Advertisement

CML cells actively evade host immune surveillance through cytokine-mediated downregulation of MHC-II expression

Anuradha Tarafdar, Lisa E. M. Hopcroft, Paolo Gallipoli, Francesca Pellicano, Jennifer Cassels, Alan Hair, Koorosh Korfi, Heather G. Jørgensen, David Vetrie, Tessa L. Holyoake and Alison M. Michie

Data supplements

Article Figures & Data

Figures

  • Figure 1.

    MHC-II and CIITA are selectively downregulated in CML stem/progenitor cells. (A) Ingenuity’s antigen presentation pathway overlaid with deregulation in CML vs normal G0 CD34+ cells (n = 5 CML, n = 2 non-CML). Deregulated members for MHC II-α and II-β (including HLA-DP, DQ, DR) are shown expanded in the bottom left. Significantly downregulated components highlighted in green. (B) GSEA analysis demonstrating significant (false discovery rate [FDR], ≤0.15) downregulation of the extended MHC-II gene family in a larger, complementary microarray data set of G0/quiescent (left) and dividing (right) CML and normal cells.14 (C) Heat map showing deregulation (as represented by logFC; see color scale below heat map) of the MHC-II gene family across multiple CML populations compared with corresponding non-CML cells from distinct microarray data sets, as indicated.14,15 The data derived from the original data set13 is highlighted by the orange bar above the corresponding columns. Average normalized MFI of surface MHC-II expression in primary (D) CD34+CD38+ CML and non-CML stem/progenitor cells or (E) CD34+CD38 CML LSCs and non-CML HSCs cultured for 48 hours in the presence or absence (UT) of IFN-γ (100 U/mL; n = 6 for CD34+CD38+; n = 4 for LSC; ± SEM). CD34+CD38 cells, when stringently gated, represent ∼1% to 5% of bulk CD34+ cells and overlap considerably with either Hstlo/Pylo or CD34+ CFSEmax populations described previously.13 (F) Average gene expression of CIITA in bulk CD34+ CML and non-CML stem/progenitor cells cultured for 48 hours ± IFN-γ (mean fold change; n = 6; ± SEM). Statistical significance was calculated between UT CML sample and all other samples, and if significant, it is indicated by asterisks above the bars. Additional comparisons between samples are indicated by lines.

  • Figure 2.

    MHC-II and CIITA downregulation occurs independent of BCR-ABL kinase activity in CML stem/progenitor cells. Primary (A) CD34+CD38+ CML cells and (B) CD34+CD38 CML LSCs were cultured for 48 hours with TKIs (5 µM NIL, 150 nM dasatinib, 5 µM IM) or no drug control (NDC) in the presence or absence of IFN-γ. Average normalized MFI of MHC-II expression was determined using flow cytometry (n = 5; ± SEM). (C) CIITA expression levels in CD34+CD38+ CML cells were analyzed by quantitative reverse transcription polymerase chain reaction (n = 5; ± SEM; calibrated to UT NDC sample). Statistical significance was calculated between UT CML sample and all other samples and, if significant, is indicated by asterisks above the bars.

  • Figure 3.

    Extended treatment of CML cells with IM does not normalize MHC-II or CIITA expression. Primary CD34+ CML were treated with 5 µM IM for 7 days. Average surface MHC-II expression levels were determined by flow cytometry, normalized to CML UT (d0) sample for (A) CD34+CD38+ CML cells and (B) CD34+CD38 CML LSCs. (C) The average expression of MHC-II encoding genes HLA-DR, HLA-DP, and HLA-DQ was determined in CD34+ CML cells by quantitative reverse transcription polymerase chain reaction (n = 7, calibrated to UT (d0) CML sample). (D) The average gene expression of CIITA was determined in CD34+ CML cells by quantitative reverse transcription polymerase chain reaction. Statistical significance was calculated between d0 UT CML sample and all other samples and, if significant, is indicated by asterisks above the bars. Additional comparisons between samples are indicated by lines. NS, not significant.

  • Figure 4.

    JAK inhibition elevates MHC-II expression in CML stem/progenitor cells. (A) Quantitative RT-PCR of RNA/cDNA generated from either non-CML or CML stem/progenitor samples for IL-4 transcripts revealed higher expression levels of IL-4 in CML cells. Data are expressed relative to the reference gene HPRT1. Primary (B) CD34+CD38+ CML cells and (C) CD34+CD38 CML LSCs were treated with IFN-γ and/or JAK inhibitor (RUX, 200 nM), as indicated. The average MHC-II expression levels were determined by flow cytometry, normalized to CML UT sample. NS, not significant. (D) Representative histograms showing MHC-II expression on CD34+CD38+ and CD34+CD38 CML cells treated with IFN-γ and/or RUX, as indicated; Primary CD34+ CML stem/progenitor cells were treated with IFN-γ and/or RUX, as indicated earlier. Thereafter, 300 cells were sorted for either (E) CD34+CD38+ CML and (F) CD34+CD38 CML LSC populations, and the average gene expression of CIITA was determined by quantitative reverse transcription polymerase chain reaction (n = 6 for CD34+CD38+, n = 3 for LSC; calibrated to UT CML sample). Statistical significance was calculated between UT CML sample and all other samples and, if significant, is indicated by asterisks above the bars.

  • Figure 5.

    Elevated MHC-II expression is associated with an increase in CD34+CML cell immunogenicity. (A) Activated and CTV labeled T cells (MNCs from healthy donors) were cocultured with bulk CD34+ CML cells for 72 hours (± treatment as indicated. Block, 10 μg/mL anti-pan HLA-class II blocking Ab). Proliferation of the responder cells was measured as a reduction in the MFI of CTV-labeled CD4+CD69+ T cells (n = 6 ± SEM; n = 3 for MHC-II blocking assays). (B) The autocrine or paracrine growth factor/cytokine signaling pathways that regulate MHC-II expression. Statistical significance was calculated between UT CML sample and all other samples and, if significant, is indicated by asterisks above the bars.

Tables

  • Table 1.

    Source of clinical samples

    Sample no.Sample IDSourceSexStageFISH/t(9:22)Figures
    1Spirit CML 0762PBMCPND1, 2, 5
    2CML 339LeukapheresisFCP+2, 4, 5
    3CML 340LeukapheresisMCP+*4, 5
    4CML 341 (Fresh)LeukapheresisMCP+*1, 2
    5CML 378LeukapheresisFCP+1, 2
    6CML 381LeukapheresisFCP+1, 2
    7CML 385LeukapheresisMCP+*1, 2
    8CML 388LeukapheresisFCP+*1, 2
    9CML 411 (Fresh)PBFCPND4, 5
    10CML 412 (Fresh)LeukapheresisMCPND2, 4, 5
    11CML 441LeukapheresisMCP+2
    12CML 442PBFCP+4, 5
    13CML 450PBMCP+5
    14CML 452LeukapheresisMCP+2
    15CML 454 (Fresh)LeukapheresisMCPND3
    16CML 456LeukapheresisFCPND3
    17CML 457 (Fresh)PBMCPND3, 4
    18CML 459LeukapheresisMCPND3
    19CML 460 (Fresh)LeukapheresisFCPND1, 3, 4
    20CML 461 (Fresh)LeukapheresisFCPND3, 5
    • CP, chronic phase; ND, not determined; PB, peripheral blood.

    • * CD34+CD38 CML LSCs were also confirmed to be Ph+ by fluorescence in situ hybridization after sorting.