Objective
Our overall objective is to find a way to prevent Type 1 diabetes (T1D). One problem is that current markers for T1D progression predict T1D after most insulin-making cells have been destroyed. We would like to find markers that predict T1D development earlier so that immune-modulating therapies and other prevention strategies can be employed before pancreatic -cells are destroyed. We think that changes in plasma levels of a new class of lipids that we discovered which are present in human tissues and blood, might serve as an early biomarker for T1D progression. These lipids are called Fatty Acid Hydroxy Fatty Acids (FAHFAs). Our preliminary data show that some FAHFAs levels are low in people with T1D. When we treat mice with autoimmune diabetes with these lipids, we markedly reduce T1D incidence and this prevention lasts for the lifetime of the mouse. Thus, we hypothesize that 1) FAHFAs might serve as early biomarkers for T1D development; 2) low levels of beneficial FAHFAs may contribute to a pro-inflammatory state and hasten T1D development; and 3) maintaining normal FAHFA levels may delay or prevent T1D in people. We will obtain plasma and blood immune cells from the TrialNet program from people who did and did not progress to T1D. We will measure 90 forms of FAHFAs in serial samples and compare the patterns to islet-antibody titers, glucose tolerance test results, C-peptide as an indicator of insulin secretion and HbA1c in the same people. We will determine whether initial plasma FAHFA levels when family members of people with T1D are first screened, or changes in levels of specific FAHFAs over time, predict T1D development. If so, plasma FAHFA levels could be used to determine when a person is likely to progress to T1D. This could allow earlier intervention to prevent T1D.
We also think that low levels of anti-inflammatory FAHFAs initially, or a reduction in levels with time, may contribute to T1D development. If so, intervening by supplementing people at risk for T1D with FAHFAs might prevent T1D. We propose studies to clarify if low FAHFA levels could contribute to T1D development so a future study could be designed to restore FAHFA levels in high-risk people. First, we need to identify which of the many forms of FAHFAs are most likely to be protective in people. In our studies in auto-immune diabetic mice, 5-PAHSA and 9-PAHSA were protective. But it could be different in people. So, the objective of Aim 2 is to determine which FAHFAs associate highly with anti-inflammatory effects in immune cells from TrialNet participants and then treat the immune cells from participants who developed T1D with those anti-inflammatory FAHFAs to test whether that reduces pro-inflammatory activation. In Aim 3, we will treat human islets with protective FAHFAs and transplant them into immune-compromised mice after we make the mice diabetic with a chemical that destroys -cells. We will continue to give FAHFAs to the mice orally daily to protect the transplanted human islets. We will use immune-compromised mice so the islets will not immediately be rejected. Once we confirm that the transplanted islets are functioning well and maintaining normal blood glucose, we will inject the mice with immune cells from other people, which will cause rejection of the human islets. We hypothesize that the FAHFAs will at least partially protect the transplanted islets. If so, we will have strong data indicating that anti-inflammatory FAHFAs can protect human islets from immune attack. We would then proceed to design a clinical trial to test whether FAHFAs can prevent T1D in people who are at high risk for T1D.
Background Rationale
Current intervention studies to prevent T1D are effective but for a limited period of time. To prevent T1D with a lasting effect, there is a great need for biomarkers that predict earlier stages of T1D development. The FAHFA class of lipids that we discovered might serve as early biomarkers. More than 100 distinct FAHFAs are normally present in human plasma. One FAHFA family, Palmitic Acid Hydroxy Stearic Acids (PAHSAs), markedly reduces the incidence and delays T1D onset in autoimmune non-obese diabetic (NOD) mice. PAHSA treatment of NOD mice attenuates autoimmune responses by lowering mature T helper cell activation and B cell number, and increasing regulatory T cell activity in islets. PAHSAs also directly protect β-cells independent of immune modulation since they reduce immune-mediated cell death, reduce cellular stress and prevent and reverse metabolic-stress-induced senescence in cultured β-cells and human islets. Senescence is similar to aging but it differs because cells survive and secrete factors that cause inflammation and damage of surrounding cells. The anti-inflammatory effects of PAHSAs are also demonstrated by the fact that they delay the onset and reduce the severity of experimental colitis in mice by decreasing gut inflammation. PAHSAs exert some of their anti-inflammatory effects by binding to receptors on the surface of many types of immune cells. In addition to PAHSAs, several other types of FAHFAs have anti-inflammatory effects including FAHFAs that contain omega-3 fatty acids.
PAHSAs also have beneficial metabolic effects. Administration of 5- or 9-PAHSA to insulin-resistant high-fat diet (HFD)-fed mice improves glucose tolerance and systemic and hepatic insulin sensitivity, in part, due to enhancement of insulin action to suppress fatty acid release from adipose tissue. PAHSAs also augment glucose-stimulated insulin and glucagon-like peptide-1 secretion in chow fed mice with insulin resistance from aging. Some PAHSA isomers dramatically increase glucose-stimulated insulin secretion from human islets and restore normal pulsatility of insulin secretion in islets from people with T2D.
Levels of specific FAHFAs are regulated in humans by factors including age, exercise, diet, short-term overfeeding and pathophysiological states. PAHSAs are present in human breast milk, and specific isomers are reduced in breast milk from women with obesity versus lean women. PAHSA levels are low in serum and subcutaneous adipose tissue of insulin-resistant people and these levels correlate highly with insulin-sensitivity. Our preliminary data show that serum levels of several FAHFA lipids from two FAHFA families predict worsening glucose tolerance and development of impaired glucose tolerance in family members of people with T2D. Hence, we know that FAHFAs can predict progression to T2D in people. In this grant, we want to determine whether that is also true in people with T1D.
In people with established T1D, an untargeted lipidomic analysis in plasma, revealed that at least two FAHFA families are lower than in non-diabetic people. These FAHFAs correlate negatively with glycated hemoglobin values and not with age, BMI, or waist circumference. Therefore, we do not expect these factors to interfere with finding differences in the TrialNet progressors and non-progressors. The study data in people with long-standing T1D combined with the predictive data in people at risk for T2D, provide a strong rationale for our study to determine whether specific FAHFA levels are altered before or during the development of islet autoimmunity and if they predict progression to overt T1D. Furthermore, the anti-inflammatory effects of FAHFAs provide a strong rationale to study relationships of FAHFAs with immune cell subtypes and their activation in PBMCs of TrialNet participants. The proposed studies may identify a plasma “FAHFA signature” that is an early biomarker for T1D progression and could be used to guide early intervention to prevent T1D.
Description of Project
Type 1 Diabetes (T1D) results from autoimmune destruction of pancreatic β-cells which make insulin. Markers are employed to track the progression from early auto-immunity to overt T1D but these indicate a relatively late disease stage after many insulin-making cells are lost. Current interventions to prevent T1D are effective but for a limited time. For lasting T1D prevention, markers that predict earlier stages of T1D development are needed. We discovered a new class of lipids called Fatty Acid Hydroxy Fatty Acids (FAHFAs). They are synthesized in human tissues. Several FAHFAs have beneficial metabolic and anti-inflammatory effects and modulate multiple types of immune responses. At least 25 FAHFA families are present in human plasma. Each family has 4-8 subtypes with a different chemical structure. This results in >100 distinct FAHFAs in human plasma.
One family, Palmitic Acid Hydroxy Stearic Acids (PAHSAs), dramatically reduces T1D incidence and delays onset in autoimmune non-obese diabetic (NOD) mice. PAHSA treatment of NOD mice diminishes autoimmune responses by lowering the activation or number of pro-inflammatory immune cells, and increasing the activity of anti-inflammatory T cells in islets. PAHSAs also directly protect β-cells independent of immune modulation since in cultured β-cells and human islets, PAHSAs reduce β-cell stress and death during immune attack. PAHSAs also directly augment glucose-stimulated insulin secretion from human islets. Our preliminary data show that people with established T1D have lower plasma PAHSA and Palmitic Acid Hydroxy Oleic Acid (PAHOA) concentrations than non-diabetic people and PAHSA levels correlate inversely with glycated hemoglobin. This raises the question: When do FAHFA levels decrease in people with T1D? Before autoimmunity occurs or during the progression from auto-antibody positivity to overt T1D? Or as a result of hyperglycemia?
Our overall goals are to determine whether initial plasma FAHFA levels, or changes in levels of specific sub-types over time, predict T1D development and whether FAHFAs could be used as biomarkers to allow earlier intervention to prevent T1D. In Aim 1, we will measure 90 regioisomers of 15 FAHFA families in serial plasma samples of TrialNet participants who did and did not progress to T1D. We will determine if changes in plasma FAHFA regioisomers are associated with changes in T1D progression indicated by anti-islet antibody titers, glucose area under the curve during OGTT, C-peptide stimulation during OGTT, and HbA1C.
In Aim 2, we will determine whether changes in plasma FAHFA levels serve as a marker for activation of specific immune cell types during T1D development and whether FAHFA treatment of immune cells in blood from TrialNet participants who developed T1D can reduce the pro-inflammatory activation. We will analyze serial serum FAHFA levels in relationship to serial RNA sequencing data from immune cells from people who did and did not progress to T1D. We will also identify the immune cell subtypes present in blood from TrialNet progressors compared to non-progressors. Then we will isolate the pro-inflammatory immune cell subsets from TrialNet progressors and determine whether treatment with known anti-inflammatory FAHFAs or those that are associated with lower inflammatory activation in the RNA analysis, will reduce activation of these immune cells. In Aim 3, we will determine whether treatment of human islets with specific FAHFAs before and after transplantation into immune-compromised diabetic mice will protect the islets from attack by human immune cells.
These studies could identify a plasma “FAHFA signature” that could guide earlier intervention to prevent T1D. They will also indicate whether reduced anti-inflammatory FAHFA levels at baseline or a decrease over time could contribute to T1D development/progression.
Anticipated Outcome
Our preliminary data show that plasma levels of certain FAHFAs in people with established T1D are lower compared to nondiabetic people. Thus, in Specific Aim 1 we anticipate that the concentrations of these and other FAHFA isomers will be altered in people who progress from early autoimmunity to stage 3 T1D compared to non-progressors, either at baseline or as T1D progresses. This would identify a “FAHFA signature” that is an early biomarker for T1D progression, and could enable earlier intervention to prevent overt T1D than is currently possible. The change in the abundance of FAHFA regioisomers in these participants over time may also provide important information about a potential role for FAHFAs in protecting against T1D development.
In Specific Aim 2, we expect to find differences in the RNA sequencing data in PBMCs from TrialNet progressors and non-progressors and we think these will associate with lower levels of anti-inflammatory FAHFAs. In particular, in the RNA sequencing we performed in islets from autoimmune NOD mice that were treated with vehicle or PAHSAs, PAHSA treatment reduced endoplasmic reticulum stress, islet cell senescence and the senescence associated secretory pathway, and increased markers of pancreatic α- and β-cell identity. We may see reduced stress and senescence in immune cells of non-progressors. We expect to see evidence of reduced activation of some cytotoxic immune cells in people with higher levels of anti-inflammatory FAHFAs. If we find associations, we will use this information to establish which FAHFAs are early biomarkers of auto-immune activity before T1D development. The association of lower levels of proinflammatory FAHFAs with more immune cell activation would also support the notion that treatment of high risk people with FAHFAs might prevent T1D development. For the studies in Aim 2 in which we will treat immune cell subsets from PBMCs from TrialNet participants with anti-inflammatory FAHFAs, we expect that FAHFA treatment will attenuate the differentiation and proliferation of cytotoxic T cells. We also expect that FAHFAs will reduce T cell polarization, pro-inflammatory cytokine and chemokine secretion, and immune cell migration, and protect islet β-cells. Such results would indicate that the beneficial effects of FAHFAs on lowering immune cell responses and attenuating T1D incidence in NOD mice are mediated, at least in part, through a reduction in diabetogenic T cell activation. These studies will determine whether reduced levels of anti-inflammatory FAHFAs at baseline or a fall in levels over time could contribute to the pro-inflammatory state during T1D development, and thereby increase the risk for, or rate of progression of, T1D.
Our data show that many FAHFAs reduce immune activation in human PBMCs, immune cell migration, and islet cell senescence, and FAHFAs potentiate insulin secretion from human islets. Therefore, in Specific Aim 3 we expect that FAHFA treatment will reduce immune attack on the human islets that we transplant into immunocompromised diabetic mice, preserve human β-cell mass, potentiate insulin secretion and lower blood glucose. In contrast, the islets transplanted into vehicle-treated diabetic immunocompromised mice will undergo immune attack and be destroyed by the allogenic PBMCs.
Overall, these results could be transformative in demonstrating that plasma FAHFA levels are early biomarkers for T1D. The data could also be used to design future studies aimed at utilizing FAHFAs for prevention and/or treatment of early T1D in people. Even if FAHFAs are not altered in T1D, administering these lipids may still have a therapeutic benefit to prevent or halt the progression of islet destruction, since, in NOD mice, PAHSAs delay the onset and dramatically decrease the incidence of T1D by attenuating islet inflammation and enhancing β-cell proliferation and survival.
Relevance to T1D
Type 1 diabetes (T1D) is an autoimmune disease characterized by the destruction of insulin-producing β cells. This results in insulin deficiency and hyperglycemia. Both genetic and environmental factors play important roles in the development of islet autoimmunity and progression to T1D. New cases of T1D are increasing worldwide and compromising patients’ quality and length of life. Markers are employed to track the progression from early auto-immunity to overt T1D but these indicate a relatively late disease stage after many insulin-making cells are lost. For lasting T1D prevention, markers that predict earlier stages of T1D development are needed. Currently, there is no cure for T1D, and diet, exercise and insulin replacement therapy are the cornerstones of disease management. Despite major advances in glucose monitoring and insulin delivery techniques, many patients with T1D can develop long-term complications. Abrogating autoimmunity and increasing β cell regeneration and proliferation in the pursuit of a cure for T1D in humans remain major challenges. Attempts to prevent T1D have included the study of antigen-specific therapies and systemic immunomodulatory and immunosuppressive agents. Many of these trials resulted in only transient β cell preservation and/or adverse effects due to generalized immune suppression. In new-onset T1D, immune modulatory agents have had heterogeneous results. Islet transplantation has also been pursued, but it has several limitations including a shortage of donors, fibrosis of transplanted islets, and side effects of immunosuppressive agents. The generation of stem cell–derived β cells is still under experimental development and requires generalized immunosuppression. Therefore, agents are needed that are safe, can induce immune tolerance to β cells, and can preserve β cells and enhance endogenous insulin production in humans at risk for, or with overt, T1D.
The fact that our novel lipids, PAHSAs/FAHFAs, modulate immune cell responses and also have direct protective effects on β-cells, makes them unique compared to other agents used to prevent or treat early T1D which are primarily immune-modulators and can be used for only short periods due to systemic immunosuppression.
Studies proposed here will contribute to a larger translational goal to eventually test a FAHFA, or a combination of FAHFAs in a human trial in islet-antibody positive people to reduce the risk of developing T1D. FAHFAs may prevent T1D in people because they effectively prevent it in mice prone to auto-immune diabetes and FAHFAs also protect human islets from cytokine attack and cellular stress. Due to the direct protective effects of FAHFAs on human islets, they may also prolong the survival of islets or human pluripotent stem cells-derived β-cells after transplantation into people with T1D and this could allow for reduction of immunosuppressive drugs which have adverse side effects.
One monogenic form of diabetes, MODY8, is caused by a gain-of-function mutation in a gene which encodes an enzyme that metabolizes FAHFAs. This mutation effectively reduces FAHFA levels in the pancreas which could remove the normal effect of FAHFAs to augment glucose-stimulated-insulin secretion. We are testing this human mutation now in mice. This supports the notion that FAHFAs are also important for the normal augmentation of insulin secretion in response to a rise in glucose.
Thus, not only do FAHFAs have the potential to be effective biomarkers for T1D progression, they could be novel therapeutic agents to prevent or treat T1D.