The transcriptional impact of treatment with interferon-alpha in CALR-mutated ET

Mutated clones in myeloproliferative neoplasms (MPN) lack cell surface markers to distinguish these cells from non-mutated hematopoietic cells, which makes it difficult to investigate the impact of MPN driver mutations in terms of their progenitor subtype identity. Available advanced methods fail to shed a light on the complexity of hematopoietic differentiation.¹

Genotyping of Transcriptomes (GoT) is a novel, advanced method that directly links mutational status with whole transcriptomes of high throughput.² Nam et al.¹ used this method to link genotypes of expressed genes to transcriptional profiling of thousands of single cells in CD34⁺ bone marrow (BM) cells, obtained from patients with essential thrombocythemia (ET) who harbored calreticulin (CALR) mutations. The study¹ revealed that the impact of CALR mutations on cell fitness increased with myeloid differentiation, and that CALR mutations skew differentiation towards megakaryocytic development through increased cell cycle status, showing the dependency on cell identity.

Based on the above study, Shira Rosenburg et al.² investigated the transcriptional impact of interferon-alpha (IFN-α) on CD34⁺ CALR-mutant and wildtype progenitor cell fates using GoT, to gain more insight on the mechanisms of action of IFN that drive clinical responses. The findings were presented at the 63rd American Society of Hematology (ASH) Annual Meeting and Exposition, and we summarize key results below.

For more information on the role of IFN in treating MPN, read our educational theme here.

Study design

• Pre- and post-treatment BM aspirates obtained from five patients with CALR-mutant ET who were treated with pegylated IFN-alpha-2a in the MPN-RC-111/112 clinical trials, were analyzed.
• To evaluate the transcriptional impact of IFN-α, batch effects were removed with the Cell Hashing method to allow comparison of samples taken in different time points from the same patient. These samples were pooled into one single-cell RNA sequencing reaction via the Cell Hashing method.
  • Fluorescence-activated cell sorting (FACS)-isolated CD34⁺ BM cells were labelled with Cell Hashing antibodies with timepoint identifying barcodes that were sequenced and linked to the rest of the single-cell RNA sequencing reaction, via these cell barcodes.
• Next, the impact of IFN-α on progenitor expansion was investigated by using serial time points for each patient sampling, after batch correction.

• The impact on pathways involved in MPN development was investigated by a differential expression analysis between pre- and post-IFN exposure in wildtype and mutated cells.
• The transcription factor regulatory networks involved in the IFN-α differentiation biases were elucidated by using a combination of GoT and single-nucleus chromatin accessibility analysis in BM samples.

Results

The transcriptional effects of IFN-α

Initially, each patient was analyzed individually. Data from the first patient is summarized below.

• Analysis of single-cell RNA sequencing data alone demonstrated that CD34⁺ BM cells clustered based on cell identity at baseline.
• Cell Hashing data on treatment status showed that, following 1 year of treatment, cells exposed to IFN clustered distinctively from baseline cells, highlighting a strong transcriptional effect of IFN-α.
• GoT showed that the wildtype and mutated progenitors co-exist throughout differentiation.

Another patient examined had been off therapy for around a month following 1 year of IFN-α treatment, and the strong transcriptional effects of IFN-α were mostly lost, shown by the co-existence of baseline and post-treatment cells. However, the earliest hematopoietic progenitors remained distinct, inferring stable changes caused by IFN-α.

Impact on hematopoietic development

The effect of IFN-α on hematopoietic differentiation was evaluated based on changes in progenitor cell frequency pre- and post-treatment. At the baseline, megakaryocytic progenitors were frequent in mutated cells. Following interferon treatment:

• There was a marked expansion of B-cell lymphoid lineage in wild type cells with a corresponding reduction in other cellular compartments, whereas mutated cells showed a less apparent expansion in immature myeloid progenitors and neutrophil progenitors.

To understand whether these changes in cell proliferation induced by IFN-α provide a basis for cell frequency shifts, the authors analysed the impact of IFN-α on cell cycle gene expression:

• There was an upregulation of cell cycle gene expression in B-cell progenitors in both wildtype and mutant samples, in line with the expansion of lymphoid progenitors following IFN-α treatment.
• On the contrary, there was a significant increase in cell cycle gene expression in immature myeloid progenitors, specifically in mutated cells.

Impact on MPN development signalling

The differential expression analysis between pre- and post-treatment mutated cells showed a robust increase in IFN-α signalling pathways, as expected. Interestingly, there was a significant downregulation of tumor necrosis factor alpha (TNF-α) and transforming growth factor beta (TGF-β) signalling pathways, which are key signalling pathways in the MPN pathobiology.

The association with transcription factor regulatory networks

Transcription factors are known to regulate hematopoietic differentiation, and Rosenberg et al.² used a combined approach to link cell identity, transcription factor activity (via chromatin accessibility), and mutation status in the same cell.

• IFN-α treatment in wildtype hematopoietic stem progenitor cells (HSPCs) resulted in the enrichment of binding sites for the transcription factors PU.1, a master regulator of both lymphoid and myeloid development, and BCL11A, which is also a transcriptional regulator of lymphoid differentiation.

• In mutated cells, activity of CEBPA, a critical regulator of myeloid differentiation, was increased with IFN-α treatment.

Conclusion

Results from this study present several possible mechanisms of IFN-α treatment that are responsible for clinical response. Firstly, IFN-α drives cell fates of CALR-mutant cells toward myeloid lineage, increasing the cell cycle gene expression, in contrast with baseline mutant cells that favour megakaryocytic fates, which are responsible for increased platelet counts. These results also support a change of differentiation of wildtype progenitors toward lymphoid lineage, which correlates with a robust increase in cell cycle gene expression. Overall, these findings suggest that IFN-α redirects hematopoietic differentiation, resulting in a clinical response despite the absence of a molecular response. Transcriptional factor networks that govern such differentiation changes, were also clarified. Finally, a reduction in TNF-α and TGF-β signalling was observed post-treatment, providing another explanation for the clinical response associated to amelioration of MPN-related signalling pathways.