Journal Club Presentation Molecular Basis of ETV6 Mediated Predisposition to Childhood Acute Lymphoblastic Leukemia

Keren Xu

2020/10/02

I presented this paper Molecular Basis of ETV6 Mediated Predisposition to Childhood Acute Lymphoblastic Leukemia PubmedID: 32693409. Slides can be found here:

Some background about this paper: GWAS analyses have identified common polymorphisms associated with ALL risk affecting: IKZF1, ARID5B, CEBPE, CDKN2A, BMI1 PIP4K2A, and TP63. Most recently, germline variants in ETV6 have been identified. ETV6 is a transcription factor, and its main function is to repress the expression of a wide spectrum of target genes, many of which are highly regulated during hematopoiesis. So what is a repressor? For example, repressor can block general transcription factors and RNA polymerase. Transcription factors include both activators and repressors. Repressor can work together with activators to turn on or turn off the transcription of the target genes.

The figure on page 3 is from one of the earlies studies that identified ETV6 variants in ALL. A screen of 23 families with autosomal dominant thrombocytopenia, high red cell mean corpuscular volume (MCV) and occurrences of B cell precursor acute lymphoblastic leukemia found two with ETV6 mutations. In panel c of this figure, they did a great job in presenting a schematic of ETV6. It is located at chromosome 12, and is composed of 8 exons. Blue regions represent protein coding sequences and yellow regions represent untranslated regions. In the lower section of the panel a schematic of ETV6 is represented with its different domains, pointed (PNT), central and ETS, which is the DNA binding domain.

On page 4, to explore the contribution of germline variation in ETV6 more broadly in childhood ALL, this study did targeted sequencing of ETV6 in 4,405 childhood ALL cases discovered 31 exonic variants (4 nonsense, 21 missense, 1 splice site, and 5 frame shift variants) that are potentially related to ALL risk in 35 cases. Exnoic variants are classified as frameshift, nonsense, missense, and splicing (green, blue, red, or orange solid circles, respectively). Each circle represents a unique individual carrying the indicated variant. Those variants that occurred in over 10 individuals were labeled with the exact frequency. Some variants have occurred in over 60 people. Fifteen (48%) of the 31 ALL-related ETV6 variants were clustered in the ETS domain and predicted to be highly deleterious. Although they identified these many ETV6 variants, the function of these variants remain largely unclear.

So in this current study, they aim to identify functions for those 34 germline ETV6 variants that have been previously discovered in ALL. They had 31 B-ALL and 2 AML samples with germline ETV6 variants from 32 unique patients. They obtained diagnostic bone marrow or peripheral blood, and also clinical remission samples from these patients. Then they extracted total RNA, and tumor DNA from these diagnostic samples. And they extracted germline DNA from these clinical remission samples. They did whole transcriptome sequencing for 22 out of 33 samples, whole genome sequencing for both germline and tumor samples for 30 patients and only tumor samples for two patients. They also did exome sequencing matched germline and tumor samples for one patient.

This figures on page 6 presents the results from luciferase reporter assay. They measured transcription repressor activity with a luciferase reporter assay. They compared the activity for each variant with the wild type. Wild type refers to a typical form of a gene as it occurs in nature. Here it means the most typical form of the ETV6 gene. They found 22 of the 34 variants tested (65%) exhibited significant impairment of transcription repression. They refer these 22 variants as damaging variants and the rest of 12 variants as WT-like variants. They additionally compared the DNA binding activity for these variants with WT in panel B and Nuclear:cytoplasmic ratio in panel C, and these 22 variants show significant loss of binding ability and nuclear localization.

The panel D on page 7 shows a summary of the functional classification for each variant as either damaging or WT-like. Most of the damaging variants were in the ETS binding domain. Damaging variants also included more nonsense, frameshift and splicing variants compared to WT-like variants.

On page 8, by using luciferase assay again, they detected a dose-response relationship – the increased amount of ETV6 variants is associated with the increased loss of activity. Panel B shows that these damaging variants did not affect much of the dimerization activity. Panel C shows that these variants caused a general reduction in the ratio of nuclear to cytoplasmic.

Taken together, these results from above pages indicate that damaging ETV6 variants lose their transcriptional repressor function due to impairment of DNA binding and nuclear localization, with minimal effects on dimerization.

Because germline ETV6 variants do not always result in malignancy, with only 25 - 30% of carriers developing leukemia, as shown in page 4, they hypothesized that additional genomic abnormalities must be acquired somatically to induce leukemia. Here they compared genomic profiles between cases with germline ETV6 variants and Cases with ETV6-RUNX1. They also compared hyperdiploidy and diploidy cases. The hyperdiploid cases exhibited recurrent somatic mutations affecting RAS pathway genes. However, diploidy cases show striking high prevalence of PAX5 and ETV6 copy number loss. These PAX5 and ETV6 deletions were largely absent in hyperdiploid ALL samples but were frequently observed in ALL harboring the somatic ETV6-RUNX1 fusion. In panel B, they used unsupervised hierarchical clustering analysis, and found that the first order cluster variable is germline ETV6 variant type. Cases were clustered based on damaging vs WT-like germline ETV6 variants, which means that ETV6 has a global effect on leukemia gene expression. Secondly, within cases with damaging ETV6 variants, the transcriptional profile was largely driven by leukemia ploidy (diploid vs. hyperdiploid) or by other genomic features that are associated with ploidy that they found from Figure A (e.g., ETV6 and/or PAX5 deletion vs. RAS pathway mutations in diploid vs. hyperdiploid cases, respectively). We can see that hypodiploidy cases were clustered together, and cases with ETV deletion and PAX5 deletion were clustered together. In panel C, they repeated the clustering analysis using whole transcriptome seq data from 231 pediatric ALL cases. Damaging ETV6 variant ALL samples with diploid karyotype grouped tightly with ETV6-RUNX1 cases, whereas ETV6 ALL with hyperdiploidy shared the expression signature seen in other hyperdiploid cases. In figure 4A. they used a cytokine-dependent growth assay to compare the impacts of ETV6 variant and wild type on oncogene-driven transformation.

In their previous analysis, they found that there are relatively high frequencies of RAS mutations in ALL with germline ETV6 variants, so they chose to use the RAS mutation as the oncogenic driver in their assay. They found that variants R359X and R399C are associated with loss of in vitro tumor suppressor activity.

Finally, they presented a comprehensive summary of the functional consequences of each germline ETV6 variant identified in this study. Circles with different colors represent different functional consequences that are associated with the ETV6 variant. Two key points of this graph are (1) Leukemia predisposition variants in ETV6 lead to dramatic loss of transcription repressor activity, mainly by disrupting DNA binding; (2) Germline ETV6 variants influence ALL transcriptional profile with a striking resemblance of ETV6 RUNX1 ALL, but unique somatic mutations.