The Ailing β-Cell in Diabetes: Insights From a Trip to the ER: The 2023 Outstanding Scientific Achievement Award Lecture Free

The ability of the pancreatic β-cell to synthesize and release insulin in response to nutritional cues is central to the regulation of glucose homeostasis and metabolism. Dysfunction of the pancreatic β-cells underlies the pathophysiology of both major forms of diabetes, and our research program over the last 15 years has focused on three overarching themes centered on β-cell biology and diabetes pathophysiology. The first theme centers on understanding the molecular and signaling pathways that drive the transition from euglycemia to hyperglycemia and diabetes. Much of this work has focused on calcium signaling within the β-cell. Second, we have an active translational research program focused on biomarker discovery and methods to detect β-cell dysfunction in the presymptomatic phase of type 1 diabetes (T1D). Finally, each of our projects is performed with an eye toward clinical translation with the goal of informing therapies to restore β-cell health in T1D.

At the very nucleus of these research themes is the ubiquitous second messenger, calcium. The discovery of the importance of calcium in cell biology is an incredible footnote in the history of science. It is also one of complete serendipity. At the center of this story is the physiologist and meticulous experimentalist, Sydney Ringer, and his technician. Dr. Ringer’s research focused on identifying the specific ions that were necessary to support cardiac contractility. Their model system included explanted frog hearts and a variety of solutions dictated by Dr. Ringer and tested in succession by his technician. Ringer’s obituary suggests that his laboratory assistant often had difficulty keeping up with his boss’s demands. One day, rather than using distilled water as mandated by Dr. Ringer, the technician substituted water from the tap. Remarkably, this substitution led to sustained cardiac contraction in the explanted hearts for the first time. Ultimately, this mistake led the team to identify calcium as the key constituent needed to maintain the electrical activity of the heart. Ringer and his technician’s work still touch modern medicine today. The intravenous lactated Ringer’s solution carries his name, and its composition still reflects discoveries made from this important series of experiments performed in the 1880s (1–5).

Analogous to the cardiovascular system, calcium plays a central role in regulating β-cell function and insulin secretion (6). The canonical pathway of insulin secretion involves cycles of glucose sensing that are coupled to mitochondrial metabolism and closure of the ATP-sensitive potassium (K_ATP) channels in the plasma membrane. Membrane depolarization leads to opening of voltage-gated calcium channels and calcium influx from the extracellular space. Release of calcium from the intracellular stores serves to amplify this response, and it is ultimately this increase in cytosolic calcium that triggers insulin granule release and exocytosis (Fig. 1).

At the core of this process of insulin secretion is the maintenance of calcium concentration gradients that are organized at both the cellular and organelle level. These gradients follow the ever-important principle of “location, location, location.” Specifically, calcium is maintained at high levels in the extracellular space and at very low levels in the cytosol, with intermediate levels found in the acidic organelles of the secretory pathway, which include the endoplasmic reticulum (ER), the Golgi apparatus, and the secretory granules (6). In addition to amplifying insulin secretion, high levels of calcium within these organelles play an important role in maintaining the heavy biosynthetic burden of insulin production, as calcium serves as a cofactor for molecular chaperones and foldases in the ER and regulates protein processing within the Golgi apparatus and secretory granules. Calcium in these compartments also regulates organelle-specific stress responses, including ER and Golgi stress (7,8). High calcium concentrations in the ER are regulated by the balance of calcium release from the inositol 1,4,5-trisphosphate receptors (IP3R) and ryanodine receptors (RYR) and calcium uptake via the sarcoendoplasmic reticulum calcium ATPase (SERCA) pumps. In addition, in response to ER calcium depletion, store-operated calcium entry is triggered to replenish calcium stores through a molecular complex that forms between stromal interaction molecule 1 (STIM1) and the calcium release–activated calcium channel protein 1 (ORA) channels on the plasma membrane (9–13) (Fig. 2).

Similar to the accidental discovery of calcium’s role in the heart by Ringer and his assistant, my interest in this aspect of cell biology had its origins in serendipity. During my postdoctoral fellowship in the laboratory of Dr. Raghu Mirmira at the University of Virginia, I had been tasked with a project aimed at defining mechanisms through which the peroxisome proliferator–activated receptor-γ (PPAR-γ) agonist pioglitazone improved β-cell health. My project involved several weeks of pioglitazone administration to db/db mice to characterize epigenetic changes within key β-cell identity genes. My neighbor in the adjoining laboratory was Dr. Craig Nunemaker, now a faculty member at Ohio University. Craig was an expert on calcium imaging and generously offered to perform this assay on islets from my pioglitazone-treated db/db mice. He found that glucose-stimulated calcium responses were significantly diminished in islets from the diabetic db/db mice and, remarkably, pioglitazone treatment nearly completely restored the calcium responses. In a second and very impactful stroke of luck, a reviewer of our manuscript suggested that we measure the expression of the three SERCA isoforms in the β-cell. Sure enough, expression of these genes was markedly reduced in vehicle-treated db/db mice, while pioglitazone restored expression of SERCA2a, SERCA2b, and SERCA3 mRNA levels (13).

Thus began my fascination with the SERCA pump as I was starting my own laboratory at Indiana University (IU). Since that time, we have studied SERCA pump dysfunction in several disease models (12–16). We have expanded our research to study how ER stress and misfolded protein accumulation lead to abnormal gating of the RYR (9,10) and how palmitate and cytokine-mediated stress lead to impaired store-operated calcium entry. In this article, I will share three recent vignettes from the laboratory that focus on this theme of ER calcium-mediated regulation of β-cell health (Fig. 3). The first of these vignettes will illustrate a role for SERCA and ER calcium in the regulation of protein trafficking and proinsulin processing within the β-cell (17). The second will highlight how alterations in β-cell calcium signaling intersect with T1D pathogenesis. Finally, I will end with a discussion of how activation of β-cell stress pathways may serve as an anchor to inform biomarker strategies in T1D (18).

Early on, we narrowed our focus to SERCA2b, which is the most highly expressed of the three SERCA isoforms found in the β-cell (16). These early studies focused on understanding the transcriptional pathways that led to impaired expression and activity of SERCA2b under a variety of different stress conditions (14–16). However, an important underpinning of these questions was whether loss of SERCA2b was sufficient to affect whole-body glucose homeostasis. To this end, we have developed several mouse models to study these relationships. We first asked whether total-body SERCA2 haploinsufficiency in mice (S2HET) affected metabolic responses to diet-induced obesity. In male mice that had been challenged with 16 weeks of high-fat diet, we found that S2HET mice exhibited impaired glucose tolerance. Consistent with the idea of a primary β-cell defect, glucose-stimulated insulin release was reduced in islets from high-fat-diet–fed S2HET mice, and isolated islets from these mice had morphological and transcriptional evidence of ER stress (12).

More recently, our laboratory developed a β-cell–specific SERCA2 knockout (βS2KO) mouse model. Because redundancy is a key feature in all biological systems, an important question we asked was whether deletion of one SERCA isoform was enough to affect steady-state calcium levels within this compartment. To address this question, we use a variety of imaging approaches to measure calcium in different cellular compartments, including a fluorescence resonance energy transfer–based approach to measure relative ER calcium levels. Using the D4ER biosensor shared with us by Dr. Patrick Gilon (19), we found that ER calcium levels were indeed reduced in islets of βS2KO mice. We went on to detect alterations in glucose-stimulated calcium responses in the cytosol that were notably reminiscent of the patterns that Craig and I had observed many years ago in db/db mouse islets (13). More detailed metabolic phenotyping revealed that βS2KO mice developed a mild age-related glucose intolerance and had evidence of reduced insulin secretion ex vivo (17).

Given the important role for calcium in patterning insulin production and maturation, we wanted to leverage our βS2KO mouse model to ask a different, but related, set of questions aimed at understanding how alterations in ER calcium levels affected the quality of insulin secretion. As a reminder, preproinsulin processing begins with cleavage of its N-terminal signal peptide to form proinsulin, a process that occurs within the lumen of the ER. This cleavage is followed by disulfide bond formation and terminal protein folding, which occur sequentially in the ER and Golgi apparatus. Finally, proinsulin is cleaved into mature insulin and C-peptide in a process regulated by the prohormone convertases 1/3 (PC1/3) and 2 (PC2) and carboxypeptidase E (CPE) within the secretory granules. β-cell dysfunction, such as that caused by inflammatory or ER stress, results in the accumulation and release of incompletely processed proinsulin (20). However, the mechanisms leading to impaired proinsulin processing remain incompletely understood.

Notably, in the serum of our aged βS2KO mice, we observed reduced insulin and increased proinsulin levels, resulting in a significant increase in the proinsulin-to-insulin ratio in the serum. This pattern was phenocopied in the pancreas, suggesting both an accumulation and increased secretion of proinsulin in our model. These findings raised an interesting set of questions regarding the mechanisms underlying this insulin processing defect. These questions sent us back to beautiful literature authored by giants in the field of prohormone processing, including Don Steiner, a recipient of this same award in 1969, as well as Halban, Hutton, Davidson, Rhodes, Lindbergh, Arvan, and others. This literature describes how proinsulin processing enzymes are synthesized as zymogens and become activated in distinct domains of the secretory pathway in a process that is spatially and temporally regulated by pH, calcium, and associated chaperone proteins (21–26).

To understand how SERCA2 deletion impacts insulin processing, we measured first the expression patterns of these prohormone convertases. We observed an interesting pattern of dysregulation characterized by loss of expression of the most active forms of PC1/3 and accumulation of the less active form of PC2, proPC2. We next measured PC1/3 and PC2 activity in a β-cell line where we had used CRISPR/Cas9 to delete SERCA2. Notably, in INS-1 cells lacking SERCA2, we found that activity of both PC1/3 and PC2 was reduced. Importantly, when we added back SERCA2 using viral overexpression, we could partially rescue PC1/3 and PC2 activity. Next, to identify mechanisms underlying this defect in proenzyme and prohormone maturation, we performed RNA sequencing. In this data set, there was significant modulation of pathways involved in protein secretion, vesicle trafficking, and processing in islets from βS2KO mice. These data led us to propose a model whereby chronic ER calcium deficiency due to SERCA2 deletion leads to a defect in anterograde trafficking in the β-cell secretory pathway, resulting in defective maturation and trafficking of the prohormone convertases and proinsulin, ultimately leading to impaired proinsulin processing. In key imaging experiments, we took advantage of an antibody shared by Iris Lindberg that specifically recognizes proPC2, and we found increased accumulation of proPC2 in the cis Golgi in the absence of SERCA2. Similarly, we observed increased accumulation of incompletely processed proinsulin in the Golgi apparatus and the intermediate compartment that lies between the ER and the Golgi apparatus.

This first vignette highlights the importance of both spatial and temporal regulation of biological processes in health and disease, and our ongoing work in this area is focused on understanding the cargo specificity and more detailed mechanisms underlying this defect in trafficking. However, as a cell biologist and physician scientist, an important goal of my work has been to always conceptualize findings from our basic science studies within the context of clinical disease. Along these lines and with Emily Sims, a close collaborator at the IU School of Medicine, we have tested proinsulin secretory characteristics in autoantibody-positive individuals who progressed to stage 3 T1D. Using samples from the TrialNet Pathway to Prevention cohort, we found that proinsulin–to–C-peptide ratios were increased in progressors to diabetes compared with nonprogressors around 12 months prior to T1D onset. Elevations in the proinsulin–to–C-peptide ratio were most prominent in very young children and could predict imminent diabetes in this group (27). In addition, we found that defects in proinsulin processing persist in established disease, as we found that most individuals with long-standing T1D retain detectable levels of proinsulin. More surprisingly, most individuals with long-standing T1D and undetectable C-peptide still had measurable proinsulin, indicating that their remaining β-cell mass is able to produce this precursor insulin molecule but is unable to process it into mature insulin and C-peptide (28). In addition, we have followed with great interest the recent identification of several monogenic forms of diabetes that are characterized by defective trafficking between the ER and Golgi apparatus, highlighting how important the fidelity of this process is to maintaining β-cell health (29,30).

We were similarly intrigued by a recent report from a Swedish dermatologist and now collaborator, Jakob Wikstrom. Jakob studies cohorts of individuals with genetic forms of ichthyosis, including individuals with Darier-White disease, which is caused by haploinsufficiency of the SERCA2 pump. A goal of his work is to identify nondermatological disorders associated with this rare disease, and he has recently found that individuals with Darier-White disease exhibit an increased risk of developing T1D (31). In the second vignette, I would like to share our laboratory’s recent journey that focused on determining how alterations in β-cell calcium signaling contribute to the development of T1D. In recent years, several pathways in the β-cell have been implicated in T1D pathogenesis and linked with increased β-cell immunogenicity. These pathways include release of extracellular vesicles containing autoantigens and inflammatory miRNAs, cellular senescence, and organelle-specific stress responses, including ER stress, which has been associated with MHC class I upregulation on the β-cell surface and neoantigen production (32).

Against this background and given Dr. Wikstrom’s important clinical observation, we developed a project to explore the relationship between SERCA2 deficiency, ER stress, and T1D development. We backcrossed the total-body SERCA2 haploinsufficient mouse onto the NOD mouse model of T1D. Interestingly, we observed an accelerated rate of diabetes development in NOD mice with SERCA2 haploinsufficiency. In these mice, the median age of diabetes onset was around 14 weeks, while it approached 18 weeks in the wild-type (WT) NOD mice. Additionally, NOD-SERCA2 haploinsufficient mice developed anti-insulin antibodies at a higher and faster rate. We confirmed that these mice had reduced ER calcium levels; however, much to our surprise and in contrast to our previous studies, there were no differences in the expression of genes involved in ER stress signaling between NOD-WT and NOD-SERCA2 haploinsufficient islets. This data suggested that, contrary to our original hypothesis, ER stress may not play a prominent role in the accelerated diabetes onset observed in these mice.

This surprising observation prompted us to take a step back and perform single-cell RNA sequencing to gain an unbiased assessment of changes in gene expression in our model. We applied this approach in islets from mice at 6 and 10 weeks of age to characterize early and late changes in gene expression, and a number of interesting patterns emerged. First, when we looked at differences between the genotypes at 6 weeks of age, we observed a significant increase in the number of islet-infiltrating immune cells in the NOD-SERCA2 haploinsufficient mice. These infiltrating cells were largely B and T cells. Even more interestingly, within the β-cell populations, there was marked upregulation of genes involved in immune activation and antigen presentation, including increased expression of several HLA class I molecules. Next, we examined β-cell gene expression changes in our NOD-SERCA2 haploinsufficient mice across time by comparing the 6- and 10-week time points. Here, we observed continued modulation of pathways related to immune activation. Interestingly, this comparison also revealed changes in several metabolic processes, including ATP synthesis and oxidative phosphorylation, both of which are regulated by the mitochondria. This finding was notable to us because although we had created a primary defect in the ER, the ER and mitochondria are intimately connected, and there is active exchange of metabolites, lipids, and calcium between these two organelles. Therefore, we reasoned that alterations in ER calcium may result in reciprocal changes in mitochondrial calcium. To test this hypothesis, we used a fluorescence resonance energy transfer probe targeted to the mitochondria and found that islets from NOD-SERCA2 haploinsufficient mice had increased levels of mitochondrial calcium with no differences in cytosolic calcium levels, thus highlighting the importance of this ER-mitochondrion interface. We measured mitochondrial function using the Seahorse platform and found that islets from NOD-SERCA2 haploinsufficient mice exhibited defects in oxygen consumption and ATP synthesis. Consistent with these findings, we observed reduced expression of ATP synthase and complex III of the electron transport chain in islets isolated from NOD-SERCA2 haploinsufficient mice. Finally, electron microscopy showed significant alterations in mitochondrial morphology in these mice. Taken together, our results suggest that SERCA2 haploinsufficiency on a background of immune activation is sufficient to reduce ER calcium levels. Surprisingly, this reduction in calcium was not associated with a prominent ER stress phenotype but rather resulted in changes in ER calcium that led to increased mitochondrial calcium levels and mitochondrial dysfunction and ultimately accelerated diabetes onset.

An important but still incompletely answered question from this unpublished work is whether SERCA2 haploinsufficiency simply leads to a weaker β-cell, which may unmask diabetes at an earlier time point. Another possibility is that we have created a primary immune phenotype, which is a hypothesis that we have not yet tested. However, at the intersection of these two scenarios lies a third possibility. Have we instead created β-cells that are intrinsically stressed in a way that leads to increased immunogenicity? Several pieces of evidence suggest that this is a possibility, as we observed HLA class I upregulation, increased immune cell infiltration into the islets, and accelerated anti-insulin antibody development. How these changes are communicated to the immune system is an important and evolving question. An intriguing possibility that we are just beginning to study is whether stressed mitochondria communicate inflammatory signals to neighboring cells. Such a model might link an intrinsic defect in mitochondria with a scenario of immune activation. However, much more work is needed to test this model.

In addition to providing mechanistic insight, our model in this second vignette offers a unique opportunity to test novel strategies aimed at improving β-cell health in T1D. In this regard, we have recently initiated a collaboration with Neurodon to test a series of SERCA allosteric activators. Notably, when we treat NOD mice with these activators from 6 to 10 weeks of age, we observe the expected improvements in ER calcium. In addition, in both WT-NOD and NOD-SERCA2 haploinsufficient mice, there is reduced insulitis and delayed diabetes development when the SERCA activator is administered, suggesting that this pathway can be targeted therapeutically, at least in preclinical models.

The last vignette centers around projects in the laboratory that are focused on leveraging our knowledge of β-cell stress pathways to inform strategies for T1D risk prediction and disease modification. Since the discovery of insulin over 100 years ago, the treatment of T1D has evolved dramatically. Individuals with T1D now have access to continuous glucose monitoring, improved insulins, and pumps and devices that allow for automated insulin delivery (33,34). As an investigator in the T1D TrialNet Network, a considerable amount of my professional effort has been focused on an additional piece of this puzzle: the possibility of using disease-modifying therapies to either delay or prevent the onset of diabetes in those at increased risk. An important step in this direction occurred in November 2022, when the U.S. Food and Drug Administration (FDA) approved teplizumab as the first disease-modifying therapy in T1D. The approval of this drug came on the heels of decades of testing in preclinical models and clinical trials, and the drug is now indicated to delay the progression from stage 2 to stage 3 T1D. This FDA decision represents a framework-shifting event in the field of T1D that has raised important questions as to how we most efficiently identify individuals who could be treated with teplizumab, how we follow cohorts of autoantibody-positive individuals identified from screening, and how we test new agents to develop additional disease-modifying therapies, including those focused on restoring β-cell health (35).

Against this background, I wanted to highlight a recent project from the laboratory that aligns with these aims, especially the goal of identifying and validating temporal biomarkers of β-cell stress. To this end, we isolated islets from NOD mice at three prediabetic time points (10, 12, and 14 weeks of age) and at diabetes onset and submitted these islets for Swath proteomics analysis followed by data curation and target validation. When we looked closely at temporally modulated and upregulated pathways from this data set, we identified pathways involved in mitochondrial dysfunction, cell-cell communication, and the unfolded protein/ER stress responses. This finding was somewhat analogous to identifying all the “usual suspects” and many of the pathways highlighted in the first two vignettes. However, when we examined these pathways further, we observed an interesting pattern of time-restricted upregulation of proteins known to play a role in mediating ER and mitochondrial stress responses. Overall, this pattern of upregulation was observed through the 14-week time point, followed by declining expression of several protective proteins that coincided with diabetes development. One intriguing protein exhibiting this pattern of expression was protein disulfide isomerase A1 (PDIA1).

This protein caught our attention for several reasons. First, PDIA1 is an ER-localized thiol oxidoreductase that plays a critical role in proinsulin folding and the regulation of ER function (36). Second, our collaborators, Mei-Ling Yang and Mark Mamula, have recently identified autoantibodies to PDIA1 in individuals with early-onset T1D (37). Third, data from other cell types have shown that PDIA1 is upregulated and secreted from stressed cells (38). To test whether PDIA1 levels were modulated in human diabetes, we stained pancreatic sections from nPOD (Network for Pancreatic Organ Donors With Diabetes) donors. Consistent with findings from the NOD mouse model, we observed upregulation of PDIA1 in islets from autoantibody-positive donors and donors with T1D compared with nondiabetic donors. To test the paradigm that PDIA1 may serve as a circulating biomarker, we developed an assay to measure PDIA1 levels in serum and plasma using the Meso Scale platform. We applied this assay in our original NOD cohort and identified increased plasma PDIA1 levels in 14-week-old NOD mice compared with nondiabetic control mice. To translate these findings to human T1D, we assayed serum samples that were collected from a small cohort of children within 48 h of diabetes onset and found that PDIA1 levels were increased in children with T1D compared with age-, sex-, and BMI category–matched controls (18).

In summary, this final vignette highlights our approach in using orthogonal methods to identify biomarkers of β-cell health. We identified PDIA1 as one such potential biomarker, and we translated our findings from preclinical and organ-based models to a small clinical cohort. Our previous work employed similar strategies that culminated in identifying proinsulin as another T1D-associated biomarker.

My hope is that the work highlighted throughout this article will help to fill in the expanding puzzle of diabetes. The research my laboratory has performed over the past 15 years has illustrated a critical role for SERCA and ER calcium in the regulation of protein trafficking, proinsulin processing, and insulin secretion within the β-cell, highlighted how alterations in β-cell calcium signaling intersect with T1D pathogenesis, and informed biomarker strategies for identifying β-cell stress pathways early in T1D development. My vision for the future is that multiple biomarkers of β-cell stress will be combined with additional immune and metabolic biomarkers to better predict risk, define mechanisms, and ultimately improve therapies in early-stage T1D (Fig. 4). This vision includes expanded efforts for disease modification in T1D that include β-cell supportive agents used in combination with immunomodulatory drugs along with complimentary improvements in diabetes technology and the idea of β-cell replacement.

Acknowledgments. Although I had the honor of delivering the Outstanding Scientific Achievement Award address in San Diego in June 2023 at the 83rd Annual Scientific Sessions of the American Diabetes Association, this award and this honor is shared with the current and past members of my laboratory. Since joining the faculty at the IU School of Medicine in 2008, I have had the opportunity to work with over 45 talented and brilliant students, fellows, technical staff, and senior scientists. I am grateful to each of these individuals for their contributions and their partnership. I want to highlight the individuals who contributed to the vignettes presented in this address: Xin Tong, Tatsu Kono, Hitoshi Iida, Emily Sims, Robert Bone, Staci Weaver, Farooq Syed, Chih-Chun Lee, Renato C.S. Branco, Madeline McLaughlin, Preethi Krishnan, Garrick Change, Wenting Wu, Divya Singhal, Jyoti Rana, Zunaira Chaudhry, and Renecia Watkins. As I prepared for the Outstanding Scientific Achievement Award talk, I had the opportunity to reflect quite a bit on my scientific and professional journey. It struck me that in many ways, a brand new assistant professor has many similarities to a nascent polypeptide, newly synthesized in the lumen of the ER. They are immature and prone to misfolding, and a good number simply do not make it. Those who do succeed have an optimal environment that supports their maturation and the benefit of many chaperones who guide them along the way, some traveling quite a distance alongside them. My career has been notable for such an environment and for a large number of mentors, collaborators, colleagues, and friends who have helped guide me. I am thankful each day for this village. A special thank you to Elaine Hylek, my first mentor at Massachusetts General Hospital, who taught me the very basics of how to ask a scientific question, and to Raghu Mirmira, my postdoctoral mentor at the University of Virginia and my early career mentor at IU, who taught me how to do basic science research and how to run a laboratory. Thank you to Linda DiMeglio for teaching me how to be a clinical researcher and to Bob Considine for his incredible partnership in running our P30 DK097512-funded Diabetes Research Center. Thank you to Wade Clapp, my department chair, for giving me the autonomy and resources to continue to grow and support diabetes research at IU. My gratitude and appreciation extend to a number of other local colleagues who have supported me along the way: Jerry Nadler, Peter Roach, Kieren Mather, Patrick Fueger, Emily Sims, Jamie Felton, Jason Spaeth, Hongxia Ren, Amelia Linnemann, Lim Kua, Heba Ismail, Teresa Mastracci, Jing Liu, Tatsu Kono, Farooq Syed, Chih-Chun Lee, and Janice Blum. In addition, I am grateful to my many external collaborators, including Bruce Verchere, Peter Arvan, Iris Lindbergh, Guido Sebastiani, Alberto Pugliese, Francesco Dotta, Mark Mamula, Decio Eizirik, Marjan Slak Rupnik, James Johnson, Richard Oram, Ele Ferrannini, Andrea Mari, Jay Sosenko, Steve Gitelman, Marcela Brissova, Huanmei Wu, Xiaowen Liu, Jennifer Van Eyk, Koen Raedschelders, Russell Dahl, and Mei-Ling Yang. Throughout my career, I have had the honor of being a member of many key diabetes networks and consortia, including the Human Islet Research Network (HIRN), the Integrated Islet Distribution Program (IIDP) with the Human Islet Phenotyping Program (HIPP) and the Human Islet Genotyping Initiative (HIGI), nPOD, the Type 1 Diabetes in Acute Pancreatitis Consortium (T1DAPC), and the Rare and Atypical Diabetes Network (Radiant). Although space does not permit me to name all the members of these groups who have touched my career, I want to especially thank Decio Eizirik, Kristin Abraham, Al Powers, and Mark Atkinson, who have served as science and career mentors. In addition, I must acknowledge and thank the clinical trial participants who are a critical component to all these efforts and the organ donors and their families for their selfless and precious gifts that help support the research of so many in our community. I also thank Emily Anderson-Baucum (IU School of Medicine) for her helpful advice and edits and Jill K. Gregory (medical illustrator) for creating the figures presented in this article. Finally, there are two remaining groups that I want to thank. The first is a special group of friends, Bruce Verchere, Richard Oram, Jessica Dunne, Teresa Mastracci, Emily Sims, Amelia Linnemann, Jamie Felton, and Jim Johnson, who have provided me with incredible support, many laughs, and several notable evenings of karaoke over the past 3 years. I am grateful for our connection. Last, but certainly not least, I want to thank my family, including my husband of 26 years, Chad Molina, our two daughters, Cameron and Olivia, and my mother Linda. Chad, thank you for taking this journey with me. I could not have done it without your love and support. Cameron and Olivia, I am incredibly proud of you both. I appreciate your patience when a grant or a manuscript submission or some perceived administrative emergency interrupted time that belonged to you, and I am so very happy you could be in San Diego with me to receive this award.

Funding. This work was supported by grants from the U.S. National Institutes of Health, the U.S. Department of Veterans Affairs, the Leona M. and Harry B. Helmsley Charitable Trust, and the JDRF and gifts and support from the Riley Children’s Foundation, the IU School of Medicine, the Sigma Beta Sorority, the Ball Brothers Foundation, the George and Frances Ball Foundation, the Luke Bracken Wiese Fund in Juvenile Diabetes, the Thomas Trust, and the Richardson Family.

Duality of Interest. C.E.-M. has served on advisory boards related to T1D research clinical trial initiatives for the following companies: Provention Bio, Avotres, DiogenyX, Isla Technologies, Neurodon, and MaiCell Technologies. C.E.-M. has in-kind research support from BMS and Nimbus Pharmaceuticals and investigator-initiated grants from Lilly Pharmaceuticals and Astellas Pharmaceuticals. C.E.-M. serves as president of the Immunology of Diabetes Society (IDS), co-executive director of nPOD, Scientific Advisor in TrialNet, and principal investigator of the National Institutes of Health Integrated Islet Distribution Program (IIDP). C.E.-M. has filed a provisional patent application for the use of PDIA1 as a diabetes biomarker and a patent for the use of extracellular vesicle RNA cargo as a diabetes biomarker. No other potential conflicts of interest relevant to this article were reported.