Published in Volume
123, Issue 10
(October 1, 2013)J Clin Invest.
Copyright © 2013, American Society for Clinical Investigation
Zinc, insulin, and the liver: a ménage à trois
1Chemistry of Life Processes Institute, Departments of Chemistry and Molecular Biosciences, Northwestern University, Evanston, Illinois, USA.
2Department of Biochemistry, University of Wisconsin-Madison, Madison, Wisconsin, USA.
3Department of Molecular Biosciences, Northwestern University, Evanston, Illinois, USA.
First published September 24, 2013
Insulin and Zn2+ enjoy a multivalent relationship. Zn2+ binds insulin in pancreatic β cells to form crystalline aggregates in dense core vesicles (DCVs), which are released in response to physiological signals such as increased blood glucose. This transition metal is an essential cofactor in insulin-degrading enzyme and several key Zn2+ finger transcription factors that are required for β cell development and insulin gene expression. Studies are increasingly revealing that fluctuations in Zn2+ concentration can mediate signaling events, including dynamic roles that extend beyond that of a static structural or catalytic cofactor. In this issue of the JCI, Tamaki et al. propose an additional function for Zn2+ in relation to insulin: regulation of insulin clearance from the bloodstream.
Zinc is an abundant and essential element that plays a number of regulatory roles in biology. Studies in model organisms indicate that zinc receptor proteins control complex networks of genes in Zn2+-responsive manners. Moreover, complex developmental events are controlled by dynamic fluctuations of billions of zinc ions between intracellular compartments and extracellular sites (1, 2). Secretory compartments enriched in Zn2+ are found in a number of cell types, including hippocampal and olfactory neurons, oocytes, and pancreatic β cells. Specific stimulation of these cells leads to Zn2+ exocytosis; however, neither the physiological nor the biochemical roles of the released Zn2+ are yet clear in these systems. In this issue of the JCI, Tamaki et al. describe striking connections among zinc compartmentalization, exocytosis, and insulin uptake by the liver (3).
A link between zinc transport and diabetes
The ZnT8 transporter (encoded by solute carrier family 30 member 8 gene; SLC30A8) is located on dense core vesicles (DCVs) in β cells and loads Zn2+ into these secretory compartments, where it binds with and stabilizes a hexameric form of insulin (4). ZnT8 is an autoantigen in type 1 diabetes (5), and a W325R SLC30A8 polymorphism is associated with T2D (6). Slc30a8 deletion in mice decreases DCV Zn2+ content (4, 7–9). A role for decreased Zn2+ in DCV functional defects is not obvious, because some species (e.g., guinea pig) produce insulin molecules that do not bind to Zn2+ but still maintain normal insulin secretion.
Tamaki et al. address contradictory findings concerning the effect of Slc30a8 deletion on insulin secretion in mice (3). Using β cell–specific deletion of Slc30a8, they report an insulin hypersecretion phenotype and suggest that functional ZnT8 facilitates autocrine and paracrine roles for the Zn2+ burst produced by β cells upon glucose stimulation. In the WT animals, the released Zn2+ bolus inhibited further insulin release, while loss of Slc30a8 resulted in sustained insulin secretion. Intriguingly, despite increased insulin secretion, peripheral insulin levels were lower and the C-peptide/insulin ratio was increased in KO mice. These data indicate that lower insulin levels might be due to increased insulin clearance from the bloodstream. Studies of proinsulin/insulin ratio and rates of insulin clearance in humans with the SLC30A8 W325R polymorphism were also consistent with this idea.
Zinc inhibition of hepatic insulin uptake
Quantitatively, the liver and the kidneys are the most important sites of insulin uptake and degradation. The liver is the first organ exposed to newly secreted insulin and can clear much of this insulin in a single pass. The authors investigated hepatic insulin clearance in mice with pancreas perfusions and pancreas-liver perfusions while measuring insulin levels in the portal vein and the inferior vena cava. The difference between the two perfusions provided a measure of hepatic extraction. Their findings indicated that Slc30a8-KO mouse livers had enhanced single-pass insulin clearance (3). A puff of Zn2+ accompanies every burst of insulin secretion (7). The authors hypothesized that Zn2+ inhibits insulin secretion and hepatic insulin uptake, and, therefore, loss of the ZnT8 transporter should relieve the inhibition of insulin secretion while simultaneously increasing hepatic insulin uptake. In support of this hypothesis, direct Zn2+ injection into the portal vein of mice or incubation with a hepatocyte cell line directly inhibited insulin uptake. In agreement with previous studies, deletion of Slc30a8 resulted in dramatic reduction of insulin crystals. Despite abnormal DCVs, more insulin was secreted, but was balanced by increased hepatic clearance. Conversely, studies in isolated islets and perfused pancreas showed that Zn2+ inhibits insulin secretion (3).
It has been proposed that β cell–derived Zn2+ acts on α cells to suppress the glucagon secretion that accompanies insulin secretion. However, Tamaki et al. did not observe any difference in glucagon secretion in the Slc30a8-KO mice. Together with prior studies (8), these data suggest that the Zn2+ cosecreted with insulin is not responsible for the suppressive effect of insulin secretion on glucagon secretion.
There is a lack of consensus on the effect of Slc30a8 deletion on insulin secretion and circulation in mice (Table 1). In all reports, the consequence of Slc30a8 deletion appears to be relatively small. Studies using different mouse colonies with whole-body Slc30a8 deletion all agreed that Slc30a8 deletion does not affect glucose homeostasis in mice fed a normal chow diet. The studies disagreed on the role of Slc30a8 on proinsulin processing, glucose-stimulated insulin secretion (GSIS), and glucose tolerance. Nicolson et al. showed that while their male KO mice process proinsulin normally, they are glucose intolerant, secrete significantly less insulin during an intraperitoneal glucose tolerance test, and have increased fasting glucose levels (9). Interestingly, GSIS was significantly elevated in isolated islets. Conversely, Pound et al. found that their Slc30a8-KO mice have normal glucose tolerance, but significantly reduced fasting plasma insulin levels, along with a 33% reduction in GSIS (10, 11). The Slc30a8-KO mice studied by Lemaire et al. lacked any significant phenotype when on normal chow (12). Another study reported that β cell–specific deletion of Slc30a8 leads to decreased proinsulin processing, significantly reduced first-phase insulin secretion, and glucose intolerance during an oral glucose tolerance test (13).
Phenotypes observed in previous Slc30a8-KO studies
Interestingly, there are fewer phenotypic differences between colonies of mice subjected to prolonged high-fat diet (HFD) feeding. The studies agreed that Slc30a8-KO mice fed HFD display greater weight gain, fasting hyperglycemia, fasting hyperinsulinemia, and glucose intolerance (9, 12, 14). The role of Slc30a8 deletion on insulin sensitivity is still debated. Nicolson et al. (9) and Hardy et al. (14) reported decreased insulin sensitivity in their Slc30a8-KO colonies after prolonged HFD feeding; however, Lemaire et al. found insulin sensitivity to be unchanged (12).
Differences in genetic background may explain phenotypic differences. Pound et al. found that Slc30a8 deletion in two genetic backgrounds renders different phenotypes. They also reported sex-specific effects on resulting Slc30a8 deletion phenotypes (10). Finally, Slc30a8 deletion in mouse α cells did not produce a visible phenotype (13).
Linking insulin, zinc, and the liver
Deletion of PCSK1, a proprotein convertase that processes proinsulin, also blocks mature insulin formation and produces β cell DCVs that are essentially devoid of aggregated insulin. The mice have hyperproinsulinemia, but apparently normal glucose tolerance (15). Thus, disruption of DCV maturation does not always lead to impaired regulated secretion. Like SLC30A8, PCSK1 has emerged as a genetic factor of T2D. Deficiency of these genes appears tolerable in animals with a normal demand for insulin secretion, but might produce a bottleneck when challenged by the increased demand posed by insulin resistance.
How might Zn2+ influence hepatic extraction of insulin? Tamaki et al. showed that Zn2+ levels did not affect the activity of the insulin-degrading enzyme. This observation is consistent with the effect of Zn2+ being restricted to insulin internalization. Blockade of insulin internalization was enhanced by pyrithione, a Zn2+ ionophore. This suggests that internalized Zn2+ is responsible for the inhibitory effect; however, the effect was independent of the Zn2+ transporter ZIP-14.
Tamaki et al. make the case that pulsatile release of Zn2+ inhibits clathrin-dependent insulin endocytosis via a complex with the insulin receptor (IR). Two physiochemical features of Zn2+ binding to proteins may provide an additional and direct antagonistic effect on IR-insulin interaction. First, Zn2+ is known to stabilize the hexameric form of insulin in plasma, which may not bind as tightly to the IR. Second, Zn2+ may be a competitive inhibitor at the insulin-binding site of IR, acting as an antagonist of insulin binding and/or internalization. In many Zn2+-dependent proteins, Zn2+ forms between three and five tight coordinate covalent bonds to side chains such as His, Glu, Asp, and Cys. Inspection of a recent IR structure (16) revealed a cluster of Zn2+-binding histidine and glutamate residues directly adjacent to the insulin-docking site (Figure 1). One of these, H710, was recently shown to be critical in the interaction with insulin (16). We note that this histidine residue is within hydrogen bonding distance to E706, both of which are in the carboxyterminal α chain (αCT) helix of the insulin-binding pocket. With minimal conformational changes, these residues, along with E97 and/or D707 or a water molecule, could form a tetrahedral coordination environment favored for Zn2+ binding (Figure 1). This raises the possibility that Zn2+ influences insulin clearance in part through direct competition with insulin for IR binding. Intriguingly, insulin monomers contain a potential Zn2+ binding site. H10 of the insulin molecule is critical for Zn2+ binding the hexameric form, but is absent in insulin variants that do not bind to Zn2+ (17). The adjacent E13 is in proximity to coordinate to the same Zn2+ ion. It is unknown whether Zn2+ occupancy at these putative sites inhibits or perhaps stabilizes the IR-insulin interaction, as is the case for other hormone-receptor interactions (1).
Putative Zn2+-binding residues at the IR-insulin interface.
Model of a putative Zn2+-binding site in the insulin-binding site of the IR. The protein coordinates are from the IR-insulin crystal structure (PDB-ID 3W11). Insulin — chains InsA (gold) and InsB (black) are shown — engages with the IR at the αCT (purple). The IR core particle β strands are also shown (light cyan). A Zn2+ ion (transparent sphere) has been modeled in the structure based on the environment and known Zn2+-coordinating abilities of the highlighted residues. IR residues (green sticks) H710 and E706 (from the αCT) and E97 form a putative Zn2+-binding site. A water molecule or InsB side chains can complete the tetracoordinated Zn2+ site after helix rotation/displacement (not shown). Intriguingly, E706 forms a H-bond to a backbone amide on InsB: this docking of the hormone with its receptor could be disrupted by binding of the zinc ion.
Additional features of Zn2+-histidine coordination chemistry are relevant to insulin biology. The histidine-rich domain of ZnT8 is essential for Zn2+ binding. Through binding to two critical histidine residues, Zn2+ activates the KATP channel and hyperpolarizes β cells, which leads to inhibition of insulin secretion (18). This potentially explains how reduced Zn2+ uptake by β cells could promote increased insulin secretion.
Answering the questions evoked by these findings will require interdisciplinary teams that can pair physical and imaging methods with receptor physiology in model systems. Can Zn2+ simultaneously bind histidine and glutamate side chains in the IR and insulin to stabilize the complex and/or to prevent internalization? Are the relative Zn2+ affinities of the receptor, hormone, and heterodimeric complex compatible with the transient concentration gradient produced by pulsed Zn2+ release into the portal vein? Does Zn2+ only block insulin internalization, or does it also block insulin signaling? Answers to this question might help distinguish binding and signaling events. Are there other receptors that bind Zn2+ and are affected by the insulin secretion–dependent Zn2+ burst? How much Zn2+ escapes hepatic clearance and albumin binding to exert actions on extrahepatic tissues?
Although fasting insulin is used as a surrogate measure of insulin resistance, it has a stronger correlation with insulin clearance, which is highly heritable (19, 20). Future human genetic studies will be important for determining whether genetic variation in SLC30A8 contributes to T2D entirely through its effect on insulin clearance and how much of the heritability of insulin clearance is due to SLC30A8.
These studies illustrate how in-depth phenotyping, which requires model organisms, can take clues from human genetics and provide mechanistic explanations for relationships between genetic variation and human disease. Results from these studies can now be used to study subphenotypes associated with diabetes susceptibility. In this case, it may motivate study of the relationship among inorganic physiology (such as the Zn2+ fluxes described here), genetic variation at the SLC30A8 locus, and insulin clearance. Most importantly, these deeper phenotypes should be present in nondiabetics, and thus can be studied independently of the disease.
Kim AM, et al. Zinc sparks are triggered by fertilization and facilitate cell cycle resumption in mammalian eggs. ACS Chem Biol. 2011;6(7):716–723.
Outten CE, O’Halloran TV. Femtomolar sensitivity of metalloregulatory proteins controlling zinc homeostasis. Science. 2001;292(5526):2488–2492.
Tamaki M, et al. The diabetes-susceptible gene SLC30A8/ZnT8 regulates hepatic insulin clearance.
J Clin Invest. 2013;123(10):4513–4524.
Chimienti F, Devergnas S, Favier A, Seve M. Identification and cloning of a β-cell-specific zinc transporter, ZnT-8, localized into insulin secretory granules. Diabetes. 2004;53(9):2330–2337.
Wenzlau JM, et al. The cation efflux transporter ZnT8 (Slc30A8) is a major autoantigen in human type 1 diabetes. Proc Natl Acad Sci U S A. 2007;104(43):17040–17045.
Sladek R, et al. A genome-wide association study identifies novel risk loci for type 2 diabetes. Nature. 2007;445(7130):881–885.
Qian WJ, Gee KR, Kennedy RT. Imaging of Zn2+ release from pancreatic β-cells at the level of single exocytotic events.
Anal Chem. 2003;75(14):3468–3475.
Cheng-Xue R, et al. Tolbutamide controls glucagon release from mouse islets differently than glucose: involvement of K(ATP) channels from both α-cells and δ-cells. Diabetes. 2013;62(5):1612–1622.
Nicolson TJ, et al. Insulin storage and glucose homeostasis in mice null for the granule zinc transporter ZnT8 and studies of the type 2 diabetes-associated variants. Diabetes. 2009;58(9):2070–2083.
Pound LD, et al. The physiological effects of deleting the mouse SLC30A8 gene encoding zinc transporter-8 are influenced by gender and genetic background. PLoS One. 2012;7(7):e40972.
Pound LD, et al. Deletion of the mouse Slc30a8 gene encoding zinc transporter-8 results in impaired insulin secretion. Biochem J. 2009;421(3):371–376.
Lemaire K, et al. Insulin crystallization depends on zinc transporter ZnT8 expression, but is not required for normal glucose homeostasis in mice. Proc Natl Acad Sci U S A. 2009;106(35):14872–14877.
Wijesekara N, et al. Beta cell-specific Znt8 deletion in mice causes marked defects in insulin processing, crystallisation and secretion. Diabetologia. 2010;53(8):1656–1668.
Hardy AB, et al. Effects of high-fat diet feeding on Znt8-null mice: differences between β-cell and global knockout of Znt8. Am J Physiol Endocrinol Metab. 2012;302(9):E1084–E1096.
Zhu X, Orci L, Carroll R, Norrbom C, Ravazzola M, Steiner DF. Severe block in processing of proinsulin to insulin accompanied by elevation of des-64,65 proinsulin intermediates in islets of mice lacking prohormone convertase 1/3. Proc Natl Acad Sci U S A. 2002;99(16):10299–10304.
Menting JG, et al. How insulin engages its primary binding site on the insulin receptor. Nature. 2013;493(7431):241–245.
Beintema JJ, Campagne RN. Molecular evolution of rodent insulins. Mol Biol Evol. 1987;4(1):10–18.
Bancila V, et al. Two SUR1-specific histidine residues mandatory for zinc-induced activation of the rat KATP channel. J Biol Chem. 2005;280(10):8793–8799.
Goodarzi MO, et al. Systematic evaluation of validated type 2 diabetes and glycaemic trait loci for association with insulin clearance. Diabetologia. 2013;56(6):1282–1290.
Goodarzi MO, Cui J, Chen YD, Hsueh WA, Guo X, Rotter JI. Fasting insulin reflects heterogeneous physiological processes: role of insulin clearance. Am J Physiol Endocrinol Metab. 2011;301(2):E402–E408.