The research, published on January 7 in Cell Stem Cell, removes a major barrier that has slowed the development, affordability, and large-scale production of cell therapies. By solving this problem, the work could help make off-the-shelf treatments more accessible and effective for conditions such as cancer, infectious diseases, autoimmune disorders, and more.

“Engineered cell therapies are transforming modern medicine,” said co-senior author Dr. Peter Zandstra, professor and director of the UBC School of Biomedical Engineering. “This study addresses one of the biggest challenges in making these lifesaving treatments accessible to more people, showing for the first time a reliable and scalable way to grow multiple immune cell types.”

The Promise and Limits of Living Drugs

Over the past several years, engineered cell therapies such as CAR-T treatments have produced dramatic, sometimes lifesaving results for people with cancers that were once considered untreatable. These therapies work by reprogramming a patient’s immune cells to recognize and destroy disease, effectively turning those cells into ‘living drugs’.

Even with their success, cell therapies remain costly, complex to manufacture, and out of reach for many patients around the world. One key reason is that most existing treatments rely on a patient’s own immune cells, which must be collected and specially prepared over several weeks for each individual.

“The long-term goal is to have off-the-shelf cell therapies that are manufactured ahead of time and on a larger scale from a renewable source like stem cells,” said co-senior author Dr. Megan Levings, a professor of surgery and biomedical engineering at UBC. “This would make treatments much more cost-effective and ready when patients need them.”

Cancer cell therapies are most effective when two types of immune cells work together. Killer T cells directly attack infected or cancerous cells. Helper T cells, which act as the immune system’s conductors — detecting health threats, activating other immune cells and sustaining the immune responses over time — play a central coordinating role.

While scientists have made progress using stem cells to create killer T cells in the lab, they have not been able to reliably generate helper T cells until now.

“Helper T cells are essential for a strong and lasting immune response,” said Dr. Levings. “It’s critical that we have both to maximize the efficacy and flexibility of off-the-shelf therapies.”

A Key Advance Toward Stem Cell Based Immune Therapies

In the new study, the UBC research team addressed this long-standing challenge by carefully adjusting biological signals that guide how stem cells develop. This approach allowed them to precisely control whether stem cells became helper T cells or killer T cells.

The scientists found that a developmental signal known as Notch plays an important but time-sensitive role in immune cell formation. Notch is necessary early in development, but if the signal stays active for too long, it blocks the formation of helper T cells.

“By precisely tuning when and how much this signal is reduced, we were able to direct stem cells to become either helper or killer T cells,” said co-first author Dr. Ross Jones, a research associate in the Zandstra Lab. “We were able to do this in controlled laboratory conditions that are directly applicable in real-world biomanufacturing, which is an essential step toward turning this discovery into a viable therapy.”

The team also confirmed that the lab-grown helper T cells functioned like real immune cells, not just in appearance but in behavior. The cells showed signs of full maturity, carried a wide variety of immune receptors, and were able to develop into specialized subtypes with distinct immune roles.

“These cells look and act like genuine human helper T cells,” said co-first author Kevin Salim, a UBC PhD student in the Levings Lab. “That’s critical for future therapeutic potential.”

Researchers say the ability to generate both helper and killer T cells, and to carefully control their balance, could greatly improve the effectiveness of stem cell-derived immune therapies.

“This is a major step forward in our ability to develop scalable and affordable immune cell therapies,” said Dr. Zandstra. “This technology now forms the foundation for testing the role of helper T cells in supporting the elimination of cancer cells and generating new types of helper T cell-derived cells, such as regulatory T cells, for clinical applications.”

The proteins at the center of the research are known as delta-type ionotropic glutamate receptors, or GluDs. These proteins are known to play an important role in how neurons communicate with each other. According to the researchers, mutations in GluDs have been linked to psychiatric disorders, including anxiety and schizophrenia. Despite this connection, scientists have struggled for years to understand exactly how these proteins work, making it difficult to design treatments that could regulate their activity.

“This class of protein has long been thought to be sitting dormant in the brain,” says Edward Twomey, Ph.D., assistant professor of biophysics and biophysical chemistry at the Johns Hopkins University School of Medicine. “Our findings indicate they are very much active and offer a potential channel to develop new therapies.”

The study describing these findings was published in Nature.

Imaging Reveals How GluDs Function

To better understand GluDs, Twomey and his team used cryo-electron microscopy, an advanced imaging technique that allows scientists to visualize proteins in fine detail. Their analysis showed that GluDs contain an ion channel at their center. This channel holds charged particles that help the proteins interact with neurotransmitters (electrical signals that allow brain cells to communicate with one another).

“This process is fundamental for the formation of synapses, the connection point where cells communicate,” says Twomey.

Implications for Movement Disorders and Mental Illness

The discovery could help accelerate the development of drugs for cerebellar ataxia, a disorder that affects movement and balance. Cerebellar ataxia can result from stroke, head injury, brain tumors, or certain neurodegenerative diseases, and it may also cause memory problems. In this condition, GluDs become “super-active” even when there is no electrical signaling in the brain. Twomey explains that a potential treatment approach would involve developing drugs that block this excessive activity.

In schizophrenia, the situation appears to be reversed. GluDs are less active than normal, and Twomey says future drugs could aim to boost their activity instead.

Potential Links to Aging and Memory Loss

The findings may also be relevant to aging and memory decline. Because GluDs help regulate synapses, drugs that target these proteins could help maintain synapse function over time. Synapses are essential for learning, memory, and the formation of thoughts.

“Because GluDs directly regulate synapses, we could potentially develop a targeted drug for any condition where synapses malfunction,” Twomey says.

Next Steps and Ongoing Research

Looking ahead, Twomey says he plans to collaborate with pharmaceutical companies to further develop this therapeutic target. His team is also studying specific GluD mutations that have been directly linked to schizophrenia, anxiety, and other psychiatric disorders. The goal is to better understand how these conditions progress and to design more precise treatments.

Other Johns Hopkins scientists who contributed to the study include Haobo Wang, Fairine Ahmed, Jeffrey Khau, and Anish Kumar Mondal.

The Johns Hopkins University has filed a patent covering the techniques used to measure electrical currents from GluDs.

Funding for the research came from the National Institutes of Health (R35GM154904), the Searle Scholars Program, and the Diana Helis Henry Medical Research Foundation.

In research published in Physical Review Letters, James Gurian and Simon May introduce a new computational tool designed to study how SIDM affects galaxy formation. Their approach makes it possible to explore types of particle interactions that were previously difficult or impractical to model accurately.

When Dark Matter Interacts With Itself

SIDM is a theoretical form of dark matter whose particles can collide with one another but do not interact with baryonic matter, the familiar matter made of protons, neutrons, and electrons. These collisions conserve energy through what physicists call elastic self-interactions. This behavior can strongly influence dark matter halos, the massive concentrations of dark matter that surround galaxies and help guide their evolution.

“Dark matter forms relatively diffuse clumps which are still much denser than the average density of the universe,” says Gurian, a Perimeter postdoctoral fellow and co-author of the study. “The Milky Way and other galaxies live in these dark matter halos.”

Heat, Energy Flow, and Core Collapse

The self-interacting nature of SIDM can trigger a process known as gravothermal collapse within dark matter halos. This phenomenon arises from a counterintuitive property of gravity, where systems bound by gravity become hotter as they lose energy rather than cooling down.

“You have this self-interacting dark matter which transports energy, and it tends to transport energy outwards in these halos,” says Gurian. “This leads to the inner core getting really hot and dense as energy is transported outwards.” Over time, this process can drive the core of the halo toward a dramatic collapse.

A Missing Link in Dark Matter Modeling

Simulating the structures formed by SIDM has long been a challenge. Existing methods work well only under certain conditions. Some simulations perform best when dark matter is sparse and collisions are rare, while others are effective only when dark matter is extremely dense and interactions are frequent.

“One approach is an N-body simulation approach that works really well when dark matter is not very dense and collisions are infrequent. The other approach is a fluid approach — and this works when dark matter is very dense and collisions are frequent.”

“But for the in-between, there wasn’t a good method,” Gurian says. “You need an intermediate range approach to correctly go between the low-density and high-density parts. That was the origin of this project.”

A Faster and More Accessible Simulation Tool

To solve this problem, Gurian and his co-author Simon May, a former Perimeter postdoctoral researcher now serving as an ERC Preparative Fellow at Bielefeld University, developed a new code called KISS-SIDM. The software bridges the gap between existing simulation methods, delivering higher accuracy while requiring far less computing power. It is also publicly available for other researchers.

“Before, if you wanted to check different parameters for self-interacting dark matter, you needed to either use this really simplified fluid model, or go to a cluster, which is computationally expensive. This code is faster, and you can run it on your laptop,” says Gurian.

Opening the Door to New Dark Matter Physics

Interest in interacting dark matter has grown in recent years, partly due to puzzling features seen in galaxies that may not fit standard models.

“There has been considerable interest recently in interacting dark matter models, due to possible anomalies detected in observations of galaxies that may require new physics in the dark sector,” says Neal Dalal, a member of the Perimeter Institute research faculty.

“Previously, it was not possible to perform accurate calculations of cosmic structure formation in these sorts of models, but the method developed by James and Simon provides a solution that finally allows us to simulate the evolution of dark matter in models with significant interactions,” Dalal says. “Their paper should enable a broad spectrum of studies that previously were intractable.”

Implications for Black Holes and Beyond

The collapse of dark matter cores is especially intriguing because it may leave observable signatures, including possible connections to black hole formation. However, how this process ultimately ends remains an open question.

“The fundamental question is, what’s the final endpoint of this collapse? That’s what we’d really like to do — study the phase after you form a black hole.”

By making it possible to explore these extreme conditions in detail, the new code represents an important step toward answering some of the deepest questions about dark matter and the structure of the universe.

One promising solution is to use existing energy supplies far more efficiently and at lower cost.

A New Approach to Power Efficiency

Researchers at the National Renewable Energy Laboratory (NREL) have developed a new silicon carbide based power module designed to dramatically improve how electricity is converted and delivered. A power module is the housing that contains power electronics, which regulate the flow of electricity between systems. This new design delivers record-breaking efficiency, higher power density, and a manufacturing process that keeps costs low.

The technology is known as NREL’s Ultra-Low Inductance Smart power module, or ULIS. By using silicon carbide semiconductors, ULIS can achieve five times the energy density of earlier designs while taking up less space. That combination allows manufacturers to build equipment that is smaller, lighter, and more energy efficient. The 1200-volt, 400-amp module is well suited for data centers, electrical grids, microreactors, and heavy-duty platforms such as next-generation aircraft and military vehicles.

Why Ultra-Low Inductance Matters

A key advantage of ULIS is its exceptionally low parasitic inductance, which refers to resistance that slows changes in electrical current and limits efficient power conversion. ULIS reduces this resistance by seven to nine times compared with today’s most advanced silicon carbide power modules.

Because the system can switch electrical current extremely quickly and efficiently, it converts more of the available electricity into usable power. That capability allows ULIS to extract significantly more value from the same energy supply, making it a strong candidate for addressing growing global energy needs.

“We consider ULIS to be a true breakthrough,” said Faisal Khan, NREL’s chief power electronics researcher and the principal investigator for the project. “It’s a future-proofed, ultrafast power module that will make the next generation of power converters more affordable, efficient, and compact.”

Built for Reliability in Extreme Conditions

ULIS is designed not only for efficiency, but also for reliability in demanding environments. According to Khan, the lightweight yet powerful module can monitor its own condition and anticipate component failures before they happen.

This feature is especially critical for high-risk applications such as aviation and military operations. For aircraft operating at 30,000 feet or vehicles navigating combat zones, early failure detection can be the difference between mission success and catastrophic loss.

“ULIS was a truly organic effort, built entirely in-house here at NREL,” Khan said. “We are very excited to demonstrate its strengths in real-world settings.”

A Radical Redesign for Lower Cost Manufacturing

Many of ULIS’ performance gains come from a completely new physical design.

Traditional power modules stack semiconductor devices inside box-like packages. ULIS instead arranges its circuitry in a flat, octagonal layout. This disk-shaped structure fits more components into a smaller footprint, reducing both size and weight. At the same time, its innovative current routing minimizes magnetic interference, which helps deliver cleaner electrical output and higher overall efficiency.

“Our biggest concern was that the device switches off and on very quickly, and we needed a layout that wouldn’t create a chokepoint within the design,” said Shuofeng Zhao, an NREL power electronics researcher who designed ULIS’ flux cancellation architecture.

Early concepts explored complex three-dimensional shapes, including designs resembling flowers or hollow cylinders. However, these ideas proved too expensive or difficult to manufacture. The breakthrough came when the team simplified the concept into a nearly two-dimensional structure. Sarwar Islam, another NREL power electronics researcher, proposed the flattened design that balanced performance, cost, and manufacturability.

“We squished it flat, like a pancake,” Zhao said, “and suddenly we had a low-cost, high-performing design that was much easier to fabricate.”

Joshua Major, also part of the NREL power electronics team, developed new fabrication methods that allowed the intricate structure to be produced using only in-house tools and facilities. The result was a design that combined the electrical advantages of three-dimensional systems with the practicality of flat manufacturing.

Flexible Materials and Wireless Control

ULIS also departs from conventional materials. Traditional power modules bond copper directly to rigid ceramic bases to conduct electricity and manage heat. While effective, this approach limits flexibility.

Instead, ULIS bonds copper to a flexible polymer called Temprion. This change produces a thinner, lighter, and more adaptable structure. The material bonds to copper using only heat and pressure, and its components can be machined with widely available equipment. As a result, manufacturing costs fall into the hundreds of dollars rather than the thousands.

Another major advance allows ULIS to operate wirelessly. The module can be controlled and monitored without physical cables, functioning as a self-contained unit. This modular, Lego-like design allows it to be integrated into a wide range of systems, from data center servers to advanced aircraft and military vehicles. A patent for the low-latency wireless communication protocol, led by Sarwar Islam, is currently pending.

Designed for Future Technologies

While ULIS currently relies on advanced silicon carbide semiconductors, the design was intentionally built to evolve. The module can be adapted for future semiconductor materials, including gallium nitride and gallium oxide, which has not yet reached commercial use.

Together, these innovations support a central goal. As societies become increasingly dependent on reliable electricity, ULIS is designed to deliver efficiency without sacrificing dependability.

Where ULIS Could Make the Biggest Difference

ULIS is expected to have broad impact across multiple sectors.

In the U.S. power grid, electricity must be converted into usable forms before it reaches consumers. This process often depends on large, low-frequency equipment that wastes energy. ULIS’ fast switching improves efficiency while its ability to tolerate high temperatures may reduce long-term maintenance costs.

In aviation, the module’s ability to move electricity quickly and conserve energy enables lighter and more powerful converters. This could help make electric vertical takeoff and landing (eVTOL) aircraft more practical and commercially viable.

ULIS could also play a role in future fusion energy systems. Although commercial fusion remains under development, these systems will require compact and reliable pulsed power components. ULIS’ ultralow inductance and durable design make it well suited for that challenge.

As industries pursue more reliable electricity, advanced artificial intelligence, and next-generation vehicles, ULIS is now available for licensing.