XPO2 Is Essential for a Subset of AML Cells and Is a Biomarker for Poor Response and Survival in Pediatric and AYA AML

Importins & exportins regulate protein transport across the nuclear membrane. Twenty nuclear transport receptors (NTRs) have been identified; six are involved in nuclear export. NTRs contribute to cancer development, through abnormal localization of protein cargos. The exportin, XPO1, is associated with high risk AML & shorter survival times. However, little is known regarding the functional importance of the related exportin, XPO2 (alias CSE1L).

We analyzed a public database of CRISPR/Cas9 depletion screens in 26 AML cell lines. XPO2 ranked in the top 1.4% of essential genes, along with XPO1 (top 1.3%) (DepMap 22Q2 Public+Score, Chronos). An RNAi depletion screen confirmed this, with XPO2 ranked in the top 0.2% of AML genes. Leukemia cells were more dependent on XPO2 vs solid tumors (P = 0.0394) & gene dependency scores indicated a greater reliance on XPO2 vs XPO1 (P <0.0001).

Increased expression of XPO2 mRNA was associated with decreased event free survival (EFS) (P = 0.0014) & overall survival (OS) (P = 0.00091) in pediatric & adolescent/young adult (AYA) AML patients (TARGET dataset). In the BeatAML dataset, increased XPO2 mRNA expression was associated with decreased OS in patients <40 years, but not in patients >40 years (P <0.0001).

We analyzed protein expression databases & identified XPO2 as the highest expressed exportin in AML cells lines & patient samples. Increased XPO2 expression was detected in 7/7 primary AML samples & 9 AML cell lines by immunoblot (HL60, MV4-11, MOML13, NB4, OCI-AML2, OCI-M2, TEX, U937 & 1254 cells), compared to normal hematopoietic cells. Expression of XPO2 was increased in younger (<40 years) vs older patients (>70 years) (P = 0.0454).

Genetic silencing of XPO2 decreased growth & viability of OCI-AML2, TEX, NB4 & 1254 cells. Gambogic acid is an inhibitor of XPO2-mediated nuclear transport. In OCI-AML2, TEX & NB4 cell lines, gambogic acid decreased viability of AML cells (TEX IC50 = 369nM, OCI-AML2 IC50 = 433nM, NB4 IC50 = 161nM).

Next, we interrogated a BioID database (PXD007976) & performed gene set enrichment analysis on XPO2 targets. Processes related to ribosome biogenesis were enriched in XPO2 cargo, whereas XPO1 cargo were enriched for snRNA transcription processes (P <0.05). A gene expression signature encompassing XPO2 & its associated cargo genes was enriched in functionally defined LSCs (P = 0.001, in patients with more primitive AML subtypes (P < 2.22 x 10 ^-16), in more primitive AML blasts as defined by FAB subtype and in high risk subsets, including WT1 mutated pediatric AML. Additionally, the XPO2 gene signature was associated with minimal-residual-disease (MRD) positive disease after induction cycle I (P < 5.1x 10 ^-6) and II (P = 0.0013), decreased complete remission (CR) rates after cycle I (P = 0.00025) and II (P = 0.02), worse EFS (P = 0.019) and worse OS (P = 0.02) in pediatric and AYA AML patients.

In summary, we identified XPO2 as an essential gene for AML . XPO2 is increased in AML cells & primary samples from younger patients. Increased XPO2 expression & increased expression of XPO2 mediated processesare associated with worse outcomes in pediatric AML & correlate with LSC enrichment & primitive AML subtypes. XPO2 may function to enhance ribosomal biogenesis, demonstrating a critical role in leukemogenesis. Thus, this work demonstrates a new dependency of pediatric AML on XPO2 highlighting a new target and biomarker for this disease.