The role of vascular adhesion protein‐1 in diabetes and diabetic complications

ABSTRACT

Vascular adhesion protein‐1 (VAP‐1) plays a dual role with its adhesive and enzymatic properties, facilitating leukocyte migration to sites of inflammation and catalyzing the breakdown of primary amines into harmful by‐products, which are linked to diabetic complications. Present in various tissues, VAP‐1 also circulates in a soluble form in the bloodstream. Diabetes is associated with several complications such as cardiovascular disease, retinopathy, nephropathy, and neuropathy, significantly contributing to disability and mortality. These complications arise from hyperglycemia‐induced oxidative stress, inflammation, and the formation of advanced glycation end‐products (AGEs). Earlier research, including our own from the 1990s and early 2000s, has underscored the critical role of VAP‐1 in these pathological processes, prompting extensive investigation into its contribution to diabetic complications. In this review, we examine the involvement of VAP‐1 in diabetes and its complications, alongside its link to other conditions related to diabetes, such as cancer and metabolic dysfunction‐associated fatty liver disease. We also explore the utility of soluble VAP‐1 as a biomarker for diabetes, its complications, and other related conditions. Since the inhibition of VAP‐1 to treat diabetic complications is a novel and promising treatment option, further studies are needed to translate the beneficial effect of VAP‐1 inhibitors observed in animal studies to clinical trials recruiting human subjects. Besides, future studies should focus on using serum sVAP‐1 levels for risk assessment in diabetic patients, identifying those who need intensive glycemic control, and determining the patient population that would benefit most from VAP‐1 inhibitor therapies.

Vascular adhesion protein‐1 (VAP‐1) has adhesive and enzymatic functions, aiding leukocyte migration to inflammation sites and producing by‐products linked to diabetic complications. This review focuses on the impact of VAP‐1 on diabetes and its complications, its association with related diseases, and the potential of soluble VAP‐1 as a biomarker.

INTRODUCTION

Vascular adhesion protein‐1 (VAP‐1) is a protein with both adhesive and enzymatic characteristics. VAP‐1 in the endothelium plays a role in mediating leukocyte rolling, adhesion, and transmigration, which are central steps during leukocyte extravasation to sites of inflammation ¹ . Besides, VAP‐1 is also a primary amine oxidase that can be inhibited by semicarbazide, and it is also called semicarbazide‐sensitive amine oxidase (SSAO) ² . The enzymatic activity of VAP‐1/SSAO can catalyze an oxidative deamination reaction of primary amines to produce aldehyde, hydrogen peroxide, and ammonia ³ . These harmful metabolites, along with inflammation, have been demonstrated to contribute to the development of diabetic complications ⁴ . VAP‐1 is expressed in endothelial cells, smooth muscle cells, and adipose tissues ⁵ . In addition to its transmembrane form, VAP‐1 also exists as a soluble form, soluble VAP‐1 (sVAP‐1), which can be detected in the circulation ⁶ . The plasma level of sVAP‐1 has been shown to be a good biomarker of various diabetic complications, including cardiovascular diseases, diabetic kidney disease, and diabetic retinopathy ⁷ , ⁸ , ⁹ , ¹⁰ . In this article, we will review the role of VAP‐1 in diabetes and diabetic complications.

DIABETES AND DIABETIC COMPLICATIONS

The global prevalence of diabetes in individuals aged 20–79 years was estimated to be 10.5% in 2021, accounting for 536.6 million people. This prevalence is predicted to increase to 12.2% (783.2 million) by the year 2045 ¹¹ . Diabetes is linked to various complications, including cardiovascular diseases, diabetic retinopathy, diabetic nephropathy, and diabetic neuropathy. According to the report of the Diabetes Control and Complications Trial (DCCT), the cumulative incidence 30 years after the onset of type 1 diabetes were 50% for proliferative retinopathy, 25% for diabetic nephropathy, and 14% for cardiovascular disease ¹² . These diabetic complications are thus the main causes of visual loss, end‐stage renal disease, disability following cardiovascular diseases, and mortality in human beings.

Hyperglycemia in diabetic patients can lead to the development of diabetic complications through several mechanisms, such as increased oxidative stress ¹³ , systemic low‐grade inflammation ¹⁴ , generation of advanced glycated end products (AGEs) ¹⁵ , and mitochondrial dysfunction ¹⁶ . In 1990s and early 2000s, many research groups, including our group, have noticed that VAP‐1 is involved in several important mechanisms linking hyperglycemia and diabetic complications. Therefore, many studies have been performed to reveal its role in diabetic complications.

DISCOVERY, EXPRESSION, AND FUNCTION OF VAP‐1

VAP‐1 was initially discovered by Salmi and Jalkanen in 1992, using a monoclonal antibody (mAb), 1B2, through the immunization of mice with stromal components derived from inflamed synovium obtained from rheumatoid arthritis patients ¹⁷ , ¹⁸ . In the study, they showed that VAP‐1 supported lymphocyte binding, and treatment with 1B2 monoclonal antibody inhibited lymphocyte binding to high endothelial venules in peripheral lymph nodes, tonsils, and synovia. These findings suggest that VAP‐1 plays a role in endothelial–lymphocyte interaction and is a new adhesion molecule that can recruit lymphocytes to sites of inflammation. In the non‐inflamed state, VAP‐1 remains as an intracellular molecule; while in an inflammatory state, VAP‐1 will translocate onto the cell surface, as shown in animal studies in pig and dog models ¹⁹ . Besides facilitating lymphocyte adhesion, VAP‐1 is also involved in granulocyte rolling and extravasation, which has been demonstrated in the acute inflammation animal model ²⁰ .

In addition to serving as an adhesion molecule, VAP‐1 also has an enzymatic function and is thus also called SSAO. The SSAO enzymatic activity catalyzes the reaction: R‐CH₂‐NH₂ + O₂ + H₂O → R‐CHO + H₂O₂ + NH₃, in which a primary amine is oxidatively deaminated to the corresponding aldehyde with a simultaneous release of hydrogen peroxide and ammonium ²¹ . The SSAO activity of VAP‐1 can enhance lymphocyte rolling, possibly through the generation of transient covalent bonds. Findings from cell studies have demonstrated that the SSAO inhibitors could reduce by >40% the number of rolling and firmly bound lymphocytes under shear ²² . Besides, the end‐products catalyzed by SSAO are highly reactive and cytotoxic, such as formaldehyde and H₂O₂, and these end‐products can modify several proteins to form AGEs, all of which are important players in the pathogenesis of diabetic complications ²³ . Indeed, we have demonstrated that the circulating sVAP‐1 concentration after glucose challenge correlated significantly with systemic AGEs and oxidative stress levels in human subjects ²⁴ .

Apart from the transmembranous VAP‐1, there is a soluble form of VAP‐1 ⁶ . sVAP‐1 retains its SSAO enzymatic function. In humans, serum sVAP‐1 is the main source of SSAO activity ²⁵ . Findings from other groups and our group have demonstrated that serum sVAP‐1 is a biomarker to predict several diabetic complications ⁷ , ⁸ , ⁹ , ¹⁰ , which will be described in detail in the following sections.

VAP‐1 AND ATHEROSCLEROTIC CARDIOVASCULAR DISEASES (ASCVD)

Diabetes is associated with an increased risk of ASCVD including coronary artery disease, cerebrovascular disease, and peripheral arterial disease ²⁶ . Oxidative stress and inflammation of the arterial wall are important in the pathophysiology of ASCVD ²⁷ , and VAP‐1 is one of the main contributors of these two important pathogenic factors, as mentioned previously ²⁸ . In the aorta, VAP‐1 expression was significantly higher in atherosclerotic plaques than in normal parts in humans, apolipoprotein E deficient mice, and New Zealand White rabbits ²⁹ , ³⁰ . In mice overexpressing human VAP‐1 in the endothelium, leukocyte binding, serum AGE production, and hepatic expression of redox‐sensitive proteins were increased ³¹ . Besides, these mice had a phenotype of accelerated atherosclerosis presenting as a similar total atheroma area but with fewer atheroma numbers. In addition, when mice were chronically fed with methylamine, a substrate of SSAO, these mice showed an increased atheroma area ³¹ . On the other hand, our group has shown that treatment with a SSAO inhibitor could reduce atherosclerosis in both the apolipoprotein E deficient mice model and in the high cholesterol‐fed New Zealand White rabbit model ²⁹ , ³⁰ . SSAO inhibition reduced several important pathophysiologic pathways of atherosclerosis, including decreasing H₂O₂ production in vessels, lowering plasma cholesterol levels, reducing the expression of adhesion molecules and inflammatory cytokines in endothelial cells, suppressing the recruitment and activation of macrophages, and inhibiting the migration and proliferation of smooth muscle cells. More importantly, the effect of SSAO inhibition on atherosclerosis in an apolipoprotein E deficient mice model is comparable to the effect of atorvastatin. Since SSAO inhibitor reduces atherosclerosis through different mechanisms from statins, these data suggest that VAP‐1 is a novel drug target of ASCVD. Currently, several VAP‐1/SSAO inhibitors have been developed, such as MDL72974A ³² and PXS‐4728A ³⁰ . Since none of the SSAO inhibitors has been tested in clinical trials for the treatment of ASCVD, this should be a reasonable and promising next step for the development of this novel therapy for ASCVD.

Several human studies have demonstrated that blood sVAP‐1 is a biomarker for cardiovascular disease. Using coronary angiography to define coronary artery disease (CAD), we have demonstrated that subjects with CAD had a higher plasma sVAP‐1 concentration than subjects without CAD, and there was a linear trend between plasma sVAP‐1 concentration and the severity of CAD, which was defined as the number of vessels or segments with clinically significant stenosis ²⁹ . Additionally, plasma sVAP‐1 concentrations were found to be elevated in patients with congestive heart failure ³³ , and there is a positive correlation between sVAP‐1 levels and arterial stiffness ³⁴ , as well as carotid intima‐medial thickness ²⁴ . In patients with type 2 diabetes, we have demonstrated that the serum sVAP‐1 concentration could independently predict both the 10 year all‐cause mortality and cardiovascular mortality ⁷ . Similarly, in a Finnish cohort study, circulating sVAP‐1 levels were found to predict the risk of major adverse cardiovascular events and cardiovascular mortality over 9 years ³⁵ . Taken together, these findings suggest that circulating sVAP‐1 is a good biomarker and predictor of ASCVD.

VAP‐1 AND DIABETIC KIDNEY DISEASE

Inflammation, oxidative stress, and the production of AGEs are linked to VAP‐1 function and are also important in the pathophysiology of diabetic kidney disease ³⁶ . Indeed, the end‐products catalyzed by VAP‐1/SSAO, i.e. formaldehyde and hydrogen peroxide, have been demonstrated to be toxic toward cultured endothelial cells ³⁷ . In a streptozotocin‐induced diabetic mice model, a significant elevation of SSAO activity was noted in the kidney ³⁷ . Besides, the overexpression of endothelial VAP‐1 in mice with induced glomerulosclerosis, marked by a reduction in capillary space, an increase in cell number, elevated extracellular matrix, and a thicker glomerular stalk ³¹ . Importantly, the expression of the receptor of AGEs in the glomerulus was up‐regulated, possibly in response to the elevated levels of AGEs generated by the oxidative deamination reaction catalyzed by SSAO. On the other hand, treatment with an SSAO inhibitor significantly reduced renal damage in a streptozotocin‐induced diabetic nephropathy mouse model. These data suggest that VAP‐1 is a drug target for diabetic kidney disease, and the development of VAP‐1/SSAO inhibitors or monoclonal antibody is a good strategy for novel treatment discovery. In 2018, an oral VAP‐1 inhibitor, ASP8232, was investigated in a phase 2 trial for the treatment of diabetic kidney disease ³⁸ . The trial enrolled patients aged 18–85 years with type 2 diabetes, an estimated glomerular filtration rate (eGFR) between 25 and 75 mL/min per 1.73 m², HbA1c less than 11.0% (97 mmol/mol), and a urinary albumin‐to‐creatinine ratio (UACR) of 200–3,000 mg/g. After 12 weeks, the UACR decreased significantly by 17.7% (95% CI 5.0–28.6) in the ASP8232 group, compared with an increase of 2.3% (−11.4 to 18.1) in the placebo group. This finding indicates that ASP8232 is effective in reducing albuminuria in patients with diabetic kidney disease.

Several reports have demonstrated that blood sVAP‐1 serves as a biomarker for kidney diseases. In non‐diabetic patients, the Taiwan Lifestyle Cohort Study has revealed that individuals with chronic kidney disease (CKD) had higher serum VAP‐1 levels compared with those without CKD, and serum sVAP‐1 levels were positively associated with the urinary albumin‐to‐creatinine ratio and negatively associated with the estimated glomerular filtration rate ³⁹ . In patients with type 2 diabetes mellitus, we have found that the serum VAP‐1 could predict the incidence of end‐stage renal disease (ESRD) during a median follow‐up time of 12.36 years ⁸ . Each 1‐standard deviation increase in serum VAP‐1 was associated with a hazard ratio of 1.55 for the risk of ESRD, after adjusting for other risk factors. In addition, serum sVAP‐1 could reclassify the risk of ESRD on top of CKD staging. These findings suggest that serum sVAP‐1 could enhance the prediction of ESRD risk in patients with type 2 diabetes. This may be particularly useful in identifying the target population that would benefit most from VAP‐1 inhibitors for treating diabetic kidney disease.

VAP‐1 AND DIABETIC RETINOPATHY

Several studies suggest that VAP‐1 is involved the pathogenesis of diabetic retinopathy ⁴⁰ , ⁴¹ . In humans, sVAP‐1 levels in vitreous fluid were higher in patients with proliferative diabetic retinopathy ¹⁰ , ⁴² . In a cell model, the release of sVAP‐1 from retinal capillary endothelial cells was increased when exposed to high glucose and treated with pro‐inflammatory cytokines ¹⁰ . In a streptozotocin‐induced diabetes rat model, the VAP‐1 inhibitor UV‐002 could significantly reduce leukocyte transmigration rate in the retina ⁴³ . Similarly, in the rat model of endotoxin‐induced ocular inflammation, the retinal expression of VAP‐1 was higher, and inhibition of VAP‐1 significantly suppressed leukocyte recruitment to the anterior chamber, vitreous, and retina ⁴¹ .

Several VAP‐1 inhibitors have been tested in animal models for the purpose of treating retinopathy. For example, RTU‐1096 has demonstrated the ability to prevent outer nuclear layer thickening caused by laser photocoagulation in mice ⁴⁴ . In streptozotocin‐induced diabetic rat model, another orally administered VAP‐1 inhibitor, 1H‐imidazol‐2‐amine, has also been shown to significantly inhibit ocular permeability, a key feature of diabetic retinopathy ⁴⁵ . However, none of these VAP‐1 inhibitors has undergone clinical trials for the treatment of diabetic retinopathy.

VAP‐1 AND CANCER

Epidemiological observations indicate that individuals with diabetes face up to a twofold increase in the risks of developing colorectal, breast, endometrial, kidney, liver, and pancreatic cancers ⁴⁶ . Cancers have become one of the main causes of death in patients with diabetes in recent decades ⁴⁷ . VAP‐1 has been demonstrated to be involved in cancers in both animal and human studies. In VAP‐1 knockout mice, the growth rate of melanoma is lower ⁴⁸ . Further exploration on the mechanism has revealed that VAP‐1 can enhance neoangiogenesis to support the growth of melanoma. The role of VAP‐1 in malignancy‐associated angiogenesis is further supported by human studies. One study demonstrated a highly significant correlation between blood SSAO activity and serum vascular endothelial growth factor levels in individuals with non‐small cell lung cancer ⁴⁹ . In addition, we have demonstrated that VAP‐1 expression is more pronounced at the invasion front of colorectal cancer ⁹ , suggesting a role for VAP‐1 in cancer invasion and metastasis.

In patients with type 2 diabetes, we have reported that serum sVAP‐1 could predict the incidence of cancers ⁵⁰ . Patients with serum sVAP‐1 in the highest tertile had a 2.95‐fold risk of cancers, compared with patients with serum sVAP‐1 in the lowest tertile. In addition, serum sVAP‐1 could also predict cancer‐related mortality and all‐cause mortality over 10 years in patients with type 2 diabetes, independent of the traditional risk factors including age, sex, smoking, history of cardiovascular disease, obesity, hypertension, hemoglobin A1c, diabetes duration, total cholesterol, use of statins, abnormal ankle‐brachial index, estimated glomerular filtration rate, and proteinuria ⁷ . Recently, we have found that the relationship between serum sVAP‐1 and cancers can also be observed in non‐diabetic patients ⁵¹ . In the Taiwan Lifestyle Cohort Study, a community‐based population, serum sVAP‐1 levels were associated with a 12 year risk of incident cancer, cancer mortality, and all‐cause mortality. Furthermore, the predictive performance of serum VAP‐1 was superior to that of traditional risk factors such as gender, smoking, body mass index, hypertension, diabetes, and estimated glomerular filtration rate. In patients with colorectal cancer, blood sVAP‐1 was found to independently predict cancer‐related mortality and all‐cause mortality independent of the cancer stage ⁹ . To sum up, these data suggest that VAP‐1 is involved in several pathogeneses of cancer development, especially invasion, neoangiogenesis, and metastasis, and serum sVAP‐1 level is a good biomarker for the prediction of cancer incidence, prognosis, and mortality.

VAP‐1 AND METABOLIC DYSFUNCTION‐ASSOCIATED STEATOTIC LIVER DISEASE

Metabolic dysfunction‐associated steatotic liver disease (MASLD), previously called non‐alcoholic fatty liver disease, commonly co‐exists in patients with type 2 diabetes. Studies have shown that MASLD could be found in up to 70% of patients with diabetes ⁵² . MASLD is a progressive disease and can present from simple steatosis or metabolic‐associated fatty liver, metabolic‐associated steatohepatitis, to liver cirrhosis or even hepatocellular carcinoma ⁵³ . As a result, MASLD stands as an important cause of hepatic failure and liver transplantation. Since inflammation and oxidative stress are shared pathogeneses of MASLD and diabetic complications, several studies have explored the role of VAP‐1 in MASLD.

The expression of VAP‐1 in liver has been shown to be higher in patients with non‐alcoholic steatohepatitis ⁵⁴ . In a CCl₄‐induced hepatic injury mouse model, both VAP‐1 knockout and VAP‐1 antibody have demonstrated the ability to attenuate hepatic fibrosis and to reduce leukocyte infiltration into the liver ⁵⁵ . Moreover, VAP‐1 knockout mice were resistant to fibrosis in a diet‐induced MASLD model ⁵⁵ and had reduced hepatic steatosis when fed with a high‐fat diet ⁵⁴ . In patients with MASLD, their blood sVAP‐1 levels were higher than control subjects matched for age and metabolic phenotype ⁵⁵ . In 2022, the VAP‐1 inhibitor TERN‐201 underwent its Phase 1b trial, but no formal scientific research report has been published yet ⁵⁶ . Moreover, given that VAP‐1 is involved in the progression of liver fibrosis, studies have explored the impact of VAP‐1 inhibitors on primary sclerosing cholangitis. The VAP‐1 inhibitor BTT1023 was subjected to a Phase II clinical trial in 2022. However, an interim assessment revealed that it failed to achieve the desired response rate, leading to the discontinuation of the trial due to futility ⁵⁷ .

VAP‐1 AND DIABETES

VAP‐1 possesses insulin‐like activities, capable of stimulating glucose uptake and mitigating hyperglycemia, as demonstrated in both in vivo and in vitro studies. Substrates of VAP‐1/SSAO have been found to normalize hyperglycemia in diabetic mouse models ⁵⁸ , ⁵⁹ , and the effect can be inhibited by VAP‐1 inhibitors ⁵⁹ . Additionally, in cell models, VAP‐1/SSAO has been shown to facilitate glucose uptake in various cell types, such as adipocytes ⁶⁰ , skeletal muscle cells ⁵⁸ , smooth muscle cells ⁶¹ , and liver tissues ⁶² .

Owing to its insulin‐mimetic effects, VAP‐1 has the potential to counteract hyperglycemia in insulin deficiency or insulin resistance. The link between sVAP‐1 and diabetes was initially revealed in type 1 diabetic patients ⁶³ , ⁶⁴ . The plasma sVAP‐1 level and SSAO activity were higher in individuals with type 1 diabetes, and there was a positive correlation of plasma sVAP‐1/SSAO with blood glucose. Later, in patients with type 2 diabetes, we showed that plasma sVAP‐1 is also elevated, compared with normal subjects ⁶⁵ . In the Taiwan Lifestyle Cohort Study, data from 600 subjects without diabetes were analyzed ⁶⁶ . The concentration of serum sVAP‐1 was higher in individuals with prediabetes compared with those with normal glycemia. Besides, serum sVAP‐1 concentration is significantly increased in response to hyperglycemia after a glucose challenge. Additionally, in patients with gestational diabetes, the level of sVAP‐1 is higher than in normoglycemic pregnant woman ⁶⁷ .

The correlation between sVAP‐1 and glucose was further substantiated in an animal study ⁶⁸ . In this research, transgenic mice engineered to produce human sVAP‐1 in their endothelial or adipose tissues showed elevated serum sVAP‐1 levels when subjected to STZ (streptozotocin) treatment for 2–5 days. STZ is known to cause the death of β‐cells in the pancreas, leading to insulin deficiency and acute hyperglycemia. The increased sVAP‐1 level after STZ treatment further establishes the correlation between hyperglycemia and sVAP‐1 levels.

In addition to its correlation with diabetes and hyperglycemia observed in cross‐sectional studies, sVAP‐1 also serves as a predictive marker for the incidence of diabetes in longitudinal follow‐up studies. Findings from the Taiwan Lifestyle Cohort Study have demonstrated that a high fasting sVAP‐1 concentration is linked to a lower incidence of diabetes during the 4.7 ± 2.6 year follow‐up period ⁶⁶ . These data appear to be contradictory when compared with previous cross‐sectional studies. However, considering the insulin‐like properties of VAP‐1 and its capability to counterbalance hyperglycemia, an increased sVAP‐1 level in prediabetes and diabetes reflects VAP‐1's effect in maintaining glucose homeostasis in addition to insulin; whereas a higher sVAP‐1 in people without diabetes suggests a better glucose‐homeostatic effect and it is thus reasonable to link it to a lower incidence of diabetes.

CONCLUSION AND FUTURE PERSPECTIVES

VAP‐1 is involved in the pathogenesis of various diabetic complications, including cardiovascular disease, diabetic kidney disease, diabetic retinopathy, and other conditions such as malignancy and MASLD (Figure 1). Findings from laboratory studies have proved that VAP‐1 is a drug target for diabetic complications and MASLD. However, only one VAP‐1 inhibitor has been tested in a clinical trial in human beings which successfully demonstrated the beneficial effect of the VAP‐1 inhibitor on diabetic kidney disease ³⁸ . Since VAP‐1 inhibition to treat diabetic complications is a novel and promising treatment option, further studies are needed to translate the beneficial effect of VAP‐1 inhibitors observed in animal studies to clinical trials recruiting human subjects. Besides, serum sVAP‐1 has been proved to be a good biomarker for diabetic complications, cancers, and MASLD (Figure 2). Further work should focus on the use of serum sVAP‐1 for risk assessment in diabetic patients, to identify the high‐risk group who need intensive glycemic control, and to find the target population that would benefit most from VAP‐1 inhibitors.

Figure 1.

Summary of the role of vascular adhesion protein‐1 (VAP‐1) in pathogenesis and as a novel treatment of cardiovascular diseases, diabetic kidney disease, diabetic retinopathy, cancer, and metabolic dysfunction‐associated fatty liver disease.

Figure 2.

Overview of circulating soluble vascular adhesion protein‐1 (sVAP‐1) as a biomarker of cardiovascular diseases, diabetic kidney disease, diabetic retinopathy, cancer, and metabolic dysfunction‐associated fatty liver disease.

DISCLOSURE

Hung‐Yuan Li is an Editorial Board member of Journal of Diabetes Investigation and a co‐author of this article. To minimize bias, he was excluded from all editorial decision‐making related to the acceptance of this article for publication.