High Human Cytomegalovirus IgG Level is Associated with Increased Incidence of Diabetic Atherosclerosis in Type 2 Diabetes Mellitus Patients

Background

T2DM is one of the most prevalent chronic diseases worldwide and accounts for 90–95% of all diabetes cases. It encompasses individuals who have insulin resistance and usually have relative rather than absolute insulin deficiency. Genetic and environmental factors are both involved in the development of T2DM. However, the underling mechanism of T2DM is poorly understood, and the present prevention and treatment of the disease are not effective. The evidence of high prevalence of human cytomegalovirus (HCMV) infection in T2DM patients and HCMV DNA existing in β cells suggest that viral infection plays an important role in the development of the T2DM [1,2].

HCMV is a member of the beta-herpesvirus family, which contains a large double-stranded DNA genome [3]. HCMV is ubiquitous, with seropositivity rates ranging from 40% to 100% in adults globally [4]. Like other herpesviruses, HCMV establishes latent infection after primary infection and remains in hosts throughout life. HCMV diseases are related to host immune status. Among immunocompetent hosts, HCMV infection is mostly an asymptomatic infection. However, as an important opportunistic pathogen, it is a major cause of morbidity and mortality among immunocompromised patients, such as patients with acquired immunodeficiency syndrome (AIDS) and recipients of solid organ and stem cell transplantation [5–7]. Congenital HCMV infection is a leading non-genetic cause of sensorineural hearing loss in children.

Previous research has found that nucleic acid sequences specific for HCMV can be located in the islets of the pancreas [2]. HCMV infection can result in inflammation by inducing immune reactions, eventually those leading to apoptosis of islets β-cells [8,9], suggesting that HCMV can infect and damage β-cells. Therefore, HCMV infection, by causing the deficit and increased apoptosis of β-cells, may be associated with T2DM [10]. However, the effect of human cytomegalovirus infection on T2DM remains unclear and controversial [1,2,11–13]. Clinical studies have shown an association between HCMV infection and T2DM [1,11]. HCMV seropositivity with detectable IgG antibodies to HCMV correspond with hemoglobin A1C (HbA1c) and non-fasting glucose levels in patients with T2DM [11]. However, other studies have not found this same correlation [12–14]. Moreover, little is known about the role of acute HCMV infection in development of T2DM.

Previous studies [15,16] confirmed that accumulated HCMV burden, which normally presents as high HCMV IgG titers, is linked to a variety of chronic diseases or symptoms, including increased rate of cognitive decline [17], functional impairment (e.g., in activities of daily living) [18], frailty in older women [19], and all-cause and cardiovascular disease mortality [19,20]. High HCMV IgG titer also has a significant association with hypertension and coronary artery disease in cardiovascular disease (CVD) [21–23]. CVD is a common and severe complication in T2DM patients. Metabolisms, hypertension, ageing, overweight, and obesity have been confirmed to be risk factors for CVD in diabetes patients [24,25]. However, the effect of high HCMV IgG titers on development of atherosclerosis in T2DM patients has received little research attention and remains undefined.

Given the high prevalence of HCMV infection in people with T2DM worldwide [1,11,13,14], and in view of the controversial and unclear relationship between HCMV infection and T2DM, we conducted a cross-sectional study in patients with T2DM. Our aims were to assess the effects of active HCMV infection on glucose regulation in people with T2DM and to study the role of HCMV infection in development of diabetic atherosclerosis among people with T2DM.

Material and Methods

PATIENTS AND DATA COLLECTION:

A total of 222 hospitalized patients with diabetes mellitus type 2 treated at the Department of Endocrinology, Wuhan General Hospital of Guangzhou Military Command were recruited into the study between March 2014 and July 2014. The diagnosis of T2DM was based on the Diagnosis and Classification of Diabetes Mellitus formulated by the American Diabetes Association in 2013 [26]. Patients with diagnosed malignant disease, autoimmune disease other than diabetes mellitus, or immunosuppression were excluded. The diagnosis of diabetic atherosclerosis was determined by ultrasound examination of arteries. Hypertension was diagnosed as a sitting systolic blood pressure ≥140 mm Hg and/or a diastolic blood pressure ≥90 mm Hg, or having medication history of using antihypertensive drugs. All participants voluntarily joined this study and provided informed consent. This study was approved by the Ethics Committee of Wuhan General Hospital of Guangzhou Military Command.

Clinical data were collected, including age, sex, duration of diabetes mellitus, body mass index (BMI), smoking, and history of hypertension. Blood samples were taken to test fasting blood glucose (FBG), postprandial blood glucose (PBG), C-peptide, hemoglobin A1C (HbAlc), total cholesterol (TC), triglycerides (TG), high-density lipoprotein cholesterol (HDL-c), and low-density lipoprotein cholesterol (LDL-c). Blood samples were also used for determining HCMV infection, including HCMV DNA and IgG antibody.

DNA EXTRACTION AND POLYMERASE CHAIN REACTION (PCR):

DNA from peripheral blood leukocytes (PBL) (separated by Lymphoprep™, Axis-Shield) was extracted according to the manufacturer’s protocols (Axygen, Hangzhou, China). The nested PCR procedure and primers for amplification of HCMV IE gene were identical to our previous study (in Chinese) [27].

Samples determined to be HCMV DNA-positive were further examined for viral load by quantitative PCR performed on a CFx96 Real-time System device (BIO-RAD, USA). Primers for the HCMV (strain AD169) IE gene amplification were designed using Primer Premier 5.0 software and synthesized by the Sangon Biotech Company (Shanghai) as primer 1(5′-tttagcacgggccttagcct-3′) and primer 2 (5′-gctgcatgatgtgagcaaggg-3′). Each reaction sample had a final volume of 15 ul, consisting of 7.5 ul of Master Mix, 0.3 ul primers, 4.9 ul of distilled water, and 2 ul of the respective DNA. PCR amplification was performed using these steps: initial denaturation at 94°C for 3 min, followed by 40 cycles of (denaturation at 94°C for 15 s, annealing at 62°C for 15 s, and elongation at 72°C for 15 s), and a final elongation at 72°C for 15 min. A standard graph was constructed and the HCMV DNA copy number was calculated according to a previously published protocol [23].

MEASUREMENT OF HCMV IGG TITERS:

HCMV IgG titers were measured by use of a commercial chemiluminescence test kit (Anti-Cytomegalovirus Antibody, immediate early, clone 6F8.2, Roche Diagnostics GmbH) according to the manufacturer’s directions. Among the 222 patients, IgG titers were measured successfully in 206 participants (16 patients did not have HCMV IgG concentration tested due to hemolysis); 204 were seropositive and the concentration of HCMV IgG was >1 U/ml (the remaining 2 participants were <1 U/ml, reflecting no prior HCMV infection).

STATISTICAL ANALYSIS:

For categorical variables, Fisher’s exact test and the chi squared test were used. For continuous variables, the t test, Kruskal-Wallis ANOVA, and Mann-Whitney U test were used, as appropriate. Binary logistic regression was used to evaluate the risk of atherosclerosis. Two-tailed p values of below 0.05 were considered significant. The analysis was done using SPSS version 20.0 statistical software (IBM Corporation, Armonk, NY). Statistical methods partially refer to this study [20,28].

Results

THE PREVALENCE OF HCMV INFECTION AND THE EFFECTS OF ACTIVE HCMV INFECTION ON GLUCOSE REGULATION IN THE T2DM POPULATION:

In the present study, the seroprevalence of HCMV IgG was 99.0% (204/206), indicating a very high prevalence of HCMV infection in the T2DM population. To determine if there is an association between HCMV infection and T2DM, we first investigated the effect of active HCMV infection on glucose regulation in T2DM. Detection of HCMV DNA in serum by nested PCR is considered as a marker of active HCMV infection. The incidence of active HCMV infection was 20.3% (45/222). The patients with active HCMV infection were significantly younger than the patients without HCMV active infection (54.1±14.9 and 59.1±13.2, respectively) (P=0.028). We also compared the duration of the T2DM, FBG, PBG, HbA1c, and C peptide or 2-h C peptide, and no statistically significant differences were observed between the 2 groups (with and without active HCMV infection) (Table 1).

THE EFFECT OF HCMV IGG ANTIBODY TITERS ON DIABETIC ATHEROSCLEROSIS IN T2DM PATIENTS:

Active HCMV infection showed no correlation with the glucose regulation according to the aforementioned results. Therefore, we further investigated whether persistent HCMV infection had an effect on the diabetic atherosclerosis of T2DM, using HCMV-specific IgG as an indicator for long-term HCMV infection. HCMV IgG titers were detected in 206 serum samples from 206 T2DM patients; 99.0% (204/206) of patients showed evidence of HCMV infection (HCMV IgG titer >1 U/ml). Patients seropositive for HCMV IgG were further categorized into 4 groups according to quartiles of HCMV antibody concentrations (U/ml): group 1 with 51 patients (HCMV antibody concentrations 1–126 U/ml), group 2 with 51 patients (127–276 U/ml), group 3 with 45 patients (277–499 U/ml), and group 4 with 57 patients (>500 U/ml).The prevalence of diabetic atherosclerosis in different HCMV IgG concentration groups was studied and we found that the prevalence of the atherosclerosis was significantly different among these 4 groups (P=0.002). We further examined the incidence of diabetic atherosclerosis within the groups, and significant differences were observed between group 1 and group 2 (P=0.014), as well as between group 1 and group 4 (P=0.0004). However, no significant difference was found between group 1 and group 3 (P=0.109). Patients with higher HCMV IgG concentrations were more likely to develop atherosclerosis (Figure 1).

ANALYSIS OF FACTORS AFFECTING DIABETIC ATHEROSCLEROSIS IN T2DM PATIENTS:

According to the results shown above, a significant difference was found in atherosclerosis morbidity among the 4 groups with different HCMV IgG titers, suggesting that HCMV IgG concentration is associated with the development of diabetic atherosclerosis. We further analyzed the potential confounding risk factors of atherosclerosis in T2DM patients with and without atherosclerosis. Therefore, clinical data (sex, age, smoking, hypertension, BMI, TC, TG, HDL-c, LDL-c, and lipoprotein) of 222 patients with T2DM were studied. The data showed that the median HCMV IgG titers were higher in the atherosclerosis group than in the group without atherosclerosis [313.6 (159.2 – >500) U/ml vs. 159.2 (60.3–366.1) U/ml, P=0.00045]. The patients with atherosclerosis (61.4±11.3 years) were older than the patients without atherosclerosis (45.7±14.6 years), (P<0.001). The group of patients with atherosclerosis had a longer duration of T2DM (P=0.001) and were more likely to have hypertension than the patients without atherosclerosis (P=0.006). In addition, BMI, smoking, TC, TG, LDL-c, lipoprotein, and sex did not differ significantly between patients with and without diabetic atherosclerosis, except for HDL-c (Table 2).

HIGH HCMV IGG TITER IS AN INDEPENDENT RISK FACTOR FOR THE DEVELOPMENT OF DIABETIC ATHEROSCLEROSIS:

To determine whether HCMV IgG titer is an independent risk factor for the development of diabetic atherosclerosis, we used binary logistic regression to adjust for potential confounding factors. According to the above results, risk factors for atherosclerosis, such as age, duration of DM, hypertension, HDL-c, and HCMV IgG, were included in a multiple forward stepwise logistic regression analysis. In this analysis, age (in years) was categorized as: group 1 (19–49), group 2 (50–58), group 3 (59–67), and group 4 (68–88), according to quartile. Duration of DM (in months) was dichotomized as group 1 (1–24), group 2 (25–72), group 3 (73–132), and group 4 (133 and above) according to quartile. The variables of HDL-c and hypertension were defined as abnormal or normal. HCMV IgG titer was initially categorized into 4 groups according to quartile: IgG titers (U/ml) <126, 127–276, 277–499, and >500.

Higher IgG titer was an independent risk factor of atherosclerosis after adjustment for confounding risk factors (except for age) compared with the lowest HCMV IgG quartile (<126 U/ml). After full adjustment (further adjusted for age), patients with HCMV IgG titers >500 U/ml and 127–276U/ml IgG titer were significantly associated with atherosclerosis. However, the p-value (0.073) was on the boundary for HCMV IgG titers of quartile 3 (277–499 U/ml) compared with the lowest quartile after adjustment for age (OR=2.6, 95%CI: 0.9–7.5). We further categorized the IgG titers into 2 groups (<126 U/ml and >127 U/ml); HCMV antibody titers were significantly associated with diabetic atherosclerosis both in the unadjusted analysis and after adjustment for all confounding factors analysis (P=0.001 and P=0.001, respectively), and patients with HCMV IgG titers >127U/ml have a median-fold of 4.6 (95%CI: 1.9–11.3) increased incidence of atherosclerosis. Therefore, titer >127 U/ml of HCMV IgG is an independent risk factor for the development of diabetic atherosclerosis (Tables 3, 4).

Discussion

HCMV is ubiquitously distributed worldwide and 40–100% of adults carry the virus with detectable HCMV IgG in their serum. After primary infection, HCMV establishes latency with sporadic reactivation or exogenous reinfection in the host throughout life. In the present study, the prevalence of HCMV infection in a Chinese T2DM population was quite high and the serological positivity of HCMV IgG reached 99.0%. Epidemiology data showed that seroprevalence of HCMV IgG was 94.9% in blood donors in the same region in China in 1991 [29]. This suggests that the prevalence of HCMV infection is probably higher in T2DM patients than in the general population. In addition, it has been reported that the prevalence of HCMV infection in T2DM patients is also high in other countries: 77% in the USA [13], 89.3% in the Netherlands [11], and 77% in the Czech Republic [14]. These findings raise the question of whether HCMV is one of the pathogenic agents causing T2DM. Previous studies showed controversial results regarding the pathogenic role of HCMV infection in T2DM. In various populations, a clear positive association between HCMV infection and T2DM has not yet been defined [1,11,13]. Researchers found that individuals with HCMV seropositivity were more likely to have T2DM than were HCMV-seronegative participants (17.2% vs. 7.9%, P=0.016) among people age 85 and older [11]. An investigation in patients undergoing hemodialysis showed that T2DM patients had a higher seroprevalence of HCMV IgG (97.6%) than in patients without T2DM [1], suggesting that HCMV infection may be associated with T2DM. However, another cross-sectional study in a multi-ethnic population (Hispanic, African-American, white, and Chinese) found no association between HCMV IgG seropositivity and occurrence of T2DM, suggesting HCMV has no etiological role in the development of T2DM [13]. In addition, most studies reported that HCMV seropositivity had no correlation with glucose regulation in T2DM patients [1,13,14,30], except for the Leiden 85-plus study, which showed that T2DM patients with HCMV seropositivity had a significantly higher level of HbA1c and higher non-fasting glucose than the sero-negative patients [11]. Therefore, information regarding the long-term clinical effect of persistent HCMV infection in T2DM patients is limited. Moreover, to the best of our knowledge, little is known about the effect of active HCMV infection on glucose regulation in T2DM patients. To date, only Liang Hao et al. have evaluated the role of active HCMV infection in development of T2DM; they reported that T2DM patients with positive HCMV DNA in their serum had a significantly lower fasting C peptide level than those with HCMV-negative DNA, but no significant differences in blood glucose and insulin level were found between the 2 groups [31]. This suggests that active HCMV infection may play a role in T2DM. In consideration of the few studies and the debatable results on the effect of active HCMV infection on T2DM, it is important to determine the role of active HCMV infection in the development of T2DM and the long-term clinical effect of persistent HCMV infection on T2DM.

In the current study, to determine if active infection indicated by sporadic reactivation and exogenous reinfection of HCMV might affect glucose regulation, we used nested PCR to amplify HCMV DNA in PBL of T2DM patients. HCMV DNA was detectable in the serum of 20.3% of patients. We found that HCMV DNA was more frequently detectable in the young group than in the older age groups. It is possible that exogenous reinfection is more easily acquired in younger populations due to lifestyle behaviors that importantly affect the epidemiology of HCMV infection. Exogenous reinfection can more easily occur when there is close contact with children less than 3 years old and when there is sexual exposure to multiple partners. Previous studies showed that HCMV viral load is highly correlated with the risk of HCMV disease in immunocompromised individuals [32–34]. In the present study, although active HCMV infection was more frequently detectable in young patients, no significant relationship was found with HCMV active infection or HCMV viral load with glucose regulation, including level of HbA1c, FBG, PBG, and C-peptide in T2DM patients. These findings suggest that active HCMV infection does not acutely impair glucose regulation in T2DM patients.

HCMV antibody may be associated with cumulative viral burden [15,16]. Previous research showed that antibody levels increase over time in individuals after viral infection [15]. Higher HCMV IgG titers are induced by frequent HCMV reactivations or long-term HCMV infection, which are generally observed in older individuals [16,35]. Our research showed that higher HCMV IgG titers in the T2DM patients aged 19–88 years had no impact on FBG, PBG, HbA1c, C peptide, or 2-h C peptide. It seems that HCMV infection is not correlated with glucose regulation in T2DM patients, and although a positive association between HCMV IgG and glucose regulation was observed in an 85-year-old individual [11], this discrepancy might be due to the differences in study populations. Cumulative viral burden is induced by long-term HCMV infection and frequent reactivation of HCMV. In addition, the impairment of pancreatic endocrine function caused by the chronic inflammation and immune response of HCMV infection also needs time to develop [36]. HCMV infection probably requires the coexistence of multiple risk factors, which is most common in the elderly; these factors importantly contribute to T2DM. Therefore, HCMV infection may more strongly influence glucose regulation in an 85-year-old than in middle-aged people with T2DM.

HCMV has a broad cell tropism and can infect endothelial cells, epithelial cells, fibroblasts cells, smooth muscle cells, and macrophages [37,38], which are considered important in pathogenesis of vascular disease [39,40]. HCMV antigen and DNA in atherosclerotic vessels were confirmed in previous studies [41–43], suggesting that HCMV infection may damage vascular cells. A previous study demonstrated that the mechanistic contributions of HCMV in atherosclerosis involve not only “pro-atherosclerotic” effects caused by the HCMV pathogen residing in the arterial wall [44,45], but also by the increasing occurrence of systemic inflammation [46–50]. HCMV may also damage the vasculature by an immune response caused by molecular mimicry that occurs because the antigens expressed by HCMV infection are homologous to peptides expressed on uninfected host cells [51–53]. Thus, HCMV infection may play a role in progression of atherosclerosis via the aforementioned mechanisms.

Cell-specific anti-cytomegalovirus response has been found to be associated with CVD related to atherosclerosis. Recent studies have reported that male patients undergoing vascular surgery for atherosclerosis had a higher HCMV IgG titer than a matched control group of patients with high cholesterol levels [54]. Moreover, high anti-HCMV IgG titer detected by enzyme-linked immunosorbent assay (ELISA) was correlated with the carotid intima-media thickness (IMT) and elastic pressure modulus in patients with carotid atherosclerosis [55]. Therefore, higher HCMV IgG titers may predict post-coronary balloon angioplasty restenosis [56] and may become a predictor for the highest quartile of carotid intima-media thickness in HIV-infected patients with low HIV viral load after anti-virus treatment [21]. However, in the study of Puz, anti-HCMV IgG titers were not associated with atherosclerotic lesions, symptomatic stenosis, or unstable plaque in the carotid arteries [57].

Although most of the aforementioned studies demonstrated a positive relationship between high HCMV IgG level and atherosclerosis in patients with CVD, little is known about whether HCMV IgG level is an independent risk factor for diabetic atherosclerosis in T2DM. Patients with T2DM are likely to have poor glycemic control, dyslipidemia, smoking history, and hypertension; these promoters have been confirmed to be associated with atherosclerotic cardiovascular disease [58–60]. Indeed, studies found that patients with dyslipidemia, such as LDL-C ≥2.6 mmol/L, TG ≥2.3 mmol/L and HDL-C ≤0.88 mmol/L, have a significantly increased incidence of CVD [24,25]. A 1997 study by Frank et al. showed that patients with diabetes mellitus type 1 or 2 with clinical manifestations of atherosclerosis had a higher prevalence of HCMV IgG and higher IgG antibody titers than patients without atherosclerosis [61].

In the present study we found that patients with higher HCMV IgG titers corresponded to the incidence of atherosclerosis in T2DM. Our positive results agree with the study of Frank et al. [61], although the ethnic groups and methods for determining HCMV IgG titer were different. Our regression analysis demonstrated that HCMV IgG titer is an independent risk factor for atherosclerosis in T2DM patients after adjustment for important covariates, such as age, duration of DM, hypertension, and HDL-c. Patients with T2DM often have hyperglycemia and dyslipidemia, which are major contributors to atherosclerosis. Our logistic regression analysis results suggest that T2DM patients with HCMV infection are more likely to develop atherosclerosis. Risk factors such as hypertension, smoking, dyslipidemia, and aging, which have been generally been the focus of endocrinologists in preventing atherosclerosis, are increasing the occurrence of diabetic atherosclerosis in T2DM. However, HCMV infection is poorly understood and should receive greater attention. Importantly, we found that titer >127 U/ml of HCMV IgG is an independent risk factor for diabetic atherosclerosis in T2DM patients (OR=4.6, 95%CI: 1.9–11.3). This could help guide preventive treatment in order to avoid the serious consequences of atherosclerosis in T2DM patients. This might have the potential to lead to a new treatment for vascular complications, directed at antiviral therapy of HCMV or prevention by vaccination in T2DM patients.

Conclusions

The present study demonstrates that active infection and reactivation of HCMV have no influence on clinical manifestations or glucose regulation in T2DM patients. However, high anti-HCMV IgG titers are associated with incidence of atherosclerosis in patients with T2DM. Titer >127 U/ml of HCMV IgG might be an independent risk factor for increasing diabetic atherosclerosis in T2DM. The interpretation of our results is limited because this was a cross-sectional study. Therefore, a large-scale prospective investigation is necessary to further confirm and extend our findings.