism and cardiac structure. 8 Besides other stimuli including natriuretic peptides, non-coding RNAs and inflammatory cytokines also contributes to the development of cardiac hypertrophy. [9][10][11] However, how the accumulation of cellular signals to initiate or exaggerate cardiomyocyte hypertrophy remains largely unknown.
Autotaxin is a secreted lysophospholipase D catalyzing the hydrolysis of lysophosphatidylcholine to lysophosphatidic acid (LPA), a pleiotropic growth factor-like phospholipid. 12 In humans, increased mRNA expression of autotaxin in fat and elevated circulating autotaxin has also been observed in obese patients with insulin resistance and glucose intolerance compared to those with normal glucose tolerance. 13 Furthermore, circulating levels of autotaxin correlate positively with insulin resistance in older humans, and are an independent predictor for non-alcoholic fatty liver disease. 14 Several insulin resistance-inducing factors, including interleukin-6 (IL-6), glucose and insulin, induce the expression and secretion of autotaxin in inflammatory status. 15 Consistently, the mRNA expression of fat autotaxin is significantly increased of both dietary and genetic obese mice, which is accompanied by increased circulating level of LPA. 16 LPA has numerous effects in almost all cell types through its G protein-coupled receptors, which plays a decisive role in the development of several diseases. 17 More importantly, a recent study identifies that autotaxin and LPA are involved in the pathobiology of cardiovascular disease. 18 LPA induces cardiomyocyte hypertrophy via stimulating the activation of Rho-mediated signals. 19 However, whether autotaxin/LPA signalling is involved in the process of obesity-related cardiomyopathy is still unknown.

| Animal experiment
The experimental procedure described here was approved by the Institutional Animal Use and Care Committee at the Wenzhou Medical University (Wenzhou, China). In brief, 6-week aged male C57BL/ 6J mice weighing~20 g were randomly assigned to normal diet (NC) with 12% of calories from fat and high fat diet (HFD, Research diets), providing 60% of calories for 8 weeks. Then the mice were randomly divided into four groups: (a) NC-fed mice that received 0.5% methylcellulose (NC+Vehicle group); (b) NC-fed mice that received PF-8380 (30 mg/kg/day) (NC+PF-8380 group); (c) HFD-fed mice that received 0.5% methylcellulose (HFD+Vehicle group); (d) HFD-fed that received PF-8380 (30 mg/kg/day) (HFD+PF-8380 group). These mice were fed STC or HFD for another 8 weeks. The serum and cardiac tissues were collected and stored in −80°C refrigerator before further analysis.

| Cardiac histological analysis
Mouse hearts were fixed in 4% paraformaldehyde for 24-hour and embedded in paraffin. 5 μm paraffin sections were prepared and stained with hematoxylin and eosin solution. To measure the histological changes, the cardiac images were observed under a light microscope (Nikon, Tokyo, Japan).

| Cell experiment
Primary mouse cardiomyocytes were isolated from C57BL/6J mice as previously described. 20

| Western blot analysis
Protein extracts (50 μg) from the mouse hearts were boiled for 10 minutes in Laemmli sample buffer and then run on 10% SDS-PAGE. The protein was then transferred to a polyvinylidene difluoride membrane (Amersham Biosciences). The membrane was blocked for 1-hour at room temperature with 10% bovine serum albumin in phosphate-buffered saline/0.05% Tween 20. The blots were incubated overnight at 4°C with anti-phosp-IκB, anti-IκB or anti-Tubulin antibody and secondary antibody (Cell Signaling). The protein expression was visualized using enhanced chemiluminescence reagents (Bio-Rad, Hercules, CA, USA). The amounts of the proteins were analyzed using Image J analysis software version 1.38e.

| ATX activity assay
Fluorescence FS-3 was used as substrate to measure ATX activity.

| Human study
We have recruited 55 individuals undergoing cardiac ultrasound measurement at the First Affiliated Hospital of Weznhou Medical University. Exclusion criteria of this study included: patients younger than 18 years or older than 80 years, with known structural heart diseases, congestive heart failure, coronary heart disease, moderate to severe valvular disease, sepsis, electrolyte imbalance, chronic obstructive pulmonary disease, history of liver or renal disease, malignancy, subclinical hyperthyroidism, history of drug abuse or pregnancy. Written informed consent was obtained from all participants and all the procedures were approved by human ethics committee of Wenzhou Medical University.

| Statistical analysis
Data were presented as mean ± SD. The Student's t test was used for comparing two groups or one-way ANOVA was used for comparing four groups. GraphPad Prism 5 (GraphPad, San Diego, CA, USA) was used to analyze the statistical significance between sets of data.
Differences were considered to be significant at P < 0.05.   22 Here, our real-time PCR results also found adipose tissue had higher mRNA level of autotaxin in both lean and obese mice ( Figure S1).
Atrial natriuretic peptide and BNP, two important secreted factors, are key characteristics of hypertrophic cardiomyopathy. 23 Therefore, we correlated the levels of autotaxin with these 2 cardiac parameters. As Figure 1B and C showed, circulating autotaxin levels were positively associated with serum ANP (r = 0.7309, P < 0.01) or BNP (r = 0.7531, P < 0.01) in mice. These results supported the possible links between autotaxin and obesity-related cardiomyopathy, especially cardiac hypertrophy.

| Autotaxin inhibitor PF-8380 attenuates HFDinduced cardiac hypertrophy and function
In order to address the role of autotaxin in the obesity-related cardiomyopathy, we next utilized PF-8380, a specific inhibitor of autotaxin, 24 to explore the cardiac changes after suppression of autotaxin activity. Administration of PF-8380 could not affect HFDinduced body weight gain (Figure 2A). Then, we measured the ratio of heart weight/tibia length, which is a key marker of cardiac hypertrophy. 25 As Figure 2B showed, obese mice were significantly increased to 2.1-fold to NC-fed mice (P < 0.001), but PF-8380 effectively decreased this increment after 8-week treatment (P < 0.01).
The H&E staining showed PF-8380 significantly attenuated HFD- We next evaluated the cardiac performance of these mice using echocardiography. As Figure 3 showed, HFD obviously damaged cardiac function, including reduction of EF ( Figure 3B, P < 0.001) and FS ( Figure 3C, P < 0.05). However, administration of PF-8380 improved cardiac systolic function ( Figure 3B and C, P < 0.05).
Echocardiographic studies also revealed that autotaxin inhibition caused a significant reduction of HFD-induced left ventricular hypertrophy, including downregulation of systolic and diastolic dimension ( Figure 3D and E, P < 0.05). All these results supported autotaxin mediated diet-induced mouse cardiac hypertrophy and dysfunction.

| Autotaxin inhibitor PF-8380 inhibits cardiac inflammatory response in obese mice
Cardiac inflammation is a key characteristic of obesity-related cardiomyopathy, including increasing secretion of inflammatory cytokines and transcriptional activities. 26 As Figure 4A and C showed, obese mouse had remarkable induction of inflammatory factors, including TNF-α, IL-6 and iNOS (P < 0.01), but administration of PF-8380 effectively decreased these upregulation in obese mice (P < 0.05). On the other hand, PF-8380 also significantly decreased the levels of systemic inflammatory cytokines, including TNF-α, IL-6 and nitric oxide ( Figure S2, P < 0.05). Nuclear factor (NF)-κB, as a key transcriptional factor, controls inflammatory response in obesityrelated cardiomyopathy. 27 As Figure 5D and E showed, PF-8380 decreased HFD-induced IκB phosphorylation and degradation (P < 0.01), as compared with vehicle-treated obese mice. Taken together, it demonstrated that autotaxin also controlled dietmediated cardiac inflammation in obese mice.

| LPA induces cellular damages in primary mouse cardiomyocytes
Previous studies have demonstrated that LPA, produced by autotaxin, regulates monocytosis and promotes inflammation. 28 To this end, we firstly measured the activity of autotaxin, which controls the production of PLA. As showed in Figure 5A, HFD increased the autotaxin activity to 1.7-fold, as compared with NC-fed mice. On the contrast, PF-8380 significantly decreased the autotaxin activities in both NC-and HFD-fed mice (P < 0.001). Consequently, the serum LPA levels were also obviously downregulated in PF-8380-treated mice ( Figure 5B, P < 0.001).
Next, we aimed to investigate the pathophysiological role of LPA in cardiomyocytes. As showed in Figure 5C   Previous studies have demonstrated that autotaxin is mainly produced in response to inflammation. For example, the high level of autotaxin persists in association with inflammatory diseases such as arthritis and inflammatory bowel disease. 29 The production of inflammatory cytokines, such as TNF-α and IL-1β, in damaged and inflamed tissue is a signal for increased expression of autotaxin. 30 In monocytic THP-1 cells, bacterial lipopolysaccharide, a well-known initiator of the inflammatory response, induces the expression of autotaxin. 31 The activity of NF-κB, a transcriptional regulator of inflammatory cytokines, is closely associated with cellular autotaxin expression. 31 Consistently, the mRNA expression of fat autotaxin is significantly increased in both dietary and genetic obese mice. 16 Mechanistically, the upregulation of circulating insulin or glucose levels may induce autotaxin production. 32 However, whether autotaxin affects inflammatory response or activity is still unknown. Present study found autotaxin inhibitor PF-8380 suppressed obesityinduced cardiac inflammation, which supported autotaxin also played feed-back effects on inflammatory response.
Adipose tissues are major source of circulating autotaxin and other inflammatory cytokines in obese status. 22 Adipose-specific deletion of autotaxin causes approximately 40% reduction in circulating LPA, suggests that adipocyte is the major source of circulating autotaxin in mice. 22 Adipocyte-specific deletion of autotaxin protects against HFD-induced metabolic disorders in mouse model. 33 Similarly, there are already amounts of studies that support the links between heart and peripheral tissues. For example, Adiponectin, an adipose-secreted adipokine, mediates the modulation of hypertrophic signals in the heart. 34 Leptin, another adipocyte-derived 16-kD peptide, induces hypertrophy in neonatal rat ventricular myocytes. 35 In our study, specific autotaxin inhibitor effectively improved HFDinduced cardiac hypertrophy, dysfunction and inflammation in mice.
More importantly, circulating levels of autotaxin were significantly correlated with parameters of cardiomyopathy in mice and human patients. All these results support that there is a close crosstalk between adipose tissues and heart, and autotaxin may participate in this tissue interaction.
However, although current study has not addressed the direct or indirect effects of adipose tissue-derived autotaxin on cardiomyocyte function, our results at least supported circulating autotaxin were closely associated with diet-induced cardiomyopathy. In future, we will investigate the cardiac changes in specific adipose tissue autotaxin deficient mouse model or mice with fat transplantation, which will solid disclose the tissue interaction.
The pathophysiological roles of autotaxin/LPA signalling in cardiovascular disease have rarely been elucidated now. Recent study shows transported autotaxin and synthesized LPA promotes inflammation and mineralization of the aortic valve. 18 In in vitro study, LPA stimulates cardiomyocyte hypertrophy, including enlargement of cell size, induction of hypertrophic genes and signalling. 19 LPA also induces cardiomyocyte hypertrophy through stimulating the expression of microRNA-23a. 36 More importantly, LPA significantly stimulates the activity of human BNP promoter, 37 which determines the cardiac hypertrophy. 23 These previous findings provide a clue that autotaxin/LPA may be involved in the process of cardiac hypertrophy. Therefore, present study investigated whether autotaxin/LPA signalling is a potential mediator in obesity-related cardiac hypertrophy. Interestingly, both mouse and human studies found the close correlation between circulating levels of autotaxin and cardiac parameters. Administration with autotaxin inhibitor effectively F I G U R E 6 Circulating autotaxin level is closely correlated with cardiac function. Plasma from 55 individuals undergoing cardiac ultrasound measurement were collected and subjected to analysis. Correlation between circulating miR-29a level and cardiac EF% (A) and plasma NT-proBNP level (B). Correlation was assessed by non-parametric Spearman's test improved HFD-induced cardiac dysfunction, hypertrophy and inflammation in obese mice. The upregulated LPA also directly cardiomyocyte inflammation and hypertrophy in vitro. These findings indicate autotaxin/LPA signalling indeed participates in the process of cardiomyopathy in obese condition.
Present study utilized PF-8380, a specific autotaxin inhibitor, to explore the pathophysiological role of autotaxin/LPA signalling in obesity-related cardiomyopathy. More importantly, this finding also supported the potential therapeutic benefits of PF-8380 in protecting against cardiomyopathy. In the past decade, dozens of autotaxin inhibitors have been developed, some of which are now in phase-II clinical trials for treatment of idiopathic pulmonary fibrosis and breast cancer. 38 For example, ONO-8430506, a tetrahydrocarboline derivative with an IC50 of 5 nmol/L for inhibition of plasma autotaxin activity can delay initial tumour growth by 60% in Balb/c mice with orthotopic 4T1 breast tumours and markedly reduced the subsequent lung and liver metastasis. 39 Therefore, our study suggested autotaxin/LPA signalling was also a potential target for improving cardiomyopathy in obese condition.
In conclusion, our study uncovered the correlation between circulating autotaxin and cardiac parameters in mice and human patients, and provided solid evidence of the therapeutic application of autotaxin inhibitor in combating obesity-mediated cardiomyopathy.

CONF LICT OF I NTEREST
The authors declare that they have no competing interests.