Awardees 2017

PI:Joseph Ciolino, MD (Mass. Eye and Ear / Schepens Eye  Research Institute)
Abstract:A New Treatment for Diabetic Macular Edema

Approximately 28% of diabetics have ocular complications.  Diabetic macular edema (DME) is one of the leading causes of blindness in the industrialized world and is caused by leaking of blood vessels. The 10-year rate of developing macular edema is 20% for type I diabetics and 25.4% for type II diabetics using insulin.  Moreover, about half of patients with DME will lose 2 or more lines of vision within 2 years.
Studies have found that anti-inflammatory therapy with corticosteroids is superior to that of antiangiogenic therapy. Current therapy consists of repeated intravitreal corticosteroid injections or surgical implantation of corticosteroid-eluting retinal implants (e.g. dexamethasone-eluting implant known as Ozudex).  The invasive nature of both intravitreal injections and implants can lead to complications and limits patient enthusiasm for such repeated procedures. Moreover, patients also develop steroid-induced increased intraocular pressure, but the drug cannot be removed once injected. Therefore, there is a large unmet need for a topical, non-invasive treatment for DME that can easily be discontinued if needed. Contact lens (CL) drug delivery has the potential to achieve this goal; however, providing sustained drug release from contact lenses has historically proven challenging. Moreover, it is unclear if topical delivery can provide therapeutic levels of dexamethasone to the retina.

We have developed an innovative drug-eluting therapeutic CL (TCL) that incorporates a thin drug-polymer film within the periphery of a standard CL (Fig. 1). In contrast to other TCL designs, our design enabled the release of large drug quantities in a controlled manner over the course of few weeks, while allowing unimpeded vision through the lens, and was composed of materials that are all FDA-approved for use on the eye. We have demonstrated sustained release of therapeutic amounts of an antibiotic (ciprofloxacin), an antifungal agent (econazole), a glaucoma medication (latanoprost), and a corticosteroid (dexamethasone) for up to a month in bench-top studies. In rabbits, a dexamethasone-eluting TCL demonstrated safety and sustained drug delivery to the retina and choroid at levels that far exceeded those of hourly drops. While it our preliminary data suggest that the TCL provides therapeutic drug levels to the retina, further research is needed to determine if the lenses are indeed as efficacious as intravitreal injections.
The underlying hypothesis of this application is that dexamethasone-eluting TCL can safely and effectively treat macular edema as much as the current standard of care, intravitreal corticosteroid injections.

            Aim 1. Evaluate safety and drug flux of dexamethasone-eluting contact lenses in vivo. We will evaluate drug flux and biocompatibility in normal rabbit eyes.
            Aim 2. Evaluate efficacy in an animal model of macular edema. We will compare the efficacy of dexamethasone-eluting contact lenses in a rabbit model that has leakage of the retinal vessels induced by intravitreal injection of VEGF A.  We have preliminary data that demonstrates that intravitreal VEGF A results in statistically significantly greater fluorescein leakage after 2 days compared to baseline.  Using fluorescein angiography, pathology, and evaluation by a fluorotron (provides quantitative analysis of fluorescein leakage), we will compare the following treatment groups:
1) Dexamethasone-eluting contact lens
2) Intravitreal dexamethasone injection
3) Hourly dexamethasone 0.1% drops (limited to 8 hours a day for humanitarian reasons)
4) No treatment


This proposed approach is both feasible and valuable. The TCL can deliver high drug concentrations to the target tissue (retina) and the device has been shown to be safe and effective in non-human primates. The TCL addresses a major compliance challenge with the current approach that requires intravitreal injections or surgical implantation of drug delivery devices


PI: Jason, Flannik, Ph.D. (Broad Institute of Harvard at MIT)
Abstract:Experimental Characterization of SLC30A8 Variants for Protection Against Type 2 diabetes

Naturally occurring genetic variants that protect from disease may suggest novel therapeutic targets. We recently identified a series of loss-of-function variants in SLC30A8, which encodes the pancreatic islet zinc transporter ZnT8, collectively associated with a three-fold decrease in type 2 diabetes (T2D) risk. However, the molecular or cellular mechanism of the protective association remains unclear, in light of non-confirmatory animal and cellular models. In the proposed work, we hypothesize that a range of SLC30A8 variants – of intermediate effect between wild type and loss-of-function – could lend insight into the mechanism of action relating SLC30A8 perturbation to disease risk. Specifically, we aim to (a) identify additional missense SLC30A8 variants from large-scale exome sequencing studies; (b) adapt established HeLa cellular models and zinc transport assays to quantify the efficiency of ZnT8 zinc transport; (c) measure the effects of a series of (10-20) ZnT8 missense variants on zinc transport; and (d) assess the extent to which molecular alterations in zinc transport correlate with physiological effects on risk of T2D or related glycemic and cardiometabolic traits. Compared to previous studies to understand the connection between SLC30A8 and T2D, and most studies to dissect the biology behind genetic associations in general, this proposal is innovative in using genetic variation – assayed via both large-scale sequencing and large-scale molecular characterizations – to link physiological, molecular, and cellular mechanisms. We will extract variants and their measured phenotypic effects from the Type 2 Diabetes Knowledge Portal, which today includes 17,000 exome sequences with plans to expand to 55,000 sequences within a year. We will characterize these variants in HeLa cells, which have successfully been used to assess ZnT8 function in the past, via previously established optical fluorescence assays and a more quantitative label-free X-ray fluorescence approach (the XRpro® assay from Icagen, Inc.) We expect that variants of more severe molecular effect will be associated with stronger protection from T2D; indeed, preliminary analysis of missense SLC30A8 variants suggest that computationally predicted damaging mutations are also associated with T2D protection, albeit with lesser effects than full loss-of-function variants. If successful, this work will provide a series of molecular “handles” to interrogate SLC30A8 function with effects that can be physiologically assessed via genetic studies – overcoming the limitations of previous approaches to validate the relevance of a cellular assay or animal model to human pathophysiology. These handles in turn could be used to assess whether inhibition of ZnT8 may be an effective therapeutic strategy for T2D.


PI: Richard Hodin, M.D (Massachusetts General Hospital)
Abstract:Development of a Novel Oral Enzyme Therapy to Prevent and Treat T1D

The pathogenesis of Type 1 Diabetes (T1D) is now understood to involve the destruction of pancreatic Beta cells through an immune/inflammatory response that likely has its origins from the intestinal microflora. Previous studies have shown that endotoxin injection substantially worsens streptozotocin (STZ)-induced T1D, whereas TLR-4 KO mice are partially protected from developing T1D. Recent studies have demonstrated that translocation of gut bacteria to pancreatic lymph nodes trigger NOD2 inflammasome activity, contributing to STZ-mediated T1D. This new understanding about the pathogenesis of T1D has been reviewed by Josef Neu and he suggests that an aberrant intestinal microbiome, a leaky gut barrier, and an altered intestinal immune response provide a “perfect storm” for the onset of T1D. Importantly, children & adolescents with T1D have high endotoxin serum levels, particularly at disease onset and in genetically predisposed infants, T1D is preceded by a decrease in gut microbiome a-diversity and an increase in inflammatory bacterial species and metabolites. Furthermore, data from the Finnish Diabetic Nephropathy Study show that serum endotoxin contributes to nephropathy in T1D and recent work (in press) has demonstrated that T1D patients have low levels of fecal IAP. Over more than two decades, we and others have established intestinal alkaline phosphatase (IAP) as a potent endogenous anti-inflammatory factor. This enzyme plays a critical role in regard to promoting the gut mucosal barrier, as well as in protecting the host from the detrimental effects related to the gut microbiota. Accordingly, we believe that orally delivered IAP could ameliorate all of the proposed T1D disease mechanisms by: (1) Improving gut barrier function and preventing bacterial translocation, (2) diminishing endotoxemia (and TLR4 activation), and (3) by maintaining a healthy gut microbiome. Our preliminary data provide convincing evidence that IAP could be an effective preventive and/or therapeutic strategy against T1D and its complications. The goals of the proposed project are to perform pre-clinical mechanistic studies that will provide the foundation for moving forward with a clinical trial of IAP in individuals at high-risk for T1D.   


PI: Margaret  Lippincott MD (Massachusetts General Hospital)

Abstract: Kisspeptin as an Insulin Secretagogue

Gestational diabetes mellitus (GDM), characterized by new onset glucose intolerance during pregnancy, affects ~8% of all pregnancies and increases the risk for diabetes with the associated health consequences in both mother and offspring. In GDM, the pancreatic β-cell is unable to produce more insulin to compensate for pregnancy-induced insulin resistance leading to post-prandial hyperglycemia. Even post-prandial hyperglycemia in mothers without the full diagnosis of GDM results in adverse outcomes for mother and fetus. Understanding how GDM develops and discovering new ways to control post-prandial hyperglycemia are both critical to improving pregnancy outcomes.

This grant presents a new factor, kisspeptin, which may increase glucose-stimulated insulin secretion. Traditionally thought of as a hypothalamic peptide, during pregnancy, kisspeptin is produced by the placenta and levels in the peripheral blood rise 7,000 fold across the trimesters of pregnancy compared to non-pregnant women. A failure to increase kisspeptin levels across pregnancy,  might contribute to less glucose-stimulated insulin secretion, post-prandial hyperglycemia, and ultimately GDM. Recent work suggests that kisspeptin levels might be lower in women with GDM.

To explore the feasibility of using kisspeptin in women to stimulate insulin secretion with a particular interest in modeling the effect of kisspeptin in pregnancy, this project will:

1: To establish the effect of kisspeptin, at levels normally found in pregnancy, on insulin secretion in healthy, normal weight women in both the fasted and fed state.

2: To establish the effect of sex steroids on kisspeptin modulation of insulin secretion in healthy, normal weight women by examining women in different phases of the menstrual cycle.

Each aim is accomplished by healthy, normal weight women undergo two paired liquid mixed meal tolerance tests during either a kisspeptin infusion or a placebo infusion after an overnight fast. These paired studies will allow for the examination of primary outcomes: changes in Disposition Index or Matsuda Insulin Sensitivity Index in response to kisspeptin.  These studies have been appropriately powered (80% power) with 10 follicular phase, and 10 luteal phase women to examine both aims. The kisspeptin infusions will approximate kisspeptin levels seen in pregnancy based on prior published literature. By defining kisspeptin’s role in glucose homeostasis in pregnancy, this grant will refine our understanding of pregnancy physiology and may inform a potential pathomechanism of gestational diabetes mellitus. The long-term objective is to develop novel treatments for gestational diabetes mellitus.

Relevance to Public Health:
Identifying novel pathways that regulate insulin secretion can help inform the how diabetes develops and how to best treat diabetes. This project studies how kisspeptin, which is naturally produced during pregnancy, can increase insulin secretion. Our goal is to better understand, and perhaps even treat, gestational diabetes which can harm both mother and baby for a lifetime.


PI: Roeland Middelbeek, (Jan-Willem) J.W., MD, Msc.
Abstract:Mechanisms Mediating the Beneficial Effects of Exercise: Novel Circulating Factors Regulating Metabolism in Human Subjects with Type 2 Diabetes

Physical exercise improves glucose homeostasis in people with type 2 diabetes (T2DM). We and others have found that factors, secreted after exercise into the circulation from skeletal muscle, adipose tissue and other tissues, may regulate glucose metabolism. It is critical to translate these animal-based findings to humans, by identifying novel exercise-stimulated factors in subjects with T2DM, and to understand their function in regulating cell metabolism. I recently spearheaded a project that generated exciting preliminary data revealing the response of 1,310 plasma proteins to an acute exercise bout in healthy lean subjects. We discovered that exercise led to highly-significant, time-specific increases in the plasma concentration of 13% of all the measured circulating proteins, of which the vast majority are novel exercise-regulated proteins. Given the importance of exercise in the treatment of T2DM, it is vital to determine the exercise-induced circulating factor response in T2DM subjects. It will next be essential to determine if these proteins can regulate glucose and fatty acid metabolism. Aims: 1) To determine the effects of a single bout of moderate intensity exercise on plasma protein concentrations of 8 subjects with T2DM.  2) To use bioinformatics to identify exercise-regulated proteins that are differentially regulated by exercise in subjects with T2DM compared to lean. 3) To select the most highly regulated proteins and determine their effects glucose and fatty acid metabolism in vitro. This study will provide critical preliminary data for an RO1 application that will focus on identifying circulating factors as mechanisms mediating the beneficial effects of exercise in subjects with and without T2DM.


PI:Christian  Rask-Madsen, MD, Ph.D. (Joslin Diabetes Center)
Abstract:Regulation of Angiogenesis by the Transcriptional Coregulator CITED2 in Obesity and Type 2 Diabetes

Patients with diabetes have an impaired angiogenic response to hypoxia resulting in worse perfusion of tissue with chronic hypoxia and poor recovery after ischemic stroke and myocardial infarction.  We have screened for insulin-regulated genes in vascular endothelial cells among 22 genes previously shown to be both regulated by the transcription factor FoxO and differentially regulated in pro-angiogenic endothelial cells.  Among these genes, insulin robustly downregulated the transcriptional corepressor CITED2 (CBP/p300 interacting transactivator 2) in endothelial cells.  We also showed that CITED2 is upregulated in arterial tissue from patients with type 2 diabetes, presumably because of insulin resistance in vascular tissue, where insulin fails to suppress CITED2 despite hyperinsulinemia.  These are particularly interesting findings because CITED2 inhibits transactivation of hypoxia-inducible factor-α (HIF-1α), a key regulator of proangiogenic genes.  Our preliminary data show that CITED2 overexpression or knockdown in endothelial cell culture inhibits or promotes, respectively, tube formation on Matrigel.  We therefore propose that upregulation of CITED2 in type 2 diabetes and other insulin resistant states can in part explain impaired angiogenesis.  In Aim 1, we will demonstrate, using endothelial cell culture, how CITED2 activates HIF responsive promoters through HIF-1α or HIF-2α, regulates HIF target genes, including VEGF and matrix metalloproteases, and influence invasion in extracellular matrix, tube formation and proliferation.  In Aim 2, we will determine whether inducible and tissue-specific knockout of CITED2 in endothelial cells can improve hindlimb angiogenesis in mice with insulin resistance due to diet-induced obesity.  Although development of antiangiogenic drugs has shown great progress, there are few targets like CITED2 which, when inhibited, promote angiogenesis.  The expected outcome of this research is to provide evidence for whether downregulation of CITED2 in endothelial cells can improve angiogenesis in insulin resistance as well as mechanistic insight into this phenomenon.


PI: Vetere,  Amedeo  PhD (Broad Institute)
Dissecting the Transcriptional Landscape of the Proliferating Adults Human Beta Cells by single-cell RNA sequencing

Diabetes mellitus is a leading cause of morbidity and mortality worldwide, predicted to affect over 500 million people worldwide by 2030. Loss of beta-cell number and function underlies much of the pathology of diabetes. Therefore, in both T1D and T2D, ways to preserve or expand pancreatic beta-cell mass could be an effective therapeutic approach. Physiologically, beta-cell mass is maintained at optimal levels through slow turnover. In particular, in humans it has been shown that beta-cell mass expands several fold from birth and during about the first three years of childhood to stop gradually in adulthood. In the past few years, attempts have been made to develop high-throughput screens (HTS) to identify small-molecule inducers of pancreatic beta-cell expansion. Among the most recent results is the identification of the small molecule 5-iodotubercidin (5-IT). 5-IT has been shown to strongly and selectively increase human beta-cell proliferation in vitro and in vivo. Similar to harmine, it inhibits DYRK1A to promote proliferation. Interestingly, the percentage of proliferating beta cells never goes above 4-5%, even with compound treatment. Although physiologically relevant to maintain glucose homeostasis, the fact that only a subset of beta cells proliferates underlines the existence of an intrinsic heterogeneity among beta cells and opens the question of what is the transcriptional profile of these proliferating beta cells.

In the present proposal, I plan to use microfluidic approaches for single-cell transcriptome profiling to understand the complexity and heterogeneity of proliferating human beta cells after treatment with 5-IT. Microfluidics-based single-cell transcriptome profiling can discriminate with great detail the cell-cycle state of a specific cell by comparing the global pattern of gene expression with a set of genes known to be enriched in each of the phases of cell cycles. Pancreatic islets cells are particularly well suited for single-cell transcriptome analysis, because each of them express high levels of cell-specific hormones, allowing for fine association of a transcriptional profile to specific cell types. Since adult human beta cells are primarily locked in the G1 phase of the cell cycle, the availability of cell-specific markers, along with the possibility to identify the proliferating status of each cell, makes microfluidics-based single-cell transcriptome profiling a very powerful tool to 1) identify beta cells that respond to 5-IT by entering the cell cycle, and 2) understand transcriptional specificity of proliferating beta cells.

BADERC Home DERC home page