Objective
Our long-term goal is to develop new drugs that reduce immune cell trafficking by targeting chemokines. In this proposal, our first goal is to develop better lab models of islet inflammation. This is needed to test any new drugs, and reduce the chance that a drug fails to work when given to patients. We will do this by designing synthetic chemokine pools using two different approaches. First we will use a method called RNA-sequencing. We will use this to determine the levels of chemokines in T1D islets, and in cytokine-stimulated islets. The RNA-sequencing data is publicly available and will incur no additional cost. Next we will use a method called Quantibody ELISA. We will use this to determine how much chemokine protein is produced by cytokine-stimulated islets. This is needed as protein does not always tally with RNA. We will use the data from these methods to guide the design of the synthetic chemokine pools. The chemokine pools will be used for cell migration experiments. The cells used will be models of T-cells, monocytes and neutrophils, which are important cell types in T1D. Synthetic chemokine pools rather than islets are used for these experiments as donor islets are scarce. Our second goal is to make further improved versions of our super-peptide so that it is even more potent. This is needed to reduce the dose that is given to patients. We will do this by adding another sulfotyrosine to the super-peptide. We will study the potency of the super-peptide in the improved lab tests. We will make improvements so that when injected the body does not destroy the super-peptide. This is needed to reduce the dose and the number of times it needs to be given. We will do this by changing the super-peptide so that enzymes in the blood don't chew them up. These changes will make what is called a "peptidomimetic". We will test the new peptidomimetic in cell migration assays that we have developed. We will also test it to see how stable it is in human blood serum.
Background Rationale
The chemokine network drives immune cell trafficking. Chemokines are a group of 46 secreted proteins, that are classified as CCL, CXCL, CX3CL or XCL. They bind G-protein coupled receptors (CCRs, CXCRs, CX3CR, XCRs) to recruit immune cells. Chemokines are an attractive therapeutic target to tackle islet inflammation. The evidence is summarized as follows:
• Chemokines are expressed in T1D patient islets
• Chemokines are expressed in transplanted donor islets
• Chemokine expression is induced in islets by cytokines
• Chemokines are expressed in T1D mouse model islets
• Chemokine receptor gene variants are connected to T1D in humans
• Chemokine gene knockouts in mice protect from diabetes
• Drugs that block chemokines protect from T1D in mice.
The promising drug studies in mice have not yet translated into clinical efficacy. Drugs such as ladaraxin and reparixin failed in T1D and in islet transplants. No other anti-chemokine approach for islet-transplantation or T1D is currently being studied.
By analysing islets from patients with T1D we have found that they produce many different chemokines. We show that blocking one or two is simply not enough. To develop therapeutics that target chemokines we took our inspiration from ticks. Ticks overcome the chemokine network by producing salivary proteins called evasins. These block multiple chemokines. This allows ticks to suppress inflammation at the bite site and feed for prolonged periods. Evasins are however not attractive as drugs. We developed a method called "multiplex-phage-display" to overcome this. Using this we identified a small part of an evasin called a "peptide". This peptide, can block several types of chemokines. Using multiplex-phage display and evolution in the lab we made this peptide even more effective. We found that this evolved peptide is effective at blocking a pool of chemokines made by T1D islets. We developed the chemokine-pool approach so that it resembles the natural mix of islet chemokines. In a mix of chemokines, there are synergistic effects that do not happen when they work alone. We feel that our pool approach is superior to the traditional approach of testing one chemokine at a time. We next used an artificial intelligence approach called "AlphaFold". We found that the evolved peptide binds to a part of the chemokine called the "sulftyrosine cleft". We felt we could make the evolved peptide better by changing the amino acid "tyrosine" to "sulfotyrosine". We made this "super-peptide". We found that it is even better than the original when tested against a chemokine pool. By blocking multiple chemokines, we think we can reduce inflammation in the islets and treat T1D.
Description of Project
Inflammation of the pancreatic islets is called "insulitis". It happens in type 1 diabetes (T1D) and leads to the destruction of insulin-producing cells. Patients with T1D thus need life-long insulin therapy. Insulitis also happens when islet cells are transplanted as part of the treatment for T1D. Scientists have shown that immune cells like T-cells and certain white blood cells cause the destruction. There is a window of opportunity to treat patients in the early stages of T1D. This is by reducing the damage caused by immune cells before islets are completely destroyed. Treatments currently in use such as teplizumab are effective but not totally so. They delay the onset of T1D by about 2 years. Similar treatments are given tp patients receiving islet transplants. They allow patients to remain free from needing insulin for about 5 years. We're looking into ways to stop immune cells from moving into the pancreatic islets and transplanted islets. These new treatments might work alongside current treatments. We know that certain molecules called chemokines play a big role in insulitis. However, attempts to treat T1D or transplant islets with drugs that block single chemokines have not worked. By analysing islets from patients with T1D we have found that they produce many different chemokines. Blocking one or two is simply not enough. We've taken inspiration from ticks. Ticks produce proteins that can block many chemokines at once. Tick proteins cannot be used as treatments. By using a special method developed in our lab, we've identified a small piece of one of these tick proteins. This piece, called a peptide, can block several types of chemokines. We've also found ways, using evolution in the lab, to make this peptide even more effective. We found that this evolved peptide is effective at blocking a pool of chemokines made by T1D islets. We next used an artificial intelligence approach called "AlphaFold". Using this we predicted that we could make the evolved peptide even better. We made this "super-peptide" and found that it is even better than the original when tested in the lab. By blocking multiple chemokines, we think we can reduce inflammation in the islets and treat type 1 diabetes. Our goals are to develop better lab tests to model islet inflammation. This is needed to reduce the chance that a drug fails to work when given to patients. We will make further improved versions of our super-peptide so that it is even more potent. This is needed to reduce the dose that is given to patients. We will study the potency of the super-peptide in the improved lab tests. We will make improvements so that when injected the body does not destroy the super-peptide. This is needed to reduce the dose and the number of times it needs to be given. Once we have a sufficiently improved super-peptide, we will ensure that it only targets islets. This is needed to prevent side effects of blocking chemokines elsewhere in the body. We will ensure that the peptide does not bind things other than chemokines, which might also cause side effects. If all goes well, we hope to test it animal models of T1D. This step is needed to reduce the chance that a drug fails to work when given to patients. It also alerts us to possible side effects. If animal studies are promising, we will plan to start trials in patients.
Anticipated Outcome
These studies will show that we have an effective peptidomimetic that will reduce human islet inflammation. Once we have this, we will ensure that it only targets islets. This is needed to prevent side effects of blocking chemokines elsewhere in the body, such as infection. We will do this by joining it to a tag developed in our colleague David Hodson's lab called Ex9. The Ex9-peptidomimetic will localize to the islets and block chemokines only in the islets. We will also ensure that the peptide does not bind things other than chemokines, which might also cause side effects. We will do this by a method called affinity-selection-mass-spectrometry". These studies will show that we have a safe and effective peptidomimetic. If all goes well, we hope to test it animal models of T1D. This step is needed to reduce the chance that a drug fails to work when given to patients. It also alerts us to possible side effects. If animal studies are promising, we will plan to start trials in patients.
Relevance to T1D
Type 1 diabetes (T1D) is an autoimmune condition marked by the destruction of pancreatic islet β-cells. The resulting metabolic dysfunction necessitates lifelong insulin therapy. This creates substantial disease and mortality burden. There are about 8.4 million people affected worldwide with T1D. Roughly 510000 new cases are diagnosed each year. Of these cases, 18% occurred in people under 20, 64% aged 20–59, and the rest 60 or older. T1D evolves silently over months to years. Stage 1 manifests with two or more autoantibodies present. Stage 2 is marked by autoantibodies and declining β-cell function. Stage 3 represents clinical onset of T1D1. Those identified at stage 1 face a 35–50% likelihood of becoming clinically diabetic within 5–6 years. A combination of risk scores and islet autoantibodies can be used to predict progression to T1D. This provides a window to act to prevent the autoimmune destruction of pancreatic islet β-cells. The immune cell types involved in this process include T-cells, macrophages, dendritic cells, and neutrophils. Cytokines involved include TNFA, IL1 and type 1 interferons. Immune mechanisms are also the major contributor to islet and β-cell loss following transplantation. The instant blood inflammation reaction causes early loss of β-cells. Further β-cell loss occurs due to infiltration of T-cells.
Different therapeutic targets in T1D and in islet transplantation have been established by scientists. Some of them are:
• T-cells. This was initially demonstrated using T-cell depleting agents in non-obese diabetic (NOD) mice. Later this was established by the well-documented clinical impact of anti-T-cell therapies. These include teplizumab which depletes T-cells and preserves β cell function. This delays progression to clinical diabetes by a median of 2 years. In human islet transplantation it is highlighted by the success of T-cell depletion or inhibition.
• Antigen presenting cells. These cells present antigens such as insulin to T-cells. Their role was demonstrated by depleting these cell types in NOD mice.
• Neutrophils. The role of these cells was shown by the efficacy of drugs blocking trafficking of these cells NOD mice. They were also effective in mouse islet transplant experiments.
• Cytokines. These molecules are important in both T1D and in islet transplantation.
The major limitation of current therapeutic approaches is that their efficacy is modest. T1D onset is delayed by a median of 2 years. Only ~50% of transplant recipients remain insulin independent at 5 years. We're looking into ways to stop immune cells from moving into the pancreatic islets and transplanted islets. These new treatments might work alongside current treatments.