In this issue, Malone and colleagues identified the nuclear mRNA export factor NXT1 as a novel genetic dependency in neuroblastoma as well as several additional pediatric cancers. These data highlight potential cell type–specific differences in the nuclear export machinery that may be exploited as cancer therapeutics.
See related article by Malone et al., p. 2282.
The regulation of gene expression is tightly controlled at multiple levels, starting from transcription initiation to RNA processing, nuclear export of RNA to cytoplasm, and translation of mRNA to protein. Each of these steps in gene expression regulation may be dysregulated in cancer to promote tumor survival and drug resistance. Alterations in transcription factors and chromatin regulators promoting oncogenic gene expression programs have been well documented, and this knowledge has resulted in a myriad of therapeutic approaches targeting protumorigenic gene expression. More recently, a number of studies have begun to identify perturbations of gene expression in cancer stemming from alterations in the physical interactions between RNA-binding proteins (RBP) and mRNA. RBPs regulate many aspects of RNA metabolism, including RNA splicing, polyadenylation, editing, modifications, and the export of RNA from the nucleus to the cytoplasm. Although genetic as well as nongenetic alterations in RNA processing have been shown to contribute to a tumorigenic state across many cancers, potential alterations in RNA processing remain to be studied in the context of neuroblastoma, an aggressive childhood cancer that continues to have a poor prognosis and limited therapeutic options.
In this issue of Cancer Discovery, Malone and colleagues identified several genetic dependencies in neuroblastomas harboring MYCN amplification (1), which occurs in 20% to 25% of patients with neuroblastoma and is associated with a lower overall survival. Using the Cancer Dependency Map (DepMap), a public database developed by the Broad Institute (2), the authors nominated 197 genes that were preferentially required for the survival of MYCN-amplified neuroblastoma cell lines. To functionally interrogate these candidates, the authors performed a broad range of CRISPR screens, including CRISPR/Cas9-mediated knockout and CRISPR interference in vitro and in vivo screens across neuroblastoma cell lines. This effort revealed several previously undescribed dependencies in neuroblastoma or other cancers including NXT1, HSPA8, and TYMS.
In this study, Malone and colleagues focused on Nuclear Transport Factor 2 Like Export Factor 1 (NXT1; also known as p15), a nuclear export factor that is essential for the transfer of most mRNAs from the nucleus to the cytoplasm (Fig. 1). NXT1 binds NXF1 (also known as TAP) to mediate mRNA export, and the NXT1:NXF1 heterodimer is conserved from yeast to humans (reviewed in ref. 3). After capping of the 5′ end of mRNA and splicing, the nuclear export complex known as Transcription and Export (TREX)/UAP56/ALYREF is recruited to the mRNA to form a mRNA ribonucleoprotein (mRNP). The NXF1/NXT1 heterodimer then binds mRNPs and facilitates their shuttling out of the nucleus to the cytoplasm where mRNA is ultimately translated into protein or degraded.
Using a degron degradation system, the authors identified that downregulation of NXT1 resulted in increased cleaved PARP and cell death in a number of neuroblastoma cell lines. To understand the basis for NXT1 dependency in neuroblastoma, the authors compared cell lines that were sensitive or resistant to NXT1 depletion. Interestingly, there was a strong inverse correlation between sensitivity to NXT1 loss and expression of NXT2, a paralog of NXT1 (4). Analysis of pan-cancer cell line DepMap data revealed that lower NXT2 expression was associated with increased sensitivity to NXT1 depletion. To further explore the relationship between NXT1 and NXT2, the authors found that overexpression of NXT2 was sufficient to stabilize NXT1. This led the authors to propose compensation of NXT1-sensitive neuroblastoma cell lines through increased expression of NXT2. CRISPR-mediated knockout of NXT2 in cell lines with high expression of NXT2 induced sensitivity to NXT1 depletion, revealing a synthetic lethal relationship between NXT2 and NXT1. Importantly, degradation of NXT1 decreased NXF1 protein levels when NXT2 levels were also low. The authors identified that simultaneous loss of NXT1 and NXT2 somehow destabilizes NXF1, which is essential for cell survival. They thereby posit that the basis for the synthetic lethal relationship between NXT1 and NXT2 was due to effects on stability of the pan-essential protein NXF1. To further explore whether a requirement for NXT1 might exist in other cancers, the authors found that additional pediatric cancers including medulloblastoma and rhabdomyosarcomas also expressed lower NXT2 levels. Consistent with the findings from neuroblastoma, genetic depletion of NXT1 in medulloblastoma and rhabdomyosarcoma also led to antitumor effects.
This study raises several exciting questions fundamental to the biology of mRNA export. For example, while the work here suggests that NXT1 and NXT2 may be paralogs with interchangeable roles, it will be important to understand if there are differences between mRNA cargoes bound by NXT1 versus NXT2. In addition, it is possible that NXT1 and NXT2 may even differ in their binding efficiencies to the same mRNA species. NXT1 and NXT2 each bind NXF1, and one of the recurrent findings in this study is that simultaneous suppression of NXT1 and NXT2 resulted in degradation of NXF1 protein. However, it is not clear how NXF1 protein stability is regulated. Furthermore, just as NXT1 and NXT2 are paralogs, paralogs of NXT1 exist (4), but their roles in cancer are currently unknown. Furthermore, it will be important to understand why NXT1 is selectively expressed in pediatric cancers compared with cancers arising in adults and how NXT1 and NXT2 expression are each regulated.
Given that mRNA export is a critical mediator of gene expression, it will be fascinating to study whether the components of the mRNA export machinery might be cell type– or tissue-specific. It is known that nuclear pore complex composition can differ between tissues and even the number of nuclear pores can vary by cell type (and this number is not directly related to nuclear envelope surface area; reviewed in ref. 5). These findings raise the question of whether cancer-specific nuclear pore and nuclear export complexes may exist.
Overall, this study proposes a model that implicates NXT1 as a strong dependency in neuroblastoma and other cancers with low NXT2 expression and that NXT2 expression could be used as a biomarker to determine sensitivity to NXT1 loss across many forms of cancer. The authors have extensively shown that the mRNA export factor NXT1 is a specific vulnerability in neuroblastoma as well as other difficult-to-treat pediatric cancers, such as medulloblastoma and rhabdomyosarcomas. Recent work from several groups, including our own laboratory (6), has revealed that therapeutic targeting of the nuclear export of macromolecules may have preferential effects on specific genetic subsets of cancers. In fact, drugs inhibiting XPO1, the main nuclear receptor responsible for protein and small nuclear RNA export are already approved by the FDA for refractory B-cell malignancies (7, 8). The work here by Malone and colleagues suggests that the machinery required for mRNA export may be an additional exciting target for chemical inhibition given tissue-specific differences in nuclear mRNA export.
No disclosures were reported.
This work was supported by NIH R01 HL128239 and R01 CA251138, the Leukemia & Lymphoma Society, and the Edward P. Evans MDS Foundation (to O. Abdel-Wahab).