SEC62 Regulates HLA-E Expression in Diffuse Large B-Cell Lymphoma to Function As a Mechanism of Immune Evasion

The expression of HLA-E, a non-classical HLA class I molecule, is an emerging mechanism of immune escape in cancer. Increased HLA-E is associated with poor prognosis in multiple malignancies and is a mechanism of resistance to immune-based therapies via an inhibitory interaction with the NKG2A/CD94 receptor on NK and T cells. We identified high expression of HLA-E in diffuse large B-cell lymphoma (DLBCL) and hypothesized that this may constitute an immune evasion mechanism that could be therapeutically targetable in this disease.

We found that DLBCL has the highest median gene expression of HLA-E when compared to 32 solid tumors and leukemias profiled in The Cancer Genome Atlas. In existing datasets, we observed that HLA-E is also upregulated in DLBCL relative to its normal cell of origin, germinal center B cells (P = 0.0001). By immunohistochemistry in DLBCL cases, HLA-E was expressed in 37 of 48 cases (77%) and present even in some cases with low or absent classical HLA class I (HLA-I). HLA-E is also abundant in transformed DLBCL and we confirmed presence of NKG2A+ cells in the tumor microenvironment in four excisional biopsies using highly multiplexed immunofluorescence. Finally, there is existing evidence that DLBCL patients have higher frequencies of NKG2A+ NK cells than healthy donors. Altogether, these findings support the notion that the HLA-E/NKG2A checkpoint may function as a mechanism of immune evasion in DLBCL.

Given that HLA-E expression can be retained when classical HLA-I is downregulated, we questioned if there may be distinct mechanisms of regulation. First, to comprehensively define the mechanisms of HLA-E regulation, we completed two genome-wide CRISPR knockout (KO) screens in germinal center B (GCB) and activated B cell (ABC) DLBCL cell lines, SU-DHL-8 and TMD8, respectively. The top and bottom 5% of HLA-E expressing cells were isolated by fluorescence activated cell sorting. Genes were ranked based on the enrichment of their corresponding sgRNAs in the top vs. bottom HLA-E expression, which identified 379 candidate regulators. As validation, we then performed a secondary screen using a custom library consisting of eight newly designed sgRNAs per gene and identified HLA-E specific regulators through parallel assessment of HLA-I. The custom library was screened in SU-DHL-8 and TMD8, plus an additional GCB (SU-DHL-4) and ABC DLBCL line (HBL1). Of the 379 primary screen hits, 337 (88.9%) validated in the original SU-DHL-8 and TMD8 cells and 284 (74.9%) validated in all cell lines (P < 0.0005).

Confirming the robustness of our system, top hits enriched in HLA-E and HLA-I low populations included known antigen presentation pathway genes, while HLA-E was the top hit in the HLA-E low population without altering HLA-I expression. Most positive and negative regulators of HLA-E were shared with HLA-I, but nine genes decreased HLA-E to a greater degree than HLA-I in three of four cell lines. These genes were all involved in protein processing and trafficking pathways, including aminopeptidases (ERAP1, HM13, and SPCS3) and regulators of transport through the endoplasmic reticulum (ER) and Golgi apparatus (SEC62 and USO1). The only shared HLA-E specific regulator in all cell lines screened was SEC62, an ER membrane protein that facilitates the post-translational import of a subset of proteins into the ER.

To further validate SEC62 as a positive regulator of HLA-E, we generated isogenic SEC62 KOs in SU-DHL-4, HBL1, and TMD8 DLBCL cell lines. We found that SEC62 KO reduced surface and whole cell HLA-E protein expression without decreasing HLA-I in flow cytometry, western blotting and immunofluorescence assays. However, HLA-E mRNA levels were unchanged, consistent with regulation via a post-transcriptional mechanism. By co-culturing isogenic SEC62 KO DLBCL cell lines with a Jurkat-NKG2A reporter system, we functionally confirmed that SEC62 KO reduced engagement with the NKG2A/CD94 receptor.

Together, our findings reveal a potential role for HLA-E in DLBCL as a mechanism of immune escape and identify SEC62 as a distinct regulator of HLA-E in this malignancy.

Doench:Tango Therapeutics: Consultancy, Current equity holder in publicly-traded company; Laboratory for Genomics Research: Other: Paid scientific advisor, funded in part by GSK; Functional Genomics Consortium: Abbvie, Bristol Myers Squibb, Janssen, and Merck: Research Funding; Innovative Genomics Institute: Other: Paid scientific advisor, funded in part by Apple Tree Partners; Pfizer: Consultancy; Servier: Consultancy; PhenomicAI: Consultancy; Microsoft Research: Consultancy; BioNTech: Consultancy. Wu:Repertoire: Membership on an entity's Board of Directors or advisory committees; BioNtech, Inc: Current equity holder in publicly-traded company; Pharmacyclics: Research Funding; Aethon Therapeutics: Membership on an entity's Board of Directors or advisory committees; Adventris: Membership on an entity's Board of Directors or advisory committees.