Functional analysis of islet cells in vitro, in situ, and in vivo

Abstract

The islet of Langerhans contains at least five types of endocrine cells producing distinct hormones. In response to nutrient or neuronal stimulation, islet endocrine cells release biochemicals including peptide hormones to regulate metabolism and to control glucose homeostasis. It is now recognized that malfunction of islet cells, notably insufficient insulin release of β-cells and hypersecretion of glucagon from α-cells, represents a causal event leading to hyperglycemia and frank diabetes, a disease that is increasing at an alarming rate to reach an epidemic level worldwide. Understanding the mechanisms regulating stimulus-secretion coupling and investigating how islet β-cells maintain a robust secretory activity are important topics in islet biology and diabetes research. To facilitate such studies, a number of biological systems and assay platforms have been developed for the functional analysis of islet cells. These technologies have enabled detailed analyses of individual islets at the cellular level, either in vitro, in situ, or in vivo.

1. Introduction

The physiological maintenance of euglycemia is achieved by the coordinated work of pancreatic islet endocrine cells. Impairment of islet β-cell function represents a causal event leading to diabetes, a disease that is on the rapid rise to reach epidemic level worldwide. In type 2 diabetes, as β-cells deteriorate, there is also a gradual loss of islet cell mass. In type 1 diabetes, insulin secretion essentially ceases when β-cells are destroyed by the autoimmunity. In both type 1 and type 2 diabetes, besides β-cells, other classes of islet endocrine cells including α-cells and δ-cells also exhibit functional abnormality which may further exacerbate glucose volatility.

Given the essential role of islet endocrine cells in hormone release and glucose control, there have been continuing efforts to develop reliable, sensitive and specific assays for the functional analysis of islet cells. The development, refinement and application of these assays have been instrumental in advancing our understanding of the biochemical and biological workings of stimulus-secretion coupling in endocrine cells, as well as in identifying molecular and cellular defects of compromised islet cells from diabetes. This review discusses these methods and their applications in primary islet cells of either rodent or human pancreas. Depending on the biological questions to be addressed, one needs to choose an appropriate biological preparation for the study. In general, there are three types of islet cell preparations that one may consider, and they can be broadly classified as in vitro, in situ, or in vivo preparation. In this concise review, I will focus on optical imaging methods that have been widely used for studying stimulus-secretion coupling. These include Ca²⁺ imaging and assays for tracking hormone secretion. Example applications of these functional assays are presented and discussed.

2. Functional analysis of islet cells in vitro

Isolated intact islets represent the most commonly used system for the in vitro assay of islet endocrine cells. Compared to the dispersed islet cells, the intact islet maintains the native cell-cell contacts and paracrine signaling among α-cells, β-cells and δ-cells. Mouse or rat islets can be routinely isolated following the established protocols [1, 2]. In recent years, human islets have become increasingly more available through sponsored islet isolation and distribution programs such as the Integrated Islet Distribution Program (IIDP), the European Consortium for Islet Transplantation (ECIT), and the IsletCore Program at the Alberta Diabetes Institute in Canada. In addition, human islets can be purchased from the Prodo Laboratories, a service company specializing in human islets isolation for research uses.

Pancreatic islets, once isolated, are normally cultured overnight to allow recovery from the harsh process of collagenase digestion and mechanical stress. Murine or rodent islets are cultured in RPMI 1640 medium with 11 mM glucose which has been reported to be optimal for the cell viability [3, 4]. Human islets are cultured in CMRL medium with 5.5 mM glucose. Isolated islets can be cultured up to a few weeks, though it is advised to analyze the islets in the following days as loss of hormone contents (degranulation) and change in islet cell function become more noticeable over prolonged cell culture [5, 6]. Compared to small islets, interior cells in the big islets tend to undergo hypoxia-induced necrosis during culture.

2.1. Biochemical assays of hormone release in whole islets

To assay hormone secretion activity at the level of whole islets, both static incubation and islet perifusion have been widely used. Static islet incubation serves as a facile approach for evaluating the accumulated insulin release over a period of time in response to insulin secretagogues, while islet perifusion is especially useful for studying the kinetics of insulin secretion with improved temporal resolution.

During static incubation, a small number of islets (~10-20) are incubated in a vial or a multi-well plate to which a secretagogue is added [7]. At the end of stimulation, the supernatant containing released hormones is removed for the assay. To facilitate liquid transfer and to avoid accidental carryover of islet cells in the supernatant, islets can be placed in a cell strainer with a mesh bottom [8]. The mesh bottom allows free diffusion of medium but retains islet cells. At the end of stimulation, the cell strainer containing islet cells is lifted from the assay medium and can be immersed in another solution for further stimulation or for the insulin content assay.

Compared to static incubation, dynamic perifusion of islets provides more accurate and dynamic information of regulated hormone release in vitro, and reveals biphasic insulin secretion of islets in response to elevated glucose [9, 10]. The procedure can be performed using an inexpensive, home-built system [11] or a commercial device (Biorep Technologies). In addition, developments of new assay formats have enabled measuring insulin release in single islets [12, 13].

2.2. Functional analysis of isolated islets at the cellular level by imaging

To analyze islet function at the cellular or even sub-cellular level, fluorescence microscopy provides a versatile and sensitive approach yielding high content information. The success of implementing these imaging assays in intact islets relies on the development of imaging probes or indicators that can provide specific readouts of biochemical processes. A number of imaging probes have been developed for tracking the spatial and temporal characteristics of hormone release and Ca²⁺ signaling.

2.2.1. Imaging hormone secretion

There are three major types of fluorescent sensors for imaging hormone secretion in islets, namely a pH-sensitive green fluorescent protein (GFP) termed pHluorin, a fluid phase fluorescent tracer Sulforhodamine B (SRB), and a fluorescent Zn²⁺ sensor ZIMIR [14]. pHluorin is a genetically encoded pH sensor [15] that can be targeted to the secretory granule when fused with a granule cargo such as phogrin or neuropeptide Y (NPY). Due to the acidic pH in the lumen of secretory vesicles, pHluorin remains weakly fluorescent. During granule fusion, alkalinization by the extracellular medium restores the fluorescence intensity of pHluorin. pHluorin is typically expressed in islet cells through adenovirus infection. In human islets, confocal imaging of NPY-pHluorin revealed preferential release sites and synchronized release events among neighboring beta cells [16].

SRB or other extracellular polar tracers have been applied to image exocytosis through activity dependent uptake [17]. Since the tracer is present in the bulk solution throughout imaging, it would be necessary to reject signals out of the focal plane in order to monitor the formation of an Ω-shaped profile during granule fusion. Two photon fluorescence microscopy offers high spatial resolution by restricting the excitation at the focal point, and two photon extracellular polar tracer imaging was developed to track the activity dependent dye uptake, which were interpreted as individual exocytotic events. An application of this two photon imaging assay in mouse islets suggested a polarized distribution of insulin release sites at the interface between the beta cell and vasculature [18].

The third type of imaging assay of insulin secretion exploits the fact that insulin granules contain a high concentration of Zn²⁺, which is co-released with insulin during granule fusion. A small synthetic fluorescent Zn²⁺ indicator for monitoring insulin release (ZIMIR) was developed for this purpose [19]. ZIMIR is an amphipathic molecule that anchors to the outer leaflet of the cell membrane when added to cells. It binds Zn²⁺ with an affinity of 0.45 μM and displays more than 70-fold fluorescence enhancement upon Zn²⁺ binding. In rodent islets, ZIMIR labeled both superficial and interior cell layers and revealed heterogenous glucose stimulated insulin secretion (GSIS) activity among islet β-cells, with small clusters of β-cells showing strong secretion, while the surrounding β-cells were inactive [19]. In human islets, ZIMIR imaging demonstrated gap junction coupling coordinated insulin secretion upon GLP-1 stimulation [20].

Thus far all three imaging assays have been predominantly used to study insulin release, though in principle they can be adapted for imaging the secretion of other hormones from different types of islet cells. Besides optical imaging, capacitance measurements by patch clamping allows highly sensitive recording of exocytosis in a single cell [21].

2.2.2. Imaging cytosolic Ca²⁺ activity ([Ca²⁺]_c) with small synthetic Ca²⁺ indicators

Since cytosolic Ca²⁺ activity ([Ca²⁺]_c) functions as an important messenger in stimulus-secretion coupling, and because numerous Ca²⁺ indicators are available for fluorescence microscopy, Ca²⁺ imaging has remained as a prime approach for studying how islet cells respond to nutrient secretagogues or agonists of cell surface receptors or ion channels. Two classes of Ca²⁺ sensors, namely small synthetic chemical indicators and genetically encoded Ca²⁺ indicators, have been widely used.

Small synthetic Ca²⁺ indicators mainly include the Fluo-family probes such as Fluo-3, Fluo-4 and Oregon Green 488 BAPTA-1 [22, 23]. Fura-2, a ratiometric Ca²⁺ sensor, yields more quantitative data on [Ca²⁺]_c but requires UV-light excitation, so its application is more restricted to dispersed islet cells. These synthetic Ca²⁺ sensors are typically loaded into cells as the AM ester in the presence of a mild detergent such as pluronic. In islets, AM ester loading is limited to the superficial cell layers, as dyes are trapped by the superficial cells once the AM ester is hydrolyzed by cellular esterases [19]. Time lapse Ca²⁺ imaging of islet cells by confocal microscopy in 3D has been instrumental in studying stimulus-secretion coupling [24], metabolic oscillation and Ca²⁺ oscillation [25], cell-cell coupling and synchronization [26], and islet β-cell heterogeneity [27], among many others.

2.2.3. Imaging [Ca²⁺]_c with genetically encoded Ca²⁺ indicators

Over the past decade, developments in genetically encoded Ca²⁺ indicators (GECIs) have greatly improved their sensitivity and dynamic range that even rival the Fluo-family of synthetic Ca²⁺ sensors [28, 29]. Transgenic mice expressing GCaMP6 [29] under the control of Cre recombinase have been developed and are commercially available (The Jackson Laboratory) [30]. Tissue- or cell-specific expression of Cre recombinase in turn switches on GCaMP6 in the corresponding cells through the Cre-lox system. Inducible expression can be achieved by fusing Cre with oestrogen receptor (CreERT2). Several mouse strains with islet cell specific expression of Cre have also been generated and are commercially available, including those expressing in mouse islet α-cells [31, 32], in β-cells [33], and in δ-cells [34, 35].

Compared to the small synthetic Ca²⁺ indicators, GECIs offers the major advantages of cell selectivity, and targetability to subcellular compartments or organelles when they are fused with the proper leader peptide sequences [36, 37]. Moreover, since most rodent islet β-cells are localized in the islet interior [38] and are inaccessible to the AM ester loading technique of synthetic dyes, GECIs offers the feasibility of tracking [Ca²⁺]_c in all cells throughout an islet. Ca²⁺ imaging of transgenic islets expressing GCaMPs has already revealed new pathways regulating Ca²⁺ signaling in islet cells. For instance, in mouse δ-cells, ghrelin was shown to activate [Ca²⁺]_c to potentiate glucose stimulated somatostatin secretion [39]; and glucose-induced [Ca²⁺]_c spikes did not correlate with membrane electrical activity but instead reflected Ca²⁺ release from the ER, suggesting the involvement of Ca²⁺ induced Ca²⁺ release in somatostatin secretion [40].

Besides using transgenic mice, GECIs can also be expressed in islet cells through adenoviral or lentiviral infection. However, similar to the AM ester delivery, viral infection is largely limited to cells at the outer layers. To facilitate introducing these probes throughout islet cells, isolated islets are dispersed into single cells by trypsin digestion before the cell infection [41]. Afterwards the infected cells are allowed to reaggregate to establish cell-cell contacts and to form pseudo-islets.

2.3. Reaggregated pseudo-islets for functional and molecular analysis

Recent years have witnessed an increasing use of pseudo-islets in the islet study. Several factors account for its increasing popularity. First, for the β-cell replacement therapy of diabetes, there have been continuing demands for high quality islet preparations of controlled size, robust GSIS activity, and hypoxia tolerance [42]. Second, molecular and functional analysis of a gene of interest can be carried out in vitro in pseudo-islets after infecting dispersed cells with siRNA or viral vectors, a process that can be achieved at high efficiency in dispersed cells prior to forming pseudo-islets. Third, there has been a recent surge of interests in studying islet cell heterogeneity [43, 44]. Islet β-cells were sorted into distinct subsets by FACS, and reaggregated into pseudoislets before functional analysis and comparison [45]. Last but not least, since isolated human islets do not always exhibit reliable functions for in vitro studies, formation of human pseudo-islets can restore their functional integrity and to prolong the duration of in vitro testing.

It has long been recognized that dispersed islet cells exhibit weak secretory activity or do not respond properly to secretagogues. Establishing cell-cell contacts by reaggregation is important for the dispersed islet cells to restore their normal secretory responses [46, 47]. A number of methods have been developed for forming pseudo-islets. Early studies established that dispersed islet cells spontaneously aggregated in culture [48], and that gentle mechanical mixing could facilitate the process [47]. The hanging drop method was later developed as a facile approach for forming pseudo-islets of uniform sizes, which were controlled simply by varying the number of cells in each hanging drop, ranging from a few hundreds of cells to over a thousand cells per pseudo-islet [49, 50]. It was reported that small islets out-perform large islets in terms of cellular insulin content and GSIS activity [51] [50]. Pseudo-islets have been shown to be comparable to intact islets in terms of cellular composition, architecture, and secretion [47, 48], while transplant studies suggested reduced immunogenicity when using pseudo-islets [52].

To facilitate cell aggregation, a centrifugal-forced-aggregation method was developed in which dispersed islet cells were spun down as small cell pellets in microwells to promote cell-cell contacts and pseudo-islet formation [42]. Centrifugation enables rapid islet assembly as it minimizes the time spent as a single cell suspension, efficiently generating uniform spherical islets with minimal loss of islet cell during the process. The resulting pseudo-islets showed much enhanced GSIS activity over that of unmodified native islets [42].

Native human islets display heterogeneity in size, cell number and composition, architecture, hormone content and secretion activity [53]. These variations may compound functional analysis and comparison of human islets in vitro. Pseudo-islets are expected to achieve a more uniform distribution of α-, β- and δ-cells and therefore to reduce islet heterogeneity and variations for in vitro testing of islet function. Moreover, pseudo-islets offer the flexibility of controlling cell composition simply by varying the ratio of sorted islet cells, providing a convenient platform to study heterotypic cell-cell interaction and paracrine signaling in vitro. In small pseudo-islets (~ 40 μm in diameter) containing less than 50 cells, glucose inhibition of glucagon secretion was examined by comparing pseudo-islets with a cell mix of α-cell/β-cell or with α-cell/δ-cell, allowing characterization of different aspects of regulated hormone secretion [54].

All the methods described above for the functional analysis of intact islets are applicable to pseudo-islets. The controllable and small size of pseudo-islets can facilitate probe delivery, cell transfection and imaging. Further, to perform molecular analysis of islet cells by genetic manipulations, more efficient gene delivery can be achieved in the dispersed islet cells prior to forming pseudo-islets. Similarly, gene inhibition or knockdown can be realized more effectively by delivery of siRNA or lentiviral shRNA to dispersed cells or small islets [55, 56], or by CRISPR-Cas9 approach to enable permanent genetic modification.

Besides hormone secretion and Ca²⁺ imaging, a number of additional optical assays are available for studying biochemical parameters important for stimulus-secretion coupling in both native and pseudo-islets. These include, among others, two photon metabolic imaging of NAD(P)H [57] and cAMP imaging [58]. Last but not least, optogenetics is a powerful tool for manipulating cell excitability with high spatial and temporal resolution, and has been increasingly utilized in studies involving islet cells [27, 37, 59, 60].

3. Functional analysis in situ

To analyze islet cells in the presence of surrounding exocrine tissues, a few biological preparations have been developed. These include perfused pancreas and pancreatic slice. The former preparation still contains an intact vasculature network for supplying nutrients and oxygen to islet cells, and hence present an experimental setting that mimics the native environment in vivo. However, analysis of islet function at the cellular level remains challenging in the intact pancreas, and pancreatic slice was developed as a compromise to provide access to islet cells and to allow imaging or electrophysiological recording of individual islet cells in situ.

3.1. Perfused pancreas

Pancreas perfusion has been used for many years for studying islet hormone secretion in situ. In this procedure, a surgery is performed on a freshly isolated intact pancreas to create an enclosed system with the celiac artery to be the only inlet to the pancreas and the hepatic portal vein to be the only outlet, with all other blood vessels ligated to prevent leakage [61, 62]. Oxygenated solution is then perfused through a pancreas, and secretagogues are introduced to stimulate hormone secretion. Compared to the perifusion of isolated islets, this methodology maintains islets in their native environment, avoids cell stress or damage due to islet isolation, and supplies ample oxygen to islet cells to maintain their physiological state and functionality. Pancreatic islets are richly vascularized, allowing their released hormones ready access to the circulation through the fenestrated capillaries. Even though pancreatic islets only constitute <2% of total pancreas mass, they maintain a capillary network that is five times denser than that of the exocrine pancreatic tissue and receive 10-15% of pancreatic blood flow [63]. Sufficient oxygen supply is critical for maintaining islet cell function, and hypoxia can compromise islet cell function and lead to necrosis.

Pancreas perfusion has been applied to a range of studies to analyze islet function and hormone secretion in both rats and mice, and to compare secretory activity of islet cells between healthy and diabetic state in response to different secretagogues, or to interventions by pharmacological agents [64]. Once established, the procedure allows detection of even small changes or impairments of secretory activity of islet cells. For example, deletion of a neutral amino acid transporter SNAT5 (Slc38s5 gene) in mouse resulted in a modest but significant reduction of the second phase glucagon secretion in response to alanine [65].

3.2. Pancreatic slice

A slice platform of mammalian pancreata was recently developed to circumvent the limitations of isolated islets, namely the cell stress from isolation and the loss of surrounding tissues and a portion of superficial islet cells and islet basement membranes [66, 67]. In this procedure, low melting point agarose is injected into the mouse pancreas through the bile duct to harden the tissue. Small pieces of injected tissue are then embedded in agarose and sliced with a vibratome to 100 μm – 200 μm thickness [68]. To slice human pancreas from organ donors, it is not necessary to harden the tissue since human pancreas, unlike mouse pancreas, is a solid organ with a defined structure that can be sliced directly. Once prepared, the pancreatic slice can be cultured for about a week during which functional analysis of both endocrine and exocrine cells can be conducted.

To perform Ca²⁺ imaging on pancreatic slices, small synthetic Ca²⁺ indicators can be loaded to cells along the surface. To image Ca²⁺ beyond the surface cell layer, transgenic mice expressing GCaMPs in the cells of interest would be a preferred model. In addition to optical imaging, prolonged patch-clamp recordings can be conducted in pancreatic slices to assess electrophysiological properties and secretion of both superficial and interior cells. A study applied patch clamping to interrogate a large number of mouse islet α-cells in freshly prepared pancreatic slices, and revealed a wide distribution in alpha-cell ion channel properties [69]. The pancreatic slice preparation could be particularly useful for studying islet endocrine cell function in which islet cell isolation remains challenging, such as pancreata of type 1 diabetes donors, or pancreata from certain diabetic mouse models. Traditional biochemical assays of cell secretion, including both endocrine hormones and exocrine amylase, are also applicable to pancreatic slices [68].

4. Functional analysis of islet cells in vivo

Owing to its desirable optical property, zebrafish has been used a model organism for studying vertebrate development for many years. Recently, Ca²⁺ imaging with GCaMP6 of islet β-cells in live zebrafish embryos has been reported [70, 71]. Functional analysis of islet cells in live mammals, especially at the cellular level or at the level of individual islets, has remained to be a challenging task due to the inaccessibility of the mammalian pancreas. Nonetheless, there have been continuing efforts to engineer new assaying platforms for studying islet biology in vivo where islet cells are supplied with physiological blood circulation, oxygenation, exocrine and nervous inputs.

Currently there are three major types of assays for studying islet cell function in live mammals: biochemical analysis of released islet hormone in the blood circulation; optical imaging via pancreas exteriorization or through an abdominal optical window; and optical imaging of implanted islets through the anterior chamber of the eye.

4.1. Blood sampling of released islet hormones

Biochemical assays of released islet hormones in systemic blood circulation is limited by its small dynamic range and poor temporal resolution. Due to issues related to hormone degradation, liver clearance, and dilution in the circulation, systemic blood sampling only provides a crude estimation of the secretory activity of islet endocrine cells and scarce dynamic information of hormone release. Portal vein blood sampling through an implanted catheter measures islet hormone release immediately downstream of pancreas before the released hormones enter liver, hence boosting the sensitivity of islet hormone detection [72]. The implanted catheter remained patent and functional for several weeks after implantation, allowing longitudinal monitoring of hormone release in vivo. This technique has been successfully applied to measure pulsatile insulin release during the fasting state in several mammals including dog [72], human [73], and rat [74].

4.2. Imaging islet cells in pancreas in vivo

To access pancreas for the optical imaging, surgical procedures have been developed to exteriorize a pancreas on a glass coverslip while the anesthetized mouse is placed on the stage of a microscope [75, 76]. Subsequent imaging by confocal microscopy or two photon excitation laser scanning microscopy locates islet cells in 3D in the intact pancreas while animals are still alive. This imaging platform or its variation has been applied to detect glucose dependent blood flow in the islet vasculature [77], to track the dynamics of the interaction between cytotoxic T lymphocytes and β-cells in a type 1 diabetes mouse model [78], to demonstrate the presence of “first responder” islets to an in vivo glycemic challenge[79], and to monitor the production of reactive oxygen species or [Ca²⁺]_c oscillation in the islet β-cell in vivo [80]. However, more often than not, pancreas exteriorization is a non-survival procedure so a mouse undergone the surgery can only be imaged once.

To enable longitudinal imaging of pancreas in vivo, a survival surgery procedure was developed by creating an optical imaging window at the mouse abdomen [81, 82]. An internal organ including spleen, kidney, pancreas or liver is then adhered to the optical window. The abdominal imaging window has been reported to image the endogenous tissue on the time scale of days to weeks, and to allow stable imaging of the same islets over time [80]. In addition to imaging endogenous pancreatic islets, the method could also be applied to image exogenous islets transplanted under the kidney capsule in mice [83]. As long as the animal viability, the stability of the window and the underneath tissue can be maintained, islets within the imaging depth ought to be imaged longitudinally using this format.

4.3. Imaging islet cells in the anterior chamber of the eye

The engraftment of exogenous islets in an ectopic location, the anterior chamber of the eye (ACE), represents a convenient model for noninvasive longitudinal imaging of islets in vivo [84, 85]. ACE is an immune privileged site with abundant autonomic nerves and blood vessels present in the iris. This provides an ideal environment to facilitate fast engraftment. Islets are transplanted into ACE via injection through the cornea, and are readily imaged through the cornea afterwards. The transplanted islets recruited blood vessels from the iris and became fully vascularized within a month, and they released insulin to maintain glucose homeostasis in the host mice that had been previously rendered diabetic by streptozotocin. To image Ca²⁺ activity of the implanted islets in vivo, synthetic small Ca²⁺ indicators are loaded to the implanted islets via perfusion of the ACE. Alternatively, islets expressing GCaMPs can be implanted. The ACE platform has enabled a number of repetitive and longitudinal Ca²⁺ imaging, and revealed that intraocular islets display significantly quicker Ca²⁺ responses to glucose stimulation than isolated islets in vitro [86]. The ACE technology has also been applied to study in vivo dynamics of autoimmune events at cellular levels relevant to diabetes, including immune cell infiltration, immune cell-β cell interaction and loss of β-cell mass and function [87-89]. Moreover, the ACE islet implantation has been extended to other species including primates [90], and human islets implanted into the athymic nude mouse were able to restore euglycemia of the recipient diabetic mice [91], demonstrating the feasibility of imaging human islet cells in an ectopic environment in vivo.

5. Summary and outlook

There have been continuing efforts to develop and optimize new technology platforms for the functional analysis of islet cells. A number of biological preparations and systems have been established for studying islet cells in vitro, in situ, and in vivo, providing us the flexibility in experimental design by choosing an appropriate assay system.

There are increasing appreciations of islet cell heterogeneity, at various levels ranging from individual cells within an islet, to inter-islet variations of the same animal, and to batch differences among islets isolated from different healthy donors. Understanding the genetic and metabolomic underpinnings of these variations would require sensitive analysis, reliable identification, and prompt separation of each subsets of cells. Thus far Ca²⁺ imaging has remained to be the prominent assay for studying islets at the cellular level, owing to the many imaging probes available. However, hormone secretion represents the most important process controlling metabolism and glucose homeostasis, and more evidences start to emerge suggesting that rise in [Ca²⁺]_c does not necessarily correlate with the secretory activity in either α-cell or β-cell [19, 92]. While several imaging assays have been developed for the cellular analysis of insulin secretion, we are still limited in our ability to track insulin release in vivo at the cellular or at the islet level, especially if we want to follow the process longitudinally. Imaging hormone secretion, including insulin, glucagon, and even somatostatin, would be a high priority task for future developments, especially in the setting of in vivo analysis. Investigating how subsets of islet cells vary in hormone release and contribute to the physiological maintenance of euglycemia, tracing the time course of their functional adaptation or impairment over the course of disease progression, and monitoring how defected cells respond to therapeutic interventions present exciting challenges and opportunities for developing the next generation of imaging assays in various biological systems.