This Article Abstract Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Copyright Permission Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Ensinck, J. W. Articles by D’Alessio, D. A. Search for Related Content PubMed PubMed Citation Articles by Ensinck, J. W. Articles by D’Alessio, D. A. Pubmed/NCBI databases Compound via MeSH Substance via MeSH

Thrittene, Homologous with Somatostatin-28_(1–13), Is a Novel Peptide in Mammalian Gut and Circulation

Division of Metabolism, Endocrinology, and Nutrition, Department of Medicine (J.W.E., D.G.B., R.E.V., E.C.L., B.H.F.) and Department of Biological Structure (D.G.B.), University of Washington, Seattle, Washington 98195; Department of Veterans Affairs Puget Sound Health Care System (D.G.B.), Seattle, Washington 98108; ZymoGenetics (R.C.H., J.D.K., M.R.S.), Seattle, Washington 98102; Vollum Institute (M.J.L., M.R., V.O.-C.), Oregon Health Sciences University, Portland Oregon 97201-3098; and Division of Endocrinology (T.P.V., D.A.D.), University of Cincinnati, Ohio 45267-0547

Address all correspondence and requests for reprints to: David D’Alessio, M.D., University of Cincinnati, Division of Endocrinology, ML 0547, Cincinnati, Ohio 45267-0547. E-mail: . david.d'alessio{at}uc.edu

	Abstract

Top Abstract Introduction Materials and Methods Results Discussion References

 
Preprosomatostatin is a gene expressed ubiquitously among vertebrates, and at least two duplications of this gene have occurred during evolution. Somatostatin-28 (S-28) and somatostatin-14 (S-14), C-terminal products of prosomatostatin (ProS), are differentially expressed in mammalian neurons, D cells, and enterocytes. One pathway for the generation of S-14 entails the excision of Arg¹³-Lys¹⁴ in S-28, leading to equivalent amounts of S-28_(1–12). Using an antiserum (F-4), directed to the N-terminal region of S-28 that does not react with S-28_(1–12), we detected a peptide, in addition to S-28 and ProS, that was present in human plasma and in the intestinal tract of rats and monkeys. This F-4 reacting peptide was purified from monkey ileum; and its amino acid sequence, molecular mass, and chromatographic characteristics conformed to those of S-28_(1–13), a peptide not described heretofore. When extracts of the small intestine were measured by RIA, there was a discordance in the ratio of peptides reacting with F-4 and those containing the C terminus of ProS, suggesting sites of synthesis for S-28_(1–13) distinct from those for S-14 and S-28. This was supported by immunocytochemistry, wherein F-4 reactivity was localized in gastrointestinal (GI) endocrine cells and a widespread plexus of neurons within the wall of the distal gut while immunoreactivity to C-terminal domains of S-14 and S-28 in these neurons was absent. Further, F-4 immunoreactivity persisted in similar GI endocrine cells and myenteric neurons in mice with a targeted deletion of the preprosomatostatin gene. We believe that these data suggest a novel peptide produced in the mammalian gut, homologous with the 13 residues of the proximal region of S-28 but not derived from the ProS gene. Pending characterization of the gene from which this peptide is derived, its distribution, and function, we have designated this peptide as thrittene. Its localization in both GI endocrine cells and gut neurons suggests that thrittene may function as both a hormone and neurotransmitter.

	Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

 
PREPROSOMATOSTATIN (PreproS) is a gene that is widely expressed in vertebrates, including mammals, birds, reptiles, amphibians, and fish. Elements of the PreproS sequence are among the most highly conserved in vertebrate evolution, suggesting important regulatory roles for its peptide products (1). In addition, there seems to have been at least two duplications of PreproS over the course of evolution (1, 2). Teleost fish and amphibians express two PreproS genes, and the recently discovered preprocortistatin is a novel mammalian gene that is thought to be a member of the somatostatin peptide family (3, 4). In mammals, PreproS is translated into the 92 residue prohormone, prosomatostatin (ProS), which includes two bioactive peptides at its C terminus: somatostatin-14 (S-14) and somatostatin-28 (S-28). These peptides are uniquely processed and differentially secreted from cells in the brain, pancreatic islets, and gastrointestinal (GI) tract, where they act through specific membrane receptors to inhibit the release of neurotransmitters and hormones within the central nervous system (CNS) and gastroenteropancreatic (G-E-P) system (5, 6).

The intracellular generation of S-28 and/or S-14 is dependent on specific endoproteases that cleave ProS at designated sites (7). The processing of ProS is differentially regulated in the cells that produce it. In intestinal endocrine cells, S-28 along with equimolar amounts of ProS_(1–64) are generated (8). By contrast, S-14 is formed in gastric and islet D cells and a subset of enteric neurons. Both S-28 and S-14 are found in specific populations of neurons in the CNS. Processing of ProS to S-14 can occur directly with the coformation of equimolar amounts of ProS_(1–76) or indirectly from cleavage of S-28. In the latter case, S-28_(1–12) is also produced (9, 10, 11, 12, 13, 14, 15, 16). Thus, the intracellular cleavages of ProS are complex, with potential for differential regulation among ProS-containing cells, leading to different products. In this setting, we have produced an antiserum (F-4) with a high avidity for S-28, but little for S-28_(1–12), and none for S-14 (S-28_(15–28)), thereby allowing the development of a sensitive RIA for the former peptide (17). Using this assay, we have found a peptide distinct from S-28 in plasma and intestinal tissues. This peptide has been purified from extracts of monkey ileum, where it occurred in highest concentration. Its immunological and physical properties are compatible with the structure of S-28_(1–13), which has not been previously described as a distinct entity. Furthermore, we have obtained evidence that S-28_(1–13) is present in the tissues of PreproS knockout mice, indicating that this peptide is not a product of ProS. Because we have been unable to find evidence, in available data banks, for genes other than PreproS that have homology with the nucleic acid sequence corresponding to S-28_(1–13), we propose that this peptide, which we have tentatively named thrittene, is a novel product derived from a hitherto unrecognized gene.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

 
Peptides and antisera
S-28, S-28_(1–12), Tyr¹⁴-S-28_(1–14), and Tyr¹-S-28_(1–14), were purchased from Peninsula Laboratories, Inc. (Belmont, CA). S-28_(1–13) was custom synthesized at ZymoGenetics (Seattle, WA). ProS, partially purified from rat brain, was a gift from Dr. Robert Benoit, the Montreal General Hospital Research Institute. The tyrosinated peptides (1–5 µg) were coupled with Na¹²⁵I (Amersham Pharmacia Biotech, Arlington Heights, IL) and purified as previously described (17). Tyr¹⁴-S-28_(1–14) was linked to Tyr residues in bovine thyroglobulin by diazotization. Approximately 30 µg peptide were emulsified in incomplete Freund’s adjuvant and injected intradermally, at 10–20 sites, into each of 10 young female New Zealand white rabbits. The animals received repeat injections at 3-month intervals, after approval of the University of Washington Animal Research Committee. The antiserum, given the trivial designation of F-4, was tested for titer, specificity, and avidity by comparing the binding by RIA of several synthetic peptides with sequential amino acid deletions from the N and C terminus of S-28 and S-28_(1–14), respectively (17). F-4 interacts predominately with the Asn⁵-Pro⁶ sequence of S-28. In addition, 2 polyclonal antisera (designated AS-10 and AS-77) and a monoclonal Ig termed MS-12 were generated, with S-14 as the antigen, under conditions previously reported (17, 18). AS-10 and MS-12 interact with the Phe⁷-Trp⁸-Lys⁹ region of S-14, whereas AS-77 binds to the Ala¹-Ser² residues of the peptide. Antiserum Kiisha, generated against a synthetic peptide containing the 14 N-terminal amino acids of ProS, was a gift from Dr. Bernard Roos, University of Miami (19). Antiserum CTP was created in rabbits using a synthetic peptide corresponding to ProS_(33–45) (17). Antiserum S-320, which reacts with S28_(1–12) (20), was kindly donated by Dr. Robert Benoit. The epitopes along ProS where these antisera interact are schematized in Fig. 1.

View larger version (12K): [in this window] [in a new window]   Figure 1. Schematic diagram of the binding to regions of ProS with the antisera used in these studies.

Human subjects
Forty lean, healthy men (age range, 22–50 yr), not taking medication, gave informed consent and signed forms approved by the University of Washington Institutional Review Board. Thirty subjects had a single 10-ml sample of blood withdrawn after an overnight fast. To prevent degradation of peptides by endo- and exopeptidases, blood was collected on ice, the plasma separated, the pH adjusted to 3, with 1 N HCl (17), and it was stored at -80 C. For routine analyses, 5-ml aliquots of acidified plasma were filtered through cartridges containing octadecylsilyl silica (Sep-Pak; Waters Corp., Milford, MA) and were washed sequentially with 5 ml H₂O and 5 ml 0.1% trifluoroacetic acid (TFA) in H₂O, and the retained peptides were eluted with 5 ml of a solution of 80% methanol and 1% TFA. The eluates were air-dried and dissolved in 2 ml assay buffer. Recoveries of S-28 and S-28_(1–13) (50–100 pg), added to 5 ml of acidified plasma, were 78 ± 2% (n = 15) and did not vary significantly from each other. Twenty milliliters of acidified plasma from six fasting volunteers were processed through Sep-Pak and underwent gel filtration chromatography (see below). Ten fasting subjects had 250 ml blood taken. The acidified plasma was pooled and then processed sequentially through Sep-Pak, an F-4 immunoadsorbent, and gel filtration chromatography, and the eluates were submitted to HPLC (see below).

Animal tissue extraction
G-E-P organs were obtained from anesthetized young monkeys (Macaca nemestrina), mature Wistar rats, and C57BL/6J wild-type (Smst +/+) and ProS knockout mice (Smst -/-) (21), following procedures approved by the University of Washington Animal Research Committees. For extraction of peptides, monkey tissue was dissected and placed on ice. The intestine was arbitrarily transected into segments corresponding to the duodenum, jejunum, ileum, and colon. The lumens of the organs were quickly washed with cold 0.15 M NaCl, and all tissues were snap-frozen in liquid nitrogen and stored at -80 C. In larger scale extraction, ileum from eight juvenile monkeys was processed as above. The frozen tissue was pulverized in a steel cylinder in liquid nitrogen and dropped into boiling 2 N acetic acid (HAc) for 5 min. After cooling and homogenization with a glass rod, the solution was centrifuged at 30,000 x g and the supernatant decanted and lyophilized. The residue was weighed, and a small portion was dissolved in water for protein determination by the Folin method or using the bicinchohonic kit technique (Pierce Chemical Co., Rockford, IL). Overnight fasted, male Wistar rats (250–300 g) were killed with ip nembutal; and tissue segments corresponding to stomach, duodenum, jejunum, and ileum were rapidly removed and the mucosal surfaces separated from the adjacent submucosa and muscularis layers by stripping with a microscopic slide. The separate layers were then immersed in boiling 2 N HAc and processed as outlined above. Smst +/+ and Smst -/- mice were killed; and pancreas, stomach, and segments of the upper and lower intestine were removed to liquid nitrogen before storage at -80 C. These tissues were also extracted in 2 N HAc and processed through Sepak cartridges, and the reconstituted eluates were passed through a 45-µm filter before HPLC.

Affinity chromatography
The F-4 antiserum was coupled to cyanogen bromide-activated Sepharose 4B as detailed elsewhere (17). For immunoadsorption of peptides from plasma or tissue extracts, reference peptides, or unknown samples, diluted in 2 ml 130-mM phosphate buffer, pH 8.5, were introduced into 5 x 1-cm syringes containing the gel in buffer, washed with borate, and the F-4 reacting peptides were eluted with 0.2 N HAc, pH 3.5, containing BSA and lyophilized.

Gel permeation chromatography
Chromatography was carried out in columns (1.5 x 90 cm) containing Bio-Gel P-10 (200–400 mesh) (Bio-Rad Laboratories, Inc., Chemical Division, Richmond, CA) equilibrated in 50 mM borate buffer, pH 8.5, containing 0.25% BSA. The columns were calibrated with blue dextran with molecular mass exceeding 1000 kDa; cytochrome c (Sigma) with molecular mass of 13 kDa; S-28 with molecular mass of 3.1 kDa; S-14 with molecular mass of 1.6 kDa; S-28_(1–13) with molecular mass of 1.4 kDa; and ³H₂O. Filtration occurred under gravity, at room temperature, at 10 ml/h; and 1-ml fractions were automatically collected.

HPLC
Chromatography of plasma, after immunoadsorption, was performed in a modular HPLC system (Spectra-Physics, San Jose, CA). Samples (0.15–1.0 ml) were injected onto a micron analytical column (Vydac C-18 5; The Separations Group, Hesperia, CA) and eluted with sequential gradients consisting of acetonitrile (CH₃CN)/0.1% TFA: first, a 0–20% solution, stepped over 10 min; followed by a 20–30% step over 10 min; then a 30–40% linear gradient for 70 min; and, lastly, a 40–90% linear gradient for 30 min. The flow rate was 0.5 ml/min. Separation of mouse tissue extracts used a similar buffer/solvent combination, with a gradient of: 5–20% over 20 min, 20–37% over 70 min, and 37–90% over 20 min.

Preparative reverse-phase (rp) HPLC of tissue extracts was performed on a 4.6 x 150-mm, PLRP-S 100 A, 5-µm column (Polymer Laboratories Inc., Amherst, MA) using an integrated HPLC system (Perspective BioSystems, Framingham, MA). A flow rate of 1 ml/min was used, and an isocratic of 1% B for 10 min, followed by a 1–41% gradient over 80 min, was initiated upon injection, where A is 2% CH₃CN + 4 mM NH₄Ac, pH 10, and B is 90% CH₃CN + 4 mM NH₄Ac, pH 10. Microbore rp HPLC was carried out on an Ultrafast Microprotein Analyzer (Michrom BioResources Inc., Auburn, CA). A 1 x 150-mm C₁₈ 300 A, 5-µm column (Vydac, Hesperia, CA) was used at a flow rate of 50 µl/min at 30 C. A gradient of 1–41% over 80 min was begun on injection, where A is 2% CH₃CN + 0.1% TFA and B is CH₃CN + 0.095% TFA.

N-terminal sequence analysis
N-terminal sequence analysis was performed on a protein sequencer (Model 476A; PE Applied Biosystems, Foster City, CA); and data analyses on a Model 610A Data Analyses System (also PE Applied Biosystems). Thirty and 50 µl of the respective microbore fractions were analyzed.

Matrix-assisted laser desorption ionization-time of flight (MALDI-TOF) mass spectrometry
MALDI-TOF mass spectra of peptides were obtained with a TofSpec SE mass spectrometer (MicroMass, Beverly, MA) equipped with a 337-nm laser. Linear positive-ion experiments were performed with the suppression mass at 400 Da and the following voltages: source, 25,000 V; extraction, 8,333 V; focus, 23,000 V; and linear detector, 3,800 V. Postsource decay (PSD) positive-ion experiments were performed with the suppression mass at 500 Da, the BN gate set to the ion of interest, and the following voltages: source, 25,000 V; extraction, 8,333 V; focus, 23,000V; reflectron voltage at 28,575 V; and reflectron detector at 1,800 V. The instrument had a 0.5% accuracy (±7 Da) for an externally calibrated sample and approximately 0.1% accuracy (±1.5 Da) for internally calibrated samples.

Immunocytochemistry
Rats and mice were anesthetized and perfused with 4% paraformaldehyde in 100 mM phosphate buffer, pH 7.4. Fresh tissue samples from rodents and monkeys were immersed in fresh fixative overnight, at 6 C, and then placed in 50 mM cold phosphate buffer containing 25% sucrose, for 24 h. Some of the tissue samples were frozen and cut, by cryostat, into 7-µm sections; others were immersed sequentially through ethanol and toluene, embedded in paraffin, and cut into 5-µm sections. Immunostaining of cryostat and paraffin sections was performed by the same procedure, except that the paraffin was removed with xylene and ethanol beforehand. For fluorescence immunostaining (22, 23), sections were placed in 10 mM phosphate buffer, pH 7.4, containing 5% normal goat serum, followed by immersion in primary antiserum, overnight, at 6 C. Primary antisera were diluted in 10 mM phosphate buffer, pH 7.4, according to the following: F-4, 1:2000; AS-10, 1:1000; MS-12, 1:1000; AS-77, 1:1000, CTP, 1:2000; and Kiisha (N-terminal ProS antiserum), 1:1000. After a rinse in cold buffer, sections were immersed for 1 h in goat antirabbit IgG (for MS-12) conjugated to Cy3 (for polyclonal antisera) or goat antimouse IgG conjugated to Cy3 (Jackson ImmunoResearch Laboratories, Inc., West Grove, PA) diluted to 1:200 in phosphate buffer. For bright-field immunostaining, sections were treated with the antisera and then processed with the Vectastain Elite ABC kit (Vector Laboratories, Inc., Burlingame, CA), using IgG-biotin-streptavidin reagents for diaminobenzidine staining. Controls for immunocytochemistry included: 1) omission of all primary and secondary antibodies; 2) omission of primary antibodies only, with substitution of normal rabbit (or mouse) serum; 3) staining of pancreatic D cells (known to express ProS); and 4) examination with fluorescence excitation and emission filters that do not excite and transmit Cy3 fluorescence. Because all of these primary antisera have been extensively characterized for specificity by RIA analyses (17), and absorption controls for AS-10 and F-4 have been previously reported (22, 23), they were not repeated in this study.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

 
Characterization of F-4 antiserum interacting with ProS and its C-terminal sequences
The binding characteristics of F-4 with ProS, S-28, S-28_(1–14), S-28_(1–13), S-28_(1–12), and S-14 are shown in Fig. 2. F-4 interacted with ProS, S-28, and S-28_(1–14) with similar avidity, permitting their detection in the femtomolar range. Noteworthy, the cross-reactivity of this antiserum with S-28_(1–12) was extremely poor, whereas S-28_(1–13) had the highest affinity of all peptides tested. S-28_(1–13) was bound to F-4 50- to 100-fold more avidly than S-28 throughout the range of concentrations tested. This implies that modifications of the C-terminal structure of the antigen, S-28_(1–14), markedly affect the binding to F-4. From previous studies using synthetic peptides with sequential amino acid deletions from the N terminus of S-28, we infer that the Asn⁵-Pro⁶ sequence is necessary for binding to F-4 (17). The lowest limits of detection of S-28_(1–13) and S-28 in the RIA were 0.2 and 2 fmol per tube, with interassay variances of 15% (n = 10) and intraassay variances of 8% and 5% (n = 10), respectively.

View larger version (21K): [in this window] [in a new window]   Figure 2. Immunologic interactions of ProS and related peptides with antiserum F-4 using 125I-Tyr-S-28(1–14) as radioligand. Data are presented as % B/B0, the amount of iodinated tracer bound at each peptide concentration relative to tracer binding without added peptide.

View larger version (8K): [in this window] [in a new window]   Figure 3. Concentrations of S-28 and all F-4 immunoreactivity, in healthy men, after an overnight fast. Peptides from plasma (5 ml) were adsorbed and eluted from Sep-Pak. S-28 was quantified by F-4 immunoadsorption and measurement of the C-terminal region by AS-10 antiserum. The F-4 immunoreactive peptides (F-4 IR) were measured directly in the Sep-Pak eluates.

View larger version (15K): [in this window] [in a new window]   Figure 4. Profiles of peptides measured by F-4 and AS-10 RIA in fasting plasma from a healthy man, separated by gel permeation chromatography. Plasma (70 ml) was adsorbed to Sep-Pak and eluted, and the eluates dissolved in borate buffer were subjected to immunoadsorption on F-4 agarose. The retained peptides were eluted with 2 N HAc, lyophilized, and reconstituted in 1.0 ml borate buffer for application to the column. Filtration was carried out in a 1.5 x 90-cm column containing BioGel P-10 in 50 mM borate buffer, pH 8.5, containing 0.25% BSA. RIA measurements were performed using S-28 as standard. Arrows, Peaks where markers of known composition were eluted; V0, void volume detected by blue dextran; cytochrome c has a molecular mass of 13 Da.

View larger version (18K): [in this window] [in a new window]   Figure 5. HPLC profile of F-4 immunoreactive peptides (F-4 IR) from a pool of 1 liter plasma from 10 healthy men after an overnight fast. The plasma was processed as described in Fig. 3. The fractions from tubes 70–110 were pooled, concentrated by Sep-Pak, and lyophilized. The peptides were applied to a C-18 rp column with a variable gradient of CH3CN in 0.1% TFA. The eluates were analyzed by both F-4 and AS-10 RIA.

Isolation and identification of S-28_(1–13) in mammalian gut
To determine the structure of the substance occurring in both plasma and gut that was detected by F-4 RIA but distinct from S-28, monkey ileum was processed as described in Materials and Methods. This entailed sequential extraction of freshly obtained gut in boiling HAc, ultracentrifugation to remove sediment, filtration on BioGel P-10, immunoadsorption on F-4 coupled to agarose, preparative rpHPLC at high pH, and microbore rpHPLC at low pH. The F-4 reactive peak from 2 independent runs of the final low-pH rpHPLC step underwent N-terminal sequence analysis and MALDI-TOF mass spectrometry. The sequence derived from the combined analyses revealed 13 residues, with identification of 8 as follows: SXNXNPAMAXXEX. Although 5 of the 13 residues could not be identified, the N-terminal Ser, Asn³, Asn⁵-Pro⁶-Ala⁷-Met⁸-Ala⁹, and Glu (12) were arranged in correct sequence for the N-terminal region of S-28. Furthermore, throughout each of the purification steps, the unknown peptide behaved identically to authentic, synthetic S-28_(1–13).

The Maldi-TOF MS analysis of the final product of our purification paradigm yielded masses of 1420 Da in 1 preparation and 1419.3 Da in the second. These masses are very similar to the isotopic average mass expected for S28_(1–13) with oxidation of the methionine residue at position 8 to methionine sulfoxide-1416.5 Da. A mass error of 0.2% is acceptable for a linear MALDI-TOF spectrum internally calibrated using the matrix peaks. A more accurate mass of 1416.9 was obtained for the peptide in reflectron mode, a mass error of only 0.03%. To confirm the identity of the peptide species present in both reactive fractions as S-28_(1–13), the masses of 1419.5 Da observed in the linear MALDI-TOF spectra were subjected to fragmentation by PSD. The PSD spectra obtained were compared with the PSD spectra obtained for pure, synthetic S-28_(1–13) in which Met⁸ was oxidized to Met sulfoxide with H₂O₂. Despite the very low signal in the PSD spectra of the reactive fractions attributable to the extremely small amount of available peptide, approximately 10 fragment peaks were above the noise and in common with the strong fragment peaks observed in the PSD spectrum of oxidized synthetic S-28_(1–13). Especially prominent were both y- and b-ions showing a neutral loss of 64 Da. A facile neutral loss of 64 Da corresponds to the loss of SOCH₄ and is indicative of Met sulfoxide in the peptide. Oxidation of the single Met residue present in S-28_(1–13) is not unexpected, because of the harsh acidic extraction conditions used in processing the monkey tissue. Because F-4 antiserum has a much higher affinity (circa 50–100 fold) for S-28_(1–13) than S-28 (Fig. 2), using this RIA with S-28 as standard led to an overestimate of the amount of material submitted for sequence and MS analyses. A reestimate suggests that 300–600 fmol were used in the N-terminal sequence analysis and 12–16 fmol for mass spectrometry. Despite the low levels of peptide, all 3 physical analyses (N-terminal sequence analysis, MALDI-TOF mass spectrometry, and PSD fragmentation) are consistent with the identity of the F-4 immunoreactive species as S-28_(1–13).

Localization of F-4-containing cells and neurons in monkey G-E-P tissues
G-E-P tissues from the monkey and rat were immunostained with F-4 to identify cells producing the peptides containing its epitope S-28 and S-28_(1–13) (Table 2). Antiserum F-4 produced bright immunofluorescence of pancreatic islet D cells compatible with the presence of ProS, and possibly S-28, within these cells (Fig. 6A). F-4 also stained enterocytes, and neural cell bodies and fibers distributed in the mucosal, submucosal, and muscularis regions of the gut wall (Fig. 6, C, E, G, and I). Although the results were qualitatively similar in the different regions of the small intestine, the ileum contained many more F-4-positive enterocytes and, particularly, neuronal elements, than the duodenum; and the numbers of F-4 positive cells gradually increased from upper to lower small intestine. In the mucosa, numerous brightly fluorescent F-4 reactive fibers coursed through and ramified into the tips of the villi (Fig. 6, C and E). In the submucosa, F-4-positive axons were distributed within the loose connective tissue and along blood vessels (Fig. 6E). In the myenteric plexus, a constellation of brightly fluorescent punctate structures resembling beaded fibers and terminals surrounded the neuronal perikarya in the myenteric ganglia (Fig. 6, G and I). Some neuronal perikarya in the myenteric and mesenteric plexuses showed cytoplasmic staining as well (Fig. 6, G and I). In contrast to the staining with F-4, antiserum AS-10, directed to a more C-terminal epitope of ProS, labeled only islet D cells (Fig. 6B) and intestinal mucosal cells (Fig. 6D), with minimal staining of neural elements (Fig. 6, F, H, and J).

View larger version (58K): [in this window] [in a new window]   Figure 6. Immunocytochemical localization of epitopes recognized by F-4 antiserum (A, C, E, G, and I) and AS-10 antiserum (B, D, F, H, and J) in pancreatic islets (A and B) and ileum (C–J) of M. nemestrina. Both antisera recognize epitopes in islet D cells (A and B) and in mucosal enterocytes (small arrows in C and D). F-4 recognizes nerve fibers in the mucosal villi (large arrows in C and small arrows in E), and neuronal cell bodies and fibers in the mesenteric (G) and myenteric (I) plexuses (examples of cell bodies indicated by arrows). However, AS-10 does not detect epitopes in any neural elements of the ileum (F, H, and J). Note that neuronal content of F-4 immunoreactivity in myenteric plexus neurons (I) (neurons indicated by arrows) is completely absent in an adjacent section immunostained with AS-10 (J). The same results were observed in jejunum and duodenum. The staining in H is nonspecific fluorescence. Controls showed no immunostaining in D cells, enterocytes, or neurons.

F-4 immunoreactivity in Smst knockout mice
To determine whether the synthesis of S-28_(1–13) is independent of PreproS gene expression, we compared the immunostaining characteristics of G-E-P cells in Smst -/- mice with those of control mice, using antisera reacting with domains throughout ProS. Wild-type mice had immunoreactivity to ProS-derived peptides in the expected regions of their pancreatic islets and GI tract (Fig. 7, Table 3). AS-10 antiserum produced bright immunofluorescence of D cells in islets (Fig. 7A) and gastric mucosa (Fig. 7E) as well as of intestinal mucosal enterocytes and neuronal fibers of intestinal villi and submucosa (Fig. 7, I and M) but with minimal staining of cellular elements in the muscle wall of the gut (Fig. 7Q). The same results were obtained with CTP and Kiisha, more N-terminally reacting ProS antisera (data not shown). AS-77, which is specific for S-14, produced results similar to those of the other antisera, with the exception that it did not stain enterocytes, similar to the results presented above in monkey tissues (data not shown). In contrast to the control animals, the Smst knockout mice had no staining with S-14-specific antibodies [Fig. 7, B, F, J, N, and R; (21)].

View larger version (153K): [in this window] [in a new window]   Figure 7. Fluorescence immunocytochemical localization of cells recognized by AS-10 and F-4 antibodies in pancreatic islets (A–D), stomach (E–H), and small intestine (I–T) of Smst +/+ (wild-type) and Smst -/- (knockout) mice. Both AS-10 and F-4 stain islet D cells (some indicated by arrows) of Smst +/+ mice (A and C) but not Smst -/- mice (B and D). In stomach, AS-10 and F-4 stain mucosal D cells (some indicated by arrows) of Smst +/+ mice (E and G). These cells are not stained by AS-10 in Smst -/- animals (F) but are stained by F-4 (H). Similarly, mucosal enterocytes throughout the small intestine are recognized by AS-10 in Smst +/+ (I), but not in Smst -/-, mice (J), whereas F-4 stains mucosal enterocytes in both Smst +/+ (K) and Smst -/- (L) mice (enterocytes are indicated by small arrows; large arrows point to stained nerve fibers). Nerve fibers and cell bodies in villi and submucosa stain positively with AS-10 in Smst +/+ mice (M) but not in Smst -/- mice (N), whereas submucosal neurons are stained by F-4 in both Smst +/+ (O) and Smst -/- (P) rodents (large arrows indicate stained nerve fibers; enterocytes are indicted by small arrows). Neurons in the myenteric plexus of the muscularis layer (some small arrows) do not stain with AS-10 in either Smst +/+ (Q) or Smst -/- (R) mice, whereas F-4 stains myenteric plexus in both Smst +/+ (S) and Smst -/- (T) mice.

Molecular characterization of F-4 immunoreactivity from Smst +/+ and Smst -/- mice
Extracts of GI tissues from Smst-/- and Smst +/+ mice were fractionated by HPLC using a gradient that provided clear separation of S-28_(1–13) and S-28 (Fig. 8). Extracts of stomach and upper and lower small intestine from the Smst-/- mice had peaks of F4-IR immunoreactivity that corresponded precisely to the elution position of synthetic S-28_(1–13), whereas there was no F4 immunoreactivity at the later elution position of S-28. Intestinal extracts from the Smst +/+ mice also had an immunoreactive peak eluting at the position of the S-28_(1–13) standard (data not shown). Samples of 0.1% TFA without added peptides, run after synthetic standard and tissue extracts, had no F4 immunoreactive peaks.

View larger version (18K): [in this window] [in a new window]   Figure 8. HPLC profile of F-4 immunoreactive peptides (F-4 IR) from acetic acid extracts of the stomach and upper and lower small intestine of Smst -/- mice. Aliquots containing 150 µg protein were applied to a C-18 rp column with a variable gradient of CH3CN in 0.1% TFA. Fractions were collected at 1-min intervals and analyzed in the F4 RIA. Synthetic S-28(1–13) eluted as a peak (detected either by absorbance of UV light or by F4 RIA) at 27 min, and S-28 at 65 min.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

We initiated these studies to explain disparities between F-4 immunoreactivity and concentrations of S-28 determined by a two-step immunoassay in plasma and tissue. In pursuing this observation, we identified an F-4 immunoreactive moiety with characteristic and reproducible chromatographic properties, whether obtained from plasma or intestinal extracts. Based on gel exclusion chromatography, this substance appeared to be larger than S-28; but, in fact, subsequent analyses showed that the F-4 immunoreactive peak immediately preceding S-28 eluted at the position identical to synthetic S-28_(1–13). We were able to confirm the implications of the first chromatographic studies by HPLC, peptide sequencing, and mass spectrometry. The partial sequence of the 13-residue peptide that was obtained from material purified from monkey intestine is consistent with 8 of the 13 interspersed residues in correct sequence with that of S-28_(1–13) and contains the epitope that we had previously shown to be required for interaction with F-4 (17). The C-terminal amino acid was not determined; thus, we cannot be absolutely certain that it is Arg. However, the mass of 1416.9 Da for the isolated peptide obtained by MALDI-TOF MS analysis and the behavior of the isolate after PSD fragmentation are precisely those predicted, and proven, for authentic S-28_(1–13) with an oxidized methionine residue. Thus, we are confident that the sequence we have assigned this peptide is correct.

Based on the inverse molar ratio of F-4 and AS-10 immunoreactivity in extracts along the GI tract, thrittene is present in highest concentrations in the distal segments of the gut, in contrast to S-14 and S-28, which we and others have shown to be most prevalent in the stomach and upper small intestine (22, 24, 25, 26, 27). Because all peptides containing the C terminus of ProS are recognized by AS-10, the relative increase in F-4 immunoreactivity cannot be explained by a preponderance of S-28 over S-14 in this region. Our immunocytochemical results amplify this point by locating F-4 reacting cells in a neuronal network in the mucosa and submucosa of the jejunum and ileum that do not stain with AS-10 and MS-12 or an antiserum highly specific for S-14 (AS-77). Though we cannot exclude the possibility that ProS and thrittene are coexpressed in some tissues, these data indicate that both endocrine and neuronal cells in the GI tract also produce thrittene independent of ProS-derived peptides. This finding is consistent with derivation of thrittene from a gene other than PreproS.

Though the nucleic acid sequence corresponding to thrittene is not known, the data presented herein support a novel gene precursor. First, we have been unable to identify a DNA sequence, other than ProS, that corresponds with the sequence of S-28_(1–13) using computer driven searches of current data banks. Second, the comparative F-4 RIA measurements in tissue extracts, coupled with the results of immunocytochemistry, suggest that thrittene is synthesized independent of the ProS-derived peptides. Third, the persistent F-4 immunoreactivity, demonstrated both by immunostaining and after extraction of peptides, in the GI endocrine and neuronal cells of mice homozygous for the Smst null allele indicates that thrittene is produced in the absence of ProS. The Smst -/- animals have a deletion of promoter sequences and the first exon of the PreproS gene that includes the transcriptional start site and encodes a major portion of the N terminus of ProS (21). Though exon 2 encoding the C terminus of ProS is present in the mice, the absence of detectable ProS mRNA, assessed either by RT-PCR or in situ hybridization, and the complete lack of ProS derived peptides, assessed either by immunocytochemistry or by RIA of tissue extracts, indicates the fidelity of the knockout of the PreproS gene in these animals (21). The distribution of F-4 immunoreactivity in GI cells in the knockout mice, similar to those that were stained in monkeys and rats and wild-type mice, indicates that thrittene synthesis is shared by both endocrine cells in stomach and gut and neurons predominating in the myenteric plexus.

The somatostatin-related peptides are ubiquitous among animals, and there is evidence that at least 1 duplication of PreproS persists in the mammalian genome (1). Anglerfish, from which ProS was first cloned, and frogs have 2 somatostatin genes (1, 28), as do several other species. Cortistatin, a recently discovered peptide that shares 11 of 14 residues with S-14, is the product of a separate gene (3, 4). Preprocortistatin has only limited homology with ProS beyond the C terminus, but these genes are considered to be related; and so, they may derive from a common ancestor. Given the significant concordance between thrittene and the proximal region of S-28, it is possible that the gene encoding thrittene is also related to PreproS. Clearly, the placement of thrittene among known families of regulatory peptides awaits the ascertainment of its DNA sequence.

To our knowledge, this paper contains the first description of thrittene as a distinct entity; nonetheless, it is possible that previous investigators of ProS and its processing may have unwittingly come upon this peptide. Two groups have described antisera that they believed to be specific for the epitopes containing the C-terminal Glu residue shared by S-28_(1–12) and ProS_(1–76) (14, 20). These antisera, designated S-320 by Benoit et al. (20) and R-109 by Patel et al. (14), were tested for cross-reactivity with several congeners of S-28. However, S-28_(1–13) was not included among the published reports; therefore, we infer that it was not evaluated. In our hands, S-28_(1–13) and S-28_(1–12) interacted equally with S-320 (data not shown). Furthermore, because the peptides differ by only one residue, they could easily be mistaken for one another unless special attention was given to methods for their separation. Thus, it is plausible that thrittene may have been included in the measurements attributed to S-28_(1–12) when antiserum S-320 was used for RIA of extracts from brain and G-E-P tissues (14, 20). This may also explain the discrepancy noted when Morrison and colleagues immunostained mammalian neocortical cells with S-320 and other immune sera interacting specifically with S-14 and S-28 (29). In these studies that were replicated by Lewis et al. (30), staining with S-320 was widespread in neuronal processes, terminals, and dendrites, whereas S-28 was localized to some cell bodies, but virtually no S-14 was observed. Although an explanation for these findings was never published, it is possible that some of the neocortical neurons that were stained with S-320 contained thrittene exclusively, analogous to our findings with F-4 in intestinal neurons. Though we have not yet examined brain tissue for immunocytochemical evidence of thrittene, it is reasonable to postulate that thrittene is also present in the CNS, because it is common for peptides expressed in the GI tract to be found also in the brain.

Based on the immunostaining of gastric and enteric endocrine-like cells in the Smst-/- mice, the major source of thrittene in human (and nonhuman primate) plasma is presumed to be endocrine cells in the stomach and distal small intestine. We have found that plasma concentrations of thrittene increased 2- to 3-fold in humans after ingestion of a mixed meal (Ensinck, J. W., unpublished observation). Because fat and, to a lesser extent, carbohydrate evoked this release, secretion of thrittene is likely to be regulated by specific nutrients similar to other GI peptides. Traditionally, peptides that are secreted from cells within GI tissues in response to food intake originate in cells of epithelial origin and act as hormones on distant targets (31). Thus, it is reasonable to deduce that the presence of thrittene in plasma reflects its secretion from GI endocrine cells. As yet, we have no clear-cut evidence for a biologic effect of this putative hormone/neurotransmitter on G-E-P functions; however, we have observed that, when instilled into the cerebral ventricles of rats, thrittene decreases food intake in a dose-dependent manner without evoking an aversive response (32).

In summary, we have immunological and physical evidence of the existence of a peptide with mass of 1416.9 Da and a partial sequence compatible with the assignment of a structure of S-28_(1–13). We isolated this peptide from the distal intestine of nonhuman primates and demonstrated, by RIA of tissue extracts and immunocytochemistry, that it is located in endocrine cells in stomach and intestine and in myenteric neurons that do not contain ProS. On the basis of persistence of immunoreactivity in GI cells of Smst-/- mice and the lack of a corresponding DNA sequence in previously reported genes, we deduce that thrittene is derived from a hitherto-unrecognized gene.

	Acknowledgments

 
We appreciate the excellent technical assistance of Joyce Murphy and Chare Vathanaprida for tissue fixation and immunostaining.

	Footnotes

 
This work was supported by NIH Grants DK-34397 (to J.W.E.), DK-02305 (to D.D.), RR00037 (University of Washington General Clinical Research Center), RR-00166 (University of Washington Regional Primate Research Center), and DK-34397 (University of Washington Diabetes and Endocrine Research Center). Additional support was provided by grants from the American Diabetes Association (to D.D.), the Juvenile Diabetes Foundation (to J.W.E.), Merit Review and Career Scientist awards from the Department of Veterans Affairs (to D.G.B.), and a grant from the Medical Research Foundation of Oregon (to M.J.L.).

¹ Present address: R. W. Johnson Pharmaceutical Research Institute, Route 202, P.O. Box 300, Raritan, New Jersey 08869-0602.

² Present address: Howard Hughes Medical Institute, University of California at San Diego, Mailstop 0314, 9500 Gilman Drive, La Jolla, California 92103-0314.

³ Present address: Argonex, Inc., 706 Forest Street, Suite 1, Charlottesville, Virginia.

⁴ Present address: Instituto de Investigaciones en Ingeneria Genetica y Biologia Molecular, Consejo Nacional de Investigaciones Científicas y Técnicas de Argentina, and Departamento de Ciencias Biológicas, Facultad de Ciencias Exactas y Naturales, University of Buenos Aires, 1428 Buenos Aires, Argentina.

Abbreviations: CH₃CN, Acetonitrile; CNS, central nervous system; G-E-P, gastroenteropancreatic; GI, gastrointestinal; HAc, acetic acid; MALDI-TOF, matrix-assisted laser desorption ionization-time of flight; PreproS, preprosomatostatin; ProS, prosomatostatin; PSD, postsource decay; rp, reverse-phase; S-14, somatostatin-14; S-28, somatostatin-28; TFA, trifluoroacetic acid.

Received July 10, 2001.

Accepted for publication March 21, 2002.

	References

Top Abstract Introduction Materials and Methods Results Discussion References

 

1. Conlon JM, Tostivint H, Vaudry H 1997 Somatostatin- and urotensin II-related peptides: molecular diversity and evolutionary perspectives. Regul Pept 69:95–103[CrossRef][Medline]
2. Argos P, Taylor WL, Minth CD, Dixon JE 1983 Nucleotide and amino acid sequence comparisons of preprosomatostatins. J Biol Chem 258:8788–8793[Abstract/Free Full Text]
3. de Lecea L, Criado JR, Prospero-Garcia O, Gautvik KM, Schweitzer P, Danielson PE, Dunlop CL, Siggins GR, Henriksen SJ, Sutcliffe JG 1996 A cortical neuropeptide with neuronal depressant and sleep-modulating properties. Nature 381:242–245[CrossRef][Medline]
4. Spier AD, de Lecea L 2000 Cortistatin: a member of the somatostatin neuropeptide family with distinct physiological functions. Brain Res Brain Res Rev 33:228–241[CrossRef][Medline]
5. Patel YC 1992 General aspects of the biology and function of somatostatin. In: Thorner MO, Muller EE, eds. Basic and clinical aspects of neuroscience. Berlin: Springer-Verlag; vol 4:1–16
6. Patel YC 1999 Somatostatin and its receptor family. Front Neuroendocrinol 20:157–198[CrossRef][Medline]
7. Patel YC, Galanopoulou A 1995 Processing and intracellular targeting of prosomatostatin-derived peptides: the role of mammalian endoproteases. Ciba Found Symp 190:26–50[Medline]
8. Beinfeld MC, Bourdais J, Kuks P, Morel A, Cohen P 1989 Characterization of an endoprotease from rat small intestinal mucosal secretory granules which generates somatostatin-28 from prosomatostatin by cleavage after a single arginine residue. J Biol Chem 264:4460–4465[Abstract/Free Full Text]
9. Gluschankof P, Gomez S, Morel A, Cohen P 1987 Enzymes that process somatostatin precursors. A novel endoprotease that cleaves before the arginine-lysine doublet is involved in somatostatin-28 convertase activity of rat brain cortex. J Biol Chem 262:9615–9620[Abstract/Free Full Text]
10. Gomez S, Gluschankof P, Lepage A, Cohen P 1988 Relationship between endo- and exopeptidases in a processing enzyme system: activation of an endoprotease by the aminopeptidase B-like activity in somatostatin-28 convertase. Proc Natl Acad Sci USA 85:5468–5472[Abstract/Free Full Text]
11. Galanopoulou AS, Seidah NG, Patel YC 1995 Direct role of furin in mammalian prosomatostatin processing. Biochem J 309:33–40
12. Galanopoulou AS, Kent G, Rabbani SN, Seidah NG, Patel YC 1993 Heterologous processing of prosomatostatin in constitutive and regulated secretory pathways. Putative role of the endoproteases furin, PC1, and PC2. J Biol Chem 268:6041–6049[Abstract/Free Full Text]
13. Patel YC 1983 A high molecular weight form of somatostatin-28 (1–12)-like immunoreactive substance without somatostatin-14 immunoreactivity in the rat pancreas. Evidence that somatostatin-14 synthesis can occur independently of somatostatin-28. J Clin Invest 72:2137–2143
14. Patel YC, O’Neil W 1988 Peptides derived from cleavage of prosomatostatin at carboxyl- and amino-terminal segments. Characterization of tissue and secreted forms in the rat. J Biol Chem 263:745–751[Abstract/Free Full Text]
15. Brakch N, Rholam M, Nault C, Boileau G, Cohen P 1991 Differential processing of hormone precursor. Independent production of somatostatins 14 and 28 in transfected neuroblastoma 2A cells. FEBS Lett 282:363–367[CrossRef][Medline]
16. Holst JJ, Baldissera FG, Bersani M, Skak-Nielsen T, Seier-Poulsen S, Nielsen OV 1988 Processing and secretion of prosomatostatin by the pig pancreas. Pancreas 3:653–661[Medline]
17. Ensinck J, Laschansky E, Vogel R, Siminowitz D, Roos B, Francis B 1989 Circulating prosomatostatin-derived peptides: differential responses to food ingestion. J Clin Invest 83:1580–1589
18. Ensinck JW, Vogel RE, Laschansky EC, Koerker DJ, Prigeon RL, Kahn SE, D’Alessio DA 1997 Endogenous somatostatin-28 modulates postprandial insulin secretion. Immunoneutralization studies in baboons. J Clin Invest 100:2295–2302[Medline]
19. Aron DC, Andrews PC, Dixon JE, Roos BA 1984 Identification of cellular prosomatostatin and nonsomatostatin peptides derived from its amino terminus. Biochem Biophys Res Commun 124:450–456[CrossRef][Medline]
20. Benoit R, Ling N, Alford B, Guillemin R 1982 Seven peptides derived from pro-somatostatin in rat brain. Biochem Biophys Res Commun 107:944–950[CrossRef][Medline]
21. Low MJ, Otero-Corchon V, Parlow AF, Ramirez JL, Kumar U, Patel YC, Rubinstein M 2001 Somatostatin is required for masculinization of growth hormone-regulated hepatic gene expression but not of somatic growth. J Clin Invest 107:1571–1580[Medline]
22. Francis BH, Baskin DG, Saunders SR, Ensinck JW 1990 Distribution of somatostatin-14 and somatostatin-28 gastrointestinal-pancreatic cells of rats and humans. Gastroenterology 99:1283–1291[Medline]
23. Baskin DG, Ensinck JW 1984 Somatostatin in epithelial cells of intestinal mucosa is present primarily as somatostatin 28. Peptides 5:615–621[CrossRef][Medline]
24. Vinik AI, Gaginella TS, O’Dorisio TM, Shapiro B, Wagner L 1981 The distribution and characterization of somatostatin-like immunoreactivity in epithelial cells, submucosa, and muscle of the rat stomach and intestine. Endocrinology 109:1921–1926[Medline]
25. Patel YC, Wheatley T, Ning C 1981 Multiple forms of immunoreactive somatostatin: comparison of distribution in neural and nonneural tissues and portal plasma of the rat. Endocrinology 109:1943–1949[Medline]
26. Penman E, Wass JA, Butler MG, Penny ES, Price J, Wu P, Rees LH 1983 Distribution and characterisation of immunoreactive somatostatin in human gastrointestinal tract. Regul Pept 7:53–65[CrossRef][Medline]
27. Baldissera FG, Holst JJ, Jensen SL, Krarup T 1985 Distribution and molecular forms of peptides containing somatostatin immunodeterminants in extracts from the entire gastrointestinal tract of man and pig. Biochim Biophys Acta 838:132–143[Medline]
28. Noe BD, Spiess J 1983 Evidence for biosynthesis and differential post-translational proteolytic processing of different (pre)prosomatostatins in pancreatic islets. J Biol Chem 258:1121–1128[Free Full Text]
29. Morrison JH, Benoit R, Magistretti PJ, Bloom FE 1983 Immunohistochemical distribution of pro-somatostatin-related peptides in cerebral cortex. Brain Res 262:344–351[CrossRef][Medline]
30. Lewis DA, Campbell MJ, Morrison JH 1986 An immunohistochemical characterization of somatostatin-28 and somatostatin-281–12 in monkey prefrontal cortex. J Comp Neurol 248:1–18[CrossRef][Medline]
31. Rehfeld JF 1998 The new biology of gastrointestinal hormones. Physiol Rev 78:1087–1108[Abstract/Free Full Text]
32. D’Alessio DA, Rushing PA, Blake K, Ensinck JW, Seeley RA 2000 Suppression of food intake by Somatostatin-28(1–13) a novel gut peptide. Diabetes 49:A260

This article has been cited by other articles:

				J. W. Ensinck, E. C. Laschansky, R. E. Vogel, and D. A. D'Alessio Effect of Ingested Nutrients on the Release of Thrittene into the Human Circulation J. Clin. Endocrinol. Metab., October 1, 2003; 88(10): 4798 - 4804. [Abstract] [Full Text] [PDF]

				K. Yaekura, R. Julyan, B. L. Wicksteed, L. B. Hays, C. Alarcon, S. Sommers, V. Poitout, D. G. Baskin, Y. Wang, L. H. Philipson, et al. Insulin Secretory Deficiency and Glucose Intolerance in Rab3A Null Mice J. Biol. Chem., March 7, 2003; 278(11): 9715 - 9721. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend