Objective
The objective of this proposal is to develop tools for accurate and reliable early diagnosis of T1D. For this purpose, we aim to expand the bank of known antibody-targets associated with the disease, and ultimately examine how well the appearance of antibodies against these targets improves our ability to predict the appearance of symptoms.
Background Rationale
Insufficient tools for early diagnosis of evolving T1D lead to its common detection only after appearance of life-threatening metabolic imbalance. Besides the need to avoid such hazardous symptoms, the requirement for markers that detect the disease during its prodrome is highlighted by the FDA’s recent approval of Teplizumab. This drug, along with others that are currently at different stages of development, are likely to delay disease onset if applied early enough. Typically, T1D is diagnosed based on detection of four autoantibodies, which appear in patients’ blood sometime during the gradual progression of the disease. However, their ability to predict the appearance of symptoms is vague, and provided in terms of years ahead. In fact, more than 10% of people with T1D have no identified autoantibodies at presentation.
The importance of early diagnosis is well appreciated by both clinical and research communities handling T1D, and is manifest in several ongoing campaigns – including by the JDRF - to screen at-risk populations for the presence of autoantibodies against any of the 4 antigens. Surprisingly, of the four types of antibodies currently assayed clinically for early diagnosis of T1D, the most recent target – ZnT8 - was discovered nearly 15 years ago (2007). This highlights the urgent need for discovery of novel antigens that raise a humoral response in T1D patients. The more diverse our bank of known antigens, the more likely we are to achieve accurate diagnosis, at an earlier stage.
To overcome the challenges associated with novel antibody-target discovery, we propose the use of novel genomic tools which will explore the antibody repertoire at the heart of the immune attack – in inflamed islets, where disease-associated antibodies are most prevalent, rather than the peripheral blood, where most antibodies target unrelated proteins.
Description of Project
Insufficient tools for early diagnosis of evolving type 1 diabetes (T1D) lead to its common detection only after appearance of life-threatening metabolic imbalance. Besides the need to avoid such hazardous symptoms, the requirement for markers that detect the disease during its prodrome is necessary to facilitate preventive treatments that are currently under different stages of development.
As an autoimmune disease, early diagnosis of T1D is based on detection of antibodies that target four proteins expressed in beta cells: Insulin, GAD65, IA-2 and Znt8. The imminence of T1D onset is correlated with the number of auto-antibodies detected, but the prediction is vague (provided in terms of years ahead) and inaccurate, and more than 10% of people with T1D have no identified autoantibodies at presentation. There is an urgent need to discover novel targets of antibodies in T1D patients in order to expand and diversify the bank of known targets, and achieve accurate diagnosis at an earlier stage. However, finding new proteins that are targeted by self-antibodies is challenging, and the most recent target used in clinical assays (Znt8) was discovered more than 15 years ago.
Here, we harness novel genomic technologies and sketch an efficient path to uncover the identity of multiple novel auto-antibody targets at the early stages of T1D. The path includes three objectives: The first objective Will use single cell RNA sequencing (scRNA-seq) to characterize B-cell populations adjacent to islets that are attacked by the immune system at the early stages of the disease. B cells are the cells which produce antibodies. Each cell produces an antibody with a unique genomic sequence, which generates the enormous diversity of targets. Once a B cell recognizes a target it divides and forms a multicellular clone. Thus, when an antibody sequence is shared by multiple cells, we can deduce that a clonal reaction had occurred and that this antibody is likely to identify a target associated with T1D. By sequencing the antibody generated by each cell, we will reveal which antibodies are most likely to be abundant and relevant to the disease. We will start the analysis with mice that develop an autoimmune disease highly similar to T1D, and proceed to human samples once the relevant protocols are well established. The second objective is to identify the proteins targeted by the novel antibodies. We will rely on the sequencing data to generate synthetic antibodies identical to those we discovered, and use them to detect their cognate targets. Finally, our third objective is to examine the clinical relevance of the detected targets. We will screen blood samples from recently diagnosed T1D patients and from healthy individuals, to test which of the novel proteins raise antibodies specifically in patients. In our preliminary experiments we have already identified an antibody which targets a membrane-associated protein expressed in beta cells, which provides evidence to the efficiency of our approach.
Altogether, the outlined experiments are expected to discover multiple antibody-targets with a high diagnostic potential. This will lay the foundations for wider studies which include first degree relatives of T1D patients, to determine how the predictive power of the novel targets compares with that of the commonly used markers. Ultimately, the novel antibodies may replace the existing markers or supplement them, and dramatically improve the accuracy and the reliability of early T1D diagnosis.
Anticipated Outcome
Altogether, the outlined experiments are expected to discover multiple antibody-targets with a high diagnostic potential. This will lay the foundations for wider studies which include first degree relatives of T1D patients, to determine how the predictive power of the novel targets compares with that of the commonly used markers. Ultimately, the novel antibodies may replace the existing markers or supplement them, and dramatically improve the accuracy and the reliability of early T1D diagnosis.
Relevance to T1D
Early detection of T1D can help avoid life threatening symptoms, and facilitate the use of prophylactic treatments under development. Current methods for diagnosis rely on the appearance of autoantibodies that target a small number of known auto-antigens in the blood, but their predictive power in the pre-symptomatic phases of the disease is limited. There is an urgent need to enhance the number of known autoantibody targets, to achieve accurate and reliable diagnosis prior to the appearance of symptoms. Here, we propose a pipeline to discover novel types of auto-antibodies which will increase the accuracy and reliability of early T1D diagnosis.