Glutathione S-transferase M1/T1 gene polymorphisms and vitiligo in a Mediterranean population

Dear Editor,

Vitiligo results from interaction between genetic, biochemical, environmental and immunological factors and events. Oxidative stress is an important pathogenetic element of vitiligo: reactive oxygen species (ROS) can impair the activity of tyrosinase and repair mechanisms and alter the structure and functions of biomolecules, ultimately leading to melanocyte malfunction/death (Glassman, 2011). To avoid such damage, enzymatic and non-enzymatic ROS-neutralizing mechanisms are present in cells.

Glutathione S-transferases (GSTs) are ubiquitous enzymes that catalyze the conjugation of reduced glutathione to electrophilic centers on various exogenous and endogenous substrates (xenobiotics, drugs, poisons, and ROS). Their importance in the human antioxidant system is confirmed by the significantly increased risk of oxidative stress-related diseases [e.g. in dermatology, melanoma (Kanetsky et al., 2001), non-melanoma skin cancer (Ramsay et al., 2001), precancerous lesions (Guarneri et al., 2010), psoriasis (Richter-Hintz et al., 2003), and allergic dermatitis (Lutz et al., 2001)], in subjects bearing a ‘null’ genotype for specific GST isoforms (i.e. homozygosity for genetic deletion), resulting in the lack of phenotypic expression of the corresponding enzyme.

The prevalence of GST polymorphisms, namely GSTM1 and GSTT1 null polymorphisms, in patients with vitiligo was investigated in Korean (Uhm et al., 2007) and Chinese (Liu et al., 2009) patients, with partially overlapping results. For this reason, and considering the widely variable frequency of GSTM1/GSTT1 null in different populations (Ada et al., 2004; Guarneri et al., 2010; Lea et al., 1998; Uhm et al., 2007), we aimed to define the prevalence of GSTM1/GSTT1 null in patients with vitiligo and controls from two Southern Italian regions, Sicily and Calabria.

The protocol was approved by the Ethics Committee of the University Hospital of Messina, Italy, where the study was carried out, and the 58 patients (28 men and 30 women with focal or generalized vitiligo, active in 27 cases and stable in the remaining 31) and 150 healthy controls (71 men, 79 women) gave their written informed consent. Buccal swabs were used to obtain two samples of oral mucosa cells from each participant. We used polymerase chain reaction (see Data S1 in Supporting Information) to define the GSTM1 or GSTT1 genotype as ‘active’ (when gene is present) or ‘null’ (total gene deletion).

Table 1 shows the distribution of the four possible combinations (‘double-active’, ‘double-null’, GSTM1 null/GSTT1 active, GSTM1 active/GSTT1 null) in our population. Among the patients with vitiligo, the double-null genotype is significantly more frequent than the double-active genotype, whereas this is not the case for either single-null genotype. Significant differences also exist between the frequency of vitiligo in the double-null group and in the rest of the population (P = 0.03, odds ratio 2.333 with 95% confidence interval 1.072–5.077; data not shown in Table 1).

Table 1.   Frequencies of GSTM1 and GSTT1 polymorphisms in patients with vitiligo and in controls. Significant P values are typed in bold

Genotype	Vitiligo (n = 58)	Controls (n = 150)	Odds ratio (95% confidence interval)	P (χ^2 test)
−/−: GSTM1 null/GSTT1 null; −/+: GSTM1 null/GSTT1 present; +/−: GSTM1 present/GSTT1 null; +/+: GSTM1 present/GSTT1 present.
GSTM1
Null (−)	35 (60.34%)	82 (54.67%)	1.262 (0.681–2.338)	0.459
Present (+)	23 (39.66%)	68 (45.33%)
GSTT1
Null (−)	22 (37.93%)	37 (24.67%)	1.866 (0.977–3.566)	0.057
Present (+)	36 (62.07%)	113 (75.33%)
GSTM1/GSTT1
−/−	14 (24.14%)	18 (12.00%)	2.541 (1.026–6.292)	0.041
−/+	21 (36.21%)	64 (42.67%)	1.072 (0.501–2.292)	0.858
+/−	8 (13.79%)	19 (12.67%)	1.375 (0.502–3.770)	0.535
+/+	15 (25.86%)	49 (32.67%)	1.00 (reference)

Our results agree with the literature (Liu et al., 2009; Uhm et al., 2007) concerning the association of the double-null genotype with a significantly increased risk of vitiligo. Conversely, differences arise when correlating vitiligo with one GST null allele: Uhm et al. (2007) show a significant association of the disease with GSTM1 null and Liu et al. (2009) with GSTT1 null, in contrast to our data, which showed neither association.

The pathogenic model of vitiligo includes three non-mutually exclusive mechanisms for melanocyte death: autoimmune (humoral and cellular), cytotoxic, and neural (Glassman, 2011). Antioxidant enzymes play a prophylactic role by neutralizing ROS and other noxious substances, thus preventing their accumulation in melanocytes and their consequent toxic effects. Recent experiments suggest that oxidative stress could also exert pathogenic effects through immunomodulation. In cultures of keratinocytes from patients with vitiligo, Kostyuk et al. (2010) observed: (i) reduced expression of GSTM1 mRNA and protein, (ii) increased levels of 4-hydroxy-2-nonenal (HNE)-protein adducts and H₂O₂, and (iii) dysregulated production of major cytokines, chemokines, and growth factors (and alteration of the corresponding plasma levels). Addition of exogenous HNE to normal keratinocytes also induces a vitiligo-like cytokine pattern and H₂O₂ overproduction but, in this case, adaptive upregulation of catalase and GSTM1 genes occurs, compensating the excess of oxidative stress (Kostyuk et al., 2010). The experiment did not consider GSTT1 expression, but a GSTM1-like behavior can be reasonably hypothesized, based on the close functional similarity of the two enzymes.

In light of these considerations, an increased risk of vitiligo in GSTM1/GSTT1 double-null subjects can be expected because of the significant reduction in the basal antioxidant potential of melanocytes and the inability to upregulate GSTM1/GSTT1 expression in response to oxidative stress. Discrepancies between the studies suggest that the role and relative importance of each GST isoform in the pathogenesis of vitiligo are variable and probably correlated with several factors. Indeed, GSTs are only a part of the system of enzymatic and non-enzymatic components that maintain the redox homeostasis of the organism. The functional equilibrium among these components is the result of evolutionary selection and adaptation to environmental conditions; consequently, multiple configurations of the system exist in different populations, reflecting the wide variability of genetic, geographical, environmental, and lifestyle factors. This view is also supported by literature data on the frequency of GSTM1 null and GSTT1 null in healthy subjects, 36–54.67% and 8–52.6%, respectively (Ada et al., 2004; Guarneri et al., 2010; Lea et al., 1998; Uhm et al., 2007).

Depending on how much of the individual redox homeostasis relies on the activity of a given GST isoform, possession of the corresponding null allele may or may not significantly increase the risk of developing vitiligo, whereas the GSTM1/GSTT1 double-null genotype invariably increases such a risk because the severe decrease in the efficiency of the antioxidant system caused by the simultaneous lack of GSTM1 and GSTT1 cannot be easily compensated by other components.

Our results shed more light on the links between genetics, oxidative stress, and vitiligo. Further research is advisable to achieve a better understanding of the pathophysiologic role of GST polymorphisms. Several (ideally all) components of the antioxidant system should be investigated simultaneously to elaborate a formula for the definition of the antioxidant potential of an organism, the contribution of each component to the overall result and, ultimately, the individual disease risk and the possible preventive measures. Finally, the study of the multiple and complex connections among oxidative, autoimmune, and neural mechanisms, only partially evaluated until now, is a critical target for the clinical and biological research of the next future.