MS4A3 Promotes Differentiation in Chronic Myeloid Leukemia By Enhancing Common β Chain Cytokine Receptor Endocytosis

Background

Chronic phase chronic myeloid leukemia (CP-CML) is characterized by overproduction of differentiating myeloid cells, while blast phase CML (BP-CML) cells exhibit a differentiation block. Tyrosine kinase inhibitors (TKIs) are effective in CP-CML, but resistance is common in BP-CML and can occur without explanatory BCR-ABL1 kinase mutations (BCR-ABL1-independent resistance). Similarly, CML stem/progenitor cells (LSPCs) are insensitive to TKIs, and residual leukemia persists in the majority of CML patients on TKI therapy. We previously reported overlap between the transcriptomes of CD34 ⁺ cells from BP-CML and TKI-naïve CP-CML patients with primary TKI resistance, pointing to commonalities between LSPC quiescence, BCR-ABL1-independent TKI resistance, and BP-CML.

Results

To identify common mechanisms, we performed a meta-analysis of published CML transcriptomes. We identified a small set of genes with consistently low expression in LSPC quiescence, BCR-ABL1-independent TKI resistance, and BP-CML, including Membrane Spanning 4-Domains A3 (MS4A3), a signaling protein previously reported to inhibit hematopoietic cell cycle progression. Low MS4A3 in CD34 ⁺ cells from TKI-naïve CP-CML patients was associated with shorter survival on subsequent TKI therapy, suggesting that MS4A3 governs TKI response.

To understand the function of MS4A3, we lentivirally introduced MS4A3 shRNA or an MS4A3 expression vector into CML CD34 ⁺ LSPCs. MS4A3 knockdown increased clonogenicity and imatinib resistance, while ectopic MS4A3 expression had opposite effects. MS4A3 KD also increased LSPC persistence ex vivo in LTC-IC assays, and in vivo in NSG mice xenografts, while modulating MS4A3 expression had no effect on normal CD34 ⁺ cells.

We next generated Ms4a3^+/+│-/-; Scl-tTA⁺; TRE-BCR-ABL1⁺ compound transgenic mice. Upon BCR-ABL1 induction, Ms4a3^-/-; Scl-tTA⁺; TRE-BCR-ABL1⁺ mice developed leukocytosis comparable to Ms4a3^+/+ controls. However, BM of Ms4a3^-/-; Scl-tTA⁺; TRE-BCR-ABL1⁺ mice showed increased short-term HSCs and multipotent progenitor cells, and reduced granulocyte-macrophage progenitors. When Lin ^- BM cells from leukemic mice were transplanted into irradiated recipients, Ms4a3^-/-; Scl-tTA⁺; TRE-BCR-ABL1⁺ cells showed increased engraftment and myeloid leukocytosis, validating our observations in human cells.

To determine how MS4A3 is downregulated in CML, we expressed BCR-ABL1in 32D-cl3 cells. p210 ^BCR-ABL1 drastically reduced Ms4a3 expression, while kinase-inactive p210 ^{BCR-ABL1-K271R} had no effect. Moreover, we found that suppression of C/EBPε by MECOM reduces MS4A3, consistent with previous reports of MECOM as a driver of TKI resistance and progression to BP. Treatment of CML CD34 ⁺ cells with a library of epigenetic pathway inhibitors revealed that MS4A3 is suppressed by both DNA methylation and PRC2/EZH2-mediated H3K27 trimethylation, which was confirmed by patch-PCR and ChIPseq. These data indicate that multi-levelled mechanisms cooperate in the suppression of MS4A3 in CML.

To determine how MS4A3 regulates clonogenicity and TKI response, we expressed MS4A3-EGFP fusion protein in LAMA-84 CML cells. We found that MS4A3 resides on the plasma membrane and in endosomes. Surface protein biotin labelling and tandem mass spectrometry ± MS4A3 KD showed that MS4A3 controls endocytosis of membrane proteins, including common β chain (βc) cytokine receptors. Specifically, MS4A3 promotes endocytosis of βc cytokine receptors upon GM-CSF/IL-3 stimulation of primary LSPCs and enhances downstream signaling and differentiation, suggesting that restoring MS4A3 expression has therapeutic efficacy. To test this, we manufactured a prototype MS4A3 protein-loaded liposomal nanoparticle (NP) using coating with the CD34 CD62L for targeted delivery to CD34 ⁺ cells. Compared to MS4A3-free NPs, MS4A3 NPs increased CD34 ⁺CD38 ⁺ and CD34 ^-CD38 ⁺ at the expense of CD34 ⁺CD38 ^- cells, reduced clonogenicity, and increased sensitivity to TKIs, mimicking ectopic MS4A3 expression.

Conclusion

MS4A3 governs response to differentiating myeloid cytokines, providing a unifying mechanism for the differentiation block characteristic of primitive LSPCs and BP-CML cells. We posit that LSPCs downregulate MS4A3 to evade βc cytokine-induced differentiation to maintain a primitive, TKI-insensitive state. MS4A3 re-expression or delivery of ectopic MS4A3 may eliminate LSPCs.

Figure 1

View largeDownload slide

Figure 1

View largeDownload slide

Close modal