The protein in question is CD37, a member of the tetraspanin family. Tetraspanins are small four-pass transmembrane proteins that do not act as classical receptors themselves. Instead, they function as molecular organizers, clustering immune receptors into microdomains on the cell surface and fine-tuning the signals those receptors deliver. On leukocytes, CD37 has been implicated in B cell signaling, phagocytosis, and inflammatory regulation, but its role in ischemic tissue injury had remained largely unexplored. The NYU team began with a simple observation drawn from human tissue: in gastrocnemius muscle samples from patients with PAD, CD37 expression was markedly elevated, hinting that this immune organizer might be part of the disease process rather than a bystander.

To understand what that elevation meant, the researchers applied deconvolution-based immune profiling to transcriptomic data from human PAD muscle. This computational approach infers the composition of immune cell populations within bulk tissue gene expression data, allowing the team to link CD37 levels to specific immune programs. The results were striking. CD37 expression correlated strongly with overall immune infiltration, with a correlation coefficient of 0.51 and a p-value of 0.0002. More tellingly, its strongest relationship was with inflammatory macrophage programs, showing a correlation of 0.48 with a p-value of 0.0003. Macrophages, the tissue-resident scavenger cells capable of both fueling and resolving inflammation, appeared to be the cell type most tightly linked to CD37 abundance in failing human muscle.

Correlation alone cannot establish causation, so the team turned to an experimental system. They used the murine hindlimb ischemia model, a standard technique in which blood flow to a mouse’s hind leg is surgically interrupted, mimicking the ischemic injury of severe PAD. They then compared wild-type mice with genetically engineered mice lacking CD37 entirely, known as Cd37 knockout mice. Over the following weeks, the researchers measured perfusion recovery using laser-based blood flow imaging, examined skeletal muscle histology under the microscope, and profiled gene expression in the recovering tissue. The differences between the two groups were immediate and substantial.

The knockout mice revascularized far faster than their normal counterparts. By the key recovery timepoint, Cd37-deficient animals had achieved 87.6 percent reperfusion on average, compared with only 62.9 percent in wild-type mice, a difference that reached statistical significance at p less than 0.04. Blood flow, however, was only part of the story. When the researchers examined the muscle itself, they found that the knockout mice had rebuilt far more of their damaged tissue. Mature muscle fibers made up 77.9 percent of the recovered tissue in CD37-deficient animals versus just 46.7 percent in wild-type controls, with a p-value below 0.02. In other words, removing CD37 did not merely restore circulation; it allowed the muscle to genuinely regenerate rather than degenerate into scar tissue and fat.

To dissect the molecular mechanisms behind this accelerated healing, the team performed RNA sequencing on post-ischemic muscle from both groups. The analysis identified 594 differentially expressed genes at an adjusted p-value threshold of less than 0.05, a substantial transcriptional rewiring driven by the absence of a single surface protein. Pathway analysis revealed a coherent pattern. In the knockout muscle, genes related to muscle mass maintenance and collagen degradation were upregulated, suggesting active tissue rebuilding and clearance of fibrotic deposits. Conversely, pathways governing mitochondrial oxidative metabolism and proinflammatory signaling were downregulated. That the metabolic machinery of the muscle shifted away from oxidative programs may seem counterintuitive, but it likely reflects a reduced burden of inflammatory stress, which is known to distort mitochondrial function in ischemic tissue.

The most compelling findings emerged when the researchers integrated their RNA module scoring with immunohistochemistry, staining tissue sections to identify specific immune cell populations. This combined approach revealed that CD37 deficiency was associated with a fundamental reprogramming of macrophages toward a resolution phenotype. The knockout muscle contained macrophages with higher proresolving signatures and lower inflammatory signatures, and it harbored increased numbers of CD163-positive macrophages, a marker classically associated with anti-inflammatory, tissue-repairing macrophage states. In essence, the immune environment of the healing muscle had flipped from one that attacks damaged tissue to one that cleans up and rebuilds it.

To confirm that this shift originated in the macrophages themselves rather than being a secondary effect of the healthier tissue environment, the team isolated bone marrow-derived macrophages from both mouse groups and stimulated them ex vivo. Macrophages lacking CD37 secreted significantly lower levels of inflammatory cytokines and chemokines, the chemical messengers that recruit and activate additional immune cells and amplify tissue-damaging inflammation. This cell-autonomous result strengthens the causal chain: CD37 on macrophages directly promotes their inflammatory behavior, and that behavior, in turn, impairs the revascularization and regeneration of ischemic muscle.

The translational implications are considerable. Current therapies for chronic limb-threatening ischemia, including surgical bypass, angioplasty, and drug-eluting stents, address the plumbing of the disease but often leave the inflammatory biology untouched. Anti-inflammatory approaches to cardiovascular disease have gained momentum in recent years, but targeting inflammation broadly carries infection risks. CD37 offers something more precise: a surface molecule on immune cells whose removal shifts macrophages toward resolution without eliminating them. Notably, CD37 is already a validated target in oncology, with anti-CD37 antibody therapies in clinical development for B cell malignancies, which means the immunomodulatory toolkit for this protein is further along than for most novel targets. Repurposing or adapting those strategies for vascular medicine is a plausible next step, though the authors and outside observers alike would caution that a genetically engineered mouse is a long way from a diabetic patient with a non-healing foot ulcer.

There are also important caveats embedded in the data. The human findings are associative, drawn from transcriptomic correlations rather than direct functional experiments in patients. The mouse studies used complete knockout animals, so the effects of partial or transient CD37 inhibition remain unknown, as do potential consequences for the immune functions CD37 serves elsewhere, including B cell development and antitumor surveillance. The study was funded by the National Institutes of Health under grant R01HL167917 and by an American Heart Association postdoctoral fellowship, and the authors report no competing interests. Still, the convergence of human association, mouse genetics, transcriptomics, histology, and ex vivo cell biology makes a unusually coherent case. If future work confirms that blocking CD37 in patients with critical limb-threatening ischemia can tip their macrophages toward resolution, a protein once studied mainly by immunologists could become an unexpected lever for saving limbs, and the millions of people living with the silent narrowing of their leg arteries may finally gain a therapy that heals muscle, not just vessels.

Subject of Research: The role of the leukocyte tetraspanin CD37 in inflammation, revascularization, and skeletal muscle recovery in peripheral artery disease

Article Title: CD37 impairs revascularization and muscle recovery in peripheral artery disease

Article References: Dai, M., Gindri Dos Santos, B., Wu, J., Ferreira, E., Li, Z., Biswas, B., Schlamp, F., Boothman, I., & Barrett, T. J. (2026). CD37 impairs revascularization and muscle recovery in peripheral artery disease. Journal of Translational Medicine. https://doi.org/10.1186/s12967-026-08799-3

Image Credits: AI Generated

DOI: 10.1186/s12967-026-08799-3

Keywords: peripheral artery disease, CD37, tetraspanin, macrophages, inflammation, hindlimb ischemia, revascularization, muscle regeneration, critical limb-threatening ischemia, immune reprogramming, translational medicine, cytokines

The protein, known as TSPAN32, belongs to the tetraspanin family, a group of four-pass transmembrane proteins that organize other surface receptors into functional microdomains on the cell membrane. Tetraspanins have long been recognized as molecular scaffolds, clustering receptors and signaling molecules to fine-tune how cells respond to their environment. What distinguishes the new work is the discovery that TSPAN32 does something particularly consequential in T cells: it assembles the interleukin-2 receptor complex at the cell surface and strengthens the signals that flow from it into the cell’s interior, thereby amplifying one of the most important growth and survival pathways in immunology.

The investigators began with a clinical observation. When they examined T cells isolated from the peripheral blood of patients with B-cell lymphoma, they found that TSPAN32 expression was consistently reduced compared with T cells from healthy donors. This depletion was intriguing because TSPAN32 had previously been implicated in immune regulation, with earlier studies documenting its modulation during T cell-mediated immune responses and in autoimmune conditions such as multiple sclerosis. Previous work by some of the same collaborators had also shown that TSPAN32 can suppress chronic myeloid leukemia by stabilizing the tumor suppressor PTEN, hinting that the protein plays meaningful roles in cancer biology from multiple angles.

To test whether restoring TSPAN32 could improve CAR-T cell performance, the team engineered T cells to co-express both a CD19-targeting chimeric antigen receptor and TSPAN32. In laboratory culture, the doubly engineered cells outperformed standard CD19 CAR-T cells on essentially every measure the researchers applied. They killed tumor cells more efficiently and produced higher quantities of cytokines, the chemical messengers through which T cells coordinate immune attacks. The enhanced activity was not a subtle effect; the addition of TSPAN32 shifted the engineered cells toward a more robust effector phenotype, the functional state associated with aggressive and sustained anti-tumor activity.

The therapeutic advantage held up in animal models. In subcutaneous tumor models, mice that received TSPAN32-enhanced CAR-T cells showed superior tumor control compared with mice receiving conventional CAR-T cells, demonstrating that the in vitro benefits translated into meaningful in vivo efficacy. Gene expression profiling using RNA sequencing provided a mechanistic clue: TSPAN32-high CAR-T cells displayed increased activation of IL-2 signaling pathways, suggesting that the tetraspanin was not merely a passive marker of T cell health but an active driver of the signaling circuits that sustain T cell proliferation and cytotoxic function.

The mechanistic studies went a step further by pinpointing the molecular partner through which TSPAN32 exerts its effects. The IL-2 receptor is composed of multiple subunits, including CD25, the alpha chain that captures interleukin-2 and initiates the signaling cascade upon ligand binding. The researchers found that TSPAN32 physically interacts with CD25 and promotes its aggregation on the T cell surface. This clustering matters because receptor aggregation can concentrate signaling components into membrane microdomains, effectively increasing the strength and efficiency of the intracellular signal that follows IL-2 engagement. In essence, TSPAN32 acts as a membrane-level organizer that ensures the IL-2 receptor complex is assembled and positioned for maximal signal transduction, allowing the T cell to extract more proliferative and activational benefit from the same amount of cytokine.

Two additional experimental systems strengthened the case that TSPAN32 is a legitimate therapeutic target rather than an artifact of overexpression. First, the team used a transgenic mouse model in which TSPAN32 was overexpressed endogenously in T cells. These animals showed increased resistance to subcutaneous tumor growth, indicating that physiological levels of elevated TSPAN32 are sufficient to confer protection. Second, and perhaps most provocatively, the researchers developed a TSPAN32-specific antibody designated FF-37. When administered, FF-37 increased TSPAN32 expression and improved the anti-tumor efficacy of CAR-T cells. The existence of an antibody that can pharmacologically upregulate the protein opens the door to approaches that do not require additional genetic engineering of the patient’s cells, which would simplify manufacturing and potentially reduce the cost and regulatory complexity associated with engineered cell therapies.

The significance of these findings lies in how they address one of the central bottlenecks in cellular immunotherapy: T cell exhaustion and functional decline. CAR-T cells often enter a state of progressive dysfunction in the tumor microenvironment, characterized by diminished cytokine production, impaired cytotoxicity, and the upregulation of inhibitory receptors. Because IL-2 signaling sits at the interface between tolerance and immunity, fine-tuning this pathway has been a long-standing goal in immunology, albeit a delicate one, since excessive IL-2 signaling can also fuel regulatory T cells that suppress immune responses or trigger systemic toxicity. By enhancing the assembly and signaling efficiency of the receptor complex at the membrane level rather than flooding the system with exogenous cytokine, TSPAN32 offers a mechanism for amplifying IL-2 pathway output in a manner that is intrinsic to the engineered cell itself.

The study, led by Yuanyuan Sun and Qiang Qiu, who contributed equally, with correspondence from Yiguo Hu of West China Hospital, Sichuan University, and supported by funding from the National Natural Science Foundation of China and regional Sichuan research programs, positions TSPAN32 as a candidate for the next generation of CAR-T cell optimization strategies. Whether genetic co-expression of TSPAN32 or antibody-mediated upregulation with FF-37 proves more practical in the clinic will depend on safety profiling and manufacturing considerations that lie ahead. But the conceptual advance is clear: the potency of engineered immune cells can be strengthened not only by changing what receptors they carry, but by reorganizing the membrane architecture through which their survival signals travel. If subsequent studies confirm these results in additional tumor models and, ultimately, in patients, a member of a protein family once regarded mainly as a structural scaffold may become one of the most sought-after enhancements in cellular cancer therapy.

Subject of Research: Enhancement of CAR-T cell anti-tumor potency through TSPAN32-mediated assembly of the IL-2 receptor complex and amplified IL-2 signal transduction.

Article Title: TSPAN32 enhances CAR-T cell potency by assembling IL-2 receptor complex and amplifying its intracellular signal transduction

Article References: Sun, Y., Qiu, Q., Wang, B., Feng, Y., Wang, J., Pan, C., Zheng, L., Qiu, H., He, W., Su, Z., Li, S., & Hu, Y. (2026). TSPAN32 enhances CAR-T cell potency by assembling IL-2 receptor complex and amplifying its intracellular signal transduction. British Journal of Cancer. https://doi.org/10.1038/s41416-026-03592-x

Image Credits: AI Generated

DOI: 10.1038/s41416-026-03592-x

Keywords: CAR-T cells, TSPAN32, tetraspanin, IL-2 signaling, CD25, B-cell lymphoma, cancer immunotherapy, FF-37 antibody, T cell exhaustion, signal transduction, CD19, cellular therapy