Harmine Plus Exendin-4 Enhances Remission of Recent-Onset Type 1 Diabetes Following Anti-CD3 Therapy

Technical Summary: Harmine Plus Exendin-4 Enhances Remission of Recent-Onset Type 1 Diabetes Following Anti-CD3 Therapy

Problem Statement
Type 1 diabetes (T1D) is characterized by the autoimmune destruction of pancreatic $\beta$-cells. While immune modulation therapies, such as anti-CD3 monoclonal antibodies (e.g., teplizumab), can delay disease progression and preserve residual $\beta$-cell function, they rarely induce durable disease reversal. This limitation suggests that immune intervention alone is insufficient to restore normoglycemia once significant $\beta$-cell mass is lost. Furthermore, emerging evidence indicates that $\beta$-cells actively contribute to disease progression by increasing their immunogenicity and susceptibility to inflammatory stress under cytokine exposure. There is a critical need for therapeutic strategies that simultaneously suppress pathogenic autoimmunity, promote $\beta$-cell regeneration, and reduce $\beta$-cell vulnerability to inflammatory attack.

Methodology
The study employed a multi-modal approach combining human islet biology, single-cell transcriptomics, and in vivo mouse models:

• Human Islet Models: Isolated human islets and EndoC-$\beta$H1 cells were exposed to a pro-inflammatory cytokine cocktail (IL-1$\beta$, TNF$\alpha$, IFN$\gamma$) with or without treatment by Harmine (H, a DYRK1A inhibitor), Exendin-4 (E, a GLP1R agonist), or the combination (H+E).
• Omics and Molecular Analysis: Single-cell RNA sequencing (scRNA-seq) was performed on treated human islets to identify transcriptomic changes in apoptosis, inflammation, stress, and immunogenicity pathways. Quantitative PCR (qPCR), ELISA, and immunostaining validated specific gene expression and protein levels (e.g., MHC-I, chemokines, stress markers).
• Mechanistic Validation: Gain- and loss-of-function studies were conducted using adenoviral vectors to overexpress or knockdown the long non-coding RNA (lncRNA) SNHG6 in human $\beta$-cells to assess its role in H+E-mediated protection.
• In Vivo Models: Diabetic NOD mice (recent-onset T1D model) were treated with a low-dose anti-CD3 antibody regimen followed by continuous infusion of H, E, or H+E. Outcomes included blood glucose monitoring, glucose tolerance tests, and histological analysis of $\beta$-cell mass, proliferation, death, and insulitis.
• Immune Profiling: Flow cytometry and Cytometry by Time-of-Flight (CyTOF) were used to characterize T-cell phenotypes (activation, exhaustion, regulatory subsets) in mouse lymphoid organs and human peripheral blood mononuclear cells (PBMCs) treated in vitro.

Key Contributions and Results

1. $\beta$-Cell Protection and Functional Recovery:

   • In human islets, H+E synergistically reduced cytokine-induced apoptosis and suppressed inflammatory signaling pathways (IFN$\gamma$, IL-1, TNF) more effectively than either agent alone.
   • H+E treatment attenuated $\beta$-cell immunogenicity by downregulating classical MHC-I molecules (HLA-A, -B, -C), antigen-processing machinery (PSMB8/9, TAP1), and T-cell chemokines (CXCL9, CXCL10, CXCL11). Conversely, it upregulated the immunomodulatory molecule HLA-E.
   • The combination reduced endoplasmic reticulum (ER) stress markers (ATF6, spliced XBP1, CHOP) and improved insulin secretory quality, evidenced by a reduced proinsulin-to-insulin ratio.

2. Identification of SNHG6 as a Mediator:

   • Transcriptomic analysis identified SNHG6 as a key lncRNA induced by H+E in $\beta$-cells under inflammatory stress.
   • Functional studies demonstrated that SNHG6 overexpression mimicked the protective effects of H+E (reducing apoptosis, CASP3, HLA-B, and CXCL10), while SNHG6 knockdown abolished these protective effects, confirming its role as a mediator of $\beta$-cell resilience.

3. In Vivo Remission in NOD Mice:

   • H+E monotherapy failed to reverse diabetes in recent-onset NOD mice.
   • However, a regimen of low-dose anti-CD3 followed by H+E administration resulted in rapid normalization of blood glucose, sustained remission, increased circulating insulin, and improved glucose tolerance.
   • This combination therapy significantly expanded $\beta$-cell mass through increased proliferation (Ki67+) and reduced apoptosis (TUNEL-), while reducing severe insulitis and islet-infiltrating CD4+ and CD8+ T cells.

4. Immunomodulatory Mechanisms:

   • The therapeutic success was not due to broad lymphocyte depletion. Instead, the combination shifted T-cell phenotypes: it reduced IFN$\gamma$-producing effector T cells, increased FoxP3+ regulatory T cells (Tregs), and upregulated exhaustion-associated markers (TIGIT, PD-1, EOMES, TOX) on T cells.
   • Similar immunomodulatory effects (reduced activation, increased Tregs and exhaustion markers) were observed in human PBMCs treated with H+E in vitro, suggesting a direct effect on human immune cells.

Significance and Claims
The paper claims to demonstrate a "dual-compartment" therapeutic model for recent-onset T1D. The authors posit that transient immune modulation (anti-CD3) creates a permissive window by dampening autoimmune pressure, which allows the regenerative and protective agents (H+E) to effectively restore $\beta$-cell mass and function.

Key claims include:

• Synergy: H+E alone is insufficient to reverse autoimmune diabetes in NOD mice; its efficacy is contingent upon prior immune modulation.
• Dual Action: The therapy addresses both the immune system (reducing pathogenic T-cell activity and enhancing regulation/exhaustion) and the $\beta$-cell compartment (promoting proliferation, reducing stress, and lowering immunogenicity).
• Novel Mechanism: The study identifies SNHG6 as a novel regulator of $\beta$-cell protection against inflammatory stress and immunogenicity.
• Clinical Potential: The findings suggest that combining anti-CD3 therapy with $\beta$-cell regenerative/protective agents could overcome the transient nature of current immunotherapies, potentially leading to durable remission or functional cure in recent-onset T1D.

The authors acknowledge limitations, including the use of mouse models that do not fully recapitulate human T1D heterogeneity, the unknown durability of remission, and the need for validation in T1D patient-derived immune cells. However, they assert that the convergent evidence from human islets, human immune cells, and NOD mice supports the therapeutic potential of this combination strategy.