Angiogenesis

Because angptl2 belongs to the angiopoietin family, it is not surprising that its first identified property concerned angiogenesis. Angptl2 was reported to induce endothelial cells sprouting, demonstrating for the first time its pro-angiogenic properties (Ref. Reference Kim1). These in vitro studies were later confirmed in vivo: in transgenic mice overexpressing angptl2 under the control of the keratinocyte-specific promoter K14 in the epidermis of skin tissues (K14- angptl2 transgenic mice), an increase in the number of blood vessels was observed (Ref. Reference Hato, Tabata and Oike26), suggesting that angptl2 promotes angiogenesis in vivo. The same team, however, reported that in K14-angptl2 transgenic mice, this angiogenic effect was less important than the pro-inflammatory effect of angptl2 (Refs Reference Tabata3, Reference Oike and Tabata12). In zebrafish, knockdown of both angptl1 and angptl2 with antisense produced severe vascular defects due to increased apoptosis of endothelial cells at the sprouting stage, and the antiapoptotic capacities of angptl2 were further confirmed in vitro (Ref. Reference Kubota9). In addition, it has been suggested that angptl2 could interfere with the apoptotic pathway via the prolyl-4-hydrolase domain 3 (PDH3), a pro-apoptotic protein that decreases angptl2 expression in macrophages (Ref. Reference Swain49). Very recently, angptl2 (through its coiled-coil domain) was shown to positively regulate vasculogenesis by promoting cell migration in isolated human cord blood derived endothelial colony forming cells, with no effect on either cell proliferation or apoptosis (Ref. Reference Richardson16). Angptl2 is acknowledged as an angiogenesis-related gene in various contexts such as pregnancy (Ref. Reference Mitko50), parturition (Ref. Reference Haddad51) and organogenesis (Ref. Reference Niki, Katsu and Yokouchi52). Recently, angptl2 (among other genes) was found to be up-regulated in the angiogenic process occurring after stroke in young rats (Ref. Reference Buga53). Angptl2 is also required for hematopoiesis: it has been questioned whether angptl2 could maintain the self-renewal of hematopoietic stem cells (Ref. Reference Oike, Yasunaga and Suda25) and demonstrated that angptl2 increased the survival and the expansion of bone marrow-derived hematopoietic stem cells (Refs Reference Zheng10, Reference Akhter27, Reference Zhang28, Reference Deng29, Reference Kubota30) and cord blood-derived hematopoietic stem cells (Ref. Reference Broxmeyer8). Accordingly, low angptl2 expression due to hypermethylation of angptl2 gene (among many other genes) was associated with a decrease in hematopoietic stem cell number and development of refractory anemia with excess blasts in myelodysplasic syndrome (Ref. Reference Lee54). The relationship between the angiogenic activity of angptl2 and its implication in the development of chronic inflammatory diseases is still not clear, with the exception of cancer (Refs Reference Odagiri7, Reference Aoi17, Reference Endo34, Reference Endo38). Altogether, these data suggest that angptl2 is a significant growth and survival factor.

Type 1A angiotensin II receptor recycling

Angptl2 has an important role in AT1A receptors recycling, but it is largely unrecognised because the team of Guo et al. (Refs Reference Guo31, Reference Guo32, Reference Guo33) that identified it did not use the name angptl2 but ARAP1, for angiotensin II receptor associated protein 1. More confusing is the fact that ARAP1 is also the abbreviated name for Centaurin-δ2 (Arf-GAP, ρ-GAP, ankyrin repeat and pleckstrin homology domain-containing proteins 1), which is entirely different from ARAP1 identified by Guo et al. (Ref. Reference Guo31). ARAP1 is therefore another name for angptl2, but it is only used in the field of angiotensin II receptors. Of note, ARAP1 is only one of several other AT1 receptor-associated proteins (Refs Reference Castrop55, Reference Higuchi56); for example, the protein ATRAP also modulates AT1 receptor surface expression, but it decreases the sensitivity of the target cell for angiotensin II, in contrast to ARAP1, which increases the sensitivity to angiotensin II (Refs Reference Castrop55, Reference Higuchi56). Two years after first identification of angptl2, employing yeast two-hybrid screening of a mouse embryo cDNA library with the carboxyl-terminal cytoplasmic domain of the murine AT1A receptor as bait, Guo et al. isolated a novel protein that they called ARAP1, that contained a coiled-coil domain and a fibrinogen-like domain, with 493 amino acid residues and a calculated molecular mass of 57 kDa (Refs Reference Guo32, Reference Guo33), and apparently identical to angptl2 (Ref. Reference Guo31). However, although angptl2 and ARAP1 are encoded by the same gene, this does not necessarily mean that the corresponding proteins are truly identical. Indeed, splicing of the gene and/or post-translational modifications may lead to different proteins. Nevertheless, Guo et al. demonstrated that ARAP1/angptl2 binds to the intracellular C-terminal domain of AT1A receptors in HEK-293 cells and that it specifically promotes AT1A recycling to the plasma membrane, with concomitant recovery of AT1A receptor signal functions (Refs Reference Guo32, Reference Guo33). Interestingly, renal proximal tubule specific overexpression of ARAP1 in mice leads to an increase in blood pressure, which is salt sensitive and can be reversed by perindopril (angiotensin converting enzyme inhibitor) or losartan (AT1 receptor antagonist), suggesting that renal ARAP1/angptl2 plays a role in the regulation of blood pressure via the renin-angiotensin system (Ref. Reference Guo31). Recently, ARAP1/angptl2 expression in the kidney was shown to be suppressed by angiotensin II, indicating that the renin-angiotensin system controls renal ARAP1/angptl2 expression (Ref. Reference Doblinger57). The down-regulation of renal ARAP1/angptl2 by angiotensin II, however, might be specific to the kidney, since its expression in the heart and aorta was not affected by angiotensin II (Ref. Reference Doblinger57). The implication of ARAP1/angptl2 in AT1A receptor recycling was also illustrated by the demonstration that ARAP1/angptl2 is down-regulated during sepsis, causing reduced renal sensitivity to angiotensin II and thus contributing to hypotension (Ref. Reference Mederle19). Because there is no clear consensus that ARAP1 is angptl2, more investigation is necessary to confirm the potential role of angptl2 in the regulation of blood pressure. Interestingly, while overexpression of ARAP1/angptl2 in renal proximal tubule leads to hypertension (Ref. Reference Guo31), angptl2^−/− mice are not hypotensive, likely due to a compensatory higher plasma renin concentration that could maintain a normal mean arterial blood pressure (Ref. Reference Mederle19). Hence, when searching in the literature databases, there are two separate worlds, that of ARAP1 (i.e. angptl2 in AT1A receptor function) and that of angptl2 (i.e. the role of angptl2 in general, except in the renin angiotensin system). Only one study combines the two worlds, although without specifying it, using ARAP1^−/− mice (Ref. Reference Mederle19), which are the angptl2^−/− mice developed by Dr Oike's group (Ref. Reference Kadomatsu24).

Circadian angptl2 expression

Interestingly, angptl2 expression shows pulsatility and a circadian rhythmicity (Ref. Reference Kitazawa58), and is strictly controlled and regulated by a complex molecular clock (CLOCK/BMAL1) (Ref. Reference Kadomatsu59). This periodicity has been proposed to be related to the physiological functions of angptl2, while a disruption in this rhythmicity may underlie chronic inflammation and diseases associated with high and sustained expression of angptl2 (Ref. Reference Kadomatsu59). Further studies are necessary to understand the exact impact of this circadian expression in the physiological functions of angptl2.

Angptl2 is antithrombotic

Very recently, it has been reported that angptl2 is highly expressed in human platelets and that angptl2 binding is partly dependent on LILRB2 (Ref. Reference Fan13), further confirming that LILRB2 is a receptor for angptl2. In addition, the authors demonstrated that recombinant purified angptl2 inhibited the aggregation of human platelets stimulated with various agonists, revealing for the first time that angptl2 possesses antithrombotic functions (Ref. Reference Fan13).

Cancer

Numerous studies reported a deleterious role for angptl2 in various types of cancer. Nevertheless, the first study published concerning the role of angptl2 in cancer suggested that angptl2 was a putative tumour suppressor (Ref. Reference Kikuchi60). Indeed, in some ovarian cancer cell lines and in primary ovarian cancer tissues, hypermethylation of angptl2 promoter was observed, leading to a loss-of-function that could contribute to abnormal cell growth and carcinogenesis (Ref. Reference Kikuchi60). In contrast, all the other studies proposed that angptl2 promotes or is associated with carcinogenesis and metastasis (Refs Reference Odagiri7, Reference Zheng10, Reference Aoi17, Reference Aoi18, Reference Endo34, Reference Sasaki35, Reference Szarc Vel Szic36, Reference Teicher37, Reference Endo38, Reference Gao39, Reference Kenney61, Reference Stockwin62). The first experimental evidence demonstrating that angptl2 contributes to both carcinogenesis and metastasis was provided by the team of Dr Oike in Japan, using a mouse model of chemically induced skin cancer in both K14-angptl2 transgenic mice and angptl2-knockout mice (Ref. Reference Aoi17). While overexpression of angptl2 per se is not sufficient to induce carcinogenesis, chemical treatment accelerated the formation and the number of skin papillomas in K14-angptl2 transgenic mice compared to wild-type mice (Ref. Reference Aoi17). Accordingly, papilloma formation was slowed and their number reduced in chemically treated angptl2^−/− mice (Ref. Reference Aoi17). In addition, the authors demonstrated that angptl2-associated chronic inflammation in skin cancer cells increased metastasis to lymph nodes and distant secondary organs, by promoting epithelial-to-mesenchymal transition and tumour angiogenesis and lymphoangiogenesis (Ref. Reference Aoi17). The critical role of angptl2 in driving metastasis was confirmed in human lung (Refs Reference Endo34, Reference Sasaki35), breast (Ref. Reference Endo38) and liver (Ref. Reference Gao39) cancer and in osteosarcoma cell lines (Ref. Reference Odagiri7). It has been proposed that angptl2 increased tumour cell metastasis through integrins α5β1, p38 mitogen activated protein kinase (MAPK) and matrix metalloproteinase (MMP)-9 activation (Ref. Reference Odagiri7). In addition, it was shown that in osteosarcoma cells, the expression of tolloid-like 1 protease, that could cleave angptl2 into inactive domain fragments, was weak, maintaining high levels of pro-inflammatory and metastatic angptl2 (Ref. Reference Odagiri7). The same team recently published that in their experimental model of skin cancer, angptl2 induced inflammation and oxidative stress, creating a microenvironment that favours methylation of gene coding for DNA repair enzymes (such as Msh2), thus lowering their expression and leading to DNA mutations and cancer (Ref. Reference Aoi18). Similarly, in human sun-exposed squamous carcinoma skin tissues, DNA repair enzyme Msh2 levels were inversely correlated with angptl2 expression (Ref. Reference Aoi18). In human with lung cancer, the tumour-derived angptl2 levels correlated with poor disease-free survival (Ref. Reference Endo34) and angptl2 mRNA expression correlated with metastasis (Ref. Reference Sasaki35). Similarly, demethylation of angptl2 gene promoter, leading to an increase in angptl2 expression, was observed in osteosarcoma cells (Ref. Reference Odagiri7). In sarcoma, up-regulated genes related to angiogenesis included angptl2 (Refs Reference Teicher37, Reference Kenney61, Reference Stockwin62). It has been suggested that the detection of higher angptl2 levels in blood samples may be useful for cancer patients (Ref. Reference Sasaki35), in particular in metastatic breast cancer patients (Ref. Reference Endo38), and serum angptl2 levels seem to have strong potential to predict diagnosis, recurrence and poor prognosis in human gastric and colorectal cancer (unpublished data from the team of Dr Kusunoki M., abstracts #009 and #417 from the 2014 Gastrointestinal Cancer Symposium). Finally, the binding of angptl2 to its LILRB2 receptor promotes leukaemia development, likely through inhibiting differentiation of acute myeloid leukaemia cells (Ref. Reference Zheng10). Paradoxically, angptl2 binding to LILRB2 also supports hematopoietic stem cells repopulation, maintaining a balance in normal stemness (Ref. Reference Zheng10). Altogether these data demonstrate the critical role for angptl2 in induction of cancer and metastasis, making angptl2 one of the multiple pro-inflammatory cancer hallmarks (Refs Reference Szarc Vel Szic36, Reference Endo38) and a therapeutic target.

Diabetes, obesity and obesity-related insulin resistance

In 2009, the team of Dr Oike published an outstanding paper demonstrating for the first time that angptl2 is a key adipocyte-derived inflammatory mediator, linking obesity to insulin resistance (Ref. Reference Tabata3). They proposed that in human, visceral fat might be one of the major sources of circulating angptl2. In human, circulating angptl2 levels were elevated in overweight or obese subjects (Refs Reference Tabata3, Reference Meng42, Reference Muramoto43) (Table 1), and correlated with inflammation (Ref. Reference Tabata3) and with the levels of insulin resistance (Refs Reference Tabata3, Reference Meng42). In addition, circulating levels of angptl2 decreased after a weight reduction in overweight men (Ref. Reference Muramoto43) or after 3 months antidiabetic treatment with pioglitazone in obese type 2 diabetic patients (Ref. Reference Tabata3). In mice, exposure to a high-fat diet increased the levels of circulating angptl2 (Ref. Reference Tabata3), indicating that angptl2 reflects adiposity (Ref. Reference Muramoto43). Adipose tissue secretes various pro-inflammatory mediators implicated in insulin resistance, and adipocyte-derived angptl2 seems to play a role in the pathogenesis of diabetes. In angptl2^−/− mice fed with a high-fat diet, insulin resistance and glucose tolerance were improved, although not normalised to levels seen in wild-type mice fed a normal chow diet; in contrast, in mice overexpressing angptl2 in adipose tissue under the control of aP2, an adipose tissue-specific promoter (aP2-angptl2 transgenic mice), inflammation and systemic insulin resistance were strong (Ref. Reference Tabata3). In addition, in liver and skeletal muscle cells from angptl2^−/− mice, insulin-signalling pathway was exacerbated, while in cells from aP2-angptl2 transgenic mice, this pathway was significantly diminished (Ref. Reference Tabata3). Angptl2 secreted by macrophages, endothelial cells and adipocytes was shown to induce local inflammation, to increase the vascular permeability and infiltration of macrophages (Refs Reference Tabata3, Reference Horio11). The data of Tabata et al. were the first to demonstrate the critical role of angptl2 in adipose tissue inflammation and systemic insulin resistance in mice (Ref. Reference Tabata3). The authors concluded that angptl2 probably influenced systemic insulin sensitivity by exacerbating adipose tissue inflammation (Ref. Reference Tabata3). This thesis of low-grade chronic inflammation is still accepted today (Ref. Reference Tateya, Kim and Tamori63). In contradiction with this conclusion, Kitazawa et al. reported that in diabetic mice treated with recombinant angptl2, insulin sensitivity, glucose tolerance and diabetes were improved, suggesting that angptl2 might be an anti-diabetic factor (Ref. Reference Kitazawa58). The discrepancies between the findings could reside in the distinct disease state, that is developing diabetes in the study of Tabata et al. and established diabetes in the study of Kitazawa et al.; angptl2 could play different roles at different stages of the pathogenesis of diabetes (Ref. Reference Kitazawa58). However, in a large study including 2164 Japanese volunteers without diabetes followed up for 7 years and an average of five times a year (the Hisayama Study), it was reported that the risk of developing diabetes was higher in the subjects in the highest serum angptl2 quartile (Table 1), demonstrating that serum angptl2 level is an independent risk factor for the development of type 2 diabetes (Ref. Reference Doi41). Accordingly, as circulating levels of angptl2 in diabetic patients correlated with albumin/creatinine ratio (Table 1), it has been proposed that levels of angptl2 could identify patients with diabetes and nephropathy at a very early point in its course (Ref. Reference Li5). This concords with the first observation of increased angptl2 levels in microvascular lesions of diabetic glomerulopathy (Ref. Reference Sun64). Altogether, these data suggest that pro-inflammatory angptl2 reflects adiposity and enhances insulin resistance, predicting the development of diabetes. The suppression of angptl2 has thus been proposed to be a strategy against obesity and obesity-related insulin resistance (Refs Reference Kadomatsu, Tabata and Oike4, Reference Oike and Tabata12).

Atherosclerosis and cardiovascular diseases

It has been said that angptl2 has the unique capacity of inducing an inflammatory response in blood vessels (Ref. Reference Kadomatsu, Tabata and Oike4), and endothelial dysfunction associated with inflammation is considered an early feature of atherosclerosis. We previously reported a significant 4-fold increase in angptl2 mRNA expression in endothelial cells isolated from arteries of active smokers with severe coronary artery disease (CAD) compared with non-smokers (Ref. Reference Farhat65). In addition, circulating levels of angptl2 were higher in Japanese (Refs Reference Tabata3, Reference Horio11) and Caucasian (Ref. Reference Farhat14) patients with CAD than in healthy subjects (Table 1), and angplt2 levels correlated with the severity of the disease (Ref. Reference Oike and Tabata12) (Table 1). Accordingly, it has been reported that plasma angptl2 levels were correlated with coronary artery stenosis in patients with CAD (Refs Reference Horio11, Reference Horio66) (Table 1) and that circulating levels of angptl2 increased with Framingham risk scores, suggesting that angptl2 concentrations could predict the development of cardiovascular disease (Ref. Reference Horio67). Very recently, the same team reported that in older subjects without cardiovascular disease history, circulating levels of angptl2 positively correlated with vascular inflammation status and with carotid intima-medial thickness and the presence of plaque, suggesting that angptl2 could predispose to future cardiovascular events (Ref. Reference Horio11). However, while it is established that angptl2 induces vascular inflammation, including in endothelial cells (Refs Reference Tabata3, Reference Horio11, Reference Farhat14), its contribution to the development of chronic endothelial/vascular inflammation leading to atherosclerosis is less understood. Only very recently, Horio et al. demonstrated that endothelial cell-derived angptl2 accelerated vascular inflammation by activating integrins α5β1/nuclear factor kappa B (NFκB) signalling and increasing macrophages infiltration, leading to endothelial dysfunction, atherogenesis and plaque instability (Ref. Reference Horio11). In their seminal study, Tabata et al. clearly demonstrated that angptl2 promoted an inflammatory response in cultured endothelial cells via the integrins α5β1/Rac1/NFκB pathway, stimulated monocytes adhesion and both endothelial cells and monocytes migration (Ref. Reference Tabata3). We also recently reported that recombinant angptl2 induced a pro-inflammatory response in freshly isolated endothelial cells and triggered leucocyte adhesion onto the native endothelium of pre-atherosclerotic severely dyslipidaemic mice (LDLr^−/−; hApoB100^+/+ mice), but not of healthy mice (Ref. Reference Farhat14). We also demonstrated that endogenous angptl2 was abundant in the blood and in the atherosclerotic lesion that spontaneously developed with age in these mice and that chronic administration of recombinant angptl2 to LDLr^−/−; hApoB100^+/+ mice strongly induced the expression of endothelial pro-inflammatory cytokines and adhesion molecules, and accelerated the formation of atherosclerotic plaques (Ref. Reference Farhat14). In addition, we reported that recombinant angptl2 reduced acetylcholine-mediated dilation of isolated mouse arteries; accordingly, in angptl2 knockdown mice, endothelial function was greater than in wild-type mice (Ref. Reference Yu23). Furthermore, after a 3-month high-fat diet, NO production was reduced in arteries isolated from high-fat diet-fed wild-type mice, whereas NO release was preserved in angptl2 knockdown mice (Ref. Reference Yu23). This suggests that angptl2 regulates endothelial function. Interestingly, although angptl2 is not acknowledged to have an impact on lipid metabolism, unlike angptl3 and angptl4 (Refs Reference Hato, Tabata and Oike26, Reference Santulli47, Reference Miida and Hirayama68), we observed that chronic administration of recombinant angptl2 to LDLr^−/−; hApoB100^+/+ mice also significantly increased total cholesterol and LDL-cholesterol levels, an effect that could synergistically contribute to the accelerated atherogenesis (Ref. Reference Farhat14). Accordingly, we observed that angptl2-knockdown mice were protected against high-fat diet-induced hypercholesterolemia and endothelial dysfunction (Refs Reference Yu22, Reference Yu23). On the other hand, in angptl2^−/− mice fed a high-fat diet compared with wild-type mice under the same diet, the lipid profile was not different (Refs Reference Tabata3, Reference Horio11) and the endothelial dysfunction associated with high expression of angptl2 in endothelial cells (Tie2-angptl2 transgenic mice) was independent from dyslipidaemia (Ref. Reference Horio11). Altogether, our data (Refs Reference Farhat14, Reference Yu22, Reference Yu23) and other reports (Refs Reference Tabata3, Reference Horio11) strongly suggest that angptl2 causally contributes to the pathogenesis of atherosclerosis through induction of vascular inflammation and endothelial dysfunction.

In addition to its implication in atherosclerosis, angptl2 has been reported to be involved in the progression of abdominal aortic aneurysm (Ref. Reference Tazume20) and in neointimal hyperplasia after endovascular injury (Refs Reference Tian21, Reference Shimizu and Walsh69). Angptl2-derived from either macrophages (Ref. Reference Tazume20) or from perivascular adipose tissues (Ref. Reference Tian21) induced vascular chronic inflammation by activating an NFκB-dependent cascade, leading to pathological vascular remodelling. Thus, angptl2 could be a new target in preventing and treating cardiovascular diseases (Refs Reference Horio11, Reference Oike and Tabata12, Reference Tazume20, Reference Tian21) and/or be a biomarker of endothelial dysfunction (Ref. Reference Santulli47), an early feature of atherosclerogenesis.

Kidney disease

Using the same large cohort to demonstrate that angptl2 was an independent risk factor for diabetes (Ref. Reference Doi41), it has been reported that elevated circulating angptl2 levels were also associated with chronic kidney disease in a general Japanese population (Ref. Reference Usui6) (Table 1). In the latter study, serum angptl2 were mainly related to albuminuria (Ref. Reference Usui6), which concords with a Chinese study where high levels of angptl2 were associated with parallel rise in albuminuria in patients with diabetic nephropathy (Ref. Reference Li5) (Table 1). Accordingly, angptl2 was increased in microvascular lesions of diabetic glomerulopathy (Refs Reference Sun64, Reference Dei Cas and Gnudi70). In mice, overexpression of angptl2 in renal proximal tubules led to hypertension and kidney hypertrophy (Ref. Reference Guo31). Angptl2 may itself contribute to kidney disease via local inflammation and local endothelial dysfunction (Ref. Reference Usui6).

Chronic inflammation in the eyes

Although there is numerous evidence showing that angptl2 mediates chronic inflammation, the first evidence reporting an acute pro-inflammatory effect of angptl2 was demonstrated in the eye (Ref. Reference Kanda71). The authors showed that angptl2 was ubiquitously expressed in the retina and increased by endotoxin treatment, leading to endotoxin-induced uveitis, associated with leucocyte adhesion to the retina vessels and leucocyte infiltration in the vitreous; all these inflammatory manifestations, which were mediated by the integrins α5β1/NFκB pathway, were significantly reduced in angptl2^−/− mice (Ref. Reference Kanda71). In addition, in a mouse model of suture-induced inflammatory corneal neovascularisation, up-regulation of angptl2 was shown to potently induce haemangiogenesis and lymphangiogenesis (Ref. Reference Toyono15). This in vivo corneal inflammation and angiogenesis were increased in K14-angptl2 transgenic mice and reduced in angptl2^−/− mice or after angptl2–siRNA subconjunctival treatment (Ref. Reference Toyono15). Unpublished data from Dr Oike's team show that the vitreous concentration of angptl2 is increased in patients with diabetic retinopathy and that angptl2 causes chronic inflammation in retinal endothelial cells (Ito Y. et al, abstract from the 2013 European Association for Vision and Eye Research Conference). All these data led to the conclusion that anti-angiogenic therapy targeting angptl2 might be useful for treating corneal transplant rejection, macular degeneration and proliferative diabetic retinopathy (Ref. Reference Ambati72).

Other chronic inflammatory diseases

In addition to its contribution in cancer, diabetes, atherosclerosis, kidney disease and corneal/retinal inflammation, angptl2 has been reported to be a key player in the chronic inflammation observed in rheumatoid arthritis (Ref. Reference Okada44), dermatomyositis (Ref. Reference Ogata45) and in the hypertrophy of the ligamentum flavum in patients with lumbar spinal canal stenosis (Ref. Reference Nakamura46). Angptl2 concentration in synovial fluid, but not in the serum (Table 1), was significantly higher in patients with rheumatoid arthritis compared with osteoarthritis (Ref. Reference Okada44). However, angptl2 circulating levels correlated with rheumatoid arthritis disease activity score, and angptl2 increased the chemotactic activity of rheumatoid arthritis synovial fluid-derived monocytes/macrophages and endothelial cells, promoting synovial inflammation (Ref. Reference Okada44). Similarly, an increase in angptl2 circulating levels was reported in patients with dermatomyositis (Table 1) and K14-angptl2 transgenic mice exhibited dermatomyositis-like manifestations and chronic inflammation in the keratinocytes through the NFκB cascade (Ref. Reference Ogata45). Finally, angptl2 may accelerate tissue fibrosis through activation of the transforming growth factor β (TGFβ)/Smad signalling cascade, leading to remodelling of elastin and collagen fibres and contributing to ligamentum flavum hypertrophy and degeneration (Ref. Reference Nakamura46). This was demonstrated using ligamentum flavum tissue, and isolated fibroblasts from patients with lumbar spinal canal stenosis exposed to mechanical stress and/or recombinant angptl2 (Ref. Reference Nakamura46). In these cells, mechanical stress-induced angptl2 expression and angptl2-siRNA prevented TGFβ expression (Ref. Reference Nakamura46). Because mechanical stress contributes to various diseases including heart diseases, and because angptl2 is largely expressed in cardiac tissues, the authors asked the pertinent question whether mechanical stress-induced angptl2 contributes to the progression of heart diseases (Ref. Reference Nakamura46), an unresolved issue.