Glucocorticoid receptor mutations and clinical sensitivity to glucocorticoid in Chinese multiple sclerosis patients

Multiple sclerosis (MS) is an inflammatory disease of central nervous system (CNS) which generally begins in early adulthood. Glucocorticoids (GC) have anti-inflammatory and immunosuppressive properties and are thus recommended as the first-line therapy in the management of acute attacks of MS [1]. However, in the clinical practice, we found individual variability in GC efficacy. Some patients were resistant to GC initially, but in some other patients, the response to GC attenuated with relapses (secondly resistant) [2]. In recent years, several large-scale databases reveal this point in real-world observational studies [3].

The mechanism governing the responsiveness of GC action remains elusive [4, 5]. Multiple factors can influence cellular glucocorticoid sensitivity at the level of the glucocorticoid receptor (GR) and its signaling pathway, including co-chaperones such as FK506-binding protein 5 (FKBP5) [6, 7]. The GR gene can give rise to multiple splice variants: GRα (the most abundant isoform), GRβ, GRγ. GRα is a transcription factor with transcriptional regulatory activity and essential to GC sensitivity regulation [8]. Researches in Caucasians report that polymorphisms of GR result in a modified transcript, which may have an impact on GC sensitivity and disease course [9‐11]. However, no research on Chinese MS patients has been conducted so far.

In this study, we detected GR gene alterations in the entire coding regions in Chinese MS patients and aimed to elucidate the association of GR gene and the GC response. We also investigated the differences in GRα and FKBP5 concentrations in GC responders and non-responders to explore possible strategies for modulation of glucocorticoid reactivity.

Whole-exome sequencing of the GR gene revealed 56 gene alterations, including 7 new unreported mutations and 49 single-nucleotide polymorphisms (SNPs), which are common in East Asian populations. All these seven mutations were located in exon 2 in the transcriptional activation region (TAR) of the GR gene, including five missense mutations and two premature termination codons (PTC) (Table 2). Six mutations were heterozygous and one mutation was homozygous. These seven mutations can cause changes in the coding amino acids and were predicted to be damaging or probably damaging by SIFT and Polyphen 2 software. Interestingly, unrelated patients No. 4 and No. 5 had an identical A > T missense mutation on chromosomal chr5:142779668.

The mutation distributions in the MS patients based on their response to GCs (responders and non-responders) are presented in Table 3. No significant difference in the mutation distribution was detected between the responders and non-responders to GC therapy (P = 0.416).

ELISA was performed to detect the serum GRα and FKBP5 levels before the GC impulse in patients with different GC sensitivities. The GRα content in the serum of the sensitive group was significantly higher than that of the insensitive group (P = 0.011, Fig. 1a). The content of FKBP5 in the sensitive group was significantly lower than the one in the resistant group (P = 0.025, Fig. 1b). This result indicated that the high expression of GRα and the low expression of FKBP5 might be related to the increased sensitivity of GC.

This study was designed to explore evidence for the possible correlation of GR gene alterations with response to GC. This is the first sequencing of entire GR coding exons in Chinese MS patients. We detected seven unreported mutations, whereas the presence or absence of mutations was independent of GC sensitivity. However, previously reported GR polymorphisms and mutations related to GC sensitivity were not found [14‐17]. To further elucidate the influencing factors of GR sensitivity, we compared the serum GRα and FKBP5 concentrations of GC-sensitive and GC-resistant groups. Our findings showed that the GC-sensitive patients had higher serum GRα and lower FKBP5 levels.

GC is the first-line therapy in acute attack of MS, and GR plays a critical role in the therapeutic effects of GC. Several GR gene polymorphisms have been reported to be associated with either glucocorticoid hypersensitivity or glucocorticoid resistance (including BclI, N363S, ER22/23EK, 9β) in various diseases [16]. For example, van Winsen et al. established that ER22/23EK polymorphism was associated with a more aggressive MS phenotype, determined both clinically and on MRI [18]. A study on Guillain-Barre syndrome found that the disease in BclI and Tth111I carriers was more severe, whereas the BclI carriers only had a better prognosis [19]. However, none of the above polymorphisms was detected in any of our 97 MS patients. A possible reason for that outcome is the considerable variations in the frequency of the genotype among various races. The frequency of the G/A genotype of the ER22/23EK polymorphism was 0.5% in a healthy Brazilian population and 0% in Asian and African subpopulations [20]. The absence of these polymorphisms in our results could be attributed to ethnic differences and the small sample size.

Many factors influence glucocorticoid sensitivity, including GC bioavailability, GR (gene polymorphisms, splice variant expression, transcriptional activity, and posttranslational modification), and the levels of chaperones and co-chaperones [6]. The GR gene is merely one of the factors that influence glucocorticoid sensitivity. On the other hand, FKBP5, as one of the most important co-chaperones, can influence the ligand binding, resulting in changes in the clinical response to glucocorticoid treatment. The degree of dexamethasone-induced expression of FKBP51 in PBMC served as a marker for the clinical response to glucocorticoids in patients with asthma or rheumatoid arthritis [21, 22]. Currently, investigations are undergoing on the application of FKBP51 in drug development [23‐25]. Inhibitor of FKBP5 may enhance GC sensitivity and serve as a new drug for GC-resistant patients with acute MS attack.

One of the limitations of this study was the incomplete number of serum samples subjected to ELISA assay; only 38 serum samples (out of 97 cases) were analyzed. Another important limitation was the fact that frozen serum samples rather than PBMCs were available for this retrospective study. Therefore, we failed to perform flow cytometer assay of GRα and FKBP5 on the cellular level. In addition, the interval from the onset to the GC impulse was variable, which might have led to bias in the GC response and efficacy. Finally, EDSS changes were hardly detectable at higher values (> 3); MS functional composite (MSFC) or MS quality of life 54(MSQoL-54) [26, 27] may be more sensitive and suggestive. However, MSFC and MSQoL-54 were unavailable in our study.