Introduction: Isochromosome 17 (i(17q)) is a rare cytogenetic abnormality reported in different myeloid neoplasms. I(17q) has been described as primary and as secondary chromosomal aberration, often acquired in the disease course. Recently, we have shown that patients with i(17q) show a distinct mutation profile with mutations in ASXL1, SETBP1 and SRSF2. Further data suggested a parallel acquisition of SETBP1 mutation and i(17q) (Meggendorfer et al., Leukemia, 2016). Of note, i(17q) results in three copies of the splicing factor SRSF2, potentially further influencing the transcriptome of affected cells.

Aim: To characterize the transcriptome in myeloid neoplasms with i(17q) and the typical mutations in ASXL1, SETBP1 and SRSF2 (A/S/S/i17pos).

Patients and Methods: In total 18 patients were selected based on the cytogenetic profile and the molecular mutations. All had the diagnosis of a myeloid neoplasm by cytomorphology according to the WHO. Chromosome banding and FISH analysis and mutation status of ASXL1, SETBP1 and SRSF2 was available in all cases. Three patient groups were defined: 1) ASXL1, SETBP1, SRSF2 mutated and a normal karyotype (A/S/S/-pos; n=5) and gain of i(17q) during follow up, 2) ASXL1, SETBP1, SRSF2 mutated and i(17q) as sole cytogenetic abnormality (A/S/S/i17pos; n=8), 3) ASXL1, SRSF2 mutated and i(17q) as sole cytogenetic abnormality (A/-/S/i17pos; n=5). In all cases RNA sequencing was performed (TruSeq RNA Sample Preparation V2, Illumina, San Diego, CA). A control RNA (Universal Human Reference RNA, Agilent Technologies, Santa Clara, CA) was investigated in triplicate for normalization and comparison. Expression analyses were performed with BaseSpace RNA Express app (Illumina, San Diego, CA).

Results: In total 23,710 genes were annotated by RNA sequencing and at least 14,773 analyzed for differential expression, showing that all three patient groups show aberrant expression in comparison to the control RNA. In mean 4,930 genes/group (range: 3,603-6,711) were differentially expressed. The differential expression ranged from -12.7 to 11.2 log2(fold change). Comparing the two groups with all three gene mutations but different i(17q) status (A/S/S/-pos and A/S/S/i17pos) showed that the presence of i(17q) changes the expression pattern with 1,596/13,218 assessed genes differentially expressed. In detail, in A/S/S/i17pos cases 790 genes were significantly over expressed while 806 genes showed reduced expression compared to A/S/S/-pos, ranging from -2.84 to 2.78 log2(fold change). The expression of SRSF2 was not affected by i(17q), although i(17q) cases show three SRSF2 gene copies. Analyzing the most strongly affected genes (2<log2(fold change)<-2, n=48) showed that differential gene expression affected mostly the homeobox (HOX) genes clustering on chromosome 7p15 (HOXA1, HOXA5 and HOXA7) as well as MEIS1, an important cofactor of HOX genes. HOX genes represent a family of transcription factors, shown to be involved in hematopoiesis. Addressing specifically the differential expression of HOX genes showed that further 4 HOXA genes (HOXA2, HOXA4, HOXA9, HOXA10) and 3 HOXB genes (HOXB2, HOXB3, HOXB4) were significantly dysregulated, with HOXB clustering on 17q21. The expression of all HOX genes was up-regulated in cases with i(17q). Interestingly, the molecular mutation pattern A/S/Spos has also been shown to associate with patients having a monosomy 7 (Meggendorfer, #1364, ASH 2013), where the HOXA gene cluster is located. Comparing the HOXA and MEIS1 gene expression in all 18 samples to human control RNA revealed a significant lower expression level in A/S/S/-pos and an increased one in A/S/S/i17pos and A/-/S/i17pos patients, clearly differentiating i(17q) carrying patients. However, the additional SETBP1 mutation did not influence the expression pattern as seen by comparing A/S/S/i17pos and A/-/S/i17pos patients (0/13,398 assessed genes differentially expressed).

Conclusion: 1) Transcriptome analysis of patients with myeloid malignancies, i(17q) and mutations in ASXL1, SETBP1 and SRSF2 show an up-regulation of HOXA genes. 2) Accompanying SETBP1 mutation does not further influence the transcriptome of A/-/S/i17pos patients. 3) The up-regulation of HOXA genes might indicate a pathogenic mechanism in patients with ASXL1, SETBP1, SRSF2 and i(17q). However, this finding has to be validated in a larger cohort and with an independent method.

Disclosures

Meggendorfer:MLL Munich Leukemia Laboratory: Employment. Nadarajah:MLL Munich Leukemia Laboratory: Employment. Haferlach:MLL Munich Leukemia Laboratory: Employment, Equity Ownership. Kern:MLL Munich Leukemia Laboratory: Employment, Equity Ownership. Haferlach:MLL Munich Leukemia Laboratory: Employment, Equity Ownership.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution