Tag: Biotech

A new study in Cell describes an AI‑enabled strategy that could accelerate the search for next‑generation CAR T cell targets—an enduring bottleneck in expanding the therapy beyond blood cancers. The work, titled “AI‑driven discovery of GPNMB CAR T cells as a multi‑cancer therapy,” was led by researchers at the Perelman School of Medicine at the University of Pennsylvania and Penn’s Abramson Cancer Center, with collaborators at the Icahn School of Medicine at Mount Sinai and RWTH Aachen University.

The Penn team developed a human‑in‑the‑loop AI framework designed to systematically nominate antigens suitable for CAR T cell therapy. Rather than replacing expert judgment, the system integrates large language models (LLMs) with single-cell RNA sequencing datasets from human skin cancer and healthy tissue to generate and refine target lists that scientists then evaluate experimentally.

The challenge is well known: while CAR T therapies have transformed treatment for several hematologic malignancies, identifying safe, selective targets in solid tumors remains slow and labor‑intensive. “Discovering a good CAR target is like trying to find a needle in a haystack, except the haystack keeps growing as more sequencing data becomes available,” said lead author Daniel Baker, PhD, who completed the work under the mentorship of Carl June, MD, and Zoltan Arany, MD, PhD. LLMs, Baker added, excel at scanning broad datasets, while human experts “go deep”—a complementary pairing the team sought to formalize.

To test the framework, the researchers focused on skin cancer, integrating four publicly available single‑cell RNA‑seq datasets with additional public resources. More than 10,000 potential antigens were filtered using criteria relevant to CAR T design, including tumor composition, tissue specificity, and clinical feasibility. Multiple LLMs then repeatedly simulated target nomination—1,000 independent runs—to reduce noise and mitigate hallucinations. The resulting consensus list was reviewed by the team, who selected Glycoprotein non-metastatic melanoma protein B (GPNMB) as the top candidate.

The researchers then engineered a GPNMB‑directed CAR T cell and validated its activity across several preclinical models. In mouse studies, the CAR T cells eliminated tumors not only in melanoma—the original focus of the dataset—but also in monoblastic leukemia and colorectal adenocarcinoma, suggesting broader therapeutic potential. These findings align with the paper’s highlight that GPNMB is expressed across a wide range of tumor types.

The full framework is included in the methods section to enable adoption by other groups. The Penn team plans to apply the approach to additional cancer types and continue advancing the GPNMB CAR T candidate toward potential clinical translation.

According to June, “this study represents one of the first uses of large language models in the field of cell and gene therapy, including CAR T cell therapy.” The framework is intentionally modular and disease‑agnostic, designed to accommodate new datasets and future LLMs as they evolve. Arany emphasized the broader implications: “This is only the tip of the iceberg, as agentic AI is on the rise.”

The post AI Framework Surfaces New CAR T Target with Multi‑Cancer Potential appeared first on GEN – Genetic Engineering and Biotechnology News.

New research headed by scientists at Baylor College of Medicine suggests that when bacteria are under antibiotic attack, it is not “every man for himself.” The team developed a genetic system in Escherichia coli to track how the cells transferred proteins between them. The results indicated that bacterial populations work as a team to survive antibiotics, pooling their resources and helping quiescent or dormant cells survive. Using different techniques, including high-resolution imaging, the team found that antibiotic treatment induced the transfer of proteins between different E. coli strains, and between E. coli and other species of bacteria.

They discovered that antibiotics stimulate bacteria to differentiate into groups of what they describe as vesicle-producing, and protein-receiving cells, and that antibiotic “persisters” with reduced protein synthesis acquire proteins released by their neighbors. The discoveries may help to explain why some bacteria are hard to eliminate, and also point to potential future approaches to improve antibiotic effectiveness.

“Antibiotics are designed to kill bacteria or stop them from growing,” said Christophe Herman, PhD, professor of molecular and human genetics and of molecular virology and microbiology at Baylor. “Yet many times, antibiotics leave behind a small group of survivors. These survivors are not genetically resistant; instead, they temporarily shut down certain parts of their metabolism, entering a dormant-like state that allows them to endure treatment and later regrow. Understanding how survivors form and remain is a major challenge in fighting persistent infections.”

Herman is senior and co-corresponding author of the team’s published paper in Science,” (“Antibiotics stimulate protein transfer to persister cells,”) in which the team further explained, “Protein uptake enhanced the antibiotic persistence of recipient cells, revealing that vesicle exchange promotes bacterial survival during antibiotic treatment.”

Scientists have long known that bacteria can help each other resist antibiotics by sharing genes that provide antibiotic resistance. But as the authors pointed out, “Whereas horizontal gene transfer is known to spread antibiotic resistance genes, far less is understood about the mechanisms and effects of horizontal protein transfer.”

Antibiotic treatment stimulates vesicle production, so for their current study, Herman and colleagues investigated whether bacteria could also directly share proteins. Previous studies had indicated that bacteria can share proteins, but the experimental evidence was not clear. “To directly measure horizontal transfer, we constructed a genetic system in Escherichia coli consisting of a donor and a recipient strain.”

First author Alice X. Wen, a Baylor McNair Scholar in the Medical Scientist Training Program (MD/PhD), working in the Herman lab, further explained, “To detect protein transfer, we designed a sensitive system using the bacterium Escherichia coli. We engineered one group of bacteria (donors) to make a special enzyme called Cre, and another group of the same bacteria (recipients) to contain a genetic ‘switch’ that could only flip if Cre protein entered the recipient.”

Using this system, investigators discovered that when donor and recipient bacteria were grown together, protein transfer occurred but was rare under normal conditions. In contrast, when the bacteria were exposed to low, non-lethal levels of antibiotics, protein transfer increased by thousands of times. “We then investigated how proteins were moving from one cell to another,” Wen said. “We found that the transfer still occurred when donor cells were removed, leaving behind only the liquid in which they had grown. This ruled out direct cell-to-cell contact and pointed to something released into the environment.”

By combining biochemical techniques and advanced microscopy, the team discovered that the proteins were transported by tiny membrane vesicles. These structures, which look like tiny bubbles, are made of bacterial membrane that pinch off from cells and float freely. “Bacterial membrane vesicles, which contain proteins, have been proposed as mediators of horizontal protein transfer,” they pointed out. “Additionally, antibiotic treatment stimulates vesicle production.”

Looking closer at their experimental system, the team found that the recipient cells showed strong signs of dormancy—these cells slowed down protein production, reduced their metabolism, and activated genes associated with persistence, such as HipA. “Recipient cells with high HipA activity were more likely to take up protein-carrying vesicles and survive antibiotic treatment,” Wen said. “When HipA was removed, both protein uptake and survival dropped.”

Protein transfer also helped dormant bacteria survive exposure to lethal antibiotic doses after vesicle transfer; that is, exposing cells to an increased concentration of vesicles before antibiotic treatment led to increased survival. “Protein uptake enhanced the antibiotic persistence of recipient cells, revealing that vesicle exchange promotes bacterial survival during antibiotic treatment,” the authors stated. The results suggested that transferred proteins helped dormant cells endure stress while their own protein production was shut down. “Uptake of key proteins, such as ribosomal components, metabolic enzymes, or DNA repair factors, from active neighbors may help persisters endure proteome-damaging stress despite reduced protein synthesis.”

Herman said, “Our study shows that antibiotics cause a genetically identical group of bacteria to differentiate into two distinct groups: donor cells that respond by releasing protein-filled vesicles, and recipient cells that become dormant but capable of taking up proteins from incoming vesicles, which helps them survive,” Herman said. “This teamwork allows vulnerable members of a bacterial population to persist in the face of a potentially deadly antibiotic attack.”

The researchers are interested in identifying the proteins in vesicles that contribute to recipient persistence. Understanding donor-recipient interactions among bacteria opens new doors in the fight against chronic and persistent infections. In conclusion, the authors stated that their work “… reveals that antibiotics stimulate the differentiation of bacteria into distinct groups of vesicle-producing and protein-receiving cells, which allows antibiotic persisters with decreased protein synthesis to acquire proteins secreted from active neighbors. New strategies to eliminate persisters could be developed by inhibiting or hijacking horizontal protein transfer.”

The post Antibiotics Trigger Protein Sharing Among Bacteria, Aiding Persister Cells appeared first on GEN – Genetic Engineering and Biotechnology News.