To appreciate why this development matters, it helps to understand the logic of targeted protein degradation. Small-molecule inhibitors block the activity of a protein, but degraders eliminate the protein itself, which can be far more effective when the target protein has scaffolding roles, non-enzymatic functions, or the ability to rebound after inhibition. Intracellular protein degradation has been revolutionized by proteolysis-targeting chimeras, or PROTACs, which hijack the ubiquitin-proteasome system to tag unwanted intracellular proteins for destruction. But an estimated 40 percent of the proteome, including most receptors, ligands, and secreted factors implicated in cancer, inflammation, and fibrosis, resides outside the cell or on its surface, beyond the reach of the proteasome. For these targets, the lysosome is the natural destination, and a family of technologies collectively known as lysosome-targeting chimeras, or LYTACs, was conceived to exploit it.

The original LYTAC concept is elegant in principle. A bifunctional molecule carries one binding arm directed at a protein of interest and another directed at a cell-surface receptor that constitutively traffics to the lysosome, such as the cation-independent mannose-6-phosphate receptor or the asialoglycoprotein receptor. By physically linking the target protein to such a receptor, the chimera tricks the cell’s endocytic machinery into internalizing the target along with the receptor. Once in the acidic environment of the lysosome, both the target protein and the chimera are degraded. This mechanism, however, carries two fundamental weaknesses that have limited the technology’s reach. First, degradation is absolutely dependent on the expression level of the shuttling receptor, which varies enormously between cell types, tissues, and disease states, and can itself be saturated or downregulated. Second, because the receptor and the chimera are destroyed along with the cargo, each round of degradation consumes degrading machinery and drug molecules alike, imposing a stoichiometric burden that demands high and repeated dosing and risks depletion of the very receptors the cell needs for its own housekeeping.

The new study set out to solve both problems simultaneously by asking a deceptively simple question: could a degrader exploit the cell’s own bulk transport pathways instead of a dedicated receptor, and could it survive the journey it asks the cell to make? The answer came in the form of a chimera built from a polyzwitterion, a synthetic polymer bearing a balanced arrangement of positive and negative charges, conjugated to a ligand that binds the protein of interest. Polyzwitterions occupy an unusual position in polymer science: their net-neutral charge surfaces resist nonspecific protein adsorption, yet certain architectures interact productively with the plasma membrane in ways that can induce membrane ruffling. The researchers harnessed this property deliberately, showing that their polyzwitterion-ligand conjugates trigger macropinocytosis, a form of endocytosis in which the cell engulfs large gulps of extracellular fluid and whatever solutes it contains, forming large vesicles called macropinosomes. Crucially, macropinocytosis is a receptor-independent process. The cell does not need to recognize a specific receptor-ligand pair; it simply drinks in the surrounding medium, and the chimera rides along with its bound target protein.

Once inside the cell, the chemistry of the endocytic pathway takes over. As endosomes acidify, dropping to a pH of around 5 to 6 in late endosomes and below 5 in lysosomes, the bond between the polyzwitterion carrier and the ligand for the protein of interest cleaves. This pH-responsive dissociation is the linchpin of the design. The target protein, released within the endolysosomal system, proceeds to the lysosome and is degraded by the acidic hydrolases waiting there. The chimera itself, however, is not condemned to the same fate. The researchers demonstrated that the liberated polyzwitterion is routed through the endoplasmic reticulum-Golgi transcytosis pathway, a cellular recycling route normally used to shuttle cargo across the cell from one membrane domain to another. Through this pathway, the intact chimera is exocytosed back into the extracellular space, ready to bind another molecule of its target protein and initiate another round of capture, internalization, release, and degradation. In effect, each McR-TAC molecule operates as a catalytic degrader, turning over many copies of its target rather than being consumed stoichiometrically with each one.

The functional consequences of this design were validated in biologically demanding settings. The team targeted programmed cell death ligand 1, better known as PD-L1, the transmembrane protein through which many tumors suppress antitumor T-cell immunity and which is a cornerstone target of the checkpoint inhibitor class of cancer immunotherapies. They also targeted macrophage migration inhibitory factor, or MIF, a secreted pro-inflammatory cytokine that promotes tumor progression, immune evasion, and metastasis. Both targets were chosen in part because they represent the two categories of proteins that LYTACs are meant to address: a cell-surface membrane protein and a soluble extracellular protein. In a triple-negative breast cancer mouse model, one of the most aggressive and difficult-to-treat breast cancer subtypes, McR-TACs directed against PD-L1 and MIF achieved durable depletion of both proteins from tumor tissue. The degradation was sustained rather than transient, a direct benefit of the recycling mechanism, and the effect translated into a significant inhibition of tumor growth in the treated animals.

The implications of receptor-independent, recyclable degradation extend well beyond these two targets. Receptor-dependent LYTACs are inherently tissue-restricted by receptor expression, which can be an advantage for targeted therapy but a severe limitation when the target cell does not express sufficient receptor or when the receptor is saturated by endogenous ligands. By invoking macropinocytosis, a process that many cells can perform, and whose activity is in fact frequently upregulated in cancer cells as part of their nutrient-scavenging metabolism, McR-TACs sidestep this dependency altogether. This may prove especially valuable in oncology, where the very cells a degrader must enter are often the most macropinocytically active cells in the body. The stoichiometric advantage is equally consequential. A degrader that recycles can achieve the same degree of target knockdown at lower administered doses and with longer effective duration of action, reducing manufacturing burden, cost, and the frequency of administration, all of which are decisive factors in whether a biological drug can reach the clinic.

The study also contributes a conceptual lesson that resonates beyond the specific chemistry involved: the cell’s natural transport pathways constitute a rich, largely untapped pharmacological toolbox. Rather than forcing cells to use artificial routes, McR-TACs recruit physiological processes, macropinocytosis for entry, endosomal acidification for release, and ER-Golgi transcytosis for exit, in a sequence that mirrors how the cell already moves material through its compartments. The authors suggest that this strategy of leveraging natural transport pathways for recyclable protein degradation could be adapted broadly, with different target-binding ligands grafted onto the recyclable polyzwitterion platform to address a wide range of membrane and extracellular proteins implicated in human disease.

Significant work remains before such molecules could approach human trials. The pharmacokinetics, immunogenicity, biodistribution, and long-term safety of polyzwitterion carriers must be characterized, and the efficiency of macropinocytosis induction will need to be validated across a broader spectrum of tissues and cell types. Dosing, formulation, and potential off-target degradation of bystander proteins drawn into macropinosomes all require careful scrutiny. Nevertheless, the demonstration that a single bifunctional molecule can repeatedly shepherd proteins to their destruction while returning intact for another cycle marks a conceptual milestone. If the recyclable paradigm generalizes, it could reshape the design logic of extracellular protein therapeutics much as catalytic turnover reshaped small-molecule drug discovery, turning what was once a stoichiometric war of attrition against disease proteins into a genuinely catalytic one.

The central question driving the research is one of the most pressing in modern tumor immunology: why do immune responses against cancer so often fade before the disease is eliminated? Durable cancer control depends not merely on generating a large army of cytotoxic CD8⁺ T cells at the moment of treatment, but on producing memory-like cells capable of persisting, self-renewing, and re-launching attacks months or years later. Central memory CD8⁺ T cells—identifiable by their co-expression of the surface markers CD62L and CD44—are the cellular custodians of long-term immunity. Yet the tumor microenvironment, a hostile mixture of metabolic depletion, suppressive cytokines, and dysfunctional antigen-presenting cells, systematically blocks the formation of these cells, leaving patients with short-lived effector responses that collapse under the relentless pressure of tumor regrowth.

Dendritic cells sit at the heart of this problem. As the body’s professional antigen-presenting cells, they are the teachers of the adaptive immune system: they capture tumor antigens, process them into peptide fragments, and present them to naive T cells in lymph nodes, effectively deciding whether those T cells become short-lived killers or long-lived memory guardians. Previous work had already implicated SHP1—short for Src homology region 2 domain-containing phosphatase 1—in fostering an immunosuppressive dendritic cell state that facilitates tumor immune escape. Earlier studies, including research showing that vitamin E can reinvigorate dendritic cells by targeting SHP1, had established the phosphatase as a kind of checkpoint molecule within these cells. What remained unknown was whether SHP1 exerted control over TCF-1, the transcription factor encoded by the Tcf7 gene that is widely regarded as indispensable for central memory CD8⁺ T cell formation and that operates in intimate conversation with canonical Wnt/β-catenin signaling.

To dissect this relationship, the research team—led by Bing Li, Huilin Lu, and Jiayi Huang of Guangzhou Medical University, with corresponding authors Ting Lei, Xiaoming Tan, and Yuan Zhang—constructed an elegant experimental system combining in vitro co-culture and in vivo genetics. They generated SHP1-deficient DC2.4 dendritic cell lines and primary bone marrow-derived dendritic cells, then co-cultured these modified cells with OT-1 T cells, a widely used laboratory T cell lineage whose T cell receptors specifically recognize ovalbumin peptide presented on MHC class I molecules. This reductionist platform allowed the researchers to precisely measure T cell proliferation, central memory differentiation, cytotoxic killing capacity, and TCF-1 expression under controlled conditions where the only variable was the presence or absence of SHP1 in the dendritic cells.

The results were striking. When SHP1 was downregulated in dendritic cells, the co-cultured CD8⁺ T cells proliferated far more vigorously, generated dramatically increased populations of CD62L⁺ CD44⁺ central memory cells, and killed B16-F10-OVA melanoma cells with markedly enhanced efficiency. All of these changes were accompanied by elevated TCF-1 expression within the T cells, suggesting that the dendritic cell phosphatase was exerting its influence through this master regulator of memory fate. To confirm that the phenomenon was not merely a petri-dish artifact, the team turned to a mouse model in which SHP1 was selectively deleted in dendritic cells—so-called SHP1 conditional knockout mice. When these animals were challenged with EO771 breast tumors, tumor growth was significantly suppressed compared with controls, and analysis of the tumor microenvironment revealed increased frequencies of both IFN-γ-producing CD8⁺ T cells—the hallmark of active cytotoxic engagement—and TCF-1⁺ CD8⁺ T cells, the memory-precursor pool from which durable antitumor responses are sustained.

The mechanistic heart of the paper lies in its dissection of the TCF-1/Wnt/β-catenin axis. TCF-1 is not an isolated actor; it functions as the nuclear endpoint of the canonical Wnt signaling cascade, a pathway in which Wnt ligands stabilize β-catenin, allowing the protein to translocate to the nucleus and partner with TCF/LEF family transcription factors to activate memory-associated gene programs. In resting cells, glycogen synthase kinase-3 phosphorylates β-catenin, tagging it for proteasomal destruction; Wnt activation halts this phosphorylation, causing both active and total β-catenin to accumulate. The researchers found that when T cells were cultured with SHP1-deficient dendritic cells, they exhibited increased levels of active β-catenin, total β-catenin, and the downstream Wnt target genes c-Myc and Cyclin D1—the latter two driving the proliferative burst characteristic of expanding memory precursors. Concurrently, the ratio of phosphorylated β-catenin to total β-catenin dropped, the molecular signature of pathway activation.

Crucially, the team performed the loss-of-function experiments needed to prove causation rather than mere correlation. When Tcf7 was silenced in the OT-1 T cells, the ability of SHP1-deficient dendritic cells to promote central memory formation was completely abrogated, establishing TCF-1 as the non-negotiable mediator of the effect. The investigators then went one step further and silenced Ctnnb1, the gene encoding β-catenin itself, in T cells. This maneuver eliminated not only the enhanced proliferation and memory generation but also the improved cytotoxic activity that SHP1-deficient dendritic cells had otherwise conferred. In other words, the entire phenomenon—proliferation, memory differentiation, and tumor-killing potency—flows through a single linear signaling route: dendritic cell SHP1 restrains TCF-1 expression and Wnt/β-catenin activation in CD8⁺ T cells, and removing SHP1 releases the pathway to drive memory formation.

The therapeutic implications are considerable. Checkpoint blockade immunotherapies such as anti-PD-1 antibodies have transformed the treatment landscape for melanoma, lung cancer, and other malignancies, but a large fraction of patients either fail to respond or relapse, in large part because their tumors lack the stem-like, TCF-1⁺ T cell populations that sustain long-term immune pressure. A growing body of literature links TCF-1⁺ CD8⁺ T cell abundance to favorable prognosis and immunotherapy response across cancer types, from microsatellite-unstable gastric cancer to lung cancer and melanoma. The new findings suggest that dendritic cell SHP1 represents an upstream, druggable node controlling whether those critical stem-like populations are generated in the first place. If pharmacological inhibition of SHP1 in dendritic cells—or strategies that mimic its absence—can be developed safely, it could convert “cold,” T cell–excluded tumors into immunologically active ones while simultaneously endowing patients with the memory reservoir needed to prevent recurrence.

There are also intriguing resonances with prior nutritional and metabolic research. A 2022 study in Cancer Discovery demonstrated that vitamin E enhances cancer immunotherapy by reinvigorating dendritic cells through targeting SHP1, hinting that the phosphatase may be modulated by lipid-soluble dietary factors. The new work provides a mechanistic explanation for how such interventions might work: by lifting SHP1’s restraint on the TCF-1/Wnt/β-catenin axis, nutrient-derived signals could indirectly promote the formation of the central memory T cells that anchor durable immune surveillance. This places SHP1 at the intersection of metabolism, dendritic cell biology, and T cell fate specification—a convergence point that immunologists are increasingly viewing as fertile ground for next-generation therapeutics.

The authors are careful to frame the work as preclinical. The experiments relied on murine models—the OT-1 transfer system, B16-F10-OVA melanoma, and EO771 mammary carcinoma—and human validation remains an essential next step. Dendritic cells are a heterogeneous family, encompassing cross-presenting cDC1 subsets, inflammatory monocyte-derived populations, and tolerogenic plasmacytoid variants, and it is not yet clear whether SHP1’s memory-suppressive function is uniform across all of these lineages or confined to particular subsets. Moreover, because Wnt/β-catenin signaling plays context-dependent roles in tumors themselves—including promoting immune exclusion when activated within cancer cells—any therapeutic strategy would need to target the pathway selectively in dendritic cell–T cell synapses rather than systemically. The study was supported by the National Natural Science Foundation of China and the Guangdong Basic and Applied Basic Research Foundation, and the animal protocols were approved by the institutional ethics committee of Qingyuan Hospital Affiliated to Guangzhou Medical University.

Even with those caveats, the paper adds a compelling new layer to the emerging picture of dendritic cells as master architects of CD8⁺ T cell fate in cancer. Rather than serving as passive antigen delivery vehicles, these cells actively calibrate the memory versus effector decision through intracellular phosphatase signaling—and SHP1, long known as a brake on immune activation, now appears to be a brake specifically on the immune system’s memory-forming machinery. For a field wrestling with why immunotherapy triumphs are so often temporary, the demonstration that a single dendritic cell-intrinsic molecule governs TCF-1 expression, Wnt/β-catenin activation, central memory formation, and ultimately tumor control in living animals offers both a conceptual advance and a concrete target. If future studies confirm the axis in human tumors and identify safe ways to inhibit dendritic cell SHP1 in patients, the promise of immunotherapies that confer not just transient tumor shrinkage but genuine, memory-anchored cancer cures will have moved a decisive step closer to reality.

Scientists in China have identified a long non-coding RNA — a molecule from the genome’s long-dismissed “dark matter” — that behaves like a factory-installed brake on lung cancer. The molecule, known as MIR99AHG, keeps tumor cells from ramping up production of the fatty building blocks they need to multiply, migrate and invade surrounding tissue. In a study published in the Journal of Cancer Research and Clinical Oncology, researchers report that MIR99AHG accomplishes this by physically associating with SCD1, a fat-synthesizing enzyme on which many tumors depend. When the RNA is lost, SCD1 protein rises and lung cancer cells become laden with lipids, faster-growing and more invasive; when the RNA is restored, that malignant behavior recedes. The work delineates what the authors describe as a MIR99AHG–SCD1 regulatory axis, a molecular circuit that suppresses lipid biogenesis and, with it, the progression of one of the world’s deadliest diseases. The discovery, published open access, adds a new name to the growing list of non-coding RNAs with mechanistically explained roles in cancer metabolism.

Lung cancer claims close to two million lives each year, and its lethality is closely tied to an ability that has fascinated researchers for a century: metabolic reprogramming. Tumor cells do not merely grow faster than healthy cells; they rebuild their entire metabolic machinery to serve that growth. Where a normal cell draws most of its energy from glucose and manufactures only the fat it needs, a cancer cell becomes an avid producer of lipids, the fatty molecules that form its membranes, stock its energy reserves and carry the signals that drive proliferation and survival. This phenomenon, known as lipid metabolic reprogramming, is now recognized as a defining feature of aggressive cancers. Yet the switches that govern it, particularly those operating at the level of RNA molecules that never become proteins, remain incompletely charted. The new study was designed to illuminate precisely that shadowy territory, asking whether the non-coding genome holds leverage over the lipid supply lines on which lung tumors depend.

The molecules at the center of the story are long non-coding RNAs, or lncRNAs: RNA transcripts longer than about 200 nucleotides that are copied from DNA but never translated into proteins. For decades after the human genome was sequenced, such transcripts were dismissed as transcriptional noise, the byproduct of a genome that reads itself far more promiscuously than biologists once imagined. That view has steadily collapsed. LncRNAs are now known to guide chemical changes to chromatin, fine-tune gene expression, scaffold multi-protein complexes and, as this study underscores, bind directly to proteins to alter their abundance or behavior. MIR99AHG, whose name reflects its identity as the genomic host gene of a small regulatory RNA, belongs to this class. When the research team, led by corresponding author Yonghui Wu of the Third Affiliated Hospital of Sun Yat-sen University, combed large public gene-expression repositories including TCGA and GEO, they found MIR99AHG consistently dialed down in lung cancer, a depletion pattern that marked the transcript as a candidate tumor suppressor worth pursuing.

A drop in a molecule’s abundance, however, does not by itself prove that the molecule matters. To establish causality, the researchers, whose first two authors, Run Chen and Ping Fang, contributed equally to the work, ran complementary gain- and loss-of-function experiments in human lung cancer cell lines. When they silenced MIR99AHG, the cells responded emphatically: they proliferated faster, formed more colonies in culture and displayed heightened migration and invasion, the two behaviors that make cancer lethal by enabling it to seed distant organs. When they forced the cells to overproduce MIR99AHG, the effect flipped. Proliferation, colony formation, migration and invasion were all restrained, painting the RNA as an active suppressor of malignancy rather than a passive correlate of it. The symmetry of the two directions, loss accelerating and gain braking, is a classic signature of a tumor-suppressive molecule, and it gave the team a solid functional foundation before they attempted to trace the mechanism underneath.

The next question was mechanistic: how does an RNA that encodes no protein exert this kind of power? To find binding partners, the team used RNA pulldown, a technique in which a specific RNA of interest serves as bait to fish associated proteins out of the crowded interior of a cell. In such assays the RNA is typically tagged with biotin, a small molecule with a voracious affinity for the protein streptavidin; the tagged transcript is introduced into cell lysate, allowed to bind its natural partners and then hauled out on beads, carrying whatever clings to it. The captured cargo was then analyzed by mass spectrometry, a method that identifies proteins by fragmenting them and reading the masses of the pieces like a barcode. Among the proteins that stayed attached to MIR99AHG was one that suddenly made biological sense of every observation so far: SCD1, the fat-building enzyme, was traveling in complex with the tumor-suppressive RNA inside lung cancer cells.

SCD1, short for stearoyl-CoA desaturase 1, is an enzyme embedded in the membrane of the endoplasmic reticulum, the cellular factory where lipids and proteins are processed. Its chemistry is deceptively simple but metabolically momentous: it inserts a double bond into saturated fatty acids, converting them into monounsaturated species such as oleate and palmitoleate. Those products are the preferred raw material for triglycerides, phospholipids and lipid droplets, and they lend growing membranes the fluidity that rapidly dividing cells demand. Cancer cells lean heavily on SCD1 to expand their membrane inventory, buffer themselves against lipotoxic stress and stockpile energy. The pivotal experiment concerned how MIR99AHG controls this enzyme. Depleting the RNA barely altered SCD1 messenger RNA levels, meaning the gene’s output at the transcript level was essentially undisturbed, yet the SCD1 protein signal, measured by fluorescence intensity, rose markedly. That divergence between transcript and protein is the fingerprint of post-transcriptional regulation: MIR99AHG restrains SCD1 not by silencing its gene but by limiting how much SCD1 protein persists inside the cell, most likely by influencing the protein’s stability.

The functional consequences followed a logical chain. Depleting MIR99AHG drove lipid accumulation inside the cells and boosted triglyceride production, the biochemical hallmarks of a tumor shifting into fat-manufacturing overdrive. The decisive test, however, was a rescue experiment, the gold standard for separating correlation from cause. If SCD1 truly executes MIR99AHG’s effects, then removing SCD1 should cancel the damage caused by losing the RNA. That is precisely what happened. When the researchers knocked down SCD1 in cells that had already lost MIR99AHG, the lipid accumulation receded, triglyceride production fell and the cells’ accelerated growth, migration and invasion were reversed. The result establishes a clean, linear pathway: MIR99AHG holds SCD1 protein in check, SCD1 drives lipid biogenesis, and lipid biogenesis fuels the malignant behaviors that make lung cancer dangerous. Release the brake and the engine roars; restore it and the machine idles. It is an unusually tidy causal story in a field where metabolic correlations abound and mechanistic proof is harder-won, and it turns the fat that accumulates in aggressive cells from an ambiguous hallmark into a readable output of a defined RNA–protein interaction.

The findings arrive at a moment of intensifying interest in both halves of the axis. SCD1 has long been coveted as a drug target in oncology because of its centrality to tumor lipid supply chains, though inhibiting an enzyme that also serves healthy tissues has complicated efforts to weaponize that interest safely. The new work suggests an alternative handle: rather than attacking the enzyme itself, future therapy could seek to restore or mimic the RNA that keeps the enzyme’s protein levels in check, exploiting a regulatory relationship that tumor cells may struggle to replace. MIR99AHG’s recurring loss in lung cancer also raises the prospect of using it as a biomarker, a measurable signal that could help identify tumors primed for aggressive, lipid-hungry growth and guide the selection of patients for metabolic therapies. Just as consequential is the conceptual shift. The study strengthens the case that the non-coding majority of the genome is not decorative but deeply wired into the metabolic logic of cancer, and that some of oncology’s most important control circuits may be written in RNA that never produces a protein at all.

The authors are careful about the boundaries of the work. The study did not involve direct recruitment of human participants, human tissue specimens or live vertebrate animals; the human data came from de-identified public datasets, and the laboratory experiments used commercially available cell lines, an approach for which the Ethics Committee of the Third Affiliated Hospital of Sun Yat-sen University waived the requirement for ethics approval and informed consent. The article itself is an early release, a peer-reviewed, accepted manuscript shared ahead of the final version of record, citable under a permanent digital object identifier but subject to further editorial edits. Substantial questions remain open, including the precise molecular route by which MIR99AHG restrains the SCD1 protein, whether the mechanism involves degradation, sequestration or interference with the protein’s lifecycle, and whether the axis operates in animal models and patient tumors as robustly as it does in laboratory culture.

The research was supported by the Jiangsu Province Traditional Chinese Medicine Science and Technology Development Program and the Xuzhou Medical Science and Technology Innovation Plan Project, with a team spanning the Southern Medical University Hospital of Integrated Traditional Chinese and Western Medicine in Guangzhou and the Third Affiliated Hospital of Sun Yat-sen University. The manuscript was received in early June, accepted in mid-July and published online on 27 August 2026, a rapid passage through peer review for a finding of this depth. For a field accustomed to hunting cancer’s weaknesses among protein-coding genes, the message is bracing: some of the most important circuitry may live in the stretches of the genome that code for nothing at all. Lung cancer’s appetite for fat has helped it claim millions of lives. This study suggests that one of the switches governing that appetite has been sitting in plain sight, written in RNA, named MIR99AHG, and waiting to be read.

Prostate cancer is among the most commonly diagnosed cancers in men, affecting approximately one in eight during a lifetime. Although many patients can be successfully treated, the disease becomes far more difficult to control after it spreads beyond the prostate. In the United States, prostate cancer remains the second-leading cause of cancer-related death in men. Most prostate tumors initially resemble normal prostate glands and retain a dependence on androgens, the male sex hormones that include testosterone. This biological dependence makes the androgen receptor an important treatment target. Drugs that block androgen production or prevent androgen receptor signaling are therefore central to the management of metastatic prostate cancer.

The initial response to androgen receptor inhibitors can be substantial, but resistance eventually develops in nearly all patients with advanced disease. Some tumors continue growing by finding alternative ways to activate androgen receptor signaling. Others take a more radical route: they reprogram their identity. Instead of maintaining the features of gland-forming prostate cells, these cancers may acquire characteristics associated with stem-like, neuroendocrine or other cellular states. This process, known as transdifferentiation, involves extensive changes in gene expression and cellular behavior. The resulting tumors are often less dependent on androgen signaling and may become far more difficult to detect and treat using conventional prostate cancer therapies.

The Michigan team focused on tumors in which two major tumor-suppressor genes, TP53 and RB1, have been lost. Previous research had connected the disappearance of these genes with prostate cancer transdifferentiation, but the molecular logic behind that association remained unclear. By comparing prostate cancer cell lines with different genetic backgrounds, the researchers found that the transition appeared to involve two coordinated processes. First, cells shut down genes associated with glandular prostate function. At the same time, they activated gene-regulatory programs linked to stem-cell-like identities and alternate developmental states. Rather than representing a single molecular switch, transdifferentiation appears to be a coordinated rewiring of the cancer cell’s regulatory system.

This distinction helped explain why an earlier therapeutic approach had only limited success. The researchers had previously shown that BET bromodomain inhibitors could interfere with the activation of alternate identity programs. BET proteins help control gene expression by recognizing acetylated histones, the proteins around which DNA is packaged. By disrupting these interactions, BET inhibitors can reduce the transcription of selected cancer-promoting programs. In the new study, however, the drugs slowed the growth of transdifferentiated prostate cancer cells without consistently killing them. The surviving cells retained enough flexibility to maintain the altered state and eventually continue progressing, suggesting that blocking the activation of new programs was not sufficient by itself.

The investigators therefore added DNMT inhibitors to the treatment strategy. DNA methyltransferases place chemical tags called methyl groups onto DNA, often reducing the activity of nearby genes. In cancer, abnormal DNA methylation can silence genes that would otherwise help maintain normal cellular identity or restrain tumor growth. DNMT inhibitors can remove or dilute some of these methylation marks as cells divide, allowing previously silenced genes to become active again. The drugs are already approved by the U.S. Food and Drug Administration for certain blood cancers, but their potential in transdifferentiated solid tumors remains under investigation. In this study, the researchers reasoned that DNMT inhibition might help restore glandular gene programs while BET inhibition suppressed the alternate programs supporting the transformed identity.

The combined treatment produced stronger effects than either drug alone in prostate cancer cell lines. According to the researchers, the two-drug regimen reduced cancer cell growth and reversed a substantial portion of the gene-expression changes associated with transdifferentiation. The results were also reproduced in mice carrying implanted tumors, where the combination slowed tumor growth more effectively than individual treatment. Notably, the researchers reported significant antitumor activity at doses lower than the recommended doses of the individual drugs, and the regimen was well tolerated by the animals. These findings suggest that the drugs may operate through complementary mechanisms: one limits the transcriptional machinery that sustains the abnormal cell state, while the other helps reactivate genes lost during the transition.

The study remains preclinical, and the results do not yet demonstrate that the combination is safe or effective in people with advanced prostate cancer. Epigenetic drugs can affect gene activity across many tissues, creating the possibility of side effects that may not be apparent in laboratory models or short-term animal experiments. The researchers are now working to determine which individual genes are responsible for the treatment response and whether molecular biomarkers can identify patients most likely to benefit. Such biomarkers could include patterns of TP53 and RB1 loss, DNA methylation signatures, or gene-expression profiles indicating that a tumor has begun adopting a stem-like or non-glandular identity.

An additional goal is to intervene before transdifferentiation becomes established. Once prostate cancer cells have fully shifted into an alternate state, they may be more adaptable and resistant to therapies designed for conventional glandular tumors. Detecting early signs of the transition could allow clinicians to use combination treatment before the cancer becomes deeply reprogrammed. The Michigan researchers also believe that the strategy may have relevance beyond prostate cancer. Similar forms of lineage plasticity and transdifferentiation are being studied in lung and pancreatic cancers, where tumor cells can escape treatment by changing their biological identity. If future studies confirm the mechanism, simultaneous targeting of epigenetic survival programs could become a broader strategy for cancers that evolve by rewriting their cellular blueprint.

Subject of Research: Animals

Article Title: Combined BET bromodomain and DNMT inhibition targets critical survival pathways in transdifferentiated prostate cancer

News Publication Date: 11-Aug-2026

Web References: https://insight.jci.org/articles/view/207543; https://doi.org/10.1172/jci.insight.207543

References: JCI Insight, “Combined BET bromodomain and DNMT inhibition targets critical survival pathways in transdifferentiated prostate cancer,” DOI: 10.1172/jci.insight.207543

Keywords: prostate cancer, metastatic prostate cancer, transdifferentiation, treatment resistance, androgen receptor inhibitors, BET bromodomain inhibitors, DNMT inhibitors, epigenetics, TP53, RB1, tumor suppressor genes, cancer cell identity, prostate cancer therapy, University of Michigan, JCI Insight

The findings, published in Cell Reports, offer an example of how computational drug discovery can be applied to targets that have resisted conventional pharmaceutical approaches. Rather than searching only for compounds that bind to the most obvious catalytic sites on a protein, the research focused on the PDZ domain, a structural region involved in protein–protein interactions. These interactions help organize signaling networks inside cells, but they are often broad, shallow or chemically difficult to occupy with conventional drugs.

GIPC1, or GAIP-interacting protein C terminus 1, functions as a scaffold that helps bring signaling molecules and membrane-associated proteins into coordinated complexes. Its PDZ domain can influence the stability, localization and activity of proteins involved in tumor biology. By disrupting this domain, the researchers sought to interfere with signaling that cancer cells use to proliferate, survive stressful conditions and resist therapy. The strategy is particularly relevant to pancreatic cancer, in which malignant cells frequently adapt to treatment and grow within a highly protective tumor environment.

To search for a suitable inhibitor, the team worked with Sravathi AI Technology for IP Sharing, a company based in Bangalore, India. Their computational screening program evaluated nearly 40,000 candidate compounds, prioritizing molecules predicted to interact with the GIPC1 PDZ domain. The process was intended to reduce the time and resources required to examine a large chemical space experimentally. After the initial selection, the researchers performed laboratory testing to determine whether the leading compound could engage GIPC1 and interfere with its biological function.

The resulting inhibitor showed activity in models of pancreatic ductal adenocarcinoma. Blocking GIPC1 reduced cancer-related growth signals and limited the ability of tumor cells to maintain aggressive behavior. In animal studies, treatment with the compound was associated with slower tumor progression and longer survival compared with untreated controls. The investigators also reported that the inhibitor enhanced the effects of gemcitabine, a commonly used chemotherapy drug for pancreatic cancer, suggesting that GIPC1 blockade may eventually become part of a combination-treatment strategy.

Pancreatic cancer remains one of the most challenging malignancies because it is often detected only after it has reached an advanced stage. Tumors can grow around blood vessels and other critical structures, while dense connective tissue can restrict the delivery of medicines. Cancer cells also display extensive molecular diversity, allowing some populations to survive chemotherapy and repopulate the tumor. These features contribute to the disease’s poor prognosis and have created an urgent need for treatments that attack pancreatic cancer through mechanisms distinct from established chemotherapy.

The study also produced early evidence that the experimental drug may affect the tumor microenvironment, the network of noncancerous cells, extracellular matrix components and signaling molecules surrounding a tumor. This environment can shelter malignant cells, suppress immune activity and reduce the penetration or effectiveness of drugs. Although the precise changes induced by GIPC1 inhibition require further investigation, the findings raise the possibility that targeting the protein could make tumors more responsive to chemotherapy or other future treatments.

Artificial intelligence is increasingly being used in drug discovery to predict how molecules may fit into protein structures, estimate binding properties and prioritize candidates for laboratory testing. Its value is especially apparent when researchers are addressing proteins without conventional enzyme pockets or other easily targeted sites. However, computational predictions do not establish that a compound will be effective or safe in a living organism. Candidate molecules must still undergo biochemical validation, cell-based testing, animal studies and, eventually, carefully controlled clinical trials.

“Our study demonstrates that AI can help us identify entirely new therapeutic opportunities against targets that have historically been considered undruggable,” says Debabrata (Dev) Mukhopadhyay, Ph.D., senior author and a cancer researcher at Mayo Clinic in Florida. He cautions that the findings remain preclinical, but says they provide a foundation for the next phase of research. The team must now determine how the inhibitor behaves in the body, establish an effective dose, assess potential toxicity and clarify which patients might benefit from treatment.

The experimental therapy is not approved for use in people and has not yet entered clinical trials. Additional studies will be needed to confirm its selectivity for GIPC1, evaluate possible effects on healthy tissues and test whether its activity can be reproduced across additional pancreatic cancer models. If those investigations are successful, the work could demonstrate that AI-guided targeting of protein–protein interaction domains is a practical route toward new cancer medicines, particularly for tumors that have remained resistant to conventional therapeutic design.

Subject of Research: AI-guided discovery of a small-molecule inhibitor targeting the GIPC1 PDZ domain for pancreatic ductal adenocarcinoma.

Article Title: AI-driven discovery and validation of a GIPC1 PDZ domain inhibitor for pancreatic ductal adenocarcinoma

Web References: Mayo Clinic; Cell Reports study; Mayo Clinic pancreatic cancer information

References: Cell Reports, “AI-driven discovery and validation of a GIPC1 PDZ domain inhibitor for pancreatic ductal adenocarcinoma,” published 31 July 2026.

Keywords: pancreatic cancer, pancreatic ductal adenocarcinoma, GIPC1, PDZ domain, artificial intelligence, AI drug discovery, cancer therapy, gemcitabine, small-molecule inhibitor, tumor microenvironment, Mayo Clinic, preclinical research

PFKL is one of the most important regulatory proteins in the pathway that converts glucose into usable cellular energy. Glycolysis takes place in the cytoplasm and breaks one molecule of glucose into two molecules of pyruvate, producing ATP while also generating metabolic intermediates needed to make nucleotides, amino acids and lipids. The reaction controlled by phosphofructokinase is especially significant because it acts as a metabolic checkpoint. Once glucose-derived carbon passes through this stage, the cell is strongly committed to processing it through glycolysis.

Cancer cells frequently alter this pathway. Many tumors consume glucose at unusually high rates even when oxygen is available, a phenomenon historically associated with the Warburg effect. This metabolic reprogramming does not simply provide energy. It gives malignant cells flexible access to the molecular building blocks required for DNA replication, membrane production and rapid division. Because glycolysis is so central to tumor biology, enzymes within the pathway have long attracted attention as possible drug targets. The challenge has been finding a way to interfere with cancer metabolism without causing unacceptable damage to healthy tissues.

The new work takes an unusual approach. Instead of blocking PFKL, the researchers developed a molecule that activates it. The compound forms a covalent interaction with the enzyme, creating a chemically stable attachment at a specific site on the protein. Covalent drugs can offer prolonged target engagement because the compound remains linked to its target after the initial binding event. That feature may be especially useful for enzymes whose activity needs to be shifted persistently rather than temporarily.

Activating PFKL can push glucose metabolism forward, but increased pathway activity does not necessarily benefit a tumor. Glycolysis is a network governed by tightly balanced flows of carbon, energy and signaling molecules. Driving one control point beyond the range that cancer cells can accommodate may create metabolic stress. Excessive glycolytic activity can alter the levels of upstream and downstream metabolites, disturb cellular energy management and intensify dependence on nutrients or pathways that tumors cannot easily replace.

The reported compound, therefore, appears to exploit a vulnerability created by cancer’s metabolic ambition. Tumor cells may be heavily invested in high-rate glucose consumption, yet that dependence can become a liability when the pathway is forcibly accelerated. A covalent PFKL activator could act like a metabolic accelerator that pushes malignant cells beyond a tolerable operating limit. Rather than starving tumors by removing glucose, the strategy aims to make their existing glucose-processing program harmful to their survival.

This concept is notable because most efforts to target cancer metabolism have focused on inhibition. Blocking glycolysis can reduce ATP production or deprive cells of biosynthetic intermediates, but normal tissues also rely on glucose metabolism, creating potential toxicity concerns. Enzyme activation offers a different therapeutic logic: selectively destabilizing the metabolic state on which tumor cells depend. The success of this approach will depend on how strongly the compound affects PFKL in cancer compared with healthy cells, as well as how different tumor types manage the resulting metabolic pressure.

The study’s title indicates that the activator suppressed tumor growth, a finding that moves the concept beyond biochemical enzyme assays. To establish whether such a molecule can become a practical therapy, researchers will need to define its selectivity, pharmacological behavior, distribution through the body and safety profile. They will also need to determine whether tumors can adapt by reducing glucose uptake, switching to alternative fuels or altering the expression of other glycolytic enzymes. Cancer cells are remarkably capable of rewiring metabolism, and resistance mechanisms will be a central question for future work.

The covalent nature of the compound also makes careful chemical characterization essential. A useful covalent drug must react efficiently with its intended protein while minimizing unwanted modification of other cellular proteins. Researchers typically examine target engagement, proteome-wide selectivity and the durability of the biological response. These studies can reveal whether the compound’s effects arise primarily from PFKL activation or from broader chemical reactivity. If the molecule demonstrates a favorable selectivity profile, it could provide a framework for developing additional covalent activators against metabolic enzymes.

The findings place PFKL in a growing category of drug targets whose therapeutic potential may lie not in shutting them down, but in pushing them into an abnormal state. By turning a central glycolytic control point against cancer cells, the researchers have highlighted a strategy that combines chemical biology with the emerging science of metabolic stress. The work does not mean that a new cancer treatment is immediately available, but it offers a provocative blueprint: sometimes the most effective way to attack a tumor’s fuel system may be to make it burn too intensely to survive.

Subject of Research: Cancer metabolism and covalent activation of phosphofructokinase L (PFKL) to suppress tumor growth

Article Title: A covalent PFKL activator suppresses tumor growth

Article References: Jiang, X., Lynch, E.M., Lyu, C. et al. “A covalent PFKL activator suppresses tumor growth.” Nature Chemical Biology (2026). https://doi.org/10.1038/s41589-026-02289-9

Image Credits: AI Generated

DOI: https://doi.org/10.1038/s41589-026-02289-9

Keywords: cancer metabolism, PFKL, phosphofructokinase, glycolysis, covalent activator, tumor growth, metabolic stress, chemical biology, cancer therapeutics

The VEGFR2 receptor plays a pivotal role in angiogenesis, the process through which new blood vessels form from existing vessels. This process is crucial for tumor growth and metastasis, as the development of a robust blood supply is often a prerequisite for tumors to thrive and expand. By inhibiting this receptor with the newly developed fusion protein, researchers aim to disrupt the blood supply to tumors, effectively starving them of the necessary oxygen and nutrients they require for survival and growth.

On the other hand, DR5 is a crucial player in apoptosis, the programmed cell death mechanism that can be harnessed to eliminate cancer cells. The fusion protein’s ability to target this receptor opens a promising avenue for enhancing the sensitivity of cancer cells to treatments that induce cell death, providing a dual attack strategy against tumors. This dual targeting aspect is the cornerstone of the researchers’ hypothesis that combining strategies can yield more potent therapeutic outcomes than traditional single-target approaches.

Utilizing multimodal microangiography, the study assessed the physiological impact of the fusion protein in a preclinical model. This imaging technique allowed for real-time visualization of microvascular changes and provided invaluable data on tumor perfusion and vascular integrity before and after the administration of the treatment. This method not only enhances the understanding of treatment effects but also aids in the early detection of therapeutic success or potential resistance.

Upon administration of the fusion protein, significant reductions in tumor volume were recorded. The data illustrated that not only did the multivalent protein incapacitate blood vessel formation, but it also initiated substantial apoptosis across various cancer cell lines tested. This combination of effects resulted in a marked improvement in survival rates for the treated subjects in the study.

The implications of these findings extend beyond the laboratory. By advancing our understanding of the complex interactions between cancer biology and therapeutic mechanisms, this study paves the way for developing more effective treatments tailored to individual patient profiles. As researchers refine this fusion protein, the potential for clinical application in human patients becomes increasingly tangible.

Moreover, the careful design of the multivalent fusion protein raises the bar for future drug development. Incorporating dual targeting systems may become a new standard in cancer therapeutics, leading to drugs that can attack tumors from multiple angles simultaneously. This paradigm shift holds promise not only for oncology but can extend to other fields where targeted therapies are crucial.

The work of Druzhkova, Orlova, Fedulova, and their team underscores the importance of collaboration in scientific research. The expertise of each author contributed to the innovative approach taken towards the development of the fusion protein, highlighting the role of multidisciplinary teams in pushing the boundaries of what is possible in medical science.

In addition to the compelling clinical implications, the detailed technical aspects of the study provide a rich source of knowledge for future researchers. The methods employed in assessing the fusion protein’s efficacy, including the elaborate protocols for multimodal microangiography, offer a framework that other scientists can build upon. This emphasis on sharing methodological insights is essential for fostering innovation and accelerating progress in the field.

As with any groundbreaking study, it is essential to approach the findings with a degree of cautious optimism. While the preclinical results are promising, further studies, including clinical trials, are required to ascertain the safety and efficacy of this multivalent fusion protein in humans. Regulatory processes will need to be navigated carefully to ensure that the advancements achieved in the laboratory translate effectively into patient care.

Peer-reviewed articles such as this one are critical for the scientific community as they catalyze discussions on new therapeutic avenues. As this research gains attention, it may inspire a new wave of studies investigating similar dual-targeting strategies, potentially leading to a renaissance in cancer treatment methodologies currently in use. Furthermore, the symbiotic relationship between research and clinical practice highlights the importance of continual exploration within the field.

In conclusion, the groundbreaking study on the multivalent fusion protein targeting VEGFR2 and DR5 represents a hopeful advancement in the ongoing battle against cancer. By demonstrating the potential of a dual-targeting approach and utilizing cutting-edge imaging technologies, this research not only contributes to scientific knowledge but also carries the promise of innovative therapies that could provide new hope for patients facing this devastating disease. As we look forward to the potential of these findings to influence future treatments, the excitement within the scientific community is palpable.

The relentless pursuit of effective cancer treatments remains a hallmark of modern medicine. As scientists continue to unravel the complexities of tumor biology and develop novel therapeutic strategies, studies like those conducted by Druzhkova and colleagues serve as beacons of hope, illuminating the path toward more effective, personalized treatment options for cancer patients worldwide.

Subject of Research: Multivalent fusion protein targeting VEGFR2 and DR5 in cancer therapy.

Article Title: Multivalent fusion protein targeting VEGFR2 and DR5 receptors: assessing the antiangiogenic and antitumor effects via multimodal microangiography.

Article References:

Druzhkova, I.N., Orlova, A.G., Fedulova, A.S. et al. Multivalent fusion protein targeting VEGFR2 and DR5 receptors: assessing the antiangiogenic and antitumor effects via multimodal microangiography. J Transl Med 23, 949 (2025). https://doi.org/10.1186/s12967-025-06859-8

Image Credits: AI Generated

DOI: 10.1186/s12967-025-06859-8

Keywords: multivalent fusion protein, VEGFR2, DR5, antiangiogenic, antitumor, multimodal microangiography, cancer therapy, apoptosis, therapeutic efficacy.

Liver cancer, particularly hepatocellular carcinoma (HCC), represents a formidable clinical challenge worldwide due to its notorious genetic heterogeneity and the scarcity of effective therapeutic targets. Conventional chemotherapeutic regimens often fall short, plagued by issues such as drug resistance and systemic toxicity. Consequently, there exists an urgent need to identify molecular targets that not only hamper tumor proliferation but also restore the normal cellular phenotype, thereby improving patient prognosis.

Central to this study is the role of Aurora kinases, a family of serine/threonine kinases known for their crucial involvement in mitotic progression and chromosomal stability. Dysregulation of Aurora kinases, especially Aurora A and Aurora B, has been implicated in tumorigenesis across various cancers, making them attractive candidates for targeted inhibition. However, their precise function in liver cancer differentiation had remained largely unexplored until now.

The investigators deployed potent Aurora kinase inhibitors, notably Alisertib and ENMD-2076, to assess their capacity to influence liver cancer cell behavior. Their experimental approach combined rigorous cellular assays with comprehensive gene expression analyses, revealing that treatment with these inhibitors not only curtailed cellular proliferation but also triggered a profound phenotypic shift. Remarkably, treated liver cancer cells exhibited transcriptional upregulation of a suite of hepatic differentiation markers, indicating a reversion toward a more differentiated, less malignant state.

This differentiation phenomenon was further characterized by a concomitant downregulation of malignancy-associated markers, underscoring the dual anti-tumorigenic effects of Aurora kinase inhibition. Importantly, these phenotypic changes persisted beyond the active presence of the drugs, maintained for several days post-withdrawal, suggesting a durable reprogramming of cancer cell identity—a feature that could translate into lasting clinical benefits.

Mechanistically, the study posits that Aurora kinase inhibitors mediate their effects through a bifurcated mechanism: the direct suppression of mitotic progression impairs unchecked cell division, while the induction of differentiation pathways reinstates hepatic cellular functions lost during oncogenesis. Transcriptomic profiling indicated activation of key hepatic transcription factors and metabolic genes, which collectively drive the maturation of malignant cells toward a more benign lineage-committed phenotype.

These findings also invite a reconsideration of how targeted therapies may be designed. Rather than exclusively striving to eradicate cancer cells via cytotoxicity, fostering differentiation represents an innovative paradigm that may mitigate adverse effects and circumvent resistance. By coaxing liver cancer cells to regain functionality akin to normal hepatocytes, Aurora kinase inhibitors could restore tissue homeostasis and inhibit tumor progression in a more physiologically congruent manner.

Furthermore, the inhibitors utilized—Alisertib and ENMD-2076—have demonstrated favorable pharmacokinetic and safety profiles in prior clinical evaluations across multiple cancer types. Their efficacy in inducing differentiation in liver cancer cells opens new vistas for clinical translation, potentially enabling combination regimens that integrate differentiation therapy with existing cytotoxic or immunotherapeutic modalities to achieve synergistic effects.

The implications of this study extend beyond the immediate clinical context as well. Understanding the molecular crosstalk between cell cycle regulation and differentiation not only enriches our comprehension of liver cancer biology but also fuels the development of next-generation therapeutics aimed at restoring cellular identity. Moreover, since Aurora kinases are universally expressed and implicated in diverse malignancies, the therapeutic concepts elucidated here may hold translational relevance across a spectrum of cancers.

Researchers emphasize that while these preclinical findings are promising, rigorous clinical investigations are imperative to evaluate the safety, optimal dosing, and long-term efficacy of Aurora kinase inhibitors in liver cancer patients. Additionally, elucidating the molecular determinants of responsiveness will be critical to stratify patients who stand to benefit most from differentiation-based therapies.

In summary, this study represents a significant leap forward in liver cancer research, revealing that targeting Aurora kinases extends beyond mere blockade of proliferation to encompass the induction of cellular differentiation. This dual action can potentially reshape therapeutic strategies aimed at this formidable malignancy, offering a beacon of hope for improved patient outcomes in the near future.

Subject of Research: Liver cancer treatment via Aurora kinase inhibition and induction of cellular differentiation.

Article Title: [Not Provided]

News Publication Date: [Not Provided]

Web References:
http://dx.doi.org/10.1007/s11427-023-2795-2

References:
[Study published in Science China Life Sciences, DOI: 10.1007/s11427-023-2795-2]

Image Credits: [Not Provided]

Keywords: Aurora kinases, liver cancer, hepatocellular carcinoma, cellular differentiation, Alisertib, ENMD-2076, targeted therapy, cancer biology, tumor proliferation, hepatic gene expression