On January 3, 2026 (Beijing time), a research team led by Zhang Fei from Xinhua Hospital, Shanghai Jiao Tong University School of Medicine (SJTUSM), in collaboration with the team of Cao Qin from the Bio-X Institutes of Shanghai Jiao Tong University, published an online research article in Cell entitled “Structure of pancreatic hIAPP fibrils derived from patients with type 2 diabetes.”

This study reports for the first time the cryo-electron microscopy (cryo-EM) structures of hIAPP fibrils derived directly from pancreatic tissue of patients with type 2 diabetes (T2D), providing critical structural insights into the pathological mechanisms of T2D and a foundation for the development of targeted therapeutic strategies.

Type 2 diabetes (T2D) is a highly prevalent metabolic disease worldwide, affecting nearly 10% of the global population and posing a major threat to human health and longevity. Its pathogenesis is complex, involving insulin resistance and progressive failure of pancreatic β-cell function. During β-cell degeneration, human islet amyloid polypeptide (hIAPP), which is co-secreted with insulin, undergoes misfolding and abnormal aggregation, forming amyloid deposits within pancreatic islets. This pathological feature is observed in more than 90% of patients with T2D and is widely considered to be closely associated with β-cell toxicity, impaired insulin secretion, and disease progression. Therefore, strategies aimed at delaying hIAPP fibrillization or clearing preformed hIAPP fibrils may hold therapeutic potential for T2D. Although previous studies have determined structures of hIAPP fibrils assembled in vitro, amyloid fibrils of the same protein can adopt distinct structures under different conditions. Whether hIAPP fibrils formed in vivo in patients share the same molecular architecture as those formed in vitro has remained unclear. The lack of structural information on pathological hIAPP fibrils has severely limited the development of hIAPP-targeted antidiabetic therapies.

A Novel Approach to Overcome the Challenge of Sample Acquisition

One of the main reasons for the absence of pathological hIAPP structures is the difficulty in obtaining pancreatic samples from patients, as diabetes treatment does not typically involve pancreatic surgery or biopsy. To overcome this obstacle, the research team adopted an alternative strategy: identifying patients who required partial pancreatectomy for other diseases (such as pancreatic cancer) and who also had T2D. With informed consent, non-tumorous adjacent pancreatic tissue from these patients was used for research. Using this approach, pancreatic tissues from six donors were collected. Given the low abundance of amyloid fibrils in pancreatic tissue and the technical challenges associated with their extraction, the team optimized a gentle yet efficient extraction and purification protocol and independently extracted fibrils from tissues of all six donors. Amyloid fibril–like structures were observed only in extracts from four donors with T2D, whereas no such fibrils were detected in samples from two non-diabetic donors, highlighting the close association between these fibrils and diabetes.

First Structural Revelation of Pathological Protein Fibrils from Patients, Providing a Structural Basis for Drug Development

After successfully isolating the pathological fibrils, the researchers employed cryo-EM to determine their structures, achieving high-resolution reconstructions from three donors at 2.9 Å, 3.0 Å, and 3.4 Å, respectively.

Structural analyses revealed that fibrils from the three donors share nearly identical architectures. Each fibril consists of two symmetrically intertwined protofilaments. The fibril core comprises residues 2–37 of hIAPP and adopts a previously unreported, distinctive “Ω-shaped” fold. Compared with previously reported in vitro hIAPP fibril structures, this in vivo structure shows substantial differences in backbone topology, side-chain packing, and intramolecular interaction networks, underscoring the unique influence of the pathological cellular environment on protein misfolding pathways. Notably, several non-proteinaceous densities were clearly identified in the structure. These densities are symmetrically distributed either within specific surface pockets or inside internal cavities of the fibrils, suggesting the binding of endogenous small-molecule ligands. The densities located within the internal cavity are surrounded by hydrophobic side chains of hIAPP, indicating that the bound ligands are likely lipids or other hydrophobic molecules. Such ligands may play roles in stabilizing fibril conformations, modulating fibril growth kinetics, or mediating cytotoxicity, offering new structural perspectives on the pathological activity of hIAPP. In addition, ligand-binding pockets on the fibril surface represent potential targets for the design of small-molecule inhibitors, antibody-based therapeutics, or interfering peptides aimed at preventing pathological fibril formation or promoting fibril disassembly, thereby contributing to therapeutic strategies for diabetes.

Revealing a Molecular Link to Alzheimer’s Disease, Providing a Theoretical Basis for Cross-Disease Diagnosis and Therapy

Clinical associations between T2D and Alzheimer’s disease (AD) have long been recognized: patients with T2D have a higher risk of developing AD, and vice versa. One hypothesis proposes that this link arises from in vivo cross-seeding between hIAPP fibrils and amyloid-β (Aβ) fibrils in AD. Specifically, hIAPP fibrils in the pancreas of T2D patients may promote Aβ fibrillization in the brain via neural connections between the pancreas and the central nervous system, and the reverse process may also occur. Through structural comparison, the research team discovered striking conformational similarities between the core folding unit of hIAPP fibrils and Aβ fibrils isolated from the brains of AD patients. An eight-residue segment in the two fibrils adopts an almost identical conformation, which may serve as a conserved structural template for cross-seeding between hIAPP and Aβ. This finding provides a structural explanation for the in vivo cross-seeding hypothesis and offers a theoretical foundation for the concepts of comorbidity, co-diagnosis, and co-treatment of T2D and AD.

This study establishes a robust technical pipeline for the extraction, purification, and structural determination of low-abundance pathological protein fibrils from complex human tissues. For the first time, it reveals the high-resolution, atomic-level structure of hIAPP amyloid fibrils formed in vivo in patients with type 2 diabetes. These findings lay a solid structural foundation for the development of therapeutic strategies targeting hIAPP amyloidosis in T2D.

The work was supported by the National Key R&D Program of the Ministry of Science and Technology of China, the National Natural Science Foundation of China, the Medical–Engineering Interdisciplinary Program of Shanghai Jiao Tong University, the Yongxin Young Scholar Program, and other funding sources.