Tumors have all sorts of tricks up their metaphorical sleeves to evade treatment. A particularly sneaky one – and a proposed new hallmark of cancer – is literally changing their cell identity by transitioning lineage states, especially under selective pressure such as chemotherapy.
One malignancy that’s infamous for lineage plasticity is prostate cancer, which is diagnosed in hundreds of thousands of men in the United States every year. In most cases, when detected early, prostate cancer is curable with existing therapies. However, drug resistance is a cause of poor outcomes for tens of thousands of patients with advanced disease.
While this lineage diversity in response to treatment currently poses a challenge, it may someday be the cancer’s Achilles’ heel. Imagine if there were a way to drug these transitions to prevent drug resistance or resensitize tumors to therapies. When one door closes, how can we open the next one?
This question fascinates Dr. Rashmi Mishra, a post-doctoral researcher in Andrew Hsieh's Lab in the Human Biology Division. She is particularly interested in how post-transcriptional gene regulation impacts lineage switching in advanced, drug-resistant prostate cancer. She is lead author on a recent study in the Journal of Clinical Investigation that expands our understanding of how RNA dynamics might be therapeutically targeted in prostate cancer.
“This study provides a unique framework in which we uncover mRNA translation regulatory mechanisms governing oncogenic lineage states,” says Dr. Mishra.
“The goal is to drug specific nodes of mRNA translation in cancer patients and determine if it is therapeutically possible and effective,” Dr. Hsieh adds.
Many prostate cancers begin as luminal cells that express luminal-specific keratins and are dependent on testosterone signaling via the androgen receptor (AR). Depriving these tumors of growth signals via hormonal therapy or AR inhibition can be quite effective—until it isn’t. Castration-resistant prostate cancers can develop resistance by lineage switching to a more basal cell type that doesn’t respond to AR inhibition.
Dr. Mishra began the study with a pharmacological screen designed to identify translation-control pathways that are particularly critical to castration-resistant tumors. She systematically tested a panel of drugs that either inhibit or enhance protein synthesis on cell lines derived from advanced stage prostate cancers to determine which translation regulators might present a therapeutic vulnerability in advanced tumors.
Through this screen, they found a very interesting hit: a drug called PF-07293623, which inhibits the ability of eukaryotic initiation factor 4e (eIF4E) to bind the chemical cap at the 5’ end of mRNA. This is a crucial stage of translation to block, as eIF4E is widely considered to be a rate-limiting step to recruit ribosomes to mRNA to initiate translation.
Interestingly, prostate cell lines that were sensitive to AR deprivation—the AR-high, more luminal subtype—were not significantly impacted by PF-07293623. In contrast, the AR-low basal cell lines that were resistant to AR deprivation were profoundly sensitive to PF-07293623, both in cell culture and in mice.
The findings raised an important question: what biological features made AR-low tumors especially dependent on eIF4E cap binding?
The drug PF-07293623 competes with mRNA to bind eIF4E, preventing the recruitment of ribosomes and reducing protein synthesis. The authors decided to measure which proteins were most affected by treatment with this drug using a novel proteomics pipeline to measure nascently made proteins developed in collaboration with the proteomics core.
“We worked closely with Philip Gafken and Chen Wei from the Mass Spectrometry Core, who were instrumental in developing, executing, and analyzing our nascent proteomics platform (HPG-TMT mass spectrometry),” explains Dr. Mishra. “This approach utilizes incorporation of the methionine analog [homopropargylglycine] HPG into newly synthesized peptides, followed by biotin-streptavidin enrichment for mass spectrometry, enabling direct, proteome-wide measurement of active mRNA translation.”
Through this, they discovered that crucial keratins vital for the basal lineage were most impacted by PF-07293623; i.e., drug treatment did not impact mRNA abundance of these genes but drastically reduced protein expression.
But why are just these keratins affected and not all proteins? To answer this, the authors focused on the mRNA structure in the 5’ untranslated region (5’UTR) using a technique called SHAPE-MaP, which maps RNA structure and flexibility to infer secondary structure.
“SHAPE-MaP was established for the first time at Fred Hutch and in the prostate cancer field by Jin Yeong Kim, a technician in the Hsieh lab, in collaboration with Assistant Professor Dr. Arnab Sengupta at Georgia State University and was subsequently applied to interrogate the role of RNA structure in lineage conversion,” says Dr. Mishra.
Through this work, they found a specific cis-regulatory element that specifically controls basal keratin expression. Importantly, another discovery was that these basal keratins are required for basal tumor survival – hence the high toxicity when their translation is inhibited.
Finally, they made another key observation: not only does eIF4E inhibition decrease the basal identity of these cells via loss of basal keratin expression, it also increases their luminal features through restoration of AR expression. This occurs in a dose dependent manner, but not through transcriptional or translational mechanisms. Instead, they found that two crucial deubiquitinases are upregulated when cells are treated with PF-07293623. Expression of these two proteins stabilizes AR expression and re-sensitizes cells to AR inhibitors.
This study “demonstrates, for the first time, that inhibition of the 5′ cap-binding function of the well-studied translation factor eIF4E drives basal-to-luminal lineage conversion,” Dr. Mishra explains. “First, targeting this domain sensitizes cells to therapy. Second, the drug-induced lineage switch restores sensitivity to androgen receptor pathway inhibitors.”
“We identify two mechanisms underlying this basal-to-luminal transition: (1) selective 5′ UTR mediated repression of basal keratins, and (2) stabilization of AR through the deubiquitinases BAP1 and OTUD3,” she continues.
Overall, this study shows an intriguing path towards a future where we can drug castration-resistant prostate cancers.
“We demonstrate that the translation initiation apparatus can control the lineage state of prostate cancer, which has implications for drug resistance,” says Dr. Hsieh.
In the future, the authors want to follow up on the mechanisms they’ve identified that drive lineage state determination.
“Why is it that specific cell states are more or less sensitive to changes in protein synthesis?” asks Dr. Hsieh.
Dr. Mishra continues: “Why are the 5′ UTRs of basal keratin transcripts particularly sensitive to inhibition of mRNA translation, specifically via disruption of the 5′ cap-binding function of eIF4E? How are the deubiquitinases (DUBs) that regulate androgen receptor (AR) stability differentially controlled across distinct lineage states?”
“My immediate focus is to study how different translation factors behave and control distinct lineage states in prostate cancer, a question I aim to pursue in my future lab,” she concludes.
The spotlighted research was funded by the Prostate Cancer Foundation, the American Cancer Society, Seattle Translational Tumor Research, the Nancy & Dick Bernheimer, Matthews Family, Stinchcomb Family, and Thomas & Patricia Wright Memorial Funds, the Larry & Virginia Gordon Endowed Chair in Prostate and Bladder Cancer Research, the Institute for Prostate Cancer Research, the Mike Slive Foundation for Prostate Cancer Research, the National Science Foundation, and the National Institutes of Health.
Fred Hutch/University of Washington/Seattle Children’s Cancer Consortium Members Drs. Eva Corey, Peter Nelson, Sita Kugel, Haolong Li, and Andrew Hsieh contributed to this work.
Mishra R, Song S, Choradia D, Rudoy D, Wladyka CL, Hoang P, Kim JY, Coleman IM, Arora S, Dobersch S, Orellana AE, Lin C, Gafken PR, Corey E, Nelson PS, Kugel S, Li H, Sengupta A, Hsieh AC. 2026. Therapeutic targeting of the eIF4E cap-binding domain reveals control of lineage fate in prostate cancer. J Clin Invest. doi: 10.1172/JCI199838
Hannah Lewis is a postdoctoral research fellow with Jim Boonyaratanakornkit’s group in the Vaccine and Infectious Disease Division (VIDD). She is developing screens to find rare B cells that produce protective antibodies against human herpesviruses. She obtained her PhD in molecular and cellular biology from the University of Washington.