What usually causes the pancreas to lose insulin-producing cells in type 1 diabetes?

An autoimmune process usually causes the loss. In most type 1 cases, a T cell-mediated attack removes beta cells from the pancreatic islets, leading to severe insulin deficiency.

Type 1 diabetes: immune recognition in tissue and long-term studies

Type 1 diabetes has an immune mechanism that researchers have examined in pancreatic tissue, isolated human cells and children followed before diagnosis. The National Institute of Diabetes and Digestive and Kidney Diseases identifies the insulin-producing beta cells of the pancreas as targets of the body's immune system. Studies have made that explanation more specific by finding immune cells that recognize islet components within the affected organ and by observing their effects in the laboratory. Other investigations have followed the appearance of immune markers over years. These records concern different stages of a process that can begin well before clinical disease becomes apparent.

Ken Coppieters and colleagues published a tissue investigation in the Journal of Experimental Medicine in 2012. Their collection included samples from 45 deceased donors with type 1 diabetes, whose disease duration ranged from a week to more than 50 years, alongside comparison groups. Across 72 available samples, the investigators examined insulin-containing cells, CD8 T-cell infiltration and expression of HLA class I proteins. They then used a staining technique capable of identifying T cells reactive to particular islet antigens in selected samples.

The investigators detected both single and multiple antigen-reactive CD8 populations within individual islets in a subset of patients, including some diagnosed up to eight years earlier. The inflammatory lesions had a multifocal distribution, with differing degrees of infiltration and beta-cell loss. Some pathological features persisted in a small portion of donors with longstanding disease. The pancreas samples therefore did not depict a uniform process affecting every islet simultaneously. They documented local immune recognition within human tissue and substantial variation in how the disease appeared after diagnosis. The donor collection supplied snapshots at different durations, rather than continuous observation of one pancreas through time.

A second 2012 investigation, led by Anna Bulek and colleagues, examined the molecular interaction underlying one such response. The group studied a CD8 T-cell clone called 1E6, originally obtained from the blood of a person recently diagnosed with diabetes. The clone recognized a fragment of preproinsulin, a precursor of insulin, presented by a particular HLA molecule. Human islet cells from three additional organ donors were susceptible to killing by the clone even without added inflammatory cytokines or an externally supplied target fragment. This was a laboratory observation involving selected cells and a defined immune receptor, rather than a demonstration that every patient's immune response used that same receptor.

The group also determined structures of the receptor, its target complex and their bound combination. Functional experiments varied the target fragment's amino acids and measured immune-cell activation. Changes in a central part of the fragment were particularly disruptive to recognition. The work connected a physical molecular interaction with an observed effect on human islet cells. Its specificity was central to the result: recognition depended on a particular fragment presented in a particular molecular setting.

The chronology before diagnosis was investigated in a 2013 study by Anette Ziegler and colleagues. They pooled prospective data from childhood cohorts in Colorado, Finland and Germany, with follow-up completed by July 2012. The children had been recruited because of increased genetic or familial risk. Among 585 who developed multiple islet autoantibodies, the estimated probability of progression to diabetes over ten years was 69.7%. Among 474 with a single autoantibody, it was 14.5%. These were follow-up estimates after antibodies appeared, rather than proportions measured once in a general childhood population. The antibodies served as markers of an immune response; the tissue and cell studies examined the T-cell activity itself.

The pooled cohorts also recorded differences in timing. Progression after multiple antibodies appeared was faster among children whose first antibody-positive measurements occurred before age three than among those with later seroconversion. Because the participants had been selected for elevated risk, the estimates described that research population. The observations established a measurable interval between immune recognition and clinical disease in many children. They did not identify a single initiating exposure common to every participant or make the later diagnosis inevitable within the same fixed number of years.

Ezio Bonifacio and colleagues studied an earlier part of that timeline in a 2021 TEDDY publication. They followed 8,556 genetically at-risk children and estimated the remaining five-year risk of developing multiple islet autoantibodies at different ages. Among children still negative at each age, that estimate declined from 4.3% at 7.5 months to 1.1% at 6.25 years. The outcome was development of multiple immune markers, rather than immediate loss of all insulin production. The influence of several susceptibility factors diminished with age within the cohort. Type 1 diabetes research thus documented variation in the emergence of autoimmunity as well as in subsequent progression, while NIDDK continued to identify genes and environmental factors as possible triggers of the immune attack on beta cells.

Type 1 diabetes: immune recognition in tissue and long-term studies

Type 1 diabetes: immune recognition in tissue and long-term studies

The Question

The Answer

Read next