Emerging Roles of Tumor Necrosis Factor-Stimulated Gene-6 in the Pathophysiology and Treatment of Atherosclerosis

Tumor necrosis factor-stimulated gene-6 (TSG-6) is a 35-kDa glycoprotein that has been shown to exert anti-inflammatory effects in experimental models of arthritis, acute myocardial infarction, and acute cerebral infarction. Several lines of evidence have shed light on the pathophysiological roles of TSG-6 in atherosclerosis. TSG-6 suppresses inflammatory responses of endothelial cells, neutrophils, and macrophages as well as macrophage foam cell formation and vascular smooth muscle cell (VSMC) migration and proliferation. Exogenous TSG-6 infusion and endogenous TSG-6 attenuation with a neutralizing antibody for four weeks retards and accelerates, respectively, the development of aortic atherosclerotic lesions in ApoE-deficient mice. TSG-6 also decreases the macrophage/VSMC ratio (a marker of plaque instability) and promotes collagen fibers in atheromatous plaques. In patients with coronary artery disease (CAD), plasma TSG-6 levels are increased and TSG-6 is abundantly expressed in the fibrous cap within coronary atheromatous plaques, indicating that TSG-6 increases to counteract the progression of atherosclerosis and stabilize the plaque. These findings indicate that endogenous TSG-6 enhancement and exogenous TSG-6 replacement treatments are expected to emerge as new lines of therapy against atherosclerosis and related CAD. Therefore, this review provides support for the clinical utility of TSG-6 in the diagnosis and treatment of atherosclerotic cardiovascular diseases.


Introduction
Atherosclerotic cardiovascular diseases, such as coronary artery disease (CAD) and stroke, are a leading cause of mortality worldwide [1]. The prevalence of traditional risk factors for CAD, such as diabetes, hypertension, dyslipidemia, obesity, and others, has been increasing worldwide, with consequent increases in the rates of coronary and cerebrovascular events [1]. Atherosclerosis is characterized by a complex interaction of vascular endothelial injury, inflammation with monocyte adhesion to endothelial cells (ECs), lipid deposition with macrophage foam cells, and the migration and proliferation of vascular smooth muscle cells (VSMCs) accompanied by extracellular matrix (ECM) remodeling [2].

Structure of TSG-6
TSG-6, which maps to human chromosome 2q23.3, was originally identified as the sixth gene product induced by tumor necrosis factor-α (TNF-α) in human fibroblasts [11], and is also called Tnfaip6 or Tnfip6 (accession numbers: CAD12353, CAD13434). TSG-6 cDNA encodes a polypeptide of 277 amino acids, including an N-terminal cleavable signal peptide of 17 amino acids. The molecular size of the mature, fully glycosylated TSG-6 protein is 31,203 Da. The N-terminal half of the TSG-6 protein shows 36-40% homology to the Link module, which is a conserved sequence in hyaluronan-binding proteins, such as CD44, cartilage link protein, and G1 domains of aggrecan and versican [12]. The C-terminal half of the molecule (CUB domain) shows 30% homology to the complement C1r A chain [12]. Human and murine TSG-6 are >94% identical [12]. The specific receptors for TSG- 6 have not yet been identified.

Expression and Regulation of TSG-6
TSG-6 is not usually expressed but is rapidly upregulated in many different cell types, including monocytes/macrophages, dendritic cells, leukocytes, ECs, VSMCs, fibroblasts, synoviocytes, chondrocytes, and proximal tubular epithelial cells upon exposure to inflammatory mediators, such as interleukin (IL)-1, interferon-γ, TNF-α, lipopolysaccharide (LPS), and prostaglandin E 2 [13][14][15] as well as growth factors, such as fibroblast growth factor, epidermal growth factor, and transforming growth factor-β, and mechanical stimuli in vitro [12,16]. Expression of TSG-6 is also known to be markedly upregulated by exposure to high glucose and fatty acid (palmitic acid) concentrations in human umbilical vein endothelial cells (HUVECs) [17]. In contrast, TSG-6 expression is suppressed by anti-inflammatory cytokines, such as IL-4 or IL-10 either directly or via inhibition of LPS/Toll-like receptor (TLR)-induced cell activation [15]. TSG-6 is released from the secretory granules of neutrophils and mast cells as well as macrophages and a wide variety of stromal cell types [12,15,[18][19][20]. Mesenchymal stem cells (MSCs) also secrete TSG-6 to repair tissue injury and wounds and reduce inflammation [21]. Several lines of evidence have shown that TSG-6 is detected in synovial fluids and joint tissues from patients with rheumatoid arthritis and osteoarthritis [13,22] and in sera of patients with bacterial sepsis, systemic lupus erythematosus, and CAD [10,23].

Roles of TSG-6
Based on structural homologies, TSG-6 binds to a large number of components of the ECM including hyaluronan, heparin, heparan sulfate, thrombospondins-1 and -2, fibronectin, and pentraxin-3 [24,25]. These interactions primarily act to stabilize or remodel the ECM [25]. Several lines of evidence have suggested that TSG-6 may play a crucial role in ECM formation, inflammatory cell migration, cell proliferation, and developmental processes [26]. TSG-6 is known as a potent inhibitor of neutrophil migration and can modulate the protease network by inhibiting plasmin [24]. TSG-6 also increases hyaluronan synthesis in airway smooth muscle cells [26]. TSG-6 has a chondroprotective effect, with reduced loss of cartilage proteoglycan and less accumulation of matrix metalloproteinase (MMP) and aggecanase-generated aggrecan fragments [27]. TSG-6 is locally coexpressed with its ligand pentraxin-3 that cooperates with TSG-6 in ECM assembly [28]. TSG-6 suppresses the inhibitory effects of pentraxin-3 on fibroblast growth factor-2-mediated angiogenesis [28]. TSG-6 has been shown to exert anti-inflammatory effects in experimental models of arthritis, corneal wounding, acute myocardial infarction, and acute cerebral infarction [29][30][31][32]. In transgenic mice, inactivation of the gene increases inflammatory responses [33], and over-expression of the gene decreases inflammatory responses [34]. Administration of recombinant TSG-6 decreases LPS-induced inflammation (IL-6 and interferon-γ), and improves arthritis and memory after traumatic brain injury in several murine models [25,29,35,36]. The half-life of TSG-6 after intravenous injection into mice is up to 0.2 h [25].
Several lines of evidence have recently shown the atheroprotective effects of TSG-6. TSG-6 exerts anti-atherosclerotic effects on all the three cellular players in the pathogenesis of atherosclerosis, such as ECs, macrophages, and VSMCs ( Figure 1). Although the receptors for TSG- 6 have not yet been clarified, it is almost certain that the receptors are present in three types of vascular cells. Next, the present review introduces the anti-atherosclerotic effects of TSG-6 on vascular cells in vitro, animal models in vivo, and human clinical data, in this order. mice, inactivation of the gene increases inflammatory responses [33], and over-expression of the gene decreases inflammatory responses [34]. Administration of recombinant TSG-6 decreases LPSinduced inflammation (IL-6 and interferon-γ), and improves arthritis and memory after traumatic brain injury in several murine models [25,29,35,36]. The half-life of TSG-6 after intravenous injection into mice is up to 0.2 h [25].
Several lines of evidence have recently shown the atheroprotective effects of TSG-6. TSG-6 exerts anti-atherosclerotic effects on all the three cellular players in the pathogenesis of atherosclerosis, such as ECs, macrophages, and VSMCs ( Figure 1). Although the receptors for TSG- 6 have not yet been clarified, it is almost certain that the receptors are present in three types of vascular cells. Next, the present review introduces the anti-atherosclerotic effects of TSG-6 on vascular cells in vitro, animal models in vivo, and human clinical data, in this order.

Effects of TSG-6 on Atherosclerotic Lesion Development in ApoE-Deficient Mice
Infusing TSG-6 (100 µg/mouse) into ApoE-deficient mice for four weeks retards the development of atherosclerotic lesions in the entire surface area of the aorta [10]. In atheromatous plaques in the aortic sinus wall, vascular inflammation (pentraxin-3) and monocyte/macrophage and VSMC contents are decreased ( Figure 2D,F,H) by TSG-6 infusion. The ratio of macrophage contents/VSMC contents (a surrogate marker of plaque instability) is reduced by TSG-6 infusion ( Figure 2I). TSG-6 also increases collagen fibers within atheromatous plaques [10]. These findings indicate that TSG-6 stabilizes atheromatous plaques. In addition, four-week infusion of anti-TSG-6 neutralizing antibody (50 µg/mouse) into ApoE-deficient mice significantly enhances the development of atherosclerotic lesions in the entire aortic surface area and plaque burden in the aortic sinus ( Figure 3A,B).
Four-week infusion of TSG-6 (100 µg/mouse) into ApoE-deficient mice decreases the inflammatory M1 phenotype and the levels of inflammasome proteins such as MCP-1, NF-κB, C-reactive protein, and apoptosis-associated speck-like protein containing a caspase recruitment domain in exudate peritoneal macrophages [10]. TSG-6 also decreases plasma levels of total cholesterol with a tendency to increase high-density lipoprotein cholesterol in ApoE-deficient mice [10]. , respectively. Hematoxylin was used for nuclear staining. Bar = 100 μm. The markers of plaque instability, such as high levels of pentraxin-3 expression and the increased ratio of macrophage contents (μm 2 )/VSMC contents (μm 2 ) within atheromatous plaques (I), were compared between 21week-old mice infused with saline (n = 13) and TSG-6 (n = 9). Data are presented as mean ± SEM. Unpaired Student's t test was used for statistical analysis. p = 0.1453. These results are unpublished data from our previous experiments [10]. Hematoxylin was used for nuclear staining. Bar = 100 µm. The markers of plaque instability, such as high levels of pentraxin-3 expression and the increased ratio of macrophage contents (µm 2 )/VSMC contents (µm 2 ) within atheromatous plaques (I), were compared between 21-week-old mice infused with saline (n = 13) and TSG-6 (n = 9). Data are presented as mean ± SEM. Unpaired Student's t test was used for statistical analysis. p = 0.1453. These results are unpublished data from our previous experiments [10].

Roles of TSG-6 in Animal Atherosclerosis and Vascular Restenosis in Wire-Injury and Vein Graft Models
Several animal studies have shown the expression of TSG-6 in atherosclerotic lesions in vivo [10,46,47]. TSG-6 is localized in rat neointima after arterial injury and rabbit carotid atherosclerotic plaques [46,47]. In our study, TSG-6 was expressed in atherosclerotic plaques in the aorta and neointimal lesions in the femoral artery after wire injury in ApoE-deficient mice ( Figure 4B,J). The expression of TSG-6 is consistent with vascular inflammation and macrophages in atheromatous plaques ( Figure 4B,D,E) and neointimal thickness (VSMCs) in obstructive arteries following injury ( Figure 4I,J).
TSG-6 inhibits the inflammatory response of transplanted vein grafts in rats and reduces vascular injury by downregulating the JNK and P38 pathways [45]. TSG-6 suppresses vascular restenosis in the venous bridge with decreasing VSMC proliferation, macrophage infiltration, and plasma levels of inflammatory cytokines, such as IL-1β, IL-6, and TNF-α, via JNK and p38 pathways [45].

Roles of TSG-6 in Animal Atherosclerosis and Vascular Restenosis in Wire-Injury and Vein Graft Models
Several animal studies have shown the expression of TSG-6 in atherosclerotic lesions in vivo [10,46,47]. TSG-6 is localized in rat neointima after arterial injury and rabbit carotid atherosclerotic plaques [46,47]. In our study, TSG-6 was expressed in atherosclerotic plaques in the aorta and neointimal lesions in the femoral artery after wire injury in ApoE-deficient mice ( Figure 4B,J). The expression of TSG-6 is consistent with vascular inflammation and macrophages in atheromatous plaques ( Figure 4B,D,E) and neointimal thickness (VSMCs) in obstructive arteries following injury ( Figure 4I,J).
TSG-6 inhibits the inflammatory response of transplanted vein grafts in rats and reduces vascular injury by downregulating the JNK and P38 pathways [45]. TSG-6 suppresses vascular restenosis in the venous bridge with decreasing VSMC proliferation, macrophage infiltration, and plasma levels of inflammatory cytokines, such as IL-1β, IL-6, and TNF-α, via JNK and p38 pathways [45].

Expression of TSG-6 in Human Arteriosclerotic Lesions and Aneurysms
Recent studies have shown the presence of TSG-6 expression in human coronary atherosclerotic plaques and abdominal aortic aneurysms [10,48]. The expression of TSG-6 is almost absent in normal and non-stenotic coronary arteries from non-CAD and CAD patients respectively [10]. In the stenotic coronary arteries from CAD patients, TSG-6 is highly expressed in the fibrous cap within atherosclerotic plaques [10]. In addition, the expression of TSG-6 is observed in the tunica media of human abdominal aortic aneurysms [48]. These findings suggest that TSG-6 is produced to counteract the progression of atherosclerotic lesions and to stabilize vulnerable plaques and aneurysms.

Potential Biomarker for CAD
Plasma TSG level was significantly higher in 135 patients with angiographically proven CAD (acute coronary syndrome) than the level in 47 non-CAD subjects [10] (Figure 5A). Based on the receiver operating characteristic (ROC) curve ( Figure 5B), 9.5 ng/mL is adopted as the cutoff value that shows a higher true-positive rate (sensitivity) with a low false-positive rate (1-specificity). The area under the curve (AUC) value is 0.68 ( Figure 4B). The sensitivity and specificity for detecting CAD using this cutoff value were 52% and 30%, respectively. Therefore, plasma TSG-6 level could be a reliable biomarker for detecting CAD.

Expression of TSG-6 in Human Arteriosclerotic Lesions and Aneurysms
Recent studies have shown the presence of TSG-6 expression in human coronary atherosclerotic plaques and abdominal aortic aneurysms [10,48]. The expression of TSG-6 is almost absent in normal and non-stenotic coronary arteries from non-CAD and CAD patients respectively [10]. In the stenotic coronary arteries from CAD patients, TSG-6 is highly expressed in the fibrous cap within atherosclerotic plaques [10]. In addition, the expression of TSG-6 is observed in the tunica media of human abdominal aortic aneurysms [48]. These findings suggest that TSG-6 is produced to counteract the progression of atherosclerotic lesions and to stabilize vulnerable plaques and aneurysms.

Potential Biomarker for CAD
Plasma TSG level was significantly higher in 135 patients with angiographically proven CAD (acute coronary syndrome) than the level in 47 non-CAD subjects [10] (Figure 5A). Based on the receiver operating characteristic (ROC) curve ( Figure 5B), 9.5 ng/mL is adopted as the cutoff value that shows a higher true-positive rate (sensitivity) with a low false-positive rate (1-specificity). The area under the curve (AUC) value is 0.68 ( Figure 4B). The sensitivity and specificity for detecting CAD using this cutoff value were 52% and 30%, respectively. Therefore, plasma TSG-6 level could be a reliable biomarker for detecting CAD. Among 135 patients with CAD, 32 patients had major adverse cardiovascular events (MACE), defined as cardiovascular death, heart failure, acute myocardial infarction, post-infarction angina, and ischemic stroke, during a period of four years. Plasma TSG-6 level at the onset of acute coronary syndrome tended to be increased in CAD patients with MACE compared with those without MACE, but there was no significant difference ( Figure 6A). Furthermore, the prevalence rate of MACE was significantly higher in CAD patients with TSG-6 ≥ 19 ng/mL (40%, 10/25 cases) than the rate in those with TSG-6 < 19 ng/mL (20%, 22/110 cases) ( Figure 6B). The finding suggests that high plasma levels of TSG-6 may induce downregulation of its receptor, leading to ineffective MACE prevention in CAD patients. Figure 6. Prediction of prognosis using plasma TSG-6 levels at onset of CAD. Among 135 patients with acute coronary syndrome, 32 patients had major adverse cardiovascular events (MACE) including cardiovascular death, heart failure, acute myocardial infarction, post-infarction angina, and ischemic stroke during a period of four years after onset. The follow-up period was 50.0 ± 0.57 months. (A) Plasma TSG-6 levels tended to increase in CAD patients with MACE compared with those without MACE. Data are presented as mean ± SEM. Unpaired Student's t test was used for statistical analysis. p = 0.6109; (B) Prevalence rate of MACE (+) was significantly higher in CAD patients with TSG-6 ≥ 19 ng/mL (40%, 10/25 cases) than those with TSG-6 < 19 ng/mL (20%, 22/110 cases). Chi-squared test was used for statistical analysis. * p = 0.0338. These results are unpublished data from our follow-up study based on our previous report [10]. Among 135 patients with CAD, 32 patients had major adverse cardiovascular events (MACE), defined as cardiovascular death, heart failure, acute myocardial infarction, post-infarction angina, and ischemic stroke, during a period of four years. Plasma TSG-6 level at the onset of acute coronary syndrome tended to be increased in CAD patients with MACE compared with those without MACE, but there was no significant difference ( Figure 6A). Furthermore, the prevalence rate of MACE was significantly higher in CAD patients with TSG-6 ≥ 19 ng/mL (40%, 10/25 cases) than the rate in those with TSG-6 < 19 ng/mL (20%, 22/110 cases) ( Figure 6B). The finding suggests that high plasma levels of TSG-6 may induce downregulation of its receptor, leading to ineffective MACE prevention in CAD patients. Among 135 patients with CAD, 32 patients had major adverse cardiovascular events (MACE), defined as cardiovascular death, heart failure, acute myocardial infarction, post-infarction angina, and ischemic stroke, during a period of four years. Plasma TSG-6 level at the onset of acute coronary syndrome tended to be increased in CAD patients with MACE compared with those without MACE, but there was no significant difference ( Figure 6A). Furthermore, the prevalence rate of MACE was significantly higher in CAD patients with TSG-6 ≥ 19 ng/mL (40%, 10/25 cases) than the rate in those with TSG-6 < 19 ng/mL (20%, 22/110 cases) ( Figure 6B). The finding suggests that high plasma levels of TSG-6 may induce downregulation of its receptor, leading to ineffective MACE prevention in CAD patients. Figure 6. Prediction of prognosis using plasma TSG-6 levels at onset of CAD. Among 135 patients with acute coronary syndrome, 32 patients had major adverse cardiovascular events (MACE) including cardiovascular death, heart failure, acute myocardial infarction, post-infarction angina, and ischemic stroke during a period of four years after onset. The follow-up period was 50.0 ± 0.57 months. (A) Plasma TSG-6 levels tended to increase in CAD patients with MACE compared with those without MACE. Data are presented as mean ± SEM. Unpaired Student's t test was used for statistical analysis. p = 0.6109; (B) Prevalence rate of MACE (+) was significantly higher in CAD patients with TSG-6 ≥ 19 ng/mL (40%, 10/25 cases) than those with TSG-6 < 19 ng/mL (20%, 22/110 cases). Chi-squared test was used for statistical analysis. * p = 0.0338. These results are unpublished data from our follow-up study based on our previous report [10]. Figure 6. Prediction of prognosis using plasma TSG-6 levels at onset of CAD. Among 135 patients with acute coronary syndrome, 32 patients had major adverse cardiovascular events (MACE) including cardiovascular death, heart failure, acute myocardial infarction, post-infarction angina, and ischemic stroke during a period of four years after onset. The follow-up period was 50.0 ± 0.57 months. (A) Plasma TSG-6 levels tended to increase in CAD patients with MACE compared with those without MACE. Data are presented as mean ± SEM. Unpaired Student's t test was used for statistical analysis. p = 0.6109; (B) Prevalence rate of MACE (+) was significantly higher in CAD patients with TSG-6 ≥ 19 ng/mL (40%, 10/25 cases) than those with TSG-6 < 19 ng/mL (20%, 22/110 cases). Chi-squared test was used for statistical analysis. * p = 0.0338. These results are unpublished data from our follow-up study based on our previous report [10].

Treatment of Atherosclerotic Diseases with TSG-6
There are several major therapeutic strategies against atherosclerosis and related diseases using TSG-6. One is the enhancement of endogenous TSG-6 by MSC infusion and the another is the supplementation with exogenous TSG-6. (1) Silencing TSG-6 in the administered MSCs results in loss of therapeutic activity, whereas administering exogenous TSG-6 rescues the therapeutic activity [21]. MSC therapeutic activities in other animal models of disease including cerebral ischemia, myocardial infarction, type 1 diabetes, peritoneal adhesions, and experimental autoimmune encephalomyelitis were observed to be dependent on TSG-6 [31,32,[49][50][51][52]. (2) For replacement therapy of exogenous TSG-6, a great amount of recombinant human TSG-6 is needed. The recombinant TSG-6 is synthesized in Escherichia coli, wheat germ, insect cells, Chinese hamster ovary (CHO) cells, and mouse myeloma cells. CHO cell-derived TSG-6 has a longer half-life compared with mouse myeloma cell-derived TSG-6 in vivo [25].
It is especially important to elucidate the putative receptor for TSG-6, which provides an investigation of its agonist as a new drug target. Otherwise, it is important to clarify a biologically active fragment that has inhibitory effects against atherosclerosis among the 260 amino acids in TSG-6 protein (excluding the signal peptide), and further to develop derivatives or analogs based on the TSG-6 protein for preventing and treating atherosclerosis. Because infusion and injection are not convenient for patients, oral administration is a more suitable method. However, TSG-6 protein is digested after oral administration. Therefore, nanocapsules have received a great amount of attention as a drug delivery method [53]. Encapsulation methods are beneficial for protecting polypeptides against the digestive environment, thereby promoting absorption and delivery to target organs [53]. Due to their small size, these particles are able to pass only through the increased endothelial gaps due to injury and blood vessel damage in atherosclerotic arteries. Nanocapsules including TSG-6 analogs would be useful for oral administration.
However, manipulating plasma TSG-6 level as a potential therapeutic strategy should be considered with caution. If the putative TSG-6 receptors are downregulated in CAD patients, simply infusing TSG-6 would not be effective. Further treatments improving TSG-6 resistance should be developed in the future.

Conclusions
The above findings indicate that TSG-6 slows the development of atherosclerotic lesions by decreasing plasma total cholesterol levels, inflammatory responses of ECs and macrophages, macrophage foam cell formation, and the migration and proliferation of VSMCs. In addition, TSG-6 contributes to plaque stability by promoting collagen production by VSMCs in the fibrous cap. Thus, endogenous TSG-6 enhancement and exogenous TSG-6 replacement treatments are expected to emerge as a new line of therapy against atherosclerosis and its related CAD. The results presented here also provide insights into the potential use of TSG-6 as a biomarker for CAD. These findings may strengthen the clinical utility of TSG-6 in the diagnosis and treatment of atherosclerotic cardiovascular diseases.  Tumor necrosis factor-stimulated gene-6 VSMC Vascular smooth muscle cell