Shared and Tissue-Specific Genetic Features of Primary Myelofibrosis, Hepatic Fibrosis, Idiopathic Pulmonary Fibrosis and Kidney Fibrosis

Fibrosis is the hallmark of primary myelofibrosis (PMF) and contributes to organ failure of other tissues, including liver, kidney, heart, lung and skin. Hence, there is a need to identify shared mechanisms across fibrotic tissues to advance the development of anti-fibrotic therapies. We previously analyzed bone marrow from three patients with PMF using a transcriptomic profiling that quantified the expression of TGF-β signaling intermediates (1). This platform identified an abnormal expression signature that includes over-expression of genes downstream from the non-canonical ERK/p38 TGF-β signaling; for example, Jun, GADD45B, EVI1 and STAT1, reinforcing TGF-β as a novel therapeutic target for this disease.

The aim of this study is to further clarify the role of TGF-β signaling in bone marrow from PMF, and to define common features of fibrosis in PMF, liver and lung. To achieve this aim, gene expression profiles in bone marrow from three additional PMF patients were compared to five non-diseased volunteers using the Illumina microarray Human HT-12_v4 Bead Chip gene expression array. These results were then compared to gene expression signatures from fibrotic tissues or cells including: activated hepatic stellate cells [100 upregulated and 100 downregulated (2); 100 upregulated (3)] and tissues from patients with idiopathic pulmonary fibrosis (100 upregulated and 100 downregulated, GSE47460) (4).

The expression profiles of the three PMF patients were more heterogeneous, owing to their different disease phenotypes (International Prognostic Score System int-1-2 and fibrosis grade 1-3), while the controls were more tightly clustered. There were 750-1105 transcripts up- or down-regulated, respectively, in the three PMF samples relative to controls. Of those, 81 and 77 were consistently up- and down-regulated in all the patients. This common signature was interrogated by pathway analysis (Fig. 1). GO biological processes enrichment identified pathways controlling cytokine production and signaling, inflammatory responses, wound healing, chemotaxis and leukocyte migration as the most enriched terms. Among transcription factors analyzed by ENCODE, STAT5A was the most enriched transcription factor; however, STAT1 and 2, JUND and FOS were also ranked highly. The enriched differentially expressed KEGG pathways included a hematopoietic stem cell lineage, toll-like receptor, and NF-kappa-B signaling. Analyses of the data using gene set enrichment analysis (GSEA) of the hallmark gene sets identified heme metabolism as the most enriched pathway (NES=2.52, p=0.000), followed by interferon (NES 2.12-2.11, p=0.000), NF-kB (NES=2.09, p=0.000) and TP53 (NES=1.51, p=0.000) signaling. In addition, we identified positive enrichment of genes with binding motifs in regions spanning up to 4 kb around the transcriptional start site for AP-1, a component of transcriptional complexes formed by Fos and Jun (TRANSFAC database ver.7.4; http://gene-regulation.com/pub/databases.html) (NES=1.71-1.23, p=0.000-0.060).

The PMF signature was compared to signatures from other fibrotic tissues: the two liver fibrosis signatures and the lung fibrosis signature, displayed modest similarity. Some common genes included activation of integrin-α9 (ITGA9) and thropomyosin α-1 (9 TPM1) when compared with liver fibrosis signatures, and ectoderm-neural cortex protein 1 (ENC1) and FRAS1-related extracellular matrix protein 1 (FREM1) when compared to the lung fibrosis signature.

In summary, this study has reinforced the presence of a prominent TGF-β signature in the bone marrow of PMF, establishing this pathway as a potential therapeutic target, and identifying interferon and NF-kB signaling as additional pathways that are modulated in PMF. The analysis has also uncovered few potential fibrosis targets, e.g., ITGA9, TPMQ, ENC1 and FREM1, shared with other fibrotic tissues.

References

1. Ciaffoni et al. Blood Cells Mol Dis 2015; 54:234-41

2. Drews et al Biochim Biophys Acta 2008; 1783: 34-48, 2008

3. Zhang et al Gut. 2016; 65:1754-64

4. Peng et al. FASEB J 2016:30:4056-70

View largeDownload slide

View largeDownload slide

Close modal