| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Departments of Medicine (D.T.O., P.E.C., C.S., R.M.S., F.R., J.S., S.K.M., M.M., G.W., L.T., K.L.H., R.R.H.) and Pharmacology (D.T.O.) and Center for Molecular Genetics (D.T.O.), University of California, Veterans Affairs San Diego Healthcare System (D.T.O., P.E.C., C.S., R.M.S., F.R., J.S., S.D.F., S.K.M., M.M., K.L.H., R.R.H.), San Diego, California 92161; and Faculty of Medicine (C.G.-Y., V.S.-M.), Virgen Macarena University Hospital, Seville, Spain
Address all correspondence and requests for reprints to: Dr. Daniel T. O’Connor, Department of Medicine (9111H), University of California School of Medicine and Veterans Affairs Medical Center, 9500 Gilman Drive, La Jolla, CA 92093-0838. E-mail: doconnor{at}ucsd.edu.
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Methods: We studied brachial arterial infusion of pancreastatin (CHGA273–301-amide at 
200 nM) on forearm metabolism of glucose, free fatty acids, and amino acids. Plasma pancreastatin was measured in obesity or type 2 diabetes. Systematic discovery of amino acid variation was performed, and the potency of one variant in the active carboxyl terminus (Gly297Ser) was tested.
Results: Pancreastatin decreased glucose uptake by approximately 48–50%; the lack of change in forearm plasma flow indicated a metabolic, rather than hemodynamic, mechanism. A control CHGA peptide (catestatin, CHGA352–372) did not affect glucose. Insulin increased glucose uptake, but pancreastatin did not antagonize this action. Pancreastatin increased spillover of free fatty acids by about 4.5- to 6.4-fold, but not spillover of amino acids. Insulin diminished spillover of both free fatty acids and amino acids, but these actions were not reversed by pancreastatin. Plasma pancreastatin was elevated approximately 3.7-fold in diabetes, but was unchanged during weight loss. Proteolytic cleavage sites for pancreastatin in vivo were documented by matrix-assisted laser desorption ionization/time of flight mass spectrometry. Three pancreastatin variants were discovered: Arg253Trp, Ala256Gly, and Gly297Ser. The Gly297Ser variant had unexpectedly increased potency to inhibit glucose uptake.
Conclusions: The dysglycemic peptide pancreastatin is specifically and potently active in humans on multiple facets of intermediary metabolism, although it did not antagonize insulin. Pancreastatin is elevated in diabetes, and the variant Gly297Ser had increased potency to inhibit glucose uptake. The importance of human pancreastatin in vivo as well as its natural variants is established.
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
-amidating monooxygenase (PAM) for activation (6, 7, 8, 9). Chromogranin A is overexpressed in human diseases such as neuroendocrine neoplasia (1, 10), but the potential role of the pancreastatin fragment in glycemic physiology or pathophysiology in humans in vivo has not been directly explored.
We therefore set out to document whether pancreastatin affects glucose metabolism in humans in vivo, cannulating the brachial arterial circulation in situ to detect effects of pancreastatin infused locally in submicromolar doses. Our results suggest that pancreastatin is highly and specifically active on both glucose and free fatty acid metabolism in humans in vivo, but not on amino acid metabolism. The plasma pancreastatin concentration was altered in disease (type 2 diabetes), but it did not respond to therapeutic intervention. We also detected naturally occurring amino acid sequence variation in pancreastatin and documented an unexpected increase in potency for the human variant Gly297Ser.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
We selected healthy, disease- and medication-free, young (age, <50 yr) males for the in vivo forearm perfusion (brachial artery cannulation) study. All were white (European ancestry) and fasted from all foods or beverages (except water) for at least 12 h before the study, which was conducted in the morning. All subjects gave written informed consent to the protocol, which was approved by the local institutional review board.
Peptides
Peptides were synthesized by the solid phase method and purified by reverse phase HPLC to at least 95% homogeneity. The identity and purity of each peptide were verified by analytical repeat HPLC and matrix-assisted laser desorption ionization (MALDI) or electrospray mass spectrometry (11). The peptides synthesized were wild-type human pancreastatin (CHGA273–301-amide; PEGKGEQEHSQQKEEEEEMAVVPQGLFRG-amide), human pancreastatin variant Gly397Ser (PEGKGEQEHSQQKEEEEEMAVVPQSLFRG-amide), and (as a control) wild-type human catestatin (CHGA352–372; SSMKLSFRARGYGFRGPGPQL). Underlined amino acids indicate varient positions (Gly297Ser). Human insulin was obtained from Novo Nordisk Pharmaceuticals, Inc. (Princeton, NJ).
Before infusion, peptides were dissolved in normal saline with 1.5% human serum albumin (Bayer Corp., Elkhart, IN), then filter-sterilized (0.22 µm pore size; Millix-GV filter, Millipore Corp., Bedford, MA). The infusate concentrations were: human pancreastatin, 20 µM; human catestatin, 100 µM; and human insulin, 0.1 mU/kg·min.
Forearm procedures
The study was performed in a temperature-controlled room at 25 C. With the subject in the supine position, an 18-gauge plastic arterial catheter (Arrow International, Reading, PA) was placed in the left brachial artery after local (skin and sc tissue) anesthesia with 2% lidocaine. A 20- to 22-gauge catheter was also placed in a deep antecubital vein in the ipsilateral arm for sampling of skeletal muscle venous drainage of the forearm.
Forearm blood flow (in milliliters per minute per 100 g forearm tissue) was measured by Hg-in-SILASTIC brand (Dow Corning, Inc., Midland, MI) strain gauge plethysmography, with an EC6 plethysmograph device and software (NIVP) from Hokanson Instruments (D. E. Hokanson, Inc., Bellevue, WA) (12, 13). The system involves sequential operation of two sets of software-driven, pneumatically inflatable, arm cuffs. First, a distal wrist cuff inflates to 200 mm Hg, to temporarily remove the anastomotic hand circulation from the forearm vascular bed; 1 sec later, an upper arm venous occlusion cuff inflates to 40 mm Hg to achieve maximal venous capacitance without affecting arterial inflow. Expansion of the arm is detected by an Hg-in-SILASTIC strain gauge, and the rate of arm expansion (in milliliters per minute per 100 g forearm, or percentage per minute) is equivalent to arterial blood inflow over the initial four or five cardiac cycles. This process can be repeated every 15 sec for frequent forearm blood flow measurements.
We evaluated the stability of such measurements of forearm blood flow over time (Table 1
). When forearm blood flow was measured repeatedly (30 times) in this fashion, the intraindividual coefficient of variation in 18 subjects during vehicle infusion into the brachial artery of the ipsilateral (same) arm was 16.4 ± 1.58% (range, 8.7–32.2%).
|
200 nM); 3) catestatin, 100 µM (target arterial concentration, 1 µM); 4) insulin, 0.1 mU/kg·min (target arterial concentration, 70 µU/ml); and 5) insulin plus pancreastatin (same concentrations as above). Target arterial concentrations were based on a peptide infusion rate of 1 ml/min into an estimated brachial arterial blood flow rate of approximately 100 ml/min. Forearm blood flow measurements were taken every 15 sec and averaged over the final 5 min of each 20-min infusion. Brachial arterial blood samples were taken before the administration of any compounds at the beginning and again at the end of the procedure. Forearm deep venous blood samples were taken at the end of each 20-min infusion.
Disease (obesity and type 2 diabetes mellitus) and therapeutic weight loss
Three subject groups were recruited (n = 6 adults/group): healthy controls [four men and two women; age, 40.2 ± 4.1 yr; body mass index (BMI), 22.0 ± 0.74 kg/m2], obese individuals (all women; age, 53.0 ± 2.6 yr; BMI, 39.2 ± 1.6 kg/m2), and obese individuals with type 2 diabetes mellitus (two men and four women; age, 54.0 ± 2.5 yr; BMI, 37.1 ± 2.4 kg/m2). The obese individuals (both diabetic and nondiabetic) underwent a program of supervised weight loss over 2 months; the nondiabetic individuals achieved 7.5 ± 1.6 kg weight loss, and the diabetic individuals achieved 6.7 ± 1.5 kg weight loss.
Pancreastatin amino acid variant discovery
Genomic DNA was isolated from 180 individuals of four ethnic/ancestry groups (by self identification; white/European, African-American, east Asian, and Mexican American) for resequencing at the CHGA locus and systematic single nucleotide polymorphism discovery, as previously described (14).
Pancreastatin amino acid variant (Gly297Ser) actions on glucose transport in isolated cells
Adipocyte isolation. Adipocytes were prepared from the epididymal fat pads of ad libitum-fed, 100- to 160-g male Wistar rats according to the method described by Rodbell (15). Fat pads were minced and then digested with collagenase at 37 C for 1 h in Krebs-Ringer buffer (113 mM NaCl, 2 mM CaCl2, 5 mM KCl, 10 mM NaH2CO3, 1.18 mM KH2PO4, and 1.18 mM MgCl2), pH 7.4, supplemented with 20 mM HEPES, 6 mM glucose, and 1% BSA. Aggregated material was removed by filtration through a mesh cloth. Isolated adipocytes were washed three times, and the packed cells were subsequently suspended in the final volume of the same buffer for metabolic experiments (105 cells/ml).
Glucose transport. Glucose transport was assayed as uptake of the nonmetabolizable glucose analog 2-deoxy-D-[2,6-3H]glucose (7 Ci/mmol), as previously described (16). Rat adipocytes were incubated in Krebs-Ringer buffer supplemented with 20 mM HEPES and 1% BSA without glucose at 37 C for 20 min in the presence or absence of 10 nM insulin. When pancreastatin was included in the experiment, it was added 4 min before insulin stimulation. Next, 0.5 µCi 2-deoxy-D-[2,6-3H]glucose was added (0.1 mM 2-deoxy-D-glucose), and the adipocytes were incubated for an additional 10 min. The assay was terminated by two rapid washes with iced PBS buffer. Cells were finally solubilized with NaOH, and radioactivity was measured by scintillation counting. Each condition was studied in triplicate.
Assays in plasma
EDTA-anticoagulated blood was kept on ice until centrifugation, and then frozen EDTA-plasma was stored at –70 C until assay. Glucose was measured with a spectrophotometric autoanalyzer. Free fatty acids were measured colorimetrically, with a normal reference range of 0.43–1.37 mM (17). -Amino acids were quantitated with a Beckman analyzer (Beckman Coulter, Fullerton, CA) after chromatographic separation (18). Pancreastatin in plasma was measured by RIA in 100-µl aliquots in duplicate, using [125I]pancreastatin in a commercial kit (Peninsula Laboratories, Belmont, CA). The molecular weight of the pancreastatin standard was 5263 g/mol, Pancreastatin (nanograms per liter) values can be divided by 5263 g/mol to yield values in nanomolar concentrations or can be divided by 5.263 to yield values in picomolar concentrations.
Pancreastatin formation in hormone storage granules in vivo: MALDI/time of flight mass spectrometry (MALDI-TOF) studies
Proteolytic cleavage sites at which pancreastatin is excised from CHGA in vivo were determined by mass spectrometry in hormone storage granule core peptides (19). Chromaffin granules (vesicles) were isolated by sucrose density gradient centrifugation of fresh homogenates of bovine adrenal medulla. After hypotonic vesicle lysis and ultracentrifugal removal of vesicle membranes, the soluble core peptides were subjected to immunoprecipitation with a rabbit anti porcine pancreastatin (Glu36-Gly49; EEETAGAPQGLFRG) primary antibody, generated by rabbit immunization with the keyhole limpet hemocyanin-conjugated peptide, as outlined previously (20, 21).
Chromaffin granule soluble core lysates were applied to Sep-Pak C18 cartridges (Waters/Millipore), eluted with 30–40% acetonitrile, lyophilized, resuspended in 500 µl immunoprecipitation buffer containing protease inhibitors [0.1% Triton X-100, 140 mM NaCl, 0.025% sodium azide, 10 mM Tris-HCl (pH 8.0), 1 mM phenylmethylsulfonylfluoride, 1 µM pepstatin, 1 mM EDTA, and 1 mM N-ethylmaleimide], and then immunoprecipitated by a modification of the protocol of Wang et al. (22). To minimize nonspecific binding results, samples were first incubated with 25 µl normal (preimmune) rabbit serum with constant rotator mixing at 4 C for 12–18 h. Sixty microliters of Protein G Plus/protein A agarose beads (33% slurry; Calbiochem, La Jolla, CA) were added, and rotational incubation was continued for 3 h, followed by centrifugation for 2 min at approximately 13,000 x g in an Eppendorf 5417 microcentrifuge, after which the pellet was discarded. To the supernatant, 20 µl rabbit antiporcine pancreastatin was added, and rotational incubation was continued for another 12–18 h. Sixty microliters of fresh Protein G Plus/Protein A agarose beads were added, and rotational incubation was continued for another 3 h, after which the beads were collected by centrifugation and washed three times with immunoprecipitation buffer, then twice with 50 mM Tris-HCl, pH 8 (to remove Na+ and detergent).
Immunoprecipitated pancreastatin peptides were eluted from the immune complexes with 20 µl trifluoroacetic acid/water/acetonitrile (1:20:20, vol/vol/vol) (22). One to 2 µl were characterized by MALDI on a Voyager-Elite mass spectrometer with delayed extraction (PerSeptive Biosystems, Framingham, MA). Samples were embedded in an -cyano-4-hydroxy-cinnamic acid matrix (23), then irradiated with a nitrogen laser at 337 nm, and the ions produced were accelerated with a deflection potential of 30,000 V. Ions were then differentiated according to their mass/charge ratio (M/Z) using a TOF mass analyzer. The mass error of this method is characteristically 0.1% or less (that is, 
1000 parts/million) (23).
Molecular weights from MALDI-TOF mass spectra were interpreted, and peptide fragments within the chromogranin A primary structure were assigned by the program PAWS (Protein Analysis WorkSheet, version 8.1.1, for Macintosh, 1997, ProteoMetrics, freeware available from http://prowl.rockefeller.edu), assigning average isotopic MH+ values for chromogranin A peptides (23).
Statistics
Forearm arterio-venous (A-V) gradients were calculated as the brachial arterial minus the antecubital deep venous concentration. A positive A-V gradient indicates uptake, whereas a negative A-V gradient indicates spillover. Forearm plasma flow was estimated as: (forearm blood flow)/(1 – hematocrit). Forearm glucose uptake was estimated as: (forearm plasma flow) x A-Vglucose. Other molecule (free fatty acid and amino acid) uptake (or spillover) rates were calculated similarly.
Multiple sequence alignment of the pancreastatin region across several species was accomplished with Clustal-W (MacVector, Accelerys, San Diego, CA).
Results are presented as the mean ± 1 SEM. Intergroup comparisons were made by t test, ANOVA, or repeated measures ANOVA, as appropriate, using SPSS (SPSS, Inc., Chicago, IL).
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Under basal conditions, there was net glucose uptake by the forearm. Pancreastatin, at a local concentration of approximately 200 nM (Fig. 1), decreased forearm A-V glucose gradient by about 50% (P = 0.049) and forearm glucose uptake by about 48% (P = 0.027). Because forearm plasma flow was unaltered (P = 0.135), the action of the peptide seemed to be metabolic, rather than hemodynamic.
|
), we analyzed the rate at which pancreastatin inhibited forearm glucose uptake in three subjects. The baseline A-V glucose gradient of 4.97 ± 0.96 mg/dl had declined to 1.23 ± 1.11 mg/dl by 16 min of pancreastatin infusion.
|
), at a local concentration of approximately 1 µM, did not alter either A-V glucose gradient or forearm glucose uptake, although catestatin did cause forearm vasodilation (by 10%; P = 0.042).
|
) substantially increased both the A-V glucose gradient (P < 0.0001) and forearm glucose uptake (P = 0.033); indeed, the changes were so marked that a logarithmic plot was required to capture the magnitude of the increase (Fig. 3
). However, pancreastatin did not reverse the effect of insulin on glucose uptake in this system. Insulin increased forearm plasma flow by about 18% (P = 0.033), and this insulin effect was not reversed by pancreastatin.
|
Under basal conditions (vehicle infusion), there was net free fatty acid spillover by the forearm into the circulation (Fig. 5), although the error terms overlapped zero (–33.7 ± 45.5 µM; –1.19 ± 1.43 µmol/min·100 g forearm). Pancreastatin augmented free fatty acid efflux from the forearm into the circulation, increasing the A-V difference (by 6.4-fold; P = 0.028) as well as the overall spillover (by 4.5-fold; P = 0.046).
|
), forearm free fatty acid metabolism transitioned from net spillover toward net uptake, although there was still substantial interindividual variability and an overlap of the error terms with zero: for A-V gradient, 13.5 ± 58.2 µM, and for uptake, 0.257 ± 21.6 nmol/min·100 g forearm. The insulin effect was significant (P = 0.022) for the A-V free fatty acid gradient, but not for overall spillover (P = 0.110). Pancreastatin did not reverse the insulin effect on the A-V free fatty acid gradient (P = 0.513).
|
Under basal conditions (vehicle infusion), 17 of the 20 natural amino acids in plasma showed net spillover from the forearm into the circulation (i.e. negative A-V gradients).
Pancreastatin had no effect, on its own, on either the A-V gradient (micromolar concentration) or the uptake (micromoles per minute per 100 g forearm) of any of the 20 natural amino acids (all P > 0.1).
As a positive control, insulin significantly diminished forearm spillover of two natural amino acids (Ala and Asp) as well as ornithine. The A-V gradient of Ala decreased by approximately 32 µM (from –33.3 ± 9.29 to –1.3 ± 11.6 µM; P = 0.0032), whereas Asp changed by about 1.5 µM (from 0.0 ± 0.471 to 1.50 ± 0.522 µM; P = 0.0212), and ornithine by about 6.2 µM (from –5.60 ± 2.60 to +0.60 ± 3.27 µM; P = 0.0176).
Pancreastatin did not significantly reverse this insulin-induced amino acid uptake in any case (all P > 0.2).
Pancreastatin in disease and treatment
Plasma pancreastatin was substantially (3.7-fold) elevated in obese subjects with type 2 diabetes (Fig. 7) compared with either controls (P = 0.009) or obese subjects without diabetes (P = 0.024). The baseline plasma pancreastatin level was 25.7 ± 3.7 ng/liter (4.9 ± 0.7 pM) in controls and 97.4 ± 22.3 ng/liter (18.5 ± 4.2 pM) in type 2 diabetics. The two obese groups (nondiabetic and diabetic) were then subjected to a 2-month course of supervised weight loss, during which they achieved comparable results (7.5 ± 1.6 vs. 6.7 ± 1.5 kg; P = 0.72). Despite this potent stimulus toward increased insulin sensitivity, plasma pancreastatin did not change (P = 0.107).
|
Likely cleavage sites of pancreastatin from CHGA in hormone storage granules were determined by MALDI-TOF mass spectrometry after antipancreastatin immunoprecipitation. The mass forms identified (Fig. 8) corresponded to bovine pancreastatin residues CHGA259–294-amide (MH+ 4042; Arg259-Gly294-amide), CHGA223–294-amide (MH+ 7735; Val223-Gly294-amide), and CHGA220–294 (MH+ 8089; Glu220-Gly294-amide).
|
Systematic resequencing (14) of the CHGA locus in 180 individuals (n = 360 chromosomes) revealed three naturally occurring variants in the pancreastatin region (Fig. 9): Arg253Trp (two heterozygotes; minor allele frequency, 0.6%), Ala256Gly (two heterozygotes and one homozygote; minor allele frequency, 1.1%), and Gly297Ser (two heterozygotes; minor allele frequency, 0.6%).
|
) reveals that the Gly397Ser variant lies in the relatively conserved carboxyl-terminal region, although the bovine residue (Val) at that position differs from the otherwise conserved Gly (in mouse, rat, porcine, and human sequences). The other two human variants toward the amino terminus (Arg253Trp and Ala256Gly) are in an unconserved region.
|
Because the biological activity of pancreastatin (human CHGA250–301) resides toward its carboxyl-terminal end (24, 25), and the carboxyl-terminal region of pancreastatin is best conserved across species (Figs. 9 and 10), we explored whether variant Gly297Ser differed in actions from the wild-type version (Fig. 11).
|
). The IC50 value for wild-type human pancreastatin was approximately 0.6 nM (600 pM), whereas that for Gly297Ser was about 0.1 nM (100 pM).
At a higher dose (
10–8 M; Fig. 11B), Gly297Ser diminished glucose uptake to 72.0 ± 3.9% of the control value, compared with 82.1 ± 4.3% by wild-type human pancreastatin. In this rat adipocyte system, rat pancreastatin diminished glucose uptake to 68.2 ± 3.2% of the control level, a value comparable to that with human Gly297Ser.
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
To understand the potential of pancreastatin for human physiology and disease, our studies had seven goals: 1) establish pancreastatin’s effect on human glucose metabolism in vivo; 2) determine whether pancreastatin’s effect on metabolism in vivo was predominantly metabolic or hemodynamic; 3) determine whether pancreastatin acts on its own, and/or antagonizes insulin’s effects in vivo; 4) determine whether pancreastatin affects additional components of intermediary metabolism in vivo, including lipid and amino acid disposition; 5) determine whether circulating pancreastatin varies in human disease states; 6) determine how pancreastatin is cleaved from CHGA in hormone storage granules; and 7) establish the functional consequences of natural human variation (polymorphism) in pancreastatin.
Pancreastatin, insulin, and human glucose metabolism
Under basal conditions, there was net glucose uptake across the human forearm vascular bed, and pancreastatin diminished this uptake by approximately 48–50%. This effect was achieved by a relatively low, submicromolar local concentration of infused pancreastatin (200 nM). The lack of change in forearm plasma flow indicated that pancreastatin’s action on glucose uptake was metabolic, rather than secondary to any changes in flow, and the catestatin peptide control was consistent with a specific effect of pancreastatin. The increase in forearm plasma flow (i.e. vasodilation) after catestatin is consistent with the actions of that peptide to diminish catecholamine secretion (4, 32) and release histamine (33).
The prompt action of pancreastatin (within minutes) is consistent with the action of a hormone upon a metabotropic G protein-coupled receptor; however, the pancreastatin receptor has not yet been isolated, nor its cDNA identified (34, 35). Likely downstream targets of pancreastatin would include glucose transporters, especially skeletal muscle GLUT4 and liver/islet GLUT2.
Although insulin increased forearm glucose uptake substantially, pancreastatin did not reverse this effect. In cultured cells, pancreastatin acts to inhibit insulin-induced glucose uptake (5). However, our local concentration of infused insulin is estimated to be approximately 70 µU/ml; although this was an effective concentration, it is likely to be in the relatively high physiological range (36) and might therefore be a difficult target for pancreastatin to overcome, especially if pancreastatin’s inhibition of insulin signaling is, to some extent, competitive with insulin as an agonist. Thus, the possibility remains open that pancreastatin might be able to antagonize less extreme, more physiological degrees of hyperinsulinemia; however, it would be difficult to test the full range of such possibilities in humans with this rather invasive brachial arterial infusion technique.
Although pancreastatin is usually described as a peptide whose actions functionally antagonize insulin effects (5), we observed that pancreastatin inhibited glucose uptake and triggered free fatty acid spillover, even in the absence of exogenous insulin. However, it should be noted that even in these fasting subjects without endogenous insulin stimulation, insulin is circulating at its basal concentration, probably in the range of approximately 5–10 µU/ml (36). Thus, under the obligate circumstances of basal circulating insulin in vivo, we cannot definitively conclude that pancreastatin would be active in the complete absence of insulin.
Pancreastatin and additional facets of human intermediary metabolism: free fatty acids and amino acids
Because insulin has actions on multiple areas of intermediary metabolism, we also tested such actions for pancreastatin in vivo. Under basal circumstances, there was net spillover of free fatty acids across the forearm vascular bed into the circulation, and this spillover was substantially (4.5- to 6.4-fold) augmented by pancreastatin. This observation is in line with the previous observation of a lipolytic effect of pancreastatin in isolated rat adipocytes (16).
Insulin reversed forearm free fatty acid metabolism, from spillover toward uptake, but once again pancreastatin did not antagonize this insulin action. Under basal conditions, 17 of 20 natural amino acids displayed net spillover from the forearm into the circulation, and this process was not affected by pancreastatin for any of the 20 naturally occurring amino acids. Although insulin significantly diminished spillover of Ala, Asp, and ornithine, these insulin actions were not affected by pancreastatin. Thus, pancreastatin has a spectrum of actions extending beyond carbohydrate metabolism into lipid metabolism, but even under these carefully controlled conditions we could not establish an effect of pancreastatin on amino acid metabolism.
Pancreastatin and human disease
Although pancreastatin was elevated approximately 3.7-fold in type 2 diabetes, it was not significantly elevated in the more modestly insulin-resistant state of obesity and did not change during substantial (7 kg) weight loss. Thus, although extreme states of insulin resistance were associated with elevations of pancreastatin, modest therapeutic changes in insulin sensitivity did not change the peptide’s concentration. The lack of decline in pancreastatin during weight loss would argue that the increase in pancreastatin in type 2 diabetes is not simply a response to the insulin resistance in that disease state, but might instead be pathophysiological in diabetes. Others have previously shown that pancreastatin is elevated in patients with type 2 diabetes (37, 38). The lack of a selective pancreastatin antagonist precludes more mechanistic studies of the role of pancreastatin in insulin-resistant states in vivo.
Although plasma pancreastatin immunoreactivity circulated at approximately 5–30 pM, we chose to infuse pancreastatin at an estimated (target) arterial concentration of about 200 nM. Our rationale was to attempt substantial (if not complete) stimulation of pancreastatin’s putative G protein-coupled receptor, because such receptors typically are stimulated by agonist threshold concentrations in the picomolar to nanomolar concentration range (39). Of note, the IC50 values for the human pancreastatins to inhibit cellular glucose uptake were approximately 100–600 pM, suggesting that even the relatively low circulating pancreastatin concentrations in humans might exert endocrine (systemic hormonal) actions.
Formation of pancreastatin in hormone storage granules in vivo
The results of pancreastatin MALDI-TOF document the formation of endogenous pancreastatin amide in hormone storage granules and suggest particular proteolytic cleavage sites for the excision of pancreastatin from bovine CHGA. Both short (CHGA259–294-amide; MH+ 4042; Arg259-Gly294-amide) and longer (CHGA220–294-amide; MH+ 8089; Glu220-Gly294-amide) pancreastatin forms were noted. At their carboxyl termini, the cleavages are consistent with the actions of basic residue-recognizing prohormone convertases (2), followed by PAM (9).
At their amino termini, the pancreastatin fragments suggest the actions of different classes of proteases. Formation of CHGA259–294-amide indicates cleavage at Gly258
Lys259, suggesting a protease that cleaves amino terminal to basic residues, such as the recently described hormone storage granule protease cathepsin L (40); indeed, we have found evidence of such cleavages in the formation of processing intermediates leading to the catestatin fragment of CHGA (41). Formation of CHGA223–294-amide by cleavage at Ala222Val223 also suggests a cathepsin-like specificity (42), whereas formation of CHGA220–294-amide by cleavage at Arg219Glu220 indicates the typical postbasic residue preference of trypsin-like enzymes (42), such as the prohormone convertases (2). These results suggest that previously isolated CHGA fragments in the pancreastatin region (6, 43) are probably formed in vivo.
Others have also observed heterogeneity in the size forms of pancreastatin isolated from neuroendocrine tissues. From a human carcinoid tumor, Schmidt et al. (43) isolated two pancreastatin forms by HPLC: human CHGA210–301-amide and CHGA273–301-amide. Funakoshi et al. (44) found human CHGA116–301-amide in human insulinoma, whereas Nakano et al. (45) isolated bovine CHGA248–294-amide from pancreas and pituitary.
Naturally occurring amino acid variants of human pancreastatin: functional consequences
Of the three nonsynonymous pancreastatin variants we discovered, one was in the likely domain of functional importance, its carboxyl-terminal region (24, 25) Gly297Ser. The crucial carboxyl-terminal recognition sites for endoproteolytic cleavage and amidation of pancreastatin (human CHGA residues Gly301Gly302Lys303) were invariant in 360 human chromosomes resequenced.
We, therefore, tested the effects of Gly297Ser (within CHGA273–301-amide) on cellular glucose uptake. Surprisingly, the potency of Gly297Ser to inhibit insulin-stimulated glucose uptake exceeded that of the wild-type peptide. The mechanism by which the Gly297Ser variant increases potency is not clear; although the three-dimensional structure of two CHGA biologically fragments, catestatin (human CHGA352–372) (46) and the antimicrobial chromofungin (bovine CHGA47–66) (47), are understood in some detail, the structure of the pancreastatin region (human CHGA250–301) has not yet been explored.
Conclusions
Pancreastatin is functional in humans in vivo, affecting both carbohydrate and lipid metabolism; indeed, its actions are potent and specific. Pancreastatin is cleaved from CHGA in hormone storage granules in vivo, and its plasma concentration varies in human disease. The pancreastatin region of CHGA gives rise to three naturally occurring human variants, one of which (Gly297Ser) occurs in the functionally important carboxyl terminus of the peptide and substantially increases the peptide’s potency to inhibit cellular glucose uptake. These observations establish a role for pancreastatin in human intermediary metabolism and disease and suggest that qualitative hereditary alterations in pancreastatin’s primary structure may give rise to interindividual differences in glucose disposition.
|
Footnotes |
|---|
First Published Online June 14, 2005
Abbreviations: A-V, Arterio-venous gradient (uptake [if +] or spillover [if –]); BMI, body mass index; CHGA, chromogranin A; FBF, forearm blood flow; MALDI-TOF, matrix-assisted laser desorption ionization/time of flight mass spectrometry; PAM, peptide -amidating monooxygenase.
Received February 25, 2005.
Accepted June 7, 2005.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
-amidating monooxygenase. Biochem Biophys Res Commun 205:282–290[CrossRef][Medline]
This article has been cited by other articles:
 |
|
 |
|
  N. R. Mahapatra Catestatin is a novel endogenous peptide that regulates cardiac function and blood pressure Cardiovasc Res, December 1, 2008; 80(3): 330 - 338. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Biswas, S. M. Vaingankar, M. Mahata, M. Das, J. R. Gayen, L. Taupenot, J. W. Torpey, D. T. O'Connor, and S. K. Mahata Proteolytic Cleavage of Human Chromogranin A Containing Naturally Occurring Catestatin Variants: Differential Processing at Catestatin Region by Plasmin Endocrinology, February 1, 2008; 149(2): 749 - 757. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. Rao, J. Wessel, G. Wen, L. Zhang, B. K. Rana, B. P. Kennedy, T. A. Greenwood, R. M. Salem, Y. Chen, S. Khandrika, et al. Renal Albumin Excretion: Twin Studies Identify Influences of Heredity, Environment, and Adrenergic Pathway Polymorphism Hypertension, May 1, 2007; 49(5): 1015 - 1031. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. O. Lillie, M. Mahata, S. Khandrika, F. Rao, R. A. Bundey, G. Wen, Y. Chen, L. Taupenot, D. W. Smith, S. K. Mahata, et al. Heredity of Endothelin Secretion: Human Twin Studies Reveal the Influence of Polymorphism at the Chromogranin A Locus, a Novel Determinant of Endothelial Function Circulation, May 1, 2007; 115(17): 2282 - 2291. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. A. Greenwood, F. Rao, M. Stridsberg, N. R. Mahapatra, M. Mahata, E. O. Lillie, S. K. Mahata, L. Taupenot, N. J. Schork, and D. T. O'Connor Pleiotropic effects of novel trans-acting loci influencing human sympathochromaffin secretion Physiol Genomics, May 16, 2006; 25(3): 470 - 479. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
