Epigenetic Loss of the HLA-DR15 Expression on Hematopoietic Stem Progenitor Cells in Patients with Acquired Aplastic Anemia Characterized By Cyclosporine Dependency: A Novel Mechanism Underlying the Immune Escape of Hematopoietic Stem Progenitor Cells
Noriaki Tsuji,Kohei Hosokawa,Ryota Urushihara,Mikoto Tanabe,Hiroyuki Takamatsu,Ken Ishiyama,Hirohito Yamazaki,Shinji Nakao
DOI: https://doi.org/10.1182/blood-2020-136893
IF: 20.3
2020-11-05
Blood
Abstract:[Background] HLA-DR15 (DR15) has been implicated in the susceptibility to immune-mediated bone marrow (BM) failure, such as acquired aplastic anemia (AA), wherein the hematopoietic function depends on cyclosporine (CsA), paroxysmal nocturnal hemoglobinuria (PNH) with BM failure, and low-risk myelodysplastic syndrome responsive to immunosuppressive therapy. However, how DR15 contributes to the development of such immune-mediated BM failure remains unclear. Although the copy-number neutral loss of heterozygosity in chromosome 6p (6pLOH) of hematopoietic stem progenitor cells (HSPCs) in AA patients sometimes involves the HLA-DRB1 region, the frequency of the resultant DR15 loss is low, suggesting little involvement of this DR allele in the escape of HSPCs from cytotoxic T-cell attack. Several studies have recently reported that acute myeloid leukemia cells that relapse after allogeneic hematopoietic stem cell transplantation often lack HLA class II expression through an epigenetic mechanism and thereby escape the graft-versus-leukemia effect. The epigenetic loss of HLA class II expression may also occur in HSPCs that survive the immune attack in AA patients in remission. [Objectives/Methods] To test this hypothesis, we determined the HLA-DR expression on HSPCs defined by lineage-CD45dimCD34+CD38+ cells as well as their subpopulations, including common myeloid progenitors (CMPs), megakaryocyte-erythroid progenitors (MEPs), and granulocyte-monocyte progenitors (GMPs), in the peripheral blood of 52 AA patients (34 with DR15 and 18 without DR15) and 20 healthy individuals using flow cytometry (FCM) with anti-pan-HLA-DR antibodies. All patients were in remission after ATG-based therapy (ATG+CsA±thrombopoietin receptor agonists [TPO-RA] or anabolic steroids [AS], n=19), CsA-based therapy (CsA±TPO-RA or AS, n=27), and others (n=6), and 18 required low-dose CsA to maintain remission. Eighteen (35%) had HLA-class I allele-lacking (HLA-class I[-]) leukocytes due to 6pLOH and/or allelic mutations while 33 (63%) had 0.003-83.8% (median 0.194%) GPI-anchored protein-deficient (GPI[-]) granulocytes. HLA-DR(-) HSPCs detected in some patients were sorted together with their HLA-DR(+) counterparts and subjected to incubation in the presence of interferon gamma (IFN-γ) to see whether or not the DR expression was restored; in addition, they were subjected to RNA sequencing to compare the gene expression profiles between DR(-) and DR(+) HSPCs. [Results] Five (9.6%) of the 52 AA patients had 28.6% to 42.3% (median 34.0%) DR(-) cell populations in HSPCs, which were not detected in either monocytes or B lymphocytes of the same patients or in HSPCs of any healthy individuals (Figure 1a). All 5 patients possessed either HLA-DRB1*15:01 (n=3), DRB1*15:02 (n=1), or DRB1*15:01/15:02 (n=1), with the other DRB1 alleles differing among individuals, and their hematopoietic function depended on CsA, except for 1 patient (Case 5) whose HSPCs consisted of 69% GPI(+) and 31% GPI(-) cells. Of particular interest, Case 5's DR(-) cells were detected in GPI(+) HSPCs but not in GPI(-) HSPCs (Figure 1b). None of the 5 patients possessed HLA-class I(-) leukocytes, which were detected in 18 (38%) of 47 patients not possessing DR(-) HSPCs. In contrast to the patients possessing DR(-) HSPCs, CsA dependency was only observed in 13 (28%) of the 47 AA patients without DR(-) HSPCs. Incubation of sorted DR(-) HSPCs in the presence of IFN-γ for 72 h resulted in full restoration of the DR expression in all HSPC subpopulations (Figure 2). A comparison of the transcriptome profile between DR(-) and DR(+) HSPCs revealed that the signature of differentially expressed genes was enriched in immune response-related genes. [Conclusions] HSPCs that lacked DR due to an epigenetic mechanism were frequently detected in AA patients with DR15 characterized by CsA dependency. Although the loss of expression occurred in both DR alleles, the fact that only DRB1*15:01 or 15:02 was an allele shared by the five patients indicates that the DR loss phenomenon targeted DR15. The DR15(-) HSPCs may escape from antigen-specific CD4+ T-cell attack, which cannot be completely abolished by CsA. As demonstrated by findings of Case 5 showing the presence of a DR(-) cell population only in GPI(+) HSPCs, the lack of GPI may be a mechanism underlying substitution for the DR15 loss. Disclosures Takamatsu: Ono pharmaceutical: Honoraria, Research Funding; SRL: Consultancy, Research Funding; Janssen Pharmaceutical: Consultancy, Honoraria, Research Funding; Bristol-Myers Squibb: Honoraria, Research Funding; Adaptive Biotechnologies: Honoraria. Ishiyama:Alexion: Research Funding; Novartis: Honoraria. Yamazaki:Kyowa Kirin: Honoraria, Research Funding; Novartis: Honoraria. Nakao:Alexion: Research Funding; Novartis: Honoraria; Kyowa Kirin: Honoraria; Symbio: Consultancy.
hematology