Engineered protein silences harmful T cells in autoimmune disease

An engineered protein turns off the kind of immune cells most likely to damage tissue as part of Type-1 diabetes, hepatitis, multiple sclerosis, shows a new study in mice.

In these autoimmune diseases, T cells mistakenly target the body's own tissues instead of invading viruses or bacteria as they would during normal immune responses. Treatments focused on T cells have been elusive because blocking their action broadly weakens the immune system and creates risk for infections and cancer.

Published online June 30 in the journal Cell, the study revealed that holding closely together two protein groups (signaling complexes) on T cells, including one found more often on T cells involved autoimmune disease, shuts down those T cells in a limited way.

Led by researchers at NYU Langone Health, the Chinese Academy of Sciences, and Zhejiang University, the study built on biology newly discovered by the team to design an antibody that attached to both T cell signaling complexes, the T cell receptor and the LAG-3 checkpoint, held them closely together, and eliminated autoimmune tissue damage in three mouse models of disease.

Antibodies are proteins made by the immune system that label specific markers on cells for notice by the immune system. Researchers learned decades ago to engineer antibodies to target certain molecules as treatments, and more recently, antibodies that attach to two targets.

Our findings reveal an intricate mechanism that enables a careful treatment approach to T-cell driven autoimmune diseases, which currently lack effective immunotherapies."

Jun Wang, PhD., co-senior study author, assistant professor, Department of Pathology at NYU Grossman School of Medicine

Held in place

The study results are based on the presence on T cells of T-cell receptors (TCRs) and checkpoints. TCRs, although shaped so that bits of invading bacteria or viruses fit into them to activate the T cell, are turned on by the body's own proteins in autoimmune diseases. Checkpoints like LAG-3 are also turned on by specific signaling partners, but when this occurs they have the opposite effect of TCRs, suppressing the T cell's activity.

Also important to the new study results is that TCR-triggering molecules must be presented to T cell receptors by another set of immune cells that "swallow" foreign (e.g., microbial) or bodily substances and display on their surfaces through protein groups called major histocompatibility complexes (MHC-II) just the small protein pieces that activate a given TCR.

"We discovered that, as a T cell's surface draws close to the MHC-II presenting its TCR trigger molecule, the T cell receptor gets particularly close to LAG-3", said co-first author Jasper Du, a third-year medical student in Dr. Wang's lab. "For the first time, we found that this proximity is central to the ability of LAG-3 to dial back T cell activity."

Mechanistically, the research team found that the proximity of LAG-3 lets it loosely stick to part of the T cell receptor called CD3ε (like two oily globs interacting). This attachment was found to pull on CD3ε enough to disrupt its interaction an enzyme called Lck, which is crucial for T cell activation. MHC-II can theoretically attach to LAG-3 and TCR at the same time, but not frequently enough to maximize LAG-3's ability to dial down T cells, the researchers said.

In addition, "checkpoints" like LAG-3 are used by the immune system to turn off T cells when the right signals, given off by normal cells, dock in to avert self-attack (autoimmunity). Cancer cells put off signaling molecules that dock into checkpoints and sabotage the ability of T cells to attack them. Therapies called checkpoint inhibitors counter this effect.

LAG-3 turns off T cells, but less easily due to its spatial requirements than another checkpoint called PD-1. This feature makes LAG-3 inhibitors weaker as anti-cancer cancer treatment than PD-1-inhibiting antibody treatments that have become a mainstay, but likely better when the immune system is overactive, and targeted T cell suppression is required for maximum safe effect.

Based on their discovery of the critical role of TCR proximity in LAG-3 function, the research team designed a molecule that enforces LAG-3/TCR proximity to achieve better LAG-3-dependent TCR inhibition and suppression of T cell responses. Their "bi-specific" antibody held LAG-3 and the T cell receptor together more strongly than MHC-II, and without depending on it.

The current authors' bispecific antibody, named the LAG-3/TCR Bispecific T cell Silencer or BiTS, potently suppressed T cell responses and lessened inflammatory damage to insulin-producing cells (insulitis) in BiTS-treated mice with a version of Type 1 diabetes. In autoimmune models of hepatitis, BiTS treatment reduced T cell infiltration and liver damage.

With the diabetes and hepatitis disease models largely driven by one type of T cells (CD8+), the team also used a mouse model of multiple sclerosis known to be driven by a second major T cell type (CD4+). The team treated mice prone to develop multiple sclerosis with short-term, preventive BiTS prior to the onset of disease symptoms, and BiTS-treated mice had reduced disease by a standard measure.

"Our study advances our understanding of LAG-3 biology and may foster more proximity-based, spatially-guided therapeutic designs like BiTS as immunotherapy for other human diseases," said co-first author Jia You, a research scientist in Dr. Wang's lab.

Along with Dr. Wang, corresponding authors of the study were Jack Wei Chen of the Department of Cell Biology and Department of Cardiology at the Second Affiliated Hospital Zhejiang University School of Medicine in China; as well Jizhong Lou of the State Key Laboratory of Epigenetic Regulation and Intervention, Institute of Biophysics, Chinese Academy of Sciences.

Also other authors from the NYU Grossman School of Medicine were Jia Liu, Qiao Lu, Connor James, Ryan Foster, and Eric Rao in the Department of Pathology at New York University Grossman School of Medicine; Meng-ju Lin and Catherine Pei-ju Lu in the Hansjörg Wyss Department of Plastic Surgery and Department of Cell Biology; and Michael Cammer at the Microscopy Core, Division of Advanced Research Technologies, and Shohei Koide of the Perlmutter Cancer Center. Also making important contributions were Hui Chen at the State Key Laboratory of Epigenetic Regulation and Intervention, Institute of Biophysics, Chinese Academy of Sciences, and Yong Zhang from University of Chinese Academy of Sciences; Wei Hu and Jie Gao at The Second Affiliated Hospital, Zhejiang University School of Medicine; and Weiwei Yin in the Key Laboratory for Biomedical Engineering of the Ministry of Education, College of Biomedical Engineering and Instrument Science, also at Zhejiang University.

The study was supported principally by a translational advancement award from the Judith and Stewart Colton Center for Autoimmunity at NYU Langone Health. Also funding the study were a Cancer Center Support Grant P30CA016087, NIH grant S10OD021727, the NYU melanoma SPORE and NIH R37CA273333, and an NIH/NIAMS T32 grant (AR069515-07). The biophysical analysis part of this work was also supported by multiple grants from National Science Foundations of China (32090044, T2394512, 32200549, and T2394511).

Dr. Wang, Du and You are listed as inventors of pending patents related to the study. NYU Langone Health and its Technology Opportunities & Ventures have formed a related startup company, Remunix Inc., with Dr. Wang as founder and shareholders, to license and commercialize the patents. In addition, Dr. Wang serves as a consultant for Rootpath Genomics, Bristol Myers Squibb, LAV, Regeneron, and Hanmi. Dr. Koide has reported interests in Aethon Therapeutics and Revalia Bio not related to this study. These relationships are managed in keeping with the policies of NYU Langone Health.