The Clinical Significance of MIF Serum-Level as an Early Predictive Marker of Cardiovascular Risk in Patients with Vitiligo

Background: In the pathogenesis of vitiligo, pro-inflammatory cytokines have been linked to insulin resistance, dyslipidemia, and atherosclerosis; which are associated with cardiovascular diseases (CD). Macrophage Migration Inhibitory Factor (MIF) is a pro-inflammatory cytokine that is well-described in the pathogenesis of several autoimmune disorders. Also, MIF has been involved in the development of atherosclerosis and has emerged as a cardiovascular risk (CR) factor. Objective: Our objective was to analyze the relationship between MIF and CR in patients with vitiligo. Method: We evaluated dyslipidemia, insulin resistance, and CR according to the Castelli Atherogenic Index, ultrasensitive C-reactive protein (usCRP), and carotid atherosclerosis by the intima-media thickness (IMT) in 50 patients and 50 healthy subjects. MIF concentrations were measured by ELISA assay. MIF tissue expression was evaluated by automated immunohistochemistry. The relationship of these cardiovascular alterations was investigated concerning the extent and activity of vitiligo. Results A higher prevalence of low HDL cholesterol (HDL-c) and moderate/high CR according to Castelli Atherogenic Index was observed in the vitiligo group. Subjects with generalized/universal vitiligo had more CR as stated in higher usCRP values, Castelli Index, Carotid IMT averages, and lower HDL-c levels. MIF serum levels were higher in patients with active disease, HOMA-IR >2.4, HDL <40mg/dL, and high-risk usCRP. Serum MIF levels were negatively correlated with variables of CR in vitiligo. It was increased MIF expression in depigmented skin. Conclusions This study provides the first evidence for the involvement of MIF in the evolution of CR in vitiligo.


Introduction
Vitiligo is a common dermatologic condition characterized by a loss of melanocytes [1]. The etiology and the exact physiopathological mechanism of this disease are unknown; however, a melanocyte death mediated by the immune system has been suggested [2].
Added to this, antibody-mediated, cell-mediated, and cytokinemediated mechanisms have all accumulated supportive evidence that the damage to the homeostasis of melanocytes is caused by autoimmune processes [3,4].
High levels of tumor necrosis factor-alpha (TNF-α), interleukin-1 alpha (IL-1α), and interleukin-6 (IL-6) have been observed in the lesions of vitiligo patients; [5]. Similarly, high granulocyte-macrophage colony-stimulating factors have been found in the serum [6]. These cytokines have also been related to the development of insulin resistance and dyslipidemia that lead to vascular damage and atherosclerosis [7,8]. Thus, it is possible that patients with vitiligo would have an increased risk of cardiovascular diseases (CDs).
The common denominator of most CDs is a chronic inflammatory pathology of the vessel wall called atherosclerosis [9].
The atherogenesis process development involves specific cellular and molecular responses that culminate into an acute event due to plaque erosion or rupture. The primary defect occurs at the vascular endothelial cells which become dysfunctional, develop pores in the junctions, and provoke inflammatory cells recruitment to the subendothelial region. A chronic inflammatory-fibroproliferative response characterizes the subsequent growth of the plaque [10].
The movement of monocytes and T cells from the bloodstream to the damaged vascular wall, together with the platelets, perpetuates the inflammatory environment through the secretion of several vasoactive chemokines and cytokines [11]. There is compelling evidence about the role of immunity in atherosclerosis, and CDs are the primary cause of death among patients with various autoimmune diseases like lupus and rheumatoid arthritis [10].

The Macrophage Migration Inhibitory Factor (MIF) is an
inflammatory cytokine with an important role in apoptosis. This cytokine has a well-established role in the onset and development of several immune-inflammatory diseases [12][13][14][15] but it has been poorly studied in dermatological disorders [16]. The powerful effects of MIF on autoimmune dermatological illness include the activation and differentiation of Th1 and Th17 cells, which are characterized by producing the major pro-inflammatory cytokines involved in its pathogenesis, such as IL-1, IL-6, IFN-γ, TNF-α, and IL-17 [17,18]. In recent years, MIF has also been broadly implicated in the atherogenesis process [19] as it can regulate the inflammatory and immune signaling in atherosclerosis, potentiate neointimal thickening by promoting the monocyte and T cell recruitment and it induces proliferation of the smooth muscle cells in vessels [20][21][22].
Several research groups have shown that high MIF serum levels might be a novel biomarker for the risk of cardiovascular disease [23][24][25]. In this study, we aimed to assess the potential of serum MIF as an early predictive marker of atherosclerosis and cardiovascular risk (CR) in patients with vitiligo.

Patients with vitiligo and control subjects
A cross-sectional study was carried out in the Dermatology Institute of Jalisco "Dr. José Barba Rubio" after authorization from the ethics committee (No.17-015) of the institution and the signing of informed consents from all participants. Our study enrolled 50 patients with vitiligo and 50 subjects with no medical record of vitiligo, autoimmune or chronic disease, named as Control Subjects (CS), matched by age, gender, and body mass index (BMI). All individuals were of Mexican origin and at least 18 years of age. Subjects with a history of a congenital or acquired condition which during its evolution had metabolic disorders and atherosclerosis; those who during the physical examination had findings compatible with diseases associated with metabolic, lipid disorders, or atherosclerosis, and those with a current consumption of hypoglycemic, lipid-lowering medications, oral steroids or other systematic pharmacological treatment, and pregnant women and smokers, were excluded from the study. males, >7 in females) (30). Ultrasensitive C-reactive protein (usCRP) was measured using the turbidimetric method. CR was classified as low (<1 mg/L), medium (1-3 mg/L), and high (>3 mg/L) [31][32][33].
Carotid atherosclerosis was evaluated using the intima-media thickness (IMT) defined as the measurement of the mean thickness over the distal 1 cm of the wall of the left and right common carotid arteries length. The measurements were done in triplicate, and the values obtained were averaged to obtain the IMT [34,35].

MIF Expression in situ by automated immunohistochemistry
A total of 20 specimens of depigmented and perilesional skin from vitiligo patients were examined. Besides, ten samples of healthy skin tissue from the surface of patients who underwent dermatological surgery for the removal of benign lesions were incorporated in the analysis. subjects included signed the written

Statistical analysis
The obtained data were analyzed with the statistical software Bonferroni correction was employed. As a measure of association, the odds ratio (OR) was determined with a 95% confidence interval (95% CI). The Spearman linear correlation analysis was performed to assess the association of MIF serum levels with the representative CR variables evaluated in the study. A p-value ≤0.05 was considered statistically significant.  In both study groups, there was a predominance of females (62%), and the average age was 45 ± 13 years (range 18-74 years).

Description of the clinical and demographic characteristics of the study subjects
The average BMI in the vitiligo group was 28 ± 6 (range  and 28   The concentration of MIF serum levels was higher in patients with vitiligo than in CS (3.1 ng/mL vs. 2.6 ng/mL, respectively), but this did not reach statistical significance (p=0.7) ( Figure 1A).

High serum levels of MIF are related to the degree of activity of vitiligo
However, when the group with vitiligo was subclassified according to the degree of activity, a significant difference was observed between active versus stable patients, regardless of the disease extent (p=0.02 for localized vitiligo, and p=0.01 for generalized vitiligo) ( Figure 1B). Moreover, patients with a disease activity index score of +1 had the highest levels in comparison with the other scores (-1, 0, +2, +3, +4) (p=0.008, Figure 1C).

Insulin resistance and dyslipidemia in patients with vitiligo
Insulin resistance was evident in 25 patients (50%) and in  (Figure 2A).    Figure 4A).

Patients with vitiligo have a higher Atherogenic Index of Castelli
Moderate/high CR was present in 40% of subjects with vitiligo and in 22% of CS (p=0.05, OR=2, 95% CI: 0.91-6) (data not shown).
There were no significant differences in the MIF serum levels among   vs. 1.750, respectively; p=0.02) ( Figure 5A); however, there were no significant differences between patients with vitiligo vs. the CS group concerning the degree of atherogenic or CR (Table 3).     Carotid IMT average between groups showed no significant differences (Table 4). Average IMT of the left carotid artery, right bulb, and left bulb was statistically higher in the group with generalized/universal vitiligo than in the group with localized vitiligo (p=0.04, p=0.02, and p=0.02 for the right bulb, left common carotid, and left bulb; respectively). We found no statistically significant differences concerning disease activity (Table 5).  (Table 1).

Serum MIF values are negatively correlated with variables of CR in patients with vitiligo
However, subsequent multivariate analysis showed only a significant correlation between MIF serum levels with the insulin serum levels and the left bulb IMT (p=0.03, and p=0.04; respectively) ( Table 6).

High levels of MIF in the patches of depigmented skin of vitiligo patients
The analysis of the optical densities for the brown staining showed that the vitiligo lesioned skin had more MIF protein expression than the perilesional skin in active vitiligo (mean difference of 847.9, optical density; p=0.006). Also, patients with active vitiligo exhibit higher MIF expression than those stables (p=0.006 for the lesioned skin, and p=0.005 for the perilesional skin). All the analyses performed showed that patients with vitiligo have more skin MIF expression in comparison with the healthy volunteers ( Figure 6).  [36], through overexpression of adhesion molecules in endothelial cells and infiltrated cells [37]. It also directs migration, since MIF exhibits chemokine-

American Journal of Biomedical Science & Research
Copy@ Lizbeth Riera-Leal like activities through interactions with the chemokine receptors CXCR2 and CXCR4 [19,36,37]. Furthermore, MIF regulates smooth muscle cell migration and proliferation and enhanced degradation of extracellular matrix proteins by overexpression of MMP [19].
On the other hand, it has been described that MIF can have a cardioprotective effect by activating AMPK that promotes cellular glucose uptake by inducing the trafficking of GLUT4-containing membrane vesicles to the membrane, inhibits pro-apoptotic cascades, and enhances pro-survival signaling [38] (Figure 7).

Discussion
The pathogenesis of vitiligo remains unknown, however, many pathways of autoimmune, genetic, and environmental origin attempt to explain the loss of melanocytes in this disease. Genetic factors possess key roles in vitiligo as the disease is characterized by incomplete penetrance and multiple susceptibility loci [39].
In this respect, we observed that one-half of our patients had a positive family history of vitiligo, whereas only 8% of CS had this condition. This can be interpreted as that having a family member with vitiligo confers an 11.5-times greater risk of presenting the disease and is agree with that reported by Zhang et al. [40] who found approximately 50% heritability in vitiligo.
Studies performed on vitiligo skin biopsies have produced interesting data implicating activated dendritic cells, Th1, and Th17 cells in the effector responses targeting melanocytes [41]. Similar findings have been found in the serum of patients, reporting an increase in the proinflammatory Th1 and Th17 profile cytokines [3,41,42].
MIF is a proinflammatory cytokine with a role not fully understood in vitiligo; however, it is known that its deletion or neutralization severely impairs the pro-inflammatory cytokine production, such as TNF-α, IL-1β, IL-6, IL-23, and IL-17, [17,18,43] all of which are important in the pathogenesis of vitiligo. In the present study, we observed that MIF serum levels were higher in patients with vitiligo than in CS, in agreement with the only three reports, to our knowledge, published to date on this issue [44][45][46]. However, unlike these authors, we did not find statistically significant differences in our population. Also, it was observed a significant positive relationship between MIF serum levels and the disease activity, as has been reported previously in Ma and Farag works, [44,46] but not by Serarslan [45] Regarding this, the pleiotropic immunologic and endocrine activities exerted by MIF suggest that its actions may be connected with complex signal transduction pathways. MIF has been associated with the pathogenesis of several chronic diseases, such as rheumatoid arthritis, [47] colitis, [48] types 1 and 2 Diabetes Mellitus, [49] chronic respiratory diseases [50,51], and cancer [52,53]. Moreover, MIF deficiency reduces chronic inflammation [54]. Since, vitiligo is an autoinflammatory disease and its activity is related to the appearance of new lesions or extension of the previous ones, grade +1 would indicate the presence of chronic and sustained inflammation. More studies are needed to find the mechanisms that regulate MIF´s chronic secretion or function and to explain why elevated levels are not observed in stages of early activity.
The increase of pro-inflammatory cytokines such as IL-1α, IL-6, and TNF-α produce a chronic subclinical inflammatory state that would lead vitiligo patients to an increased predisposition for metabolic disorders, atherosclerosis, and cardiovascular damage [5,6] [30]. In disagreement with Karadag et al. [29] we found no significant differences between groups concerning the insulin resistance and the elevated LDL levels, possibly because these authors did not pair patients and controls according to the BMI.
Based on the Framingham study, the Castelli Atherogenic Index and the usCRP are prognostic indicators of the atherosclerotic processes and identifies individuals with risk of heart disease [49][50][51][52][53][54][55][56][57]. In this work, a higher average of the Castelli Atherogenic Index and the usCRP were observed in the subjects with vitiligo. Carotid atherosclerosis measured via IMT did not demonstrate significant differences between groups. Nonetheless, we noted that the average the IMT of the right carotid in subjects with vitiligo was 0.99 mm, which alerts us to the fact that there is considerable damage at this level. Carotid IMT showed a positive relationship with the extent of the vitiligo. This could indicate that the greater the area of vitiligo, the higher the inflammatory state mediated by pro-inflammatory cytokines, leading to more significant damage to the vascular walls [58][59][60].
Considering that MIF has also been related to inflammation, atherosclerosis, and CDs, [38,61,62] this study also evaluated its possible role as an early predictor of these diseases in vitiligo.
There was a significant association between MIF levels and the MIF serum levels are typically found between 2 and 10 ng/mL in human plasma [73]. Notably, the mean value of MIF in our group was 3.02 ng/mL. If we also consider that our patients had an average disease evolution of 13 years, it would seem possible to hypothesize that the negative correlation of MIF with cardiovascular study risk variables supports the cardioprotective role of this cytokine in acute events. Besides, different genetic variants in the MIF gene may be masking its effect on our study group, as reported by De la Cruz-Mosso et al. [14] in the population from Western Mexico.
Another relevant fact is that cardiomyocytes express and store MIF at a high level. During a myocardial infarction, the expression of MIF by circulating mononuclear cells remained unchanged, [74] indicating a cardiac source of circulating MIF in the acute phase of the pathology [24]. Thus, according to that observed in the skin of vitiligo patients, it is possible that the most important source of MIF concentrations could be the lesioned skin of the patients.
Unfortunately, we were not able to correlate the optical densities of MIF tissue immunopositivity with serum levels of the protein because they were not equivalent samples. According to our results, the only study in the literature conducted by Ma et al. found

American Journal of Biomedical Science & Research
Copy@ Lizbeth Riera-Leal that MIF mRNA levels were significantly higher in lesioned than in healthy skin [44].
In this research, the study group with vitiligo was stratified as a strategy that allows us to simplify and understand the most complex problems. In particular, in our study, we identify the set of parameters that could influence the relationship between MIF and vitiligo. However, the relatively small sample in each subgroup analysis could be an important limitation in the interpretation of our results. Greater efforts are required to demonstrate the causal relationship between MIF, vitiligo, and its possible complications.

Conclusion
In conclusion, our study supports the presence of a chronic inflammatory state in vitiligo that makes patients more vulnerable to cardiovascular disease and links to MIF as a master regulator of the immune response within the pathophysiology of the disease.
Also, this work questions the role of MIF as a CR factor, since previous studies associate it more with acute CDs, leaving the doubt as to whether MIF is a predisposing factor or a protective-repairing mechanism of acute and severe heart injury. To our knowledge, the present study provides the first experimental evidence for the direct involvement of MIF in the evolution of CR variables in vitiligo patients.

Declaration of competing interests
There is no conflict of interest of any of the authors with the results of this study.

Funding
This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

Ethical approval
The ethics committee of The Dermatological Institute of Jalisco Dr. José Barba Rubio approved this study (No. 17-015).

Guarantor
LRL is the guarantor of this work.