Serum semaphorin 7A is associated with the risk of acute atherothrombotic stroke

Abstract Semaphorin 7A (Sema7A), a neural guidance cue, was recently identified to regulate atherosclerosis in mice. However, the clinical relevance of Sema7A with atherosclerotic diseases remains unknown. The aim of this study was to investigate the association between serum Sema7A and the risk of acute atherothrombotic stroke (AAS). We measured serum concentrations of Sema7A in 105 newly onset AAS cases and 105 age‐ and sex‐matched controls, showing that median Sema7A level in AAS cases was over three times of that in controls (5.86 vs 1.66 ng/mL). Adjusted for hypertension, body mass index, fasting blood glucose, total cholesterol, triglyceride, high‐density lipoprotein (HDL)‐cholesterol, low‐density lipoprotein (LDL)‐cholesterol, current smoking and alcohol consumption, multivariate logistic regression showed that higher Sema7A was independently associated with the odds of AAS (OR = 6.40, 95% CI: 2.88‐14.25). Each 1‐standard deviation increase in Sema7A was associated with a threefold higher odds of AAS (OR = 3.42, 95% CI: 1.84‐6.35). Importantly, adding Sema7A to a multivariate logistic model containing conventional cardiovascular risk factors improved the area under receiver operating characteristic curves from 0.831 to 0.891 for the association with AAS. In conclusion, elevated serum Sema7A is independently associated with the risk of AAS, suggesting that it may play a potential role in AAS.


| INTRODUC TI ON
Stroke is one of the leading causes of mortality and disability worldwide. Acute atherothrombotic stroke (AAS), the most common subtype, is a result of the atherosclerosis of cerebral arteries accounting for an overwhelming majority of deaths in China. 1,2 The proatherogenic milieu consists of dysfunctional endothelium, activated leucocytes, platelets and inflammatory factors that facilitate atheroprogression and cardiovascular events. 3,4 These molecules, reflecting systemic or focal inflammation, have shed light on the risk stratification of AAS. 5,6 To date, emerging evidence suggests the roles of semaphorins, a protein family primarily detected on immunocyte membranes and signalling through plexins or integrins to exert versatile functions, 7 underlying atherosclerotic diseases since their initial demonstration as axonal guidance cues. [8][9][10] Among this family, semaphorin 7A (Sema7A) is a membrane protein encoded by SEMA7A gene on human chromosome 15 (25 kb) or mouse chromosome 9B (22 kb) and shows an important role in immune response, neural development, bone homeostasis, cancer and inflammation. [11][12][13][14] Up-regulation of Sema7A expression has been associated with various inflammatory diseases, including interstitial lung disease, 15 multiple sclerosis (MS), 16 rheumatoid arthritis 17 and airway hypersensitivity. 18 Recently, we demonstrated a causal role of Sema7A in experimental atherosclerosis via eliciting endothelial dysfunction and vascular inflammation, 19 whereas the clinical relevance of Sema7A in atherosclerotic disease remains to be elucidated. Therefore, we conducted this case-control study to investigate the association of serum Sema7A with AAS.

| Study population
We randomly selected 105 cases from a large cohort of patients with AAS from CATIS (China Antihypertensive Trial in Acute Ischemic Stroke). 20 In CATIS, patients with a systolic BP ≥220 mm Hg or diastolic BP ≥120 mm Hg, severe heart failure, acute myocardial infarction or unstable angina, atrial fibrillation, aortic dissection, cerebrovascular stenosis or resistant hypertension; those in a deep coma and those treated with intravenous thrombolytic therapy were excluded. The randomization process was conducted as follows: There were a total of 3170 patients with atherothrombotic stroke and with serum samples in CATIS. Among all these atherothrombotic stroke patients, we selected 105 patients as cases by means of simple random sampling method.
Controls free of stroke from a cross-sectional study in China were matched to the cases by 1:1 based on age (birth at the same year) and sex by SAS software. 21 Subjects with conditions known to affect Sema7A levels, including rheumatoid arthritis, interstitial lung disease, idiopathic pulmonary fibrosis (IPF), polycystic kidney disease, alcoholic liver disease, MS and bone fracture were excluded. The protocol of this case-control study conforms to the ethical guidelines of the Declaration of Helsinki 22 and has been approved by the institutional review boards and ethics committees at Soochow University in China and all participating hospitals.
Written informed consent was obtained from all study participants included in the study or their immediate family members.

| Statistical analysis
Continuous variables with normal and skewed distribution were presented as means with standard deviation (SD) and medians with interquartile ranges (IQRs) respectively. Means and medians in variables between the two groups were compared using the paired Student's t test or Wilcoxon rank-sum test. Comparisons of percentages for categorical variables between the two groups were performed using the χ 2 test. Differences in Sema7A levels were evaluated between AAS cases and controls. Median Sema7A level (3.20 ng/mL) was served as cut-off and the study population was divided into low Sema7A (<3.20 ng/mL) and high Sema7A (≥3.20 ng/mL) groups.
Both univariate and multivariate logistic regression analyses were

List of main topics
• Serum Sema7A is increased in patients with AAS.
• Higher serum Sema7A is independently associated with the risk of AAS.
conducted to calculate the odds ratios (ORs) and 95% confidence intervals (95% CIs) of AAS associated with high Sema7A levels. In multivariate logistic analyses, adjustment variables included hypertension, BMI, current smoking, alcohol consumption, FBG, total cholesterol (TC), triglyceride (TG), low-density lipoprotein cholesterol (LDL-C) and high-density lipoprotein cholesterol (HDL-C). In addition, we assessed the improvement value of Sema7A for prediction of AAS risk by computing the area under receiver operating characteristic curves (AUC) and compared the model including Sema7A and other conventional risk factors to a model including only other conventional risk factors. 24 The association between serum Sema7A and high-sensitivity C-reactive protein (hsCRP) was evaluated using linear correlation and regression. A two-tailed P < 0.05 was considered statistically significant. Statistical analyses were performed using

| RE SULTS
A total of 105 AAS patients and 105 age-and sex-matched controls were included in the analyses. As shown in Table 1 (P < 0.001), compared with low Sema7A. Additionally, each 1-SD increase of Sema7A was associated with a three times higher odds of AAS (adjusted OR = 3.42, 95% CI: 1.84-6.35; P < 0.001) ( Table 2).
As displayed in Figure 2, the AUC for the model containing

| D ISCUSS I ON
In this 1:1 matched case-control study, we found that serum Sema7A level was significantly higher in AAS patients than that in healthy controls. Furthermore, in multivariate logistic analyses, increased  studies used only simple statistical tests to explore the association between Sema7A and diseases without controlling for potential confounders. 15,17,25,27 In contrast, the stringent criteria for selecting cases and controls in our study ensure comparability between AAS and control groups and minimize the interference from potential confounders. We further adjusted for some potential confounders in multivariate logistic analyses. We demonstrated an important result that elevated Sema7A was highly associated with AAS, which provided strong support to our previous findings in animal models of atherosclerosis.
The mechanisms underlying the observed association of serum Sema7A with AAS have been proposed in our previous study 19 and other studies. It is well documented that Sema7A and its receptors are expressed in leucocytes, endothelial cells (ECs), neurons and platelets, [28][29][30] all of which play essential roles in atherosclerosis.
The increased Sema7A observed in AAS may be originated from injured EC, 19 activated leucocytes, 31 platelets and erythrocytes. 28,30 Furthermore TA B L E 2 Odds ratios and 95% confidence intervals of acute atherothrombotic stroke risk associated with serum Sema7A

ACK N OWLED G EM ENTS
This work was supported by grants from the Natural Science

CO N FLI C T O F I NTE R E S T
The authors confirm that there are no conflicts of interest.