0013-7227/04/$15.00/0 Endocrinology 145(11):5129–5140 Printed in U.S.A. Copyright © 2004 by The Endocrine Society doi: 10.1210/en.2004-0912

Identity and Regulation of Stored and Secreted Progastrin-Derived in Sheep

ADRIENNE C. PATERSON, SHARON M. LOCKHART, JOSEPHINE BAKER, GREG NEUMANN, GRAHAM S. BALDWIN, AND ARTHUR SHULKES Department of Surgery, University of Melbourne, Austin Health, Melbourne, Victoria 3084, Australia Downloaded from https://academic.oup.com/endo/article/145/11/5129/2500504 by guest on 25 September 2021

(Amidated and nonamidated progastrin-derived peptides have antral (CTFP, 710 ؎ 62 pmol/liter; Gamide, 211 ؎ 35 pmol/liter distinct biological activities that are mediated by a range of and jugular (CTFP, 308 ؎ 16 pmol/liter; amide, 32 ؎ 3 receptor subtypes. The objective was to determine the nature pmol/liter) veins. Alteration of gastric acidity in sheep by iv of the stored and secreted progastrin-derived peptides and to infusion of a H/K-adenosine triphosphatase inhibitor or so- investigate whether progastrin release is regulated by gastric matostatin or by intragastric infusion of HCl demonstrated acidity. Using an antiserum directed to the C terminus of that the CTFP concentrations changed, although to a lesser progastrin for identification and to monitor purification, C- extent than the changes in circulating gastrin amide. We con- terminal flanking peptides (CTFP) of progastrin (prog76–83, clude that the CTFP of progastrin is the major stored and prog77–83, and prog78–83 in approximately equivalent amounts) circulating species of the gastrin gene, and that it is secreted were isolated and identified from extracts of sheep antrum in a regulated fashion rather than constitutively. Because using ion exchange, HPLC, and mass spectrometry. Only trace full-length progastrin is bioactive, but is only a minor antral amounts of full-length progastrin were present. Progastrin and secreted form, determination of the biological activity of CTFP was the predominant progastrin-derived in the the C-terminal flanking peptides will be important for a com- Sim- plete understanding of gastrin endocrinology. (Endocrinology .[3 ؍ (antrum [progastrin CTFP/gastrin amide (Gamide ilarly, progastrin CTFP was the major circulating form in the 145: 5129–5140, 2004)

ASTRIN, A CLASSICAL gastrointestinal pro- and colorectal in vitro and in vivo (15–18). The re- G duced by antral G cells, is the principal stimulant of ceptors mediating the effects of Ggly and progastrin have not secretion (1, 2). Like many other peptide hor- yet been characterized, but, based on responses to specific mones, gastrin is synthesized as a large precursor molecule receptor antagonists, are distinct from the CCK-2 and CCK-1 (101 amino acids in man and 104 amino acids in most other receptors that interact with amidated gastrin and CCK (6). species) that is converted to progastrin (80 and 83 amino Depending on the extent of posttranslational processing acids in man and sheep, respectively) by cleavage of the and the source (antrum, duodenum, or tumor), the tissues

NH2-terminal signal peptide. Subsequent processing results can contain variable amounts of progastrin, progastrin- in the generation of glycine-extended (Ggly) (G34gly processing intermediates, Ggly, and Gamide (19). Colorectal and G17gly), with the final step being amidation to gastrin carcinomas express progastrin, but processing is attenuated, amide (Gamide; G34amide and G17amide; Fig. 1) (3, 4). so that much of the synthesized progastrin is apparently not Until recently, amidated gastrin was thought to be the only fully processed to the amidated end product (20–22). biologically active form of gastrin. A major reassessment of In and other species, Gamide is generally ac- our understanding of gastrin biology occurred with the rec- cepted to be the predominant gastrin form found in the ognition that the precursor forms and the amidated end antrum, whereas full-length progastrin and Ggly are less product have distinct biological activities mediated by dif- than 5% as abundant as Gamide (13, 23–25). However, an- ferent receptors (5, 6). Gamide, acting through the cholecys- tisera to the carboxyl-terminal sequence of progastrin have tokinin-2 (CCK-2) receptor (previously known as the gas- shown high concentrations of C-terminal flanking peptides trin/CCK-B receptor) is the major hormonal regulator of (CTFP) in approximately equal concentration to Gamide (26, gastric acid secretion (1) and is a mitogen for normal gastric 27). This is true for the and a number of species, epithelium and some gastric cancers in vitro and in vivo (5–8). including the antrum of pig, dog, ferret, and rat (13, 28). In In contrast, the precursor forms and intermediates, such as the human, the predominant form was progastrin75–80;in Ggly, have little direct effect on gastric function, but may other species, it was progastrin75–83 (13, 27). 75 was potentiate the effects of Gamide (9–14). Ggly and progastrin phosphorylated to varying extents in the different species independently stimulate proliferation of the colonic mucosa (13, 27, 28). Fragments lacking serine 75 are present in the antrum of pig and dog, but at much lower concentrations

than the nonapeptide progastrin75–83 (13). Abbreviations: CCK, Cholecystokinin; CTFP, C-terminal flanking There are few studies of the regulation of secretion of peptide; CTI, C-terminal immunoreactivity; Gamide, gastrin amide; Ggly, glycine-extended gastrin; TFA, trifluoroacetic acid. Gamide precursors. The release of Ggly from the antrum Endocrinology is published monthly by The Endocrine Society (http:// appears to be controlled in a similar fashion to Gamide (29, www.endo-society.org), the foremost professional society serving the 30). However, the ratio of Ggly to Gamide in the circulation endocrine community. is higher than that in the antrum, suggesting a differential

5129 5130 Endocrinology, November 2004, 145(11):5129–5140 Paterson et al. • Identity and Regulation of Progastrin-Derived Peptides Downloaded from https://academic.oup.com/endo/article/145/11/5129/2500504 by guest on 25 September 2021

FIG. 1. Structure of sheep progastrin and progastrin-derived peptides and specificity of antibodies. Preprogastrin (104 amino acids) is converted to progastrin (83 amino acids) by removal of the signal peptide. The sequential action of prohormone convertases (endopeptidases and carboxypeptidase B-like ) converts progastrin to glycine-extended forms and CTFP by cleavage at paired basic residues. The C terminus of Ggly is then amidated by peptidyl ␣-amidating monooxygenase. For simplicity, only the 17- forms of Ggly and Gamide are depicted. The epitopes for the different antisera are shown. secretion (24, 31, 32). This may relate to the heterogeneity of a purse string suture into the antral region of the abomasum. Gastric the secretory granules, which contain different proportions juice was collected by first clearing the cannula of any contents, then of progastrin and its processing products (6, 33). The pro- aspirating approximately 5 ml gastric secretion. Cannulas were also inserted into the jugular veins, as described for the conscious sheep cessing of progastrin occurs within the secretory granules studies. Basal samples were not collected until 30 min after the surgery and involves cleavage, , , and ami- was completed. dation. The extent of processing is dependent on residence times and the strength and duration of the stimulation (34). Studies in conscious sheep However, the regulated secretion of progastrin has not been examined in vivo. The studies were performed in Merino-Corriedale-cross sheep, weighing 35–45 kg. After a local anesthetic (2 ml 1% lignocaine) had been Progastrin has recently been demonstrated to have administered, a hypodermic needle was inserted into the jugular vein, growth-promoting effects on normal colonic mucosa and directed toward the head (up). A polyvinyl cannula (inside diameter, colorectal cancers (15, 17, 18). These studies were performed 0.76 mm; outside diameter, 1.65 mm) was then passed through the using full-length progastrin, either as a recombinant peptide needle into the vein. The same procedure was followed for the con- tralateral jugular vein, except that the cannula was directed toward the or overexpressed as a transgene in the liver and secreted heart (down). Agonists and antagonists were infused through the can- constitutively. Although recombinant progastrin is relatively nula directed downward, and blood was collected from the cannula stable in the circulation (35), a comparison of the antral and directed upward. A 24-h recovery period was allowed, and animals were secreted forms of progastrin has not been reported. We, housed in metabolism cages for the duration of the experiments. therefore, determined the structure of the stored and secreted forms of ovine progastrin and compared the regulation of Studies in anesthetized sheep antral secretion of progastrin to that of Gamide. Protocol 1: effect of iv omeprazole and intragastric HCl infusions. After a 30-min basal period, omeprazole was administered via the jugular can- Materials and Methods nula pointing toward the heart as a bolus (0.65 mg/kg), followed im- Animal preparation mediately by an infusion of 0.90 mg/kg⅐h at 20 ml/h for 180 min. During the last hour of omeprazole infusion, 0.1 m HCl was administered via All experiments conformed with the National Health and Medical the gastric cannula to return the pH to between 1.5 and 2.5. This required Research Council of Australia code of practice for the care and use of an average of 120 ml over 30 min. Blood samples were obtained from animals for scientific purposes and were approved by the Austin health the jugular, antral, and fundic cannulas at regular intervals before, animal ethics committee. during, and after the infusion. Gastric pH was also measured at these times. Studies in anesthetized sheep The model for collecting antral and fundic blood has been reported Studies in conscious sheep previously (36). Merino-Corriedale-cross sheep. weighing 35–45 kg, were used. Anesthesia was induced by injection of pentothal into the Protocol 2: effect of omeprazole infusion. After a 30-min basal period, ome- jugular vein and was maintained throughout the experiment with a prazole was administered via the jugular cannula pointing toward the 1–2% fluothane-oxygen mixture (ICI Australia Operations Pty. Ltd., heart as a bolus (0.65 mg/kg), followed immediately by an infusion of Heidelberg, Australia). Using a midline incision, cannulas (inside di- 0.90 mg/kg⅐h at 20 ml/h for 120 min. Blood samples were obtained from ameter, 0.76 mm; outside diameter, 1.65 mm; Dow Corning Corp., Mid- the upward-pointing cannula at regular intervals before, during, and land, MI) were inserted into the gastric vein, draining blood from the after the infusion. fundus, and the gastroepiploic vein, draining blood from the antrum. The vein connecting the antrum to the fundus was ligated to prevent Protocol 3: effect of somatostatin infusion. After a 30-min basal period, contamination of the samples by blood flow from the fundic to the antral somatostatin-14 (5 ␮g/kg⅐h; Bachem, Bubendorf, Switzerland) was in- vein. The was cannulated by the insertion of a catheter through fused via the jugular cannula pointing toward the heart at 20 ml/h for Paterson et al. • Identity and Regulation of Progastrin-Derived Peptides Endocrinology, November 2004, 145(11):5129–5140 5131

120 min. Blood samples were obtained from the upward-pointing can- Amidated gastrin. Tissue extracts, plasma, and chromatography fractions nula at regular intervals before, during, and after the infusion. were measured with antiserum 1296 as previously reported (38). This antiserum detects all amidated C-terminal fragments greater than the pentapeptide and does not cross-react with CCK or Ggly. Laboratory methods Glycine-extended gastrin. The antral tissue extract and sizing chromatog- Tissue extracts. Antral tissue was collected immediately after a lethal dose raphy fractions were assayed with antiserum 109-2 using a previously of barbiturate, washed with ice-cold 0.9% saline, snap-frozen in liquid described RIA (32). This antiserum detects Ggly forms and CCK-glycine, nitrogen, and stored at Ϫ70 C until extraction. Antral mucosae from four but does not cross-react with Gamide (29). sheep were extracted for gastrin peptides using either a two-step boiling Somatostatin. Plasma somatostatin was measured in ethanol-extracted water/acetic acid extraction or an organic solvent extraction. The water/ plasma, as described previously (39). Extracted plasma was assayed acid procedure involved boiling the specimen in 5 vol water for 5 min, against a somatostatin-14 standard curve with antiserum 8402, which followed by mechanical homogenization (Polytron, Brinkmann Instru- detects somatostatin-14 and somatostatin-28 to an equal extent. [125I]Tyr- ments, Lucerne, Switzerland). The sample was boiled for an additional somatostatin-14 was prepared using the chloramine-T method and pu- 5 min, then microfuged at 10,000 rpm for 15 min at 4 C. The supernatant rified by reverse phase HPLC. The 50% inhibitory dose was 8 fmol/ml, Downloaded from https://academic.oup.com/endo/article/145/11/5129/2500504 by guest on 25 September 2021 Ϫ (water extract) was collected and stored at 20 C until assay. The pellet and the inter- and intraassay coefficients of variation were less than 5% was resuspended in 5 vol boiling 3% acetic acid, boiled for 10 min, and and 12%, respectively. centrifuged as before. The supernatant (acid extract) was collected and stored at Ϫ20 C until assay. The organic solvent extract was obtained by Sizing chromatography. Antral plasma and antral tissue extracts were pulverizing frozen tissue with a mortar and pestle on dry ice and ho- chromatographed on a calibrated Sephadex G-50 superfine column mogenizing (Polytron) in 10 vol 80% acetonitrile. Cellular debris was (10 ϫ 1200 mm; Pharmacia Biotech, Uppsala, Sweden) eluted with 0.02 pelleted by centrifugation (20,400 ϫ g, 4 C, 10 min). The supernatant was m veronal, 0.05% BSA, and 0.005% sodium azide, pH 8.7, at4Cataflow collected, and organic solvent was removed by air stream. Particulates rate of 6.3 ml/h in 1-ml fractions. Blue dextran (Pharmacia Biotech) and 125 were removed by centrifugation (3,000 ϫ g, 4 C, 10 min), and the Na- I (ICN, Sydney, Australia) were used to determine void and total supernatant was stored at Ϫ20 C. volumes, respectively. Ion exchange chromatography. Antral extract (16 ml of an organic solvent Concentrating plasma. Antral plasma samples from seven sheep infused extract, containing ϳ5.1 nmol progastrin CTI, in a final concentration of with omeprazole and HCl (protocol 1) were pooled and applied to a 0.02 m triethanolamine, pH 8) was filtered, loaded onto a Pharmacia reverse phase C18 Sep-Pak (Waters Corp., Milford, MA) according to the Mono Q HR 5/5 column equilibrated with 0.02 m triethanolamine, pH manufacturer’s instructions. Briefly, 2 ml plasma were mixed with 4 vol 8, and eluted with a gradient from 0.02 m triethanolamine, pH 8 to 8% 0.05% trifluoroacetic acid (TFA) and 60 ␮l 10% TFA (final pH ϳ3), and acetic acid in 0.02 m triethanolamine, pH 2 over 40 min. passed through an activated reverse phase C Sep-Pak three times. 18 Reverse phase HPLC. A portion of the pooled immunoreactive peaks from Unbound material was removed by washing with 10 ml 0.05% TFA, and ␮ peptide was eluted with 6 ml 80% acetonitrile. Eluates from five Sep- ion exchange chromatography was applied to a C18 Bondapak Radial Paks were combined and lyophilized by air stream. Samples were stored Pak cartridge with a gradient from 0–30% acetonitrile in 0.05% TFA over at Ϫ20 C until RIA and chromatography. 45 min at 1 ml/min. A single immunoreactive peak eluting at 9.3% acetonitrile was recovered and used for mass spectrometry. To compare stored and circulating forms of ovine progastrin CTI, RIAs. The epitopes for the different antisera are shown in Fig. 1. A new organic antral extract containing 77 pmol progastrin CTI was subjected antiserum was generated to measure the C terminus of ovine progastrin. to reverse phase HPLC as described above. Concentrated plasma was The decapeptide (TyrSerAlaGluGluGlyAspGlnHisPro), which corre- reconstituted in 0.05% TFA, filtered (0.45 ␮m), and subjected to reverse sponds to the ovine C-terminal-flanking peptide progastrin (CTFP) 75–83 phase HPLC as described above immediately after the antral extract. with an additional N-terminal tyrosine (37), was custom-synthesized by Chiron (Melbourne, Australia), with sequence and purity (Ͼ95%) de- Mass spectrometry. Electrospray ionization mass spectrometry was per- termined by mass spectrometry and HPLC. This peptide was used for formed on a Sciex API-300 triple quadrupole mass spectrometer conjugation and radiolabeling. For antibody production, the N terminus (PerkinElmer Life Sciences, Melbourne, Australia) fitted with a micro- of the peptide was conjugated with glutaraldehyde to keyhole limpet ion spray ion source (flow rate, 0.2 ␮l/min), previously calibrated to an hemocyanin in a molar ratio of 600:1:6000 (peptide/keyhole limpet accuracy of Ϯ 0.01% using singly charged (polypropylene glycol) ref- hemocyanin/glutaraldehyde). The conjugate was emulsified in erence ions. Portions from the reverse phase HPLC fractions were con- Freund’s complete adjuvant (Sigma-Aldrich Corp., Castle Hill, Austra- centrated 6-fold by freeze-drying. Three to 5 ␮l of the concentrated lia) in a 1:1 ratio (vol/vol), and 2 ml emulsified conjugate (which con- fractions were mixed with acetonitrile/0.2% formic acid (1:1) before tained about 100 nmol CTFP) were injected sc into each of four New analysis. MS/MS spectra (Q3 scans) were obtained using nitrogen col- Zealand White Cross rabbits. For subsequent booster injections (at 6- to lision gas (4 millitorr pressure, 20.7 cm cell length) and optimized col- 8-wk intervals), the conjugate was emulsified using Freund’s incomplete lision energies of 32–64 eV. Signal-averaged raw mass spectra were adjuvant (Sigma-Aldrich Corp.) in a 1:1 ratio (vol/vol). After several analyzed manually and transformed to a true mass scale using the booster injections, one rabbit (no. 152) generated high titer, high affinity PE-Sciex BioMultiview program Biospec Reconstruct (PerkinElmer Life antibodies against the CTFP. There was insignificant (Ͻ0.001%) cross- Sciences). Peptide sequences inferred from observed masses were con- reactivity with Ggly, Gamide, or human Tyr-progastrin71–80 even at firmed by analysis of MS/MS spectra manually and using the BioMul- relatively high concentrations (10 ␮m). Because this antiserum detects tiview programs Predict Sequence and Peptide Fragments (PerkinElmer both intact progastrin and the CTFP of progastrin, the results are ex- Life Sciences). pressed as progastrin C-terminal immunoreactivity (progastrin CTI) for clarity. Assay incubations were made in veronal buffer (1 ml) containing Statistical analyses 0.1% BSA, pH 8.7. The antibody was used at a final dilution of 1:40,000 125 Results are expressed as the mean Ϯ se. The integrated gastrin/ with [ I]Tyr-progastrin75–83, prepared by the chloramine-T method, as the label. The calculated 50% inhibitory dose was 12.2 fmol/tube, and progastrin CTI response was calculated as the area under the concen- the intra- and interassay variations were less than 6% and less than 9% tration-time curves as reported previously (36). Statistical comparison ␮ between more than two groups was made by one-way ANOVA, fol- respectively. Progastrin CTI was measured in tissue extracts (50–100 l Ͻ of a 1:1000 dilution), plasma (50 ␮l), and chromatography fractions lowed by Dunnett’s test. P 0.05 was considered statistically significant. ␮ (50–800 l) against a Tyr-progastrin75–83 standard curve. Solutions of the decapeptide were prepared gravimetrically using a microbalance, Results and their concentrations were checked by absorbance at 280 nm before Antral tissue concentrations use as the standard for the RIA. RIA of various dilutions of antral extracts and plasma samples showed the samples diluted in parallel to the Progastrin CTI, as measured with antiserum 152, was the standard curve. major antral form of progastrin-derived peptide with a mean 5132 Endocrinology, November 2004, 145(11):5129–5140 Paterson et al. • Identity and Regulation of Progastrin-Derived Peptides concentration of 1961 Ϯ 806 pmol/g. The concentration of separation medium, because synthetic ovine C-terminal Ϯ Gamide (antiserum 1296) was about 3-fold less at 735 157 flanking peptide (SAEEGDQHP, prog75– 83; molecular pmol/g, whereas the Ggly concentration (antiserum 109-2) mass, 1132 Da) also eluted in the same position as Ϯ was 74 18 pmol/g, an order of magnitude less than G17amide. Sizing chromatography anomalies such as these Gamide. have also been encountered in previous studies charac- terizing progastrin forms (25, 26, 40). The elution profiles Sizing chromatography of antral progastrin- from both the water and organic solvent extracts were derived peptides identical (data not shown). Antral extracts from several sheep were examined by size exclusion chromatography, and a representative chro- Peptide purification matograph of a water extract is shown in Fig. 2. Com- parison of the peak area of progastrin CTI (Fig. 2A) and The organic solvent antral extract was fractionated on an Downloaded from https://academic.oup.com/endo/article/145/11/5129/2500504 by guest on 25 September 2021 Gamide (Fig. 2B) confirms that progastrin CTI is the ion exchange column. Two immunoreactive peaks were predominant immunoreactive form in the antrum. The identified (Ͼ90% of amount recovered; Fig. 3A). The frac- elution profile indicated that ovine progastrin CTI was tions encompassing these two peaks were pooled and re- considerably smaller than the recombinant human pro- chromatographed by reverse phase HPLC (Fig. 3B). A single gastrin marker. Interestingly, ovine progastrin CTI eluted major immunoreactive peak eluting at 9.3% acetonitrile was in a similar position to G17amide (2098 Da; Fig. 2). How- obtained and subjected to mass spectrometry for molecular ever, this appears to be the result of an interaction with the weight determination and sequence analysis.

FIG. 2. Sizing chromatography of water extract from ovine antrum. Fractions were assayed with antiserum 152 for pro- gastrin CTI (A) and antiserum 1296 for Gamide (B). A similar profile was ob- served with an organic solvent extract (data not shown). The elution positions of human progastrin6–80 (hPG), G17amide (G17), and human and ovine C-terminal flanking peptides (hCTFP and oCTFP), run on separate occasions, are indicated by arrows. Paterson et al. • Identity and Regulation of Progastrin-Derived Peptides Endocrinology, November 2004, 145(11):5129–5140 5133 Downloaded from https://academic.oup.com/endo/article/145/11/5129/2500504 by guest on 25 September 2021

FIG. 3. Purification of ovine progastrin CTI. Organic solvent extract from ovine antrum was subjected to ion exchange chromatography (A). The peak of progastrin CTI (fractions 35–39) was additionally purified by reverse phase HPLC (B), as described in Materials and Methods. The peak from the latter chromatography (fractions 18–20) was used for molecular mass and sequence determination by mass spectrometry. Dashed and dotted lines represent eluant gradient and optical absorbance at 214 nm, respectively.

Mass spectrometry TABLE 1. Mass spectrometry of antral progastrin CTI

The immunoreactive fractions from the reverse phase Peptide Observed Theoretical Progastrin peptidea HPLC were concentrated by freeze-drying. Electrospray ion- sequence massb (Da) massc (Da) Ϯ ization mass spectrometry identified three peaks with mo- Progastrin76–83 AEEGDQHP 881.35 0.3 881.85 Ϯ lecular masses of 681.3, 810.3, and 881.4 Da in approximately Progastrin77–83 EEGDQHP 810.31 0.3 810.78 Progastrin EGDQHP 681.27 Ϯ 0.3 681.66 equal amounts (Table 1). These molecular masses corre- 78–83 a sponded precisely with those expected for three peptides Sequence numbering is based on ovine progastrin1–83. b Monoisotopic mass Ϯ machine uncertainty. (EGDQHP, EEGDQHP, and AEEGDQHP, respectively) de- c Calculated using “compute pl/MW” tool (www.expasy.ch/tools/ rived from the known ovine progastrin sequence (Table 1). pi_tool.html). In the case of the largest peptide, AEEGDQHP, the sequence was confirmed by MS/MS collisional fragmentation. Similar linear applied gradient of 0–50% acetonitrile/0.1% TFA over mass spectrometric results were obtained when antral ex- 150 min). ␮ tracts were purified by chromatography on a C18 Bondapak Radial Pak cartridge (as described in Materials and Methods), Basal concentrations

followed by purification of the immunoreactive peak on a C18 In the anesthetized animals, Gamide concentrations in the microbore column (1 mm ϫ 250 m; 40 ␮l/min flow rate; antral vein were significantly higher than those in the fundic 5134 Endocrinology, November 2004, 145(11):5129–5140 Paterson et al. • Identity and Regulation of Progastrin-Derived Peptides and peripheral circulations (Table 2). Progastrin CTI was also were similar to the stored antral concentrations, with pro- higher in the antral vein than in the peripheral circulation gastrin CTI being the major form. and was by far the major circulating form (Table 2). The progastrin CTI to Gamide ratio increased from 3.4 in the Studies in anesthetized sheep antral circulation to 9.6 in the peripheral circulation, sug- gesting that Gamide was being cleared more rapidly than Protocol 1: effect of iv omeprazole and intragastric HCl infusion. progastrin CTI. Similar results were seen in the peripheral Gamide concentrations in the antral and peripheral circula- circulation of the nonanesthetized animals (Table 2). The tions increased by 30% after2hofomeprazole infusion and Ϯ Ϯ relative amounts of circulating progastrin-derived peptides reached maximums of 291 73 and 48 11 pmol/liter for antral and jugular concentrations, respectively (Fig. 4). In- tragastric HCl rapidly reversed this increase to below basal TABLE 2. Basal plasma concentrations of progastrin CTI and levels, reaching minimums at 180 min of 184 Ϯ 62 and 27 Ϯ Gamide (picomoles per liter) Downloaded from https://academic.oup.com/endo/article/145/11/5129/2500504 by guest on 25 September 2021 5 pmol/liter for antral and jugular concentrations, respec- Antral Fundic Jugular tively. On cessation of both omeprazole and intragastric HCl, vein vein vein there was a large rebound increase in plasma Gamide (Fig. Anesthetised animals (n ϭ 7) 4). Progastrin CTI concentrations also changed in response to Progastrin CTI 710 Ϯ 62 311 Ϯ 15 308 Ϯ 16 the alterations in gastric pH, although not to the same extent Ϯ Ϯ Ϯ Gamide 211 35 24 2323 or as rapidly as plasma Gamide. Thus, as shown in Fig. 5, the Ratio 3.4 13.0 9.6 Conscious animals omeprazole infusion caused a progressive increase in antral Progastrin CTI 403 Ϯ 12 and peripheral progastrin CTI concentrations that was mod- Gamide 26 Ϯ 2 erated by the administration of intragastric HCl. The extent Ratio 15.5 of these changes for Gamide and progastrin CTI are dem- Values are the mean Ϯ SEM (n ϭ 7). onstrated with the integrated outputs, which measure the net

FIG. 4. Effect of iv infusion of omeprazole (0–120 min) and then of omeprazole plus intragastric HCl on Gamide in the antral (A) and jugular (B) veins of anesthetized sheep (n ϭ 7). Paterson et al. • Identity and Regulation of Progastrin-Derived Peptides Endocrinology, November 2004, 145(11):5129–5140 5135 Downloaded from https://academic.oup.com/endo/article/145/11/5129/2500504 by guest on 25 September 2021

FIG. 5. Effect of iv infusion of omeprazole (0–120 min) and then of omeprazole plus intragastric HCl on plasma pro- gastrin CTI in the antral (A) and jugular (B) veins of anes- thetized sheep (n ϭ 7).

changes compared with control, rather than the absolute in antral vein somatostatin from 170 Ϯ 44 to a maximum of changes. These are shown for the antrum only (Fig. 6). Antral 519 Ϯ 114 pmol/liter (Fig. 7). Gamide output in response to the omeprazole-induced in- Ϯ ⅐ crease in gastric pH was 0.87 0.29 pmol/ml min, reaching Studies in conscious sheep a nadir of Ϫ0.56 Ϯ 0.71 pmol/liter after 1 h of intragastric HCl, with a rebound increase after cessation of acid admin- Protocol 2: effect of iv omeprazole. As in the anesthetized ani- istration. Antral progastrin CTI output was higher than an- mals, omeprazole infusion resulted in a sustained increase in tral Gamide, with a maximum output of 1.99 Ϯ 0.57 pmol/ circulating Gamide and progastrin CTI. Although, as noted ml⅐min after 2-h omeprazole infusion. The progastrin CTI previously, basal levels of Gamide and progastrin CTI were output remained at or below this level during intragastric quite different, the net increase was similar for both peptides, acid administration, with a rebound increase to 3.89 Ϯ 1.41 with a maximum increase of about 0.5 pmol/liter⅐min pmol/ml⅐min in the 30 min after cessation of acid adminis- (Fig. 8). tration (Fig. 6). Basal somatostatin was highest in the fundic vein (327 Ϯ Protocol 3: effect of somatostatin infusion. Somatostatin de- 62 pmol/liter), followed by levels in the antral vein (186 Ϯ creased Gamide concentrations, and the decrease was fol- 25 pmol/liter) and peripheral circulation (79 Ϯ 5 pmol/liter). lowed by a rebound increase after the cessation of the infu- Omeprazole with or without concurrent intragastric HCl had sion (Fig. 9A). Somatostatin also appeared to inhibit no effect on jugular or fundic somatostatin concentrations progastrin CTI secretion, although the change was not sta- (data not shown). Omeprazole alone had no effect on antral tistically significant. After the somatostatin infusion was vein somatostatin concentrations. However, omeprazole stopped, there was a significant rebound increase in plasma plus intragastric HCl resulted in a rapid and large increase progastrin CTI (Fig. 9B). 5136 Endocrinology, November 2004, 145(11):5129–5140 Paterson et al. • Identity and Regulation of Progastrin-Derived Peptides Downloaded from https://academic.oup.com/endo/article/145/11/5129/2500504 by guest on 25 September 2021

FIG. 7. Effect of iv infusion of omeprazole (0–120 min) and then of omeprazole plus intragastric HCl on plasma somatostatin in the an- tral vein of anesthetized sheep (n ϭ 7).

FIG. 6. Antral integrated outputs for Gamide (A) and progastrin CTI (B) in response to omeprazole and omeprazole plus intragastric HCl administered to anesthetized sheep (n ϭ 7). *, P Ͻ 0.05 compared with basal output (Ϫ30 to 0 min).

Comparison of stored and circulating forms of ovine progastrin CTI Size exclusion chromatography demonstrated that antral and plasma progastrin CTI eluted with a similar Kav of ap- proximately 0.65 (Fig. 10, A and B). Amidated gastrin from both antrum and plasma eluted in the same position for reverse phase HPLC (Fig. 10, C and D, dotted line), indicating that conditions were the same between consecutive HPLC runs. However, the plasma progastrin CTI had several peaks, two eluting slightly earlier than the stored form (17 and 20 min compared with 22 min) and two eluting later (56 and 66 min; Fig. 10, C and D). Trace amounts of these latter peaks FIG. 8. Peripheral integrated outputs for Gamide (A) and progastrin were also detected in the antral extracts (Fig. 10C). CTI (B) in response to omeprazole infusion to the conscious sheep. *, P Ͻ 0.05 compared with basal output (Ϫ30 to 0 min). Discussion The present report describes the isolation and character- spond to peptides derived from the progastrin C terminus, ization of the predominant progastrin-derived peptides in namely progastrin76–83, progastrin77–83, and progastrin78–83. sheep antrum. The molecular masses and sequence corre- The three peptides were present in approximately equivalent Paterson et al. • Identity and Regulation of Progastrin-Derived Peptides Endocrinology, November 2004, 145(11):5129–5140 5137

only trace amounts of full-length progastrin (13, 25). An increased proportion of intact progastrin is found in gastri- nomas (ϳ40%) (26) and in colorectal cancers (ϳ30%) (22), consistent with attenuated processing in the hypersecretory gastrinoma cell and the nonendocrine gastrin-containing cell in colorectal (5, 19). Nevertheless, even in these cir- cumstances, the most abundant progastrin-derived peptide is the CTFP, not intact progastrin (20, 22, 26). C-Terminal flanking peptides of the related peptide, CCK, which also start with the sequence Ser-Ala-Glu, have been isolated from pig (44) and rat (45, 46) in similar concentrations to

amidated CCK. These peptides were not active in bioassay Downloaded from https://academic.oup.com/endo/article/145/11/5129/2500504 by guest on 25 September 2021 systems that respond to amidated CCK (44, 45). Previous studies of a number of species, including pig, dog, ferret, and rat (13, 28, 47), have reported that the sum of Gamide and its immediate precursor, Ggly, is approxi- mately equal to the CTFP concentration. Because all three peptides are derived from the same precursor, this stoichi- ometry is expected. However, in the human antrum the CTFP concentration is about 40% higher than the Gamide concen- tration, and in gastrinomas the proportion ranges from 0.5- to 18-fold (26). In the current study we found that the CTFP concentration was about 2.5-fold higher than the sum of the concentrations of Gamide and Ggly. The excess of CTFP in man and sheep is not the result of reduced secretion, because the antral vein concentration of CTFP was more than 3-fold higher than Gamide. The presence of nonimmunoreactive metabolites offers a possible explanation for the apparent excess of CTFP, because variable amounts of N-terminal fragments of G17, such as G13, or of G17 extended by Gly- Arg or Gly-Arg-Arg at the C terminus have been reported in the antrum and in gastrinomas (23, 24, 48, 49). Alternative explanations that require testing include the presence of non- immunoreactive breakdown products of Gamide or Ggly, heterogeneity of the secretory granules, differences in pro- FIG. 9. Peripheral integrated outputs for Gamide (A) and progastrin cessing within secretory granules, the influence of phosphor- CTI (B) in response to somatostatin infusion to the conscious sheep. ylation on precursor cleavage, and preferential sorting of Ͻ Ϫ Ͻ *, P 0.05 compared with basal output ( 30 to 0 min); #, P 0.05 CTFP into secretory granules (6, 33, 41). compared with output at 90–120 min. Our findings in the sheep agree with previous reports on the presence of substantial amounts of CTFP in the antrum amounts. A peptide of similar chromatographic character- of other species, but differ on the lengths and relative istics was the major secreted form, and its release was reg- amounts of the CTFPs present. Processing of progastrin in- ulated by gastric pH and somatostatin, suggesting that the volves cleavage after three dibasic sites (Arg36Arg37, CTFP was stored in secretory granules. Indeed, the dibasic Lys53Lys54, and Arg73Arg74). Gly72 then provides the donor residues adjacent to the CTFP appear to be essential for for the C-terminal amidation of gastrin (4, 6). In human, dog, sorting of progastrin into secretory granules and to the reg- pig, and ferret, the major CTFP commences with Ser75 and is ulated secretory pathway (41). present in both phosphorylated and nonphosphorylated Nonamidated forms of gastrin, such as progastrin and forms (13, 26, 28, 47). Smaller amounts of nonserine-contain- Ggly, are now known to be biologically active. Studies with ing peptides are also detected. We found three different transgenic animals overexpressing progastrin or Ggly or af- CTFP peptides: AEEGDQHP (progastrin76–83), EEGDQHP ter infusion of synthetic peptides have shown that nonami- (progastrin77–83), and EGDQHP (progastrin78–83), none of dated gastrins can stimulate colonic proliferation, accelerate which contained Ser75. This appears to be a true species the development of colon cancer, and potentiate gastric acid difference and not an artifact of extraction, because the tis- secretion (9, 15, 16, 42, 43). To render these findings infor- sues were rapidly removed and frozen, and two different mative, it is essential that relevant molecular forms of pro- extraction procedures, namely, boiling water, as used pre- gastrin are examined. However, the studies described above viously (13), and cold organic solvent extraction, were used used full-length progastrin or Ggly. Depending on the assay with the same result. Furthermore, two different purification system, full-length progastrin comprises only 1–10% of gas- procedures were used before mass spectroscopy for molec- trin immunoreactivity in humans (19, 20, 26, 27). As found ular mass and sequence determination. in the present study in sheep, the pig, ferret, and dog have The present study is the first to report that very high 5138 Endocrinology, November 2004, 145(11):5129–5140 Paterson et al. • Identity and Regulation of Progastrin-Derived Peptides Downloaded from https://academic.oup.com/endo/article/145/11/5129/2500504 by guest on 25 September 2021

FIG. 10. Comparison of antral and circulating forms of progastrin CTI by reverse phase HPLC. Organic solvent antral extract (A and C) and concentrated plasma (B and D) were subjected to sizing chromatography (A and B) and reverse phase HPLC (C and D). Fractions were assayed with antiserum 152 for progastrin CTI (solid line) and antiserum 1296 for Gamide (dotted line). The elution positions for sizing chromatography of human progastrin6–80 (hPG), G17amide (G17), and human and ovine C-terminal flanking peptides (hCTFP and oCTFP) are indicated by arrows. concentrations of CTFP are present in the circulation, and from the Gamide-containing granules being more tightly that CTFP release is regulated by similar factors that control regulated. Omeprazole also increased the CTFP concentra- amidated gastrin. Sizing and reverse phase chromatography tion in the conscious animal, showing that the changes were of plasma confirmed that the progastrin C-terminal immu- not related to the anesthetic. The final demonstration of dif- noreactivity was not intact progastrin, but a CTFP of similar ferential regulation of CTFP secretion was that somatostatin chromatographic characteristics to the antral forms. The infusion into conscious sheep reduced CTFP secretion to a CTFP was the predominant circulating progastrin-derived lesser extent than Gamide secretion. After cessation of the peptide. The CTFP to Gamide ratio in the antral vein was somatostatin infusion, there was a rebound increase above about 3:1, whereas in the peripheral circulation it was 10:1, basal concentrations for both gastrin forms resulting from the demonstrating that not only was CTFP being secreted at a removal of the direct inhibitory effect on the coupled higher rate, but apparently it was also more stable than with the still elevated pH (50). Taken together with the pH Gamide after secretion. effects on CTFP secretion, these findings indicate that CTFP Our studies in anesthetized sheep used a previously re- is regulated, but in a different way from Gamide. ported model that allows antral vein concentrations (secre- Although there have been no previous reports on the tion) and peripheral concentrations (equilibrium values) to quantity and regulation of circulating CTFP, antral CTFP is be measured simultaneously (36). An increase in gastric pH known to be influenced by alterations in the gastric luminal by inhibition of the H/K-adenosine triphosphatase with environment. Fasting of rats for 48 h caused a 60% decrease omeprazole increased circulating CTFP, although the re- in CTFP content and decreased the proportion of Ser75 phos- sponse was delayed compared with Gamide. pH reversal by phorylation (28). In dogs with surgically excluded antra, the intragastric HCl prevented any additional increase in CTFP, concentration of CTFP increased 4-fold, and the concentra- but reversed the increase in Gamide. These findings dem- tion of Ggly increased 8-fold, whereas Gamide increased onstrate that CTFP was regulated by gastric pH, but to a only 3-fold compared with control antrum. Intact progastrin lesser extent than Gamide, and suggest that different pop- remained about 1% of the total immunoreactivity, indicating ulations of secretory granules were involved, with secretion that endopeptidase cleavage of progastrin was normal, Paterson et al. • Identity and Regulation of Progastrin-Derived Peptides Endocrinology, November 2004, 145(11):5129–5140 5139 whereas conversion to Gamide was suppressed (40). The 16. Koh TJ, Dockray GJ, Varro A, Cahill RJ, Dangler CA, Fox JG, Wang TC 1999 available data imply that the processing of progastrin is Overexpression of glycine-extended gastrin in transgenic mice results in in- creased colonic proliferation. J Clin Invest 103:1119–1126 regulated, but the initial endopeptidase cleavage of progas- 17. Baldwin GS, Hollande F, Yang Z, Karelina Y, Paterson A, Strang R, Fourmy trin is virtually complete and is not a rate-limiting step. D, Neumann G, Shulkes A 2001 Biologically active recombinant human There are few reports on the biological activity of the progastrin(6–80) contains a tightly bound calcium ion. J Biol Chem 276:7791– 7796 CTFP. Short-term (5-min) CTFP infusions did not alter acid 18. Singh P, Lu X, Cobb S, Miller BT, Tarasova N, Varro A, Owlia A 2003 secretion or modify the acid response to Gamide (13). CTFP Progastrin1–80 stimulates growth of intestinal epithelial cells in vitro via stimulated histamine release, but at 1% the potency of high-affinity binding sites. Am J Physiol 284:G328–G339 19. Rehfeld JF, van Solinge WW 1994 The tumor biology of gastrin and chole- Gamide (14). However, by analogy with Ggly, long-term cystokinin. Adv Cancer Res 63:295–347 administration may be required to demonstrate a biologi- 20. Ciccotosto GD, McLeish A, Hardy KJ, Shulkes A 1995 Expression, processing, cally relevant effect (9, 10, 12, 51). The long-term effects of and secretion of gastrin in patients with colorectal carcinoma. Gastroenterol- ogy 109:1142–1153

CTFP on gastric acidity and gastrointestinal mucosal prolif- 21. Van Solinge WW, Nielsen FC, Friis-Hansen L, Falkmer UG, Rehfeld JF 1993 Downloaded from https://academic.oup.com/endo/article/145/11/5129/2500504 by guest on 25 September 2021 eration, either alone or in combination with other gastrin Expression but incomplete maturation of progastrin in colorectal carcinomas. peptides, have not been reported. Because CTFP is the major Gastroenterology 104:1099–1107 22. Nemeth J, Taylor B, Pauwels S, Varro A, Dockray GJ 1993 Identification stored and secreted form of progastrin-derived peptide, and of progastrin derived peptides in colorectal carcinoma extracts. Gut 34: nonamidated forms of gastrin have been implicated in the 90–95 acceleration of gastric and development (43, 23. Del Valle J, Sugano K, Yamada T 1987 Progastrin and its glycine-extended posttranslational processing intermediates in human gastrointestinal tissues. 52), studies of the biological activity of CTFP will be of great Gastroenterology 92:1908–1912 interest. 24. Jensen S, Borch K, Hilsted L, Rehfeld JF 1989 Progastrin processing during antral G-cell hypersecretion in humans. Gastroenterology 96: 1063–1070 Acknowledgments 25. Desmond H, Dockray GJ, Spurdens M 1985 Identification by specific radio- immunoassay of two novel peptides derived from the C-terminus of porcine We thank Ms. Lisa Hogan for expert technical assistance. Antibody preprogastrin. Regul Pept 11:133–142 1296, raised against gastrin amide, 109–2 (gastrin gly), and 8402 (so- 26. Pauwels S, Desmond H, Dimaline R, Dockray GJ 1986 Identification of matostatin), was provided by CURE/Digestive Diseases Research Cen- progastrin in gastrinomas, antrum, and duodenum by a novel radioimmu- ter, Antibody/RIA Core. noassay. J Clin Invest 77:376–381 27. Varro A, Desmond H, Pauwels S, Gregory H, Young J, Dockray GJ 1988 The Received July 15, 2004. Accepted August 6, 2004. human gastrin precursor. Characterization of phosphorylated forms and frag- Address all correspondence and requests for reprints to: Dr. Arthur ments. Biochem J 256:951–957 28. Varro A, Nemeth J, Bridson J, Lee C, Moore S, Dockray GJ 1990 Processing Shulkes, Department of Surgery, University of Melbourne, Austin Health, of the gastrin precursor. Modulation of phosphorylated, sulfated, and ami- Heidelberg, Victoria 3084, Australia. E-mail: [email protected]. dated products. J Biol Chem 265:21476–21481 This work was supported by the National Health and Medical Research 29. Azuma T, Taggart RT, Walsh JH 1987 Effects of on the release of Council of Australia, the Austin Hospital Medical Research Foundation, glycine-extended progastrin (gastrin G) in rat antral tissue culture. Gastroen- and National Institutes of Health Grant DK-41301. terology 93:322–329 30. Sugano K, Park J, Dobbins WO, Yamada T1987 Glycine-extended progastrin processing intermediates: accumulation and cosecretion with gastrin. Am J References Physiol G502–G507 1. Hersey SJ, Sachs G 1995 Gastric acid secretion. Physiol Rev 75:155–189 31. Hilsted L, Hansen CP 1988 Corelease of amidated and glycine-extended antral 2. Walsh JH 1994 Gastrin. In: Walsh JH, Dockray GJ, eds. Gut peptides. New gastrins after a meal. Am J Physiol 255:G665–G669 York: Raven Press; 75–122 32. Ciccotosto GD, Shulkes A 1992 Pharmacokinetics and specific metab- 3. Dockray GJ, Varro A, Dimaline R 1996 Gastric endocrine cells: gene expres- olism of glycine-extended and amidated gastrin in sheep. Am J Physiol 263: sion, processing, and targeting of active products. Physiol Rev 76:767–798 G802–G809 4. Rehfeld JF 1998 The new biology of gastrointestinal . Physiol Rev 33. Rahier J, Pauwels S, Dockray GJ 1987 Biosynthesis of gastrin. Localization of 78:1087–1108 the precursor and peptide products using electron microscopic-immunogold 5. Baldwin GS, Shulkes A 1998 Gastrin, gastrin receptors and colorectal carci- methods. Gastroenterology 92:1146–1152 noma. Gut 42:581–584 34. Dockray GJ 1999 Topical review. Gastrin and gastric epithelial . 6. Dockray GJ, Varro A, Dimaline R, Wang T 2001 The gastrins: their production J Physiol 518:315–324 and biological activities. Annu Rev Physiol 63:119–139 35. Paterson AC, Baldwin GS, Shulkes A 2002 Metabolism of recombinant pro- 7. Iwase K, Evers BM, Hellmich MR, Guo Y, Higashide S, Kim HJ, Townsend gastrin in sheep. Am J Physiol 283:E449–E456 CMJ 1997 Regulation of growth of human gastric cancer by gastrin and 36. Zavros Y, Fleming WR, Hardy KJ, Shulkes A 1998 Regulation of fundic glycine-extended progastrin. Gastroenterology 113:782–790 and antral somatostatin secretion by CCK and gastrin. Am J Physiol 274: 8. Koh TJ, Chen D 2000 Gastrin as a growth factor in the . G742–G750 Regul Pept 93:37–44 37. Moore C, Jie R, Shulkes A, Baldwin GS 1997 Molecular cloning and sequence 9. Chen D, Zhao CM, Dockray GJ, Varro A, Van Hoek A, Sinclair NF, Wang of the ovine gastrin gene. DNA Seq 8:39–44 TC, Koh TJ 2000 Glycine-extended gastrin synergizes with gastrin 17 to stim- 38. Shulkes A, Read M 1991 Regulation of somatostatin secretion by gastrin- and ulate acid secretion in gastrin-deficient mice. Gastroenterology 119:756–765 acid-dependent mechanisms. Endocrinology 129:2329–2334 10. Dickinson CJ, Marino L, Yamada T 1990 Inhibition of the a amidation of 39. Read MA, Read DM, Kapuscinski M, Shulkes A 1992 Achlorhydria induced gastrin; effects on gastric acid secretion. Am J Physiol 258:G810–G814 ϩ ϩ changes in gastrin, somatostatin, H /K -ATPase and carbonic anhydrase in 11. Palnaes Hansen C, Stadil F, Rehfeld JF 1996 Metabolism and influence of the sheep. Regul Pept 40:13–27 glycine-extended gastrin on gastric acid secretion in man. Digestion 57:22–29 40. Varro A, Nemeth J, Bridson J, Lonovics J, Dockray GJ 1990 Modulation of 12. Higashide S, Gomez G, Greeley Jr GH, Townsend Jr CM, Thompson JC 1996 Glycine-extended gastrin potentiates gastrin-stimulated gastric acid secretion posttranslational processing of gastrin precursor in dogs. Am J Physiol 258: in rats. Am J Physiol 270:G220–G224 G904–G909 13. Desmond H, Varro A, Young J, Gregory H, Nemeth J, Dockray GJ 1989 The 41. Bundgaard JR, Birkedal H, Rehfeld JF 2004 Progastrin is directed to the constitution and properties of phosphorylated and unphosphorylated C- regulated secretory pathway by synergistically acting basic and acidic motifs. terminal fragments of progastrin from dog and ferret antrum. Regul Pept J Biol Chem 279:5488–5493 25:223–233 42. Aly A, Shulkes A, Baldwin GS 2001 Short term infusion of glycine-extended 14. Sandvik AK, Dockray GJ 1999 Biological activity of carboxy-terminal gastrin gastrin17 stimulates both proliferation and formation of aberrant crypt foci in analogs. Eur J Pharmacol 364:199–203 rat colonic mucosa. Int J Cancer 94:307–313 15. Wang TC, Koh TJ, Varro A, Cahill RJ, Dangler CA, Fox JG, Dockray GJ 1996 43. Singh P, Velasco M, Given R, Varro A, Wang TC 2000 Progastrin expression Processing and proliferative effects of human progastrin in transgenic mice. predisposes mice to colon carcinomas and adenomas in response to a chemical J Clin Invest 98:1918–1929 carcinogen. Gastroenterology 119:162–171 5140 Endocrinology, November 2004, 145(11):5129–5140 Paterson et al. • Identity and Regulation of Progastrin-Derived Peptides

44. Eng J, Gubler U, Raufman JP, Chang M, Hulmes JD, Pan YC, Yalow RS 1986 48. Power DM, Dimaline R, Balaspiri L, Dockray GJ 1988 A novel gastrin- Cholecystokinin-associated COOH-terminal peptides are fully sulfated in pig processing pathway in mammalian antrum. Biochim Biophys Acta 954:141–147 brain. Proc Natl Acad Sci USA 83:2832–2835 49. Kothary PC, Mahoney WC, Vinik AI 1987 Identification of gastrin molecular 45. Varro A, Young J, Gregory H, Cseh J, Dockray GJ 1986 Isolation, structure and variants in gastrinoma syndrome. Regul Pept 17:71–84 properties of the C-terminal flanking peptide of preprocholecystokinin from 50. Shulkes A, Hardy KJ 1982 Effect of somatostatin on basal concentrations of rat brain. FEBS Lett 204:386–390 gastrin and pancreatic polypeptide in the fetal sheep. Biol Neonate 42:249–256 46. Adrian TE, Domin J, Bacarese-Hamilton AJ, Bloom SR 1986 Is the C- 51. Kaise M, Muraoka A, Seva C, Takeda H, Dickinson CJ, Yamada T 1995 Glycine- terminal flanking peptide of rat cholecystokinin double sulphated? FEBS extended progastrin processing intermediates induce Hϩ,Kϩ-ATPase ␣-subunit Lett 196:5– 8 gene expression through a novel receptor. J Biol Chem 270:11155–11160 47. Dockray GJ, Varro A, Desmond H, Young J, Gregory H, Gregory RA 1987 52. Fox JG, Rogers AB, Ihrig M, Taylor NS, Whary MT, Dockray G, Varro A, Post-translational processing of the porcine gastrin precursor by phosphory- Wang TC 2003 Helicobacter pylori-associated gastric cancer in INS-GAS mice is lation of the COOH-terminal fragment. J Biol Chem 262:8643–8647 gender specific. Cancer Res 63:942–950

Endocrinology is published monthly by The Endocrine Society (http://www.endo-society.org), the foremost professional society serving the endocrine community. Downloaded from https://academic.oup.com/endo/article/145/11/5129/2500504 by guest on 25 September 2021