Integrated Metabolomics and Proteomics Analyses in the Local Milieu of Islet Allografts in Rejection versus Tolerance.

An understanding of the immune mechanisms that lead to rejection versus tolerance of allogeneic pancreatic islet grafts is of paramount importance, as it facilitates the development of innovative methods to improve the transplant outcome. Here, we used our established intraocular islet transplant model to gain novel insight into changes in the local metabolome and proteome within the islet allograft's immediate microenvironment in association with immune-mediated rejection or tolerance. We performed integrated metabolomics and proteomics analyses in aqueous humor samples representative of the graft's microenvironment under each transplant outcome. The results showed that several free amino acids, small primary amines, and soluble proteins related to the Warburg effect were upregulated or downregulated in association with either outcome. In general, the observed shifts in the local metabolite and protein profiles in association with rejection were consistent with established pro-inflammatory metabolic pathways and those observed in association with tolerance were immune regulatory. Taken together, the current findings further support the potential of metabolic reprogramming of immune cells towards immune regulation through targeted pharmacological and dietary interventions against specific metabolic pathways that promote the Warburg effect to prevent the rejection of transplanted islets and promote their immune tolerance.

In vivo imaging of type 1 diabetes immunopathology using eye-transplanted islets in NOD mice.

AIMS/HYPOTHESIS: Autoimmune attack against the insulin-producing beta cells in the pancreatic islets results in type 1 diabetes. However, despite considerable research, details of the type 1 diabetes immunopathology in situ are not fully understood mainly because of difficult access to the pancreatic islets in vivo. METHODS: Here, we used direct non-invasive confocal imaging of islets transplanted in the anterior chamber of the eye (ACE) to investigate the anti-islet autoimmunity in NOD mice before, during and after diabetes onset. ACE-transplanted islets allowed longitudinal studies of the autoimmune attack against islets and revealed the infiltration kinetics and in situ motility dynamics of fluorescence-labelled autoreactive T cells during diabetes development. Ex vivo immunostaining was also used to compare immune cell infiltrations into islet grafts in the eye and kidney as well as in pancreatic islets of the same diabetic NOD mice. RESULTS: We found similar immune infiltration in native pancreatic and ACE-transplanted islets, which established the ACE-transplanted islets as reliable reporters of the autoimmune response. Longitudinal studies in ACE-transplanted islets identified in vivo hallmarks of islet inflammation that concurred with early immune infiltration of the islets and preceded their collapse and hyperglycaemia onset. A model incorporating data on ACE-transplanted islet degranulation and swelling allowed early prediction of the autoimmune attack in the pancreas and prompted treatments to intercept type 1 diabetes. CONCLUSIONS/INTERPRETATION: The current findings highlight the value of ACE-transplanted islets in studying early type 1 diabetes pathogenesis in vivo and underscore the need for timely intervention to halt disease progression.

Operational immune tolerance towards transplanted allogeneic pancreatic islets in mice and a non-human primate.

AIMS/HYPOTHESIS: Patients with autoimmune type 1 diabetes transplanted with pancreatic islets to their liver experience significant improvement in quality of life through better control of blood sugar and enhanced awareness of hypoglycaemia. However, long-term survival and efficacy of the intrahepatic islet transplant are limited owing to liver-specific complications, such as immediate blood-mediated immune reaction, hypoxia, a highly enzymatic and inflammatory environment and locally elevated levels of drugs including immunosuppressive agents, all of which are injurious to islets. This has spurred a search for new islet transplant sites and for innovative ways to achieve long-term graft survival and efficacy without life-long systemic immunosuppression and its complications. METHODS: We used our previously established approach of islet transplant in the anterior chamber of the eye in allogeneic recipient mouse models and a baboon model of diabetes, which were treated transiently with anti-CD154/CD40L blocking antibody in the peri-transplant period. Survival of the intraocular islet allografts was assessed by direct visualisation in the eye and metabolic variables (blood glucose and C-peptide measurements). We evaluated longitudinally the cytokine profile in the local microenvironment of the intraocular islet allografts, represented in aqueous humour, under conditions of immune rejection vs tolerance. We also evaluated the recall response in the periphery of the baboon recipient using delayed-type hypersensitivity (DTH) assay, and in mice after repeat transplant in the kidney following initial transplant with allogeneic islets in the eye or kidney. RESULTS: Results in mice showed >300 days immunosuppression-free survival of allogeneic islets transplanted in the eye or kidney. Notably, >70% of tolerant mice, initially transplanted in the eye, exhibited >400 days of graft survival after re-transplant in the kidney without immunosuppression compared with ~30% in mice that were initially transplanted in the kidney. Cytokine and DTH data provided evidence of T helper 2-driven local and peripheral immune regulatory mechanisms in support of operational immune tolerance towards the islet allografts in both models. CONCLUSIONS/INTERPRETATION: We are currently evaluating the safety and efficacy of intraocular islet transplantation in a phase 1 clinical trial. In this study, we demonstrate immunosuppression-free long-term survival of intraocular islet allografts in mice and in a baboon using transient peri-transplant immune intervention. These results highlight the potential for inducing islet transplant immune tolerance through the intraocular route. Therefore, the current findings are conceptually significant and may impact markedly on clinical islet transplantation in the treatment of diabetes.

Correction to: In vivo imaging of type 1 diabetes immunopathology using eye-transplanted islets in NOD mice.

Unfortunately there is a mistake in the presentation of the affiliations in this paper.

Apolipoprotein CIII links islet insulin resistance to ß-cell failure in diabetes.

Insulin resistance and ß-cell failure are the major defects in type 2 diabetes mellitus. However, the molecular mechanisms linking these two defects remain unknown. Elevated levels of apolipoprotein CIII (apoCIII) are associated not only with insulin resistance but also with cardiovascular disorders and inflammation. We now demonstrate that local apoCIII production is connected to pancreatic islet insulin resistance and ß-cell failure. An increase in islet apoCIII causes promotion of a local inflammatory milieu, increased mitochondrial metabolism, deranged regulation of ß-cell cytoplasmic free Ca(2+) concentration ([Ca(2+)]i) and apoptosis. Decreasing apoCIII in vivo results in improved glucose tolerance, and pancreatic apoCIII knockout islets transplanted into diabetic mice, with high systemic levels of the apolipoprotein, demonstrate a normal [Ca(2+)]i response pattern and no hallmarks of inflammation. Hence, under conditions of islet insulin resistance, locally produced apoCIII is an important diabetogenic factor involved in impairment of ß-cell function and may thus constitute a novel target for the treatment of type 2 diabetes mellitus.

Young capillary vessels rejuvenate aged pancreatic islets.

Pancreatic islets secrete hormones that play a key role in regulating blood glucose levels (glycemia). Age-dependent impairment of islet function and concomitant dysregulation of glycemia are major health threats in aged populations. However, the major causes of the age-dependent decline of islet function are still disputed. Here we demonstrate that aging of pancreatic islets in mice and humans is notably associated with inflammation and fibrosis of islet blood vessels but does not affect glucose sensing and the insulin secretory capacity of islet beta cells. Accordingly, when transplanted into the anterior chamber of the eye of young mice with diabetes, islets from old mice are revascularized with healthy blood vessels, show strong islet cell proliferation, and fully restore control of glycemia. Our results indicate that beta cell function does not decline with age and suggest that islet function is threatened by an age-dependent impairment of islet vascular function. Strategies to mitigate age-dependent dysregulation in glycemia should therefore target systemic and/or local inflammation and fibrosis of the aged islet vasculature.

The Different Faces of the Pancreatic Islet.

Type 1 diabetes (T1D) patients who receive pancreatic islet transplant experience significant improvement in their quality-of-life. This comes primarily through improved control of blood sugar levels, restored awareness of hypoglycemia, and prevention of serious and potentially life-threatening diabetes-associated complications, such as kidney failure, heart and vascular disease, stroke, nerve damage, and blindness. Therefore, beta cell replacement through transplantation of isolated islets is an important option in the treatment of T1D. However, lasting success of this promising therapy depends on durable survival and efficacy of the transplanted islets, which are directly influenced by the islet isolation procedures. Thus, isolating pancreatic islets with consistent and reliable quality is critical in the clinical application of islet transplantation.Quality of isolated islets is important in pre-clinical studies as well, as efforts to advance and improve clinical outcomes of islet transplant therapy have relied heavily on animal models ranging from rodents, to pigs, to nonhuman primates. As a result, pancreatic islets have been isolated from these and other species and used in a variety of in vitro or in vivo applications for this and other research purposes. Protocols for islet isolation have been somewhat similar across species, especially, in mammals. However, given the increasing evidence about the distinct structural and functional features of human and mouse islets, using similar methods of islet isolation may contribute to inconsistencies in the islet quality, immunogenicity, and experimental outcomes. This may also contribute to the discrepancies commonly observed between pre-clinical findings and clinical outcomes. Therefore, it is prudent to consider the particular features of pancreatic islets from different species when optimizing islet isolation protocols.In this chapter, we explore the structural and functional features of pancreatic islets from mice, pigs, nonhuman primates, and humans because of their prevalent use in nonclinical, preclinical, and clinical applications.

Noninvasive in vivo model demonstrating the effects of autonomic innervation on pancreatic islet function.

The autonomic nervous system is thought to modulate blood glucose homeostasis by regulating endocrine cell activity in the pancreatic islets of Langerhans. The role of islet innervation, however, has remained elusive because the direct effects of autonomic nervous input on islet cell physiology cannot be studied in the pancreas. Here, we used an in vivo model to study the role of islet nervous input in glucose homeostasis. We transplanted islets into the anterior chamber of the eye and found that islet grafts became densely innervated by the rich parasympathetic and sympathetic nervous supply of the iris. Parasympathetic innervation was imaged intravitally by using transgenic mice expressing GFP in cholinergic axons. To manipulate selectively the islet nervous input, we increased the ambient illumination to increase the parasympathetic input to the islet grafts via the pupillary light reflex. This reduced fasting glycemia and improved glucose tolerance. These effects could be blocked by topical application of the muscarinic antagonist atropine to the eye, indicating that local cholinergic innervation had a direct effect on islet function in vivo. By using this approach, we found that parasympathetic innervation influences islet function in C57BL/6 mice but not in 129X1 mice, which reflected differences in innervation densities and may explain major strain differences in glucose homeostasis. This study directly demonstrates that autonomic axons innervating the islet modulate glucose homeostasis.

Rearrangement of microtubule network under biochemical and mechanical stimulations.

Cells are constantly under the influence of various external forces in their physiological environment. These forces are countered by the viscoelastic properties of the cytoskeleton. To understand the response of the cytoskeleton to biochemical and mechanical stimuli, GFP-tubulin expressing CHO cells were investigated using scanning laser confocal microscopy. Cells treated with nocodazole revealed disruption in the microtubule network within minutes of treatment while keeping the cell shape intact. By contrast, trypsin, a proteolytic agent, altered the shape of CHO cells by breaking the peptide bonds at adhesion sites. CHO cells were also stimulated mechanically by applying an indentation force with an atomic force microscope (AFM) and by shear stress in a parallel plate flow chamber. Mechanical stimulation applied using AFM showed two distinct cytoskeletal responses to the applied force: an immediate response that resulted in the depolymerization and displacement of the microtubules out of the contact zone, and a slower response characterized by tubulin polymerization at the periphery of the indented area. Flow chamber experiments revealed that shear force did not induce formation of new microtubules in CHO cells and that detachment of adherent cells from the substrate occurred independent from the flow direction. Overall, the experimental system described here allows real-time characterization of dynamic changes in cell cytoskeleton in response to the mechano-chemical stimuli and, therefore, provides better understanding of the biophysical and functional properties of cells.

High-resolution, noninvasive longitudinal live imaging of immune responses.

Intravital imaging emerged as an indispensible tool in biological research, and a variety of imaging techniques have been developed to noninvasively monitor tissues in vivo. However, most of the current techniques lack the resolution to study events at the single-cell level. Although intravital multiphoton microscopy has addressed this limitation, the need for repeated noninvasive access to the same tissue in longitudinal in vivo studies remains largely unmet. We now report on a previously unexplored approach to study immune responses after transplantation of pancreatic islets into the anterior chamber of the mouse eye. This approach enabled (i) longitudinal, noninvasive imaging of transplanted tissues in vivo; (ii) in vivo cytolabeling to assess cellular phenotype and viability in situ; (iii) local intervention by topical application or intraocular injection; and (iv) real-time tracking of infiltrating immune cells in the target tissue.

Supervised Parametric Learning in the Identification of Composite Biomarker Signatures of Type 1 Diabetes in Integrated Parallel Multi-Omics Datasets.

BACKGROUND: Type 1 diabetes (T1D) is a devastating autoimmune disease, and its rising prevalence in the United States and around the world presents a critical problem in public health. While some treatment options exist for patients already diagnosed, individuals considered at risk for developing T1D and who are still in the early stages of their disease pathogenesis without symptoms have no options for any preventive intervention. This is because of the uncertainty in determining their risk level and in predicting with high confidence who will progress, or not, to clinical diagnosis. Biomarkers that assess one's risk with high certainty could address this problem and will inform decisions on early intervention, especially in children where the burden of justifying treatment is high. Single omics approaches (e.g., genomics, proteomics, metabolomics, etc.) have been applied to identify T1D biomarkers based on specific disturbances in association with the disease. However, reliable early biomarkers of T1D have remained elusive to date. To overcome this, we previously showed that parallel multi-omics provides a more comprehensive picture of the disease-associated disturbances and facilitates the identification of candidate T1D biomarkers. METHODS: This paper evaluated the use of machine learning (ML) using data augmentation and supervised ML methods for the purpose of improving the identification of salient patterns in the data and the ultimate extraction of novel biomarker candidates in integrated parallel multi-omics datasets from a limited number of samples. We also examined different stages of data integration (early, intermediate, and late) to assess at which stage supervised parametric models can learn under conditions of high dimensionality and variation in feature counts across different omics. In the late integration scheme, we employed a multi-view ensemble comprising individual parametric models trained over single omics to address the computational challenges posed by the high dimensionality and variation in feature counts across the different yet integrated multi-omics datasets. RESULTS: the multi-view ensemble improves the prediction of case vs. control and finds the most success in flagging a larger consistent set of associated features when compared with chance models, which may eventually be used downstream in identifying a novel composite biomarker signature of T1D risk. CONCLUSIONS: the current work demonstrates the utility of supervised ML in exploring integrated parallel multi-omics data in the ongoing quest for early T1D biomarkers, reinforcing the hope for identifying novel composite biomarker signatures of T1D risk via ML and ultimately informing early treatment decisions in the face of the escalating global incidence of this debilitating disease.

A Composite Biomarker Signature of Type 1 Diabetes Risk Identified via Augmentation of Parallel Multi-Omics Data from a Small Cohort.

Background: Biomarkers of early pathogenesis of type 1 diabetes (T1D) are crucial to enable effective prevention measures in at-risk populations before significant damage occurs to their insulin producing beta-cell mass. We recently introduced the concept of integrated parallel multi-omics and employed a novel data augmentation approach which identified promising candidate biomarkers from a small cohort of high-risk T1D subjects. We now validate selected biomarkers to generate a potential composite signature of T1D risk. Methods: Twelve candidate biomarkers, which were identified in the augmented data and selected based on their fold-change relative to healthy controls and cross-reference to proteomics data previously obtained in the expansive TEDDY and DAISY cohorts, were measured in the original samples by ELISA. Results: All 12 biomarkers had established connections with lipid/lipoprotein metabolism, immune function, inflammation, and diabetes, but only 7 were found to be markedly changed in the high-risk subjects compared to the healthy controls: ApoC1 and PON1 were reduced while CETP, CD36, FGFR1, IGHM, PCSK9, SOD1, and VCAM1 were elevated. Conclusions: Results further highlight the promise of our data augmentation approach in unmasking important patterns and pathologically significant features in parallel multi-omics datasets obtained from small sample cohorts to facilitate the identification of promising candidate T1D biomarkers for downstream validation. They also support the potential utility of a composite biomarker signature of T1D risk characterized by the changes in the above markers.

Fluorescence Angiography with Dual Fluorescence for the Early Detection and Longitudinal Quantitation of Vascular Leakage in Retinopathy.

BACKGROUND: Diabetic retinopathy (DR) afflicts more than 93 million people worldwide and is a leading cause of vision loss in working adults. While DR therapies are available, early DR development may go undetected without treatment due to the lack of sufficiently sensitive tools. Therefore, early detection is critically important to enable efficient treatment before progression to vision-threatening complications. A major clinical manifestation of early DR is retinal vascular leakage that may progress from diffuse to more localized focal leakage, leading to increased retinal thickness and diabetic macular edema (DME). In preclinical research, a hallmark of DR in mouse models is diffuse retinal leakage without increased thickness or DME, which limits the utility of optical coherence tomography and fluorescein angiography (FA) for early detection. The Evans blue assay detects diffuse leakage but requires euthanasia, which precludes longitudinal studies in the same animals. METHODS: We developed a new modality of ratiometric fluorescence angiography with dual fluorescence (FA-DF) to reliably detect and longitudinally quantify diffuse retinal vascular leakage in mouse models of induced and spontaneous DR. RESULTS: These studies demonstrated the feasibility and sensitivity of FA-DF in detecting and quantifying retinal vascular leakage in the same mice over time during DR progression in association with chronic hyperglycemia and age. CONCLUSIONS: These proof-of-concept studies demonstrated the promise of FA-DF as a minimally invasive method to quantify DR leakage in preclinical mouse models longitudinally.

Exploring Computational Data Amplification and Imputation for the Discovery of Type 1 Diabetes (T1D) Biomarkers from Limited Human Datasets.

BACKGROUND: Type 1 diabetes (T1D) is a devastating disease with serious health complications. Early T1D biomarkers that could enable timely detection and prevention before the onset of clinical symptoms are paramount but currently unavailable. Despite their promise, omics approaches have so far failed to deliver such biomarkers, likely due to the fragmented nature of information obtained through the single omics approach. We recently demonstrated the utility of parallel multi-omics for the identification of T1D biomarker signatures. Our studies also identified challenges. METHODS: Here, we evaluated a novel computational approach of data imputation and amplification as one way to overcome challenges associated with the relatively small number of subjects in these studies. RESULTS: Using proprietary algorithms, we amplified our quadra-omics (proteomics, metabolomics, lipidomics, and transcriptomics) dataset from nine subjects a thousand-fold and analyzed the data using Ingenuity Pathway Analysis (IPA) software to assess the change in its analytical capabilities and biomarker prediction power in the amplified datasets compared to the original. These studies showed the ability to identify an increased number of T1D-relevant pathways and biomarkers in such computationally amplified datasets, especially, at imputation ratios close to the "golden ratio" of 38.2%:61.8%. Specifically, the Canonical Pathway and Diseases and Functions modules identified higher numbers of inflammatory pathways and functions relevant to autoimmune T1D, including novel ones not identified in the original data. The Biomarker Prediction module also predicted in the amplified data several unique biomarker candidates with direct links to T1D pathogenesis. CONCLUSIONS: These preliminary findings indicate that such large-scale data imputation and amplification approaches are useful in facilitating the discovery of candidate integrated biomarker signatures of T1D or other diseases by increasing the predictive range of existing data mining tools, especially when the size of the input data is inherently limited.

Challenges in stem cell-derived islet replacement therapy can be overcome.

In this Commentary, we echo the conclusions of a recent review titled "The promise of stem cell-derived islet replacement therapy," which highlighted recent advances in producing glucose responsive "islets" from stem cells and the benefits of their use in islet transplant therapy in type 1 diabetes (T1D). The review also outlined the status of clinical islet transplantation and the challenges that have prevented it from reaching its full therapeutic promise. We agree with the conclusions of the review and suggest that the identified challenges may be overcome by using the eye anterior chamber as an islet transplant site. We anticipate that the combination of stem cell-derived islets and intraocular transplant could help this promising T1D therapy reach full fruition.

Parallel Multi-Omics in High-Risk Subjects for the Identification of Integrated Biomarker Signatures of Type 1 Diabetes.

BACKGROUND: Biomarkers are crucial for detecting early type-1 diabetes (T1D) and preventing significant ß-cell loss before the onset of clinical symptoms. Here, we present proof-of-concept studies to demonstrate the potential for identifying integrated biomarker signature(s) of T1D using parallel multi-omics. METHODS: Blood from human subjects at high risk for T1D (and healthy controls; n = 4 + 4) was subjected to parallel unlabeled proteomics, metabolomics, lipidomics, and transcriptomics. The integrated dataset was analyzed using Ingenuity Pathway Analysis (IPA) software for disturbances in the at-risk subjects compared to controls. RESULTS: The final quadra-omics dataset contained 2292 proteins, 328 miRNAs, 75 metabolites, and 41 lipids that were detected in all samples without exception. Disease/function enrichment analyses consistently indicated increased activation, proliferation, and migration of CD4 T-lymphocytes and macrophages. Integrated molecular network predictions highlighted central involvement and activation of NF-κB, TGF-ß, VEGF, arachidonic acid, and arginase, and inhibition of miRNA Let-7a-5p. IPA-predicted candidate biomarkers were used to construct a putative integrated signature containing several miRNAs and metabolite/lipid features in the at-risk subjects. CONCLUSIONS: Preliminary parallel quadra-omics provided a comprehensive picture of disturbances in high-risk T1D subjects and highlighted the potential for identifying associated integrated biomarker signatures. With further development and validation in larger cohorts, parallel multi-omics could ultimately facilitate the classification of T1D progressors from non-progressors.

HGAL inhibits lymphoma dissemination by interacting with multiple cytoskeletal proteins.

Human germinal center-associated lymphoma (HGAL) is an adaptor protein specifically expressed in germinal center lymphocytes. High expression of HGAL is a predictor of prolonged survival of diffuse large B-cell lymphoma (DLBCL) and classic Hodgkin lymphoma. Furthermore, HGAL expression is associated with early-stage DLBCL, thus potentially limiting lymphoma dissemination. In our previous studies, we demonstrated that HGAL regulates B-cell receptor signaling and cell motility in vitro and deciphered some molecular mechanisms underlying these effects. By using novel animal models for in vivo DLBCL dispersion, we demonstrate here that HGAL decreases lymphoma dissemination and prolongs survival. Furthermore, by using an unbiased proteomic approach, we demonstrate that HGAL may interact with multiple cytoskeletal proteins thereby implicating a multiplicity of effects in regulating lymphoma motility and spread. Specifically, we show that HGAL interacts with tubulin, and this interaction may also contribute to HGAL effects on cell motility. These findings recapitulate previous observations in humans, establish the role of HGAL in dissemination of lymphoma in vivo, and explain improved survival of patients with HGAL-expressing lymphomas.

Temperature modulation of integrin-mediated cell adhesion.

In response to external stimuli, cells modulate their adhesive state by regulating the number and intrinsic affinity of receptor/ligand bonds. A number of studies have shown that cell adhesion is dramatically reduced at room or lower temperatures as compared with physiological temperature. However, the underlying mechanism that modulates adhesion is still unclear. Here, we investigated the adhesion of the monocytic cell line THP-1 to a surface coated with intercellular adhesion molecule-1 (ICAM-1) as a function of temperature. THP-1 cells express the integrin lymphocyte function-associated antigen-1 (LFA-1), a receptor for ICAM-1. Direct force measurements of cell adhesion and cell elasticity were carried out by atomic force microscopy. Force measurements revealed an increase of the work of de-adhesion with temperature that was coupled to a gradual decrease in cellular stiffness. Of interest, single-molecule measurements revealed that the rupture force of the LFA-1/ICAM-1 complex decreased with temperature. A detailed analysis of the force curves indicated that temperature-modulated cell adhesion was mainly due to the enhanced ability of cells to deform and to form a greater number of longer membrane tethers at physiological temperatures. Together, these results emphasize the importance of cell mechanics and membrane-cytoskeleton interaction on the modulation of cell adhesion.

NOD Mice-Good Model for T1D but Not Without Limitations.

The nonobese diabetic (NOD) mouse model of type 1 diabetes (T1D) was discovered by coincidence in the 1980s and has since been widely used in the investigation of T1D and diabetic complications. The current in vivo study was originally designed to prospectively assess whether hyperglycemia onset is associated with physical destruction or functional impairment of beta cells under inflammatory insult during T1D progression in diabetes-prone female NOD mice. Prediabetic 16- to 20-wk-old NOD mice were transplanted with green fluorescent protein (GFP)-expressing reporter islets in the anterior chamber of the eye (ACE) that were monitored longitudinally, in addition to glycemia, with and without immune modulation using anti-CD3 monoclonal antibody therapy. However, there was an early and vigorous immune reaction against the GFP-expressing beta cells that lead to their premature destruction independent of autoimmune T1D development in progressor mice that eventually became hyperglycemic. This immune reaction also occurred in nonprogressor NOD recipients. These findings showed a previously unknown reaction of NOD mice to GFP that prevented achieving the original goals of this study but highlighted a new feature of the NOD mice that should be considered when designing experiments using this model in T1D research.

A machine learning approach to predict pancreatic islet grafts rejection versus tolerance.

The application of artificial intelligence (AI) and machine learning (ML) in biomedical research promises to unlock new information from the vast amounts of data being generated through the delivery of healthcare and the expanding high-throughput research applications. Such information can aid medical diagnoses and reveal various unique patterns of biochemical and immune features that can serve as early disease biomarkers. In this report, we demonstrate the feasibility of using an AI/ML approach in a relatively small dataset to discriminate among three categories of samples obtained from mice that either rejected or tolerated their pancreatic islet allografts following transplant in the anterior chamber of the eye, and from naïve controls. We created a locked software based on a support vector machine (SVM) technique for pattern recognition in electropherograms (EPGs) generated by micellar electrokinetic chromatography and laser induced fluorescence detection (MEKC-LIFD). Predictions were made based only on the aligned EPGs obtained in microliter-size aqueous humor samples representative of the immediate local microenvironment of the islet allografts. The analysis identified discriminative peaks in the EPGs of the three sample categories. Our classifier software was tested with targeted and untargeted peaks. Working with the patterns of untargeted peaks (i.e., based on the whole pattern of EPGs), it was able to achieve a 21 out of 22 positive classification score with a corresponding 95.45% prediction accuracy among the three sample categories, and 100% accuracy between the rejecting and tolerant recipients. These findings demonstrate the feasibility of AI/ML approaches to classify small numbers of samples and they warrant further studies to identify the analytes/biochemicals corresponding to discriminative features as potential biomarkers of islet allograft immune rejection and tolerance.