HLA-E circulating and genetic determinants in schizophrenia and bipolar disorder

Abstract

Schizophrenia (SZ) and bipolar disorders (BD) are severe mental illnesses that lack reliable biomarkers to guide diagnosis and management. As immune dysregulation is associated with these disorders, we utilized the immunoregulatory functions of the natural killer cell inhibitory HLA-E locus to investigate the relationships between HLA-E genetic and expression diversities with SZ and BD risk and severity. Four hundred and forty-four patients meeting DSM-IV criteria for SZ (N = 161) or BD (N = 283) were compared to 160 heathy controls (HC). Circulating levels of the soluble isoform of HLA-E molecules (sHLA-E) were measured and HLA-E*01:01 and HLA-E*01:03 variants genotyped in the whole sample. sHLA-E circulating levels were significantly higher in both SZ and in BD patients compared to HC (pc < 0.0001 and pc = 0.0007 for SZ and BD, respectively). High sHLA-E levels were also observed in stable SZ patients and in acute BD patients experiencing depressive episodes when comparisons were made between the acute and stable subgroups of each disorder. sHLA-E levels linearly increased along HLA-E genotypes (p = 0.0036). In conclusion, HLA-E variants and level may have utility as diagnostic biomarkers of SZ and BD. The possible roles of HLA diversity in SZ and BD etiology and pathophysiology are discussed.

Figure 1

Analysis of the distribution of circulating sHLA-E levels among SZ, BD patients and healthy controls. Statistically significant increase in both patient subgroups (mean sHLA-E: 316.6 vs. 189.3, pc < 0.0001 in SZ vs. HC and 260.9 vs. 189.3, pc = 0.0007 in BD vs. HC) and between SZ and BD patients (mean sHLA-E: 316.6 vs. 260.9, pc = 0.0009 in SZ vs. BD). p value corrected using the Bonferroni test.

Figure 2

Relationship between circulating levels of sHLA-E, depressive scores measured by MADRS and functioning measured by GAF scores (p = 0.017 and p = 0.022 respectively).

Figure 3

Analysis of sHLA-E distribution according to acute and stable phase of SZ or BD (265.14 vs. 387.05, pc = 0.025 in acute and stable SZ patients respectively) and (278.9 vs. 232.48, pc = 0.024 in acute BD episode and euthymic phase respectively). p value was corrected using the Bonferroni test.

Figure 4

Analysis of sHLA-E levels according to the genetic distribution of the HLA-E rs1264457 AA, AG and GG genotypes in the whole cohort of patients and in the subgroup of SZ patients. (A) In the whole cohort of patients including SZ and BD, sHLA-E levels increase in a linear manner along HLA-E genotypes (AA = 218.5, AG = 270.27, GG = 315.28, overall p = 0.0036). (B) In patients with SZ, sHLA-E levels are also observed to increase along HLA-E genotypes (AA = 265.65, AG = 298, GG = 364.58, overall p = 0.0147). (C) In patients with BD, no difference was observed concerned the distribution of sHLA-E according to the HLA-E rs1264457 AA, AG and GG genotypes (AA = 202.1, AG = 259.9, GG = 277.1, overall p = 0.1356). All figures were generated using GraphPad Prism version 7.00 for Windows, GraphPad Software, La Jolla California USA, www.graphpad.com.