MX2 Gene Expression Tends to be Downregulated in Subjects with HLA-DQB1*0602

Abstract

Objective:

There is a close association between narcolepsy and the human leukocyte antigen (HLA)-DQB1*0602. The detailed influence and function of this specific HLA allele with regard to narcolepsy have not yet been elucidated. Our previous report identified the myxovirus resistance 2 (MX2) gene as a narcolepsy-specific dysregulated gene; however, the report had a limitation—the control groups were not HLA matched. In this study, we examined the possibility of an association between MX2 expression and HLA haplotypes.

Designs:

The expression levels of the MX2 gene in 3 groups (24 narcolepsy with cataplexy patients; 24 age-, sex-, and HLA-DQB1 genotype-matched controls; and 24 age- and sex-matched controls without the HLA-DQB1*0602 allele) were measured by quantitative real-time RT-PCR.

Results:

The expression level of the MX2 gene tended to be downregulated in subjects carrying HLA-DQB1*0602, compared with that of the control subjects without this allele. There was no difference in the MX2 expression level between the narcolepsy subjects and the HLA-DQB1 genotype-matched control subjects.

Conclusion:

Our previous finding—the narcolepsy-specific reduction of MX2 gene expression—was not replicated in this follow-up study. The expression level of the MX2 gene in white blood cells was found to be lower in subjects with the HLA-DQB1*0602 than in subjects without this allele, suggesting that there exists a relationship between the HLA-DQB1*0602 allele and MX2 gene expression. This might be a possible explanation for the strong HLA association observed in narcolepsy.

Citation:

Tanaka S; Honda Y; Honda M. MX2 gene expression tends to be downregulated in subjects with HLA-DQB1*0602. SLEEP 2008;31(5):749-751.

NARCOLEPSY-CATAPLEXY IS A COMMON CHRONIC SLEEP DISORDER CHARACTERIZED BY EXCESSIVE DAYTIME SLEEPINESS AND CATAPLEXY.¹ THE human leukocyte antigen (HLA) that comprises the major histocompatibility complex (MHC) molecules might be involved in the pathophysiology of narcolepsy. Almost all cases with narcolepsy-cataplexy share a common HLA haplotype, i.e., DRB1*1501/DQB1*0602,² suggesting an autoimmune basis; however, there is still no evidence regarding this hypothesis.³

Recently, we attempted to identify differentially expressed genes in the white blood cells of patients with narcolepsy by a differential display method⁴ and found downregulation of the myxovirus resistance 2 (MX2) gene. However, the report had a limitation—the control group subjects were not HLA matched (HLA-DQB1*0602 was detected in only 3 out of the 20 control subjects in the validation study).

In this study, we compared the expression levels of MX2 gene between a new set of narcolepsy subjects and age-, sex-, and HLA genotype-matched control subjects, as well as HLA-DQB1*0602-negative control subjects, to evaluate the contribution of the HLA allele in MX2 transcription. We used multiple housekeeping genes identified by the geNorm analysis⁵ instead of a single housekeeping gene as a normalization factor in order to increase the stringency of relative mRNA measurements.

METHODS

This research was approved by the ethics committees of the collaborative institutes, and written informed consents were obtained from all participants (all participants were Japanese). All patients were clinically diagnosed with narcolepsy with cataplexy according to the International Classification of Sleep Disorders, 2^nd edition¹, at the Neuropsychiatric Research Institute. HLA typing for the HLA-DQB1 loci was performed at the NPO HLA Laboratory (Kyoto, Japan). Twenty-four patients with narcolepsy with cataplexy (all positive for the HLA-DQB1*0602 allele); 24 age-, sex-, and HLA-DQB1 genotype-matched control subjects; and 24 age- and sex-matched control subjects without the HLA- DQB1*0602 allele were selected for the study (12 men and 12 women in each group). The distribution of the other HLA-DQB1 alleles besides DQB1*0602 was also matched between the narcolepsy and HLA-matched control groups. In each group, we selected 3 cases with DQB1*0301, 1 with *0302, 4 with *0303, 4 with *0401, 2 with *0501, 2 with *0503, 4 with *0601, 2 with *0604, and 2 cases homozygous for *0602; the frequency of these alleles corresponds to the DQB1 allele frequency in the Japanese general population.

The preparation of cDNA from blood samples was described previously.⁴ To increase sensitivity, reproducibility, and stringency, we used the geNorm analysis and selected the following most-stable 3 housekeeping genes from 7 candidate genes in white blood cells—β-actin (ACTB), glyceraldehyde-3-phosphate dehydrogenase (GAPDH), and peptidylprolyl isomerase B (PPIB).⁶ We used the geometric means of these 3 genes as a normalization factor in our experiments. The distribution of relative MX2 gene expression was compared among the narcolepsy group (n = 24), HLA-matched control group (n = 24), and HLA-negative control group (n = 24) using the Student t test. The mean and standard error (SEM) were calculated. Thompson rejection test was conducted to check outliers. A value of P < 0.05 was considered to be significant.

RESULTS

The expression level of MX2 tended to be downregulated in patients with narcolepsy and in the HLA-DQB1*0602-matched control subjects, compared with the HLA-DQB1*0602-negative control subjects (Figure 1). However the differences in relative MX2 gene levels among the narcolepsy subjects (mean ± SEM, 2.64 ± 0.19), the HLA-DQB1*0602-negative control subjects (mean ± SEM, 3.24 ± 0.26), and the HLA-DQB1*0602-positive control subjects (mean ± SEM, 2.73 ± 0.19) failed to reach the statistical significance. No significant differences in relative MX2 gene levels were observed between subjects homozygous (mean ± SEM, 2.22 ± 0.27; n = 4) and heterozygous (mean ± SEM, 2.73 ± 0.16; n = 44) for the HLA-DQB1*0602 allele.

Figure 1.

Quantitative real-time RT-PCR analysis for the MX2 gene. Each dot corresponds to the relative value of the MX2 gene expression level in each subject. Horizontal lines and squares indicate the mean ± SEM of 3 groups. Matched control: HLA-DQB1*0602-positive control subjects; No-matched control: HLA-DQB1*0602-negative control subjects.

Thompson rejection test judged 2 patients with narcolepsy, 2 HLA-DQB1*0602-positive control subjects, and 1 HLA-negative subject as outliers in each group (P < 0.01). We would like to note that, if we had excluded outliers based on rejection test, there would have been a significant difference in relative MX2 gene levels between the narcolepsy subjects (mean ± SEM, 2.43 ± 0.13) and the HLA-DQB1*0602-negative control subject group (mean ± SEM, 3.07 ± 0.22) (p < 0.05), the HLA-DQB1*0602-positive control subjects (mean ± SEM, 2.52 ± 0.14) and the HLA-DQB1*0602-negative group (p < 0.05). However, we could not find any common demographic factors explaining the MX2 gene-expression change in these outliers.

DISCUSSION

Our previous finding—the significant reduction of MX2 gene expression in HLA-DQB1*0602 positive narcolepsy subjects—did not reflect the specificity of narcolepsy but was apparently related to the HLA-DQB1*0602 positivity. MX2 gene expression levels tended to be downregulated in the white blood cells of subjects carrying this HLA allele, compared with the control group subjects without this allele; however, it failed to reach the statistical significance (P = 0.075). The possible reason why the difference remained nonsignificant in this follow-up study could be the variation of MX2 gene expression itself or the existence of outliers. The expression levels of MX2 are expected to be controlled by various factors such as interferon (INF), infection, or comorbid disease, which we did not control for in the sample selection in our study. MX2 gene expression level in subjects homozygous for the HLA-DQB1*0602 allele also tended to be low, compared with that in subjects heterozygous for HLA-DQB1*0602 allele. This tendency toward MX2 transcriptional downregulation suggests the relationship between the HLA-DQB1*0602 allele and MX2 gene expression; this might be a possible explanation for the strong HLA association observed in narcolepsy.

MHC class II genes are regulated in a cell type-specific manner and are constitutively expressed in B lymphocytes, macrophages, dendritic cells, and thymic epithelia.^7,8 However, in response to IFN-γ, other cell types can be induced to express MHC class II genes. The regulation of MHC class II genes is coordinated by a group of conserved sequence elements, termed W/Z, X1, X2, and Y boxes, that are located at a promoter proximal region upstream of all MHC-II genes. It has also been shown that IFN-γ induces MX2 gene expression.⁹ We identified IFN-stimulus responsive elements and the X2 box sequence (GTCATCT) at position −135 to −141 in the upstream region of the human MX2 gene. We speculated that specific HLA alleles responded differently to some factors, including IFN stimulus, resulting in the alteration of HLA-linked gene expression.

In conclusion, we observed the tendency toward reduction of MX2 transcription in subjects carrying HLA-DQB1*0602 that is closely associated with narcolepsy. Further investigation is required not only to study the relationship between MX2 gene and the immune system, including HLA-DQB1*0602 molecules in narcolepsy, but also to explore differentially expressed genes using HLA-matched control subjects in order to identify the factors that are involved in the development of narcolepsy in HLA-DQB1*0602-positive subjects.