- Email: [email protected]
Antral content, secretion and peripheral metabolism of N-terminal progastrin fragments
Regulatory Peptides 133 (2006) 47 – 53 www.elsevier.com/locate/regpep
Antral content, secretion and peripheral metabolism of N-terminal progastrin fragments Jens Peter Goetze a,*, Carsten Palnæs Hansen b, Jens F. Rehfeld a a
Department of Clinical Biochemistry, Rigshospitalet, University of Copenhagen, DK-2100 Copenhagen, Denmark b Gastrointestinal Surgery, Rigshospitalet, University of Copenhagen, DK-2100 Copenhagen, Denmark Received 4 July 2005; received in revised form 4 September 2005; accepted 9 September 2005 Available online 5 October 2005
Abstract Objectives: In addition to the acid-stimulatory gastrins, progastrin also release N-terminal fragments. In order to examine the cellular content, secretion and peripheral metabolism of these fragments, we developed an immunoassay specific for the N-terminal sequence of human progastrin. Results: The concentration of N-terminal progastrin fragments in human antral tissue was 6.7 nmol/g tissue (n = 5), which was only half of that of acid-stimulatory gastrins (12 nmol/g tissue). Gel chromatography of antral extracts showed that the progastrin fragment 1 – 35 and 1 – 19 constitute the major part of the N-terminal progastrin fragments. The basal concentration of N-terminal fragments in normal human plasma was almost 30fold higher than that of the amidated, acid-stimulatory gastrins (286 pmol/l versus 9.8 pmol/l, n = 26, P < 0.001). In contrast, the concentration of N-terminal fragments in hypergastrinemic plasma was only 2.7-fold higher than the concentration of amidated gastrins (540 pmol vs. 198 pmol/l, P = 0.02). During meal stimulation, the plasma concentrations of N-terminal progastrin fragments and amidated gastrins increased in a correlated manner (r = 0.97, P = 0.005). The half life for progastrin 1 – 35 in circulation was 30 min, and a pig model revealed the kidneys and the vasculature to the head as the primary sites of degradation. Conclusion: The cellular and circulatory concentration profiles of N-terminal progastrin fragments differ markedly from those of the acidstimulatory gastrins. The high basal plasma concentrations of N-terminal progastrin fragments cannot be explained by differences in elimination. D 2005 Elsevier B.V. All rights reserved. Keywords: Elimination; Gastrin; Hypergastrinemia; Progastrin; Radioimmunoassay; Secretion
1. Introduction Gastrin is the major regulator of gastric acid secretion and gastric mucosal growth [1] Gastrin is synthesized in antroduodenal G-cells by elaborate posttranslational processing into acid-stimulatory peptides of which gastrin-17 and gastrin-34 are the most abundant ([2 –6], for review see Ref. [7]). The bioactive site of the gastrins is the evolutionary conserved Cterminal tetrapeptide amide, Trp – Met – Asp – Phe – NH2. In addition to the acid stimulatory gastrins, progastrin also release other fragments. The N-terminal fragments have for instance been identified in extracts from gastrinoma tissue [8– 10], and in normal human antral tissue [11]. Hence, in addition to the large amidated gastrin-71, fragments 1 –35, 6– 35, and 20 –35 from the N-terminal part of progastrin were also present in G* Corresponding author. Tel.: +45 3545 4640; fax: +45 3545 8323. E-mail address: [email protected] (J.P. Goetze). 0167-0115/$ - see front matter D 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.regpep.2005.09.016
cells with fragment 1 – 35 being the most abundant after purification [11]. So far, the biology of N-terminal progastrin fragments has not been examined. But recently, we reported that the Nterminal fragment 1– 35 is not required for cellular sorting to the secretory pathway and subsequent regulated secretion of gastrins [12]. Other reports have suggested that the intact progastrin molecule in itself can stimulate colonic cell growth [13 – 17]. However, the active site within progastrin has not always been defined, which hampers a molecular interpretation of the observed effects. Furthermore, the normal and pathophysiological occurrence of the N-terminal progastrin derivatives is virtually unknown. In order to examine the secretion and metabolism of N-terminal progastrin fragments, we therefore developed a sensitive and specific radioimmunoassay for the N-terminus of human progastrin. Our results reveal a surprisingly different cellular and circulatory profile compared to gastrin.
48
J.P. Goetze et al. / Regulatory Peptides 133 (2006) 47 – 53
2. Methods and materials 2.1. Peptides Human progastrin fragments 1– 35, 1 –19 and 6 – 35 were custom synthesized by Cambridge Research Biochemicals Ltd (Cheshire, UK). Also, progastrin 1 –10 extended C-terminally with a tyrosyl residue and fragment 1 – 10 extended Cterminally with cysteine were synthesized for tracer and standard, and for immunization, respectively. The structure and identities of all synthetic peptides were verified by reversed-phase HPLC, amino acid analysis and mass spectrometry before use. 2.2. Radioimmunoassay A radioimmunoassay directed against sequence 1 –10 of human progastrin was developed using 10 mg of the 1 –10 fragment extended C-terminally with cysteine and coupled to 20 mg bovine serum albumin using the m-maleimidobenzoylN-hydroxysuccinimide ester conjugation method [18]. The coupled product was dissolved in 15 ml of distilled water (conjugate solution). The antigen solution (2 ml) was mixed with 2.5 ml isotonic saline and emulsified with an equal volume of complete Freund’s adjuvant (The Danish Serum Institute, Copenhagen, Denmark) and used for the first immunization. For booster injections, 1 ml of the antigen solution was mixed with 4 ml of saline and an equal volume of incomplete Freund’s adjuvant. Eight random-bred white Danish rabbits were immunized subcutaneously over the lower back at 8-week intervals. Twenty millimeter of blood was collected from an ear vein 14 days after each immunization, and the serum was stored at 20 -C for evaluation. For tracer use, the tyrosine-extended 1 – 10 fragment (4.5 nmol) was monoiodinated using the chloramine-T method as previously described [19] and purified on reversed-phase HPLC (Pierce C8 column, 4.6 220 mm) eluted by a linear ethanol gradient (5 – 25%) in 1% trifluoroacetic acid. The gradient was selected to ensure separation of the non-labeled peptide from the iodinated tracer. The specific tracer radioactivity was determined by self-displacement [20]. For measurement of gastrins, we used a radioimmunoassay based on antiserum no. 2604, which is specific for the C-terminal octapeptide sequence of the aamidated gastrins [21]. This assay measures all amidated gastrins with equimolar potency and the reactivity with the related hormone cholecystokinin is negligible. 2.3. Human antral tissue Tissue biopsies of human antral mucosa (n = 5) were obtained from patients undergoing surgery for pancreatic cancer (Whipple’s operation). The use of human tissue was approved by the local ethics committee (KF01-352/96) and informed consent was obtained from all patients. The tissue biopsies were immediately frozen in liquid nitrogen and stored at 80 -C until extraction. While frozen, the tissues were
sliced into small pieces and immersed in boiling water (10 ml/g tissue) for 20 min, homogenized and centrifuged at 10000 g for 30 min. The supernatant was stored and the pellet extracted in 0.5 mol/l acetic acid (10 ml/g tissue), homogenized and centrifuged as above. 2.4. Gel filtration chromatography The molecular composition of N-terminal progastrin fragments in tissue was examined by gel chromatography (Sephadex G-50 Superfine, 10 1000 mm, Pharmacia, Sweden). Columns were calibrated with synthetic progastrin 1– 35 and 1 –19. A barbital buffer containing 0.1% bovine albumin was used as eluent at a flow rate of 4 ml/h (4 -C). The buffer was supplemented with 0.05 M NaCl to avoid protein binding. Void and total volumes were determined by eluting 125Ialbumin and 22NaCl, respectively. 2.5. Plasma Blood samples were collected from 26 healthy and fasting volunteers. Six of the volunteers were later served at proteinrich meal, and blood samples were obtained from a cubital vein at set time intervals ( 35, 5, 0, 10, 20, 30, 40, 50, 70, 95 min). Plasma was stored at 20 -C until analysis. In addition, plasma from 48 patients with hypergastrinemia due to gastrinoma or achlorhydria was collected from the local gastrin analysis facility in our department. 2.6. Elimination of progastrin 1– 35 in humans Synthetic progastrin 1 – 35 (100 pmol/kg) was injected intravenously as a bolus into fasting, healthy subjects (n = 6). Blood samples were collected from the opposite arm (0, 2, 4, 6, 8, 10, 15, 20, 25, 30, 40, 50 and 60 min after injection, and then every 15 min until 180 min from the time of bolus injection). After plasma measurement, the pharmacokinetics was analyzed according to a two-compartment open model: C t = Ae at + Be bt , where A and B is the zero intercepts with the ordinate of the individual exponential terms and a and b are the slopes of the curves. 2.7. Organ extraction of progastrin 1 –35 in pigs To further elucidate the metabolic fate of progastrin 1– 35, we used a previously described animal model [22]. Eight pigs (Danish landrace – Yorkshire breed, 30 –40 kg) were anesthetized and intubated. Isotonic saline was infused at 10 ml/kg/ h during the whole experiment, and donor blood from siblings was additionally infused in volumes equal to that removed during sampling. Cardiac output was measured regularly with a Swan – Ganz thermodilution catheter (Swan – Ganz pediatric 5F; Baxter Health Care, CA, USA). Polyethylene catheters for blood sampling were placed in the thoracic aorta via the left carotid artery and cephalically in the left internal jugular vein. Via the external jugular veins, two angiography catheters were introduced into a major hepatic vein and the left renal vein.
J.P. Goetze et al. / Regulatory Peptides 133 (2006) 47 – 53
Blood from a peripheral lung artery was drawn from the Swan – Ganz catheter. The positions of all catheters were verified on fluoroscopy. Of note, neither the hepatic nor the pulmonary catheter was wedged to avoid mixing with portal and pulmonary venous blood when collecting blood samples. After a midline laparotomy, a catheter was placed at the entrance of the portal vein via the splenic vein and another catheter was positioned in one of the major tributaries of the mesenteric vein draining the small intestine. Blood from a femoral vein was drawn from an inserted catheter after exposure of the vessels in the groin. Progastrin 1– 35 and indicator substance were infused into catheters with their tips placed in the right atrium. After completion of the surgical procedure, a 1 h stabilization period was allowed before starting the experiments. Bloodflow in the renal, femoral and portal vein was measured with an electromagnetic flowmeter (Nycotron, Oslo, Norway) and pulmonary flow was regarded as equivalent to cardiac output. Blood flow was converted to plasma flow based on the hematocrit. Hepatic plasma flow was measured by continuous infusion of indocyanine green (Cardiogreen, Becton and Dickinson, MD, USA) at 0.130 Amol/min. After calibration, simultaneous samples from the aorta and the hepatic vein were drawn at set time intervals. The concentration of indocyanine in plasma was measured spectrophotometrically, and hepatic plasma flow was calculated according to Fick’s principle. The elimination of progastrin 1 – 35 was determined as the concentration gradient across the vascular beds during infusion of 100 pmol/kg/h of the peptide. Arterial samples were taken from the aorta every 15 min during infusion. After 90 min, venous samples were drawn every 15 min for 1 h from a lung artery and the femoral, renal, portal, hepatic, internal jugular, and cranial mesenteric veins. The fractional extraction was calculated as: E R = (C i C o) / C i, where C i and C o are the inflow and outflow concentrations in plasma, respectively. 2.8. Elimination of progastrin 1 –35 in pigs When the infusion of progastrin 1 –35 was terminated after 170 min, aortic blood was sampled at regular intervals for 90 min to determine the half-life in circulation. A two-compartment model was used to determine the half-lives as above, but as data were obtained after a continuous infusion, the values of A and B were calculated as: A ¼ RðaT Þ= 1 eaT ; B ¼ S ðbT Þ= 1 ebT where R and S represent the intercepts with the ordinate at the start of the post-infusional period, and T is the period of infusion [23].
49
samples. Correlations were performed on logarithmically transformed data prior to regression analysis. A two-tailed level for significance was set a 0.05. 3. Results 3.1. Radioiodination The incorporation of 125I in the tyrosine-extended progastrin 1– 10 varied between 60 –95%. The non-specific binding was <5% and labeled tracer was completely separated from the non-labeled peptide. Dilution curves for labeled and unlabeled antigen were parallel. The specific radioactivity of the iodinated tracer peptide was 1.08 Ci/Amol. 3.2. Antiserum evaluation The characteristics of the three best antisera are shown in Table 1. Antiserum no. 94023 demonstrated the highest avidity expressed by the effective equilibrium constant, K 0eff = 0.39 1012 l/mol [24]. This antiserum was therefore chosen for further characterization and measurements of gel filtration fractions, tissue extracts and plasma samples. The index of heterogeneity according to Sips was 1.02, indicating that the ligand binding is highly homogenous and that the diluted antiserum acts like a solution of monoclonal antibodies [25]. A high sensitivity, expressed as the analytical detection limit, was achieved for several antisera (Table 1). Specificity expressed as the ID50 of the N-terminal truncated progastrin 6 –35 peptide compared to the ID50 of progastrin 1– 10 Tyr was found to be < 0.0001. In contrast, the binding of both progastrin 1 –19 and 1 –35 did not differ from the binding of progastrin 1 – 10 Tyr. Antiserum no. 94023 was also tested for reactivity with N-terminal progastrin fragments in porcine antral extracts. Notably, the N-terminus of porcine progastrin is homologous to the human sequence at the first 4 N-terminal amino acid residues but with substitutions in position 5 and 6 (Arg–Ser to Gly– Phe, Fig. 1). No binding to porcine fragments could be demonstrated. The FASTA computer program was then used to search available databases for amino acid sequences resembling the N-terminal sequence 1 –10 of human progastrin. No relevant sequences other than the N-terminal progastrin regions from other mammals were found. Finally, dilution curves for tissue extracts (antral mucosa) were parallel with the calibrator Table 1 Antiserum characteristics Antiserum no. (rabbit and bleeding)
Titera (103)
K 0effb 1012 (l/mol)
ac
Detection limit (pmol/l)
2.9. Statistical analysis
94023/4 94024/4 94027/4
570 450 160
0.39 1.32 2.40
1.02 0.80 0.83
0.07 0.04 0.04
Results are expressed as means T SE, except for the measurements in hypergastrinemic plasma with a non-Gaussian distribution. Comparative analyses were performed either by a paired Student’s t-test or the Wilcoxon’s test for paired
a: Antiserum dilution at which 35% of the tracer is bound at equilibrium. b: Effective equilibrium constant according to Ekins and Newman [24]. c: Index of heterogeneity according to Sips [25]. d: Detection limit with ten replicates of 0-calibrator, one-sided 95% confidence interval.
50
J.P. Goetze et al. / Regulatory Peptides 133 (2006) 47 – 53
Fig. 1. Amino acid sequence of N-terminal progastrin 1 – 10 in mammals. Note the homology of the first N-terminal residues, and the basic arginyl residue in position 5.
curves (data not shown). The recovery of progastrin 1– 35 in plasma kept at room temperature was unaffected after 8 h but reduced to 27 T 3% after 24 h. 3.3. N-terminal progastrin fragments in human antrum The total concentration of N-terminal progastrin immunoreactivity in antral mucosa was lower than that of the amidated gastrins with only ¨60% N-terminal progastrin fragments compared to the gastrin contents (Table 2). Gel chromatography revealed distinct elution peaks corresponding to progastrin fragments 1 – 35 and 1– 19 (Fig. 2). Also, a larger molecular form was detected, which may correspond to intact progastrin.
Fig. 2. Gel chromatography of human antral tissue. The positions of the calibrator peptides, progastrin fragments 1 – 35 and 1 – 19, were determined in separate experiments. The first immunoreactive peak (K D ¨ 0.25) presumably corresponds to intact progastrin.
correlation between the N-terminal progastrin and the gastrin concentrations.
3.4. N-terminal progastrin fragments in normal plasma
3.6. Elimination of progastrin fragment 1 – 35 in man
The basal concentration of N-terminal progastrin immunoreactivity in healthy individuals was 285 T 20 pmol/l, which was almost 30-fold higher than the concentration of amidated gastrins (9.8 T 0.8 pmol/l, P < 0.0001). Meal stimulation increased both N-terminal progastrin fragments and amidated gastrins (r = 0.97, P = 0.005) with gastrin concentrations reaching a plateau 30 min after stimulation. The concentration of Nterminal progastrin fragments remained elevated throughout the entire time period studied (90 min) after stimulation (Fig. 3). Gel chromatography of pooled plasma 30 min after meal stimulation revealed only one peak corresponding to the calibration position of progastrin fragment 1 – 35 (data not shown).
The elimination of progastrin 1 – 35 after a bolus injection (112 T 12 pmol/kg) was biexponential with half-lives: T 1 / 2a = 7.0 T 1.2 and T 1 / 2b = 30.5 T 8.2 min (Fig. 5). Clearance and V dss (apparent volume of distribution at steady state) were
3.5. N-terminal progastrin fragments in hypergastrinemic plasma The N-terminal progastrin concentration in hypergastrinemic plasma was 2.7-fold higher than the gastrin concentration (median 198 pmol/l (range 18 –37000) vs. 540 pmol/l (130 – 2070), P = 0.02, Fig. 4). Of note, there was no significant Table 2 N-terminal progastrin and amidated gastrin in human antral tissue extracts Antrum no. N-terminal progastrin Amidated gastrin [N-terminal progastrin] (pmol/g tissue) (pmol/g tissue) [gastrin] 1 2 3 4 5 Mean
5718 7061 14411 3871 2630 6738
9813 10948 28890 6324 4320 12059
0.58 0.64 0.50 0.61 0.61 0.59
Fig. 3. N-terminal progastrin fragments (upper panel) and gastrin concentrations (lower panel) in normal subjects (n = 26) during a protein-rich meal. The results are shown as means T SEM.
J.P. Goetze et al. / Regulatory Peptides 133 (2006) 47 – 53
51
Table 3 Fractional extraction of progastrin fragment 1 – 35
Kidney Liver Gut Head Hindlimb Lungs
Progastrin 1 – 35
Gastrin-17
0.16 T 0.03* 0.04 T 0.04 0.03 T 0.01 0.11 T 0.02* 0.02 T 0.01 0.03 T 0.02
0.40 T 0.04* 0.07 T 0.03 0.13 T 0.03* 0.32 T 0.04* 0.42 T 0.04* 0.01 T 0.03
Values are means T SEM; n = 8, *P < 0.05. The gastrin-17 results have been published previously [22].
3.7. Organ extraction of progastrin fragment 1– 35 in pigs
Fig. 4. N-terminal progastrin fragments and amidated gastrin concentrations in hypergastrinemic plasma. The concentrations were transformed (Log10) prior to plotting and statistical analysis. The horizontal lines represent median concentrations.
calculated to 6.8 T 0.6 ml/kg per min and 126.0 T 1.3 ml/kg. Analysis according to a two-compartment model revealed that the fractions of elimination during the two phases, f a and f b , A=a B=b expressed as: fa ¼ AU C and fb ¼ AU C , were 0.29 and 0.71, indicating that most of the peptide was eliminated in the hphase.
The infused dose rate of human progastrin fragment 1– 35 was 52 T 6 pmol/kg/h. Extraction of the peptide was recorded over the kidney (metabolic clearance rate, MCR = 0.9 T 0.2 ml/ kg/min) and head (MCR = 1.1 T 0.1 ml/kg/min), whereas no significant extraction was observed across the lungs, liver, gut or hind limb (Table 3). The elimination kinetics of progastrin 1– 35 in pigs was then calculated from post-infusion data with biexponential half-lives: T 1 / 2a = 2.7 T 0.7 and T 1 / 2b = 35.1 T1.9 min. The total MCR and the V dss were 3.7 T 0.2 ml/kg/min and 130.8 T 12.7 ml/kg, respectively. The fractions of elimination were f a = 0.37 and f b = 0.63, implying that total progastrin 1– 35 elimination in pig occurred mainly during the h-phase. Gel chromatography on pooled blood after the infusion period (t 135) showed identical elution profiles from all organ veins, and only immunoreactivity eluting in a position corresponding to progastrin 1– 35 was detected (data not shown). 4. Discussion
Fig. 5. Elimination of progastrin fragment 1 – 35 in humans. A bolus dose (100 pmol/kg) was intravenously injected into healthy subjects (n = 6). Plasma concentrations are shown after subtraction of baseline values. Rate constants were: k 12 = 0.032 T 0.008, k 21 = 0.066 T 0.090 and k 10 = 0.057 T 0.003.
This study describes the development of an assay for measurement of human N-terminal progastrin fragments. Such assay has not been reported previously. In human antral tissue, the mono-specific immunoassay revealed large quantities of progastrin 1 – 35 and a smaller fragment corresponding to progastrin 1 – 19. Nevertheless, the antral contents of Nterminal progastrin fragments were only half of that of the amidated gastrins. In circulation, the opposite situation was observed with N-terminal progastrin fragments present in almost 30-fold higher concentrations. Examination of both stimulated secretion and metabolic fate of progastrin fragment 1– 35 did not explain the discrepancies. The elaborate post-translational processing of progastrin in G-cells includes several endoproteolytic cleavages at monoand bibasic sites. Two major cleavage sites are located immediately N-terminal of gastrin-34 and gastrin-17, which in turn releases the two dominant gastrins (Fig. 6). Besides the gastrins, progastrin processing also gives rise to complementary N-terminal flanking fragments, some of which have already been identified [11]. In this early report, however, the fragments were identified after multistep purification, which may have led to loss of peptides during the procedures. Thus, the tissue concentrations of the different fragments were not established. The present study therefore adds to the prevailing
52
J.P. Goetze et al. / Regulatory Peptides 133 (2006) 47 – 53
Fig. 6. Schematic figure of human progastrin (top), the N-terminal fragments (black bars) and the gastrins (white bars). The evolutionary preserved progastrin 1 – 5 sequence (lower left) and the pentapeptide-sequence required for gastrin bioactivity (lower right) are also shown.
knowledge in several ways. First, a progastrin fragment 1 –19 also seems to be present in normal antral tissue (Figs. 2 and 6). This fragment represents the complementary N-terminal flanking fragment to gastrin-52 after cleavage at the arginyl residue in position 19 of progastrin. Moreover, the dominant fragment in antral G-cells was progastrin 1 –35, which accordingly also may be the dominant form released to circulation. Finally, we did not detect any larger N-terminal fragments, as for instance a complementary N-terminal fragment to gastrin-17, i.e., progastrin 1 – 52. Although gel filtration of antral tissue extracts did reveal a larger molecular form, this chromatographic peak is more likely to represent intact progastrin (Fig. 2). As neither intact progastrin or progastrin 1 –52 was available for column calibration, it was not possible to determine the molecular identity of this larger form. The immunoassay also allowed us to compare the molar relationship between the N-terminal progastrin fragments and the gastrins. In normal antral tissue extracts, there were consistently lower contents of N-terminal progastrin fragments (Table 2). This may partly be due to endoproteolytic cleavage at the arginyl residue in position 5 (Fig. 1). In fact, some evidence corroborates this possibility as large quantities of a truncated progastrin fragment 6 –35 also have been identified [11]. Another plausible mechanism for the low cellular contents could be that the N-terminal fragments are secreted in a way different from the gastrins. Meal-stimulated G-cell secretion, however, disclosed almost parallel increases in Nterminal progastrin fragments and gastrins (Fig. 3). The Nterminal progastrin fragments are therefore, at least to some extent, co-released during meal-stimulated secretion. The high basal concentration of N-terminal progastrin fragments compared to amidated gastrin in normal plasma is intriguing. A feasible explanation would be major differences in the metabolic handling. However, the half-life of progastrin fragment 1 – 35 of ¨30 min in humans (Fig. 5) is similar to the half-life of the dominant gastrin-34 form in circulation [6,26]. Further studies in pigs consolidated the half-life of progastrin 1 – 35, which was found to be ¨35 min after infusion. Consequently, pharmacokinetics does not explain the high basal plasma concentrations. In pigs, progastrin 1– 35 was extracted across the kidney and in the vasculature to the head
(Table 3). Interestingly, this extraction pattern closely resembles that of gastrin-52 [22]. Although this type of study does not reveal the underlying molecular mechanism of peptide degradation, it is nevertheless captivating to speculate that a common sequence in progastrin 1 – 35 and gastrin-52, i.e., amino acids 20– 35, may target both peptides for degradation in the kidneys and vasculature to the head. In some support of this, it has been shown that recombinant progastrin 6 –80 is stable in vitro but degraded in vivo [27]. Thus, it may be worthwhile to specifically examine the organ extraction of progastrin fragment 20 –35. A more feasible explanation may be that the vascular endothelium in the kidneys and to the head expresses an aminopeptidase responsible for cleavage at the N-terminus. In turn, such N-terminal truncation would lead to diminished antibody binding and measurement of both peptides. In hypergastrinemic plasma, the concentration of N-terminal progastrin fragments was only 2.7-fold increased compared to gastrin (Fig. 4). Moreover, there was no correlation between the peptide concentrations. This shift in the ratio of N-terminal progastrin fragments and gastrins are likely to be caused by changes in the secretion from the affected cells. As a result, Gcells overexpressing the gastrin gene may change the otherwise regulated release into a more constitutive-like secretion. An alternative mechanism may be that N-terminal progastrin fragments are increasingly processed at the monobasic arginyl residue in position 5 (Fig. 1). Either way, measurement of Nterminal progastrin fragments in hypergastrinemic plasma does not merely reflect increased gastrin concentrations. Rather, it may hold a new type of information in patients suspected for a gastrin-producing malignancy. A promising perspective could be that measurement of both N-terminal progastrin fragments and gastrins in plasma may improve the diagnostic separation between conditions with increased antral G-cell secretion and uncontrolled secretion from malignant gastrinoma cells. Precursor-derived peptides have gained interest in recent years. Rather than being considered as a mere byproduct from post-translational precursor processing, N-terminal fragments have been demonstrated to be intricately involved in both intracellular and endocrine mechanisms. For instance, Nterminal fragments from some prohormones contain important regulatory motifs for the biosynthetic process including intracellular sorting and transportation [28,29]. From a diagnostic point of view, N-terminal precursor fragments have also proven useful as plasma markers of disease. In this respect, a timely example of such function seems to be cardiac-derived pro-B-type natriuretic peptide fragments, which are likely to become the most frequently measured peptide in modern medicine [30]. In contrast, elucidation of the N-terminal progastrin fragments has only just begun. As the fragments do not seem to be required for intracellular processing and sorting of gastrin [12], it may be worthwhile to look for biological functions outside the G-cells. Some recent reports have in fact suggested that progastrin can impose gastrinindependent stimulatory effects on colonic cell growth [13 – 17]. Notably, increased colonic proliferation has been demonstrated in mice overexpressing glycine-extended gastrin [31].
J.P. Goetze et al. / Regulatory Peptides 133 (2006) 47 – 53
However, the precise stimulatory region within progastrin is not always defined, which hampers a molecular interpretation of some of the observed effects. Furthermore, the basic physiology behind the experimental findings, i.e., the normal and pathophysiological occurrence of the stimulatory peptides, is often not known. Whether some of the effects could be attributable to N-terminal progastrin fragments remains to be determined. Acknowledgements The expert technical assistance of Lone Olsen, Letty Klarskov and Mette Olesen is gratefully acknowledged. The study was supported by grants from the Danish Cancer Union and the P. A. Messerschmidt Foundation. References [1] Rehfeld JF. The new biology of gastrointestinal hormones. Physiol Rev 1998;78:1087 – 108. [2] Gregory RA, Tracy HJ. The constitution and properties of two gastrins extracted from hog antral mucosa. Gut 1964;46:103 – 14. [3] Yalow RS, Berson SA. Size and charge distinctions between endogenous human plasma gastrin in peripheral blood and heptadecapeptide gastrins. Gastroenterology 1970;58:609 – 15. [4] Gregory RA, Tracy HJ. Isolation of two ‘‘big gastrins’’ from Zollinger – Ellison tumour tissue. Lancet 1972;2:797 – 9. [5] Rehfeld JF, Stadil F, Vikelsoe J. Immunoreactive gastrin components in human serum. Gut 1974;15:102 – 11. [6] Walsh JH, Isenberg JI, Ansfield J, Maxwell V. Clearance and acidstimulating action of human big and little gastrins in duodenal ulcer subjects. J Clin Invest 1976;57:1125 – 31. [7] Rehfeld JF, Bundgaard JR, Goetze JP, Friis-Hansen L, Hilsted L, Johnsen AH. Naming progastrin-derived peptides. Regul Pept 2004;120:177 – 83. [8] Reeve Jr JR, Walsh JH, Tompkins RK, Hawke D, Shively JE. Amino terminal fragments of human progastrin from gastrinoma. Biochem Biophys Res Commun 1984;123:404 – 9. [9] Desmond H, Pauwels S, Varro A, Gregory H, Young J, Dockray GJ. Isolation and characterization of the intact gastrin precursor from a gastrinoma. FEBS Lett 1987;210:85 – 188. [10] Huebner VD, Jiang RL, Lee TD, Legesse K, Walsh JH, Shively JE, Chew P, Azumi T, Reeve Jr JR. Purification and structural characterization of progastrin-derived peptides from a human gastrinoma. J Biol Chem 1991;266:12223 – 7. [11] Rehfeld JF, Johnsen AH. Identification of gastrin component I as gastrin71. The largest possible bioactive progastrin product. Eur J Biochem 1994;223:765 – 73. [12] Bundgaard JR, Birkedal H, Rehfeld JF. Progastrin is directed to the regulated secretory pathway by synergistically acting basic and acidic motifs. J Biol Chem 2004;279:5488 – 93.
53
[13] Singh P, Velasco M, Given R, Varro A, Wang TC. Progastrin expression predisposes mice to colon carcinomas and adenomas in response to a chemical carcinogen. Gastroenterology 2000;119:162 – 71. [14] Baldwin GS, Hollande F, Yang Z, Karelina Y, Paterson A, Strang R, Fourmy D, Neumann G, Shulkes A. Biologically active recombinant human progastrin(6 – 80) contains a tightly bound calcium ion. J Biol Chem 2001;276:7791 – 6. [15] McQueen K, Kovac S, Ho PK, Rorison K, Pannequin J, Neumann G, Shulkes A, Baldwin GS. Preparation of biologically active recombinant human progastrin(1 – 80). J Protein Chem 2002;21:465 – 71. [16] Wu H, Owlia A, Singh P. Precursor peptide progastrin(1 – 80) reduces apoptosis of intestinal epithelial cells and upregulates cytochrome c oxidase Vb levels and synthesis of ATP. Am J Physiol Gastrointest Liver Physiol 2003;285:G1097 – 110. [17] Ottewell PD, Watson AJ, Wang TC, Varro A, Dockray GJ, Pritchard DM. Progastrin stimulates murine colonic epithelial mitosis after DNA damage. Gastroenterology 2003;124:1348 – 57. [18] Liu FT, Zinnecker M, Hamaoka T, Katz DH. New procedures for preparation and isolation of conjugates of proteins and a synthetic copolymer of D-amino acids and immunochemical characterization of such conjugates. Biochemistry 1979;18:690 – 3. [19] Stadil F, Rehfeld JF. Preparation of 125 I-labelled synthetic human gastrin I for radioimmunoanalysis. Scand J Clin Lab Invest 1972;30:361 – 8. [20] Morris BJ. Specific radioactivity of radioimmunoassay tracer determined by self-displacement: a re-evaluation. Clin Chim Acta 1976;73:213 – 6. [21] Rehfeld JF, Stadil F, Rubin B. Production and evaluation of antibodies for the radioimmunoassay of gastrin. Scand J Clin Lab Invest 1972;30: 221 – 32. [22] Hansen CP, Goetze JP, Stadil F, Rehfeld JF. The metabolism of gastrin-52 and gastrin-6 in pigs. Am J Physiol Gastrointest Liver Physiol 2000;279:G552 – 60. [23] Loo JC, Riegelman S. Assessment of pharmacokinetic constants from postinfusion blood curves obtained after I. V. infusion. J Pharm Sci 1970;59:53 – 5. [24] Ekins R, Newman B. Theoretical aspects of saturation analysis. Acta Endocrinol Suppl (Copenh) 1970;147:11 – 36. [25] Sips R. On the structure of a catalyst surface. J Chem Phys 1948;16: 490 – 5. [26] Walsh JH, Debas HT, Grossman MI. Pure human big gastrin. Immunochemical properties, disappearance half time, and acid-stimulating action in dogs. J Clin Invest 1974;54:477 – 85. [27] Paterson AC, Baldwin GS, Shulkes A. Metabolism of recombinant progastrin in sheep. Am J Physiol 2002;283:E449 – 56. [28] Cool DR, Fenger M, Snell CR, Loh YP. Identification of the sorting signal motif within pro-opiomelanocortin for the regulated secretory pathway. J Biol Chem 1995;270:8723 – 9. [29] Baertschi AJ, Monnier D, Schmidt U, Levitan ES, Fakan S, Roatti A. Acid prohormone sequence determines size, shape, and docking of secretory vesicles in atrial myocytes. Circ Res 2001;89:E23 – 9. [30] Goetze JP. Biochemistry of pro-B-type natriuretic peptide-derived peptides: the endocrine heart revisited. Clin Chem 2004;50:1503 – 10. [31] Koh TJ, Dockray GJ, Varro A, Cahill RJ, Dangler CA, Fox JG, Wang TC. Overexpression of glycine-extended gastrin in transgenic mice results in increased colonic proliferation. J Clin Invest 1999;103:1119 – 26.
Antral content, secretion and peripheral metabolism of N-terminal progastrin fragments Jens Peter Goetze a,*, Carsten Palnæs Hansen b, Jens F. Rehfeld a a
Department of Clinical Biochemistry, Rigshospitalet, University of Copenhagen, DK-2100 Copenhagen, Denmark b Gastrointestinal Surgery, Rigshospitalet, University of Copenhagen, DK-2100 Copenhagen, Denmark Received 4 July 2005; received in revised form 4 September 2005; accepted 9 September 2005 Available online 5 October 2005
Abstract Objectives: In addition to the acid-stimulatory gastrins, progastrin also release N-terminal fragments. In order to examine the cellular content, secretion and peripheral metabolism of these fragments, we developed an immunoassay specific for the N-terminal sequence of human progastrin. Results: The concentration of N-terminal progastrin fragments in human antral tissue was 6.7 nmol/g tissue (n = 5), which was only half of that of acid-stimulatory gastrins (12 nmol/g tissue). Gel chromatography of antral extracts showed that the progastrin fragment 1 – 35 and 1 – 19 constitute the major part of the N-terminal progastrin fragments. The basal concentration of N-terminal fragments in normal human plasma was almost 30fold higher than that of the amidated, acid-stimulatory gastrins (286 pmol/l versus 9.8 pmol/l, n = 26, P < 0.001). In contrast, the concentration of N-terminal fragments in hypergastrinemic plasma was only 2.7-fold higher than the concentration of amidated gastrins (540 pmol vs. 198 pmol/l, P = 0.02). During meal stimulation, the plasma concentrations of N-terminal progastrin fragments and amidated gastrins increased in a correlated manner (r = 0.97, P = 0.005). The half life for progastrin 1 – 35 in circulation was 30 min, and a pig model revealed the kidneys and the vasculature to the head as the primary sites of degradation. Conclusion: The cellular and circulatory concentration profiles of N-terminal progastrin fragments differ markedly from those of the acidstimulatory gastrins. The high basal plasma concentrations of N-terminal progastrin fragments cannot be explained by differences in elimination. D 2005 Elsevier B.V. All rights reserved. Keywords: Elimination; Gastrin; Hypergastrinemia; Progastrin; Radioimmunoassay; Secretion
1. Introduction Gastrin is the major regulator of gastric acid secretion and gastric mucosal growth [1] Gastrin is synthesized in antroduodenal G-cells by elaborate posttranslational processing into acid-stimulatory peptides of which gastrin-17 and gastrin-34 are the most abundant ([2 –6], for review see Ref. [7]). The bioactive site of the gastrins is the evolutionary conserved Cterminal tetrapeptide amide, Trp – Met – Asp – Phe – NH2. In addition to the acid stimulatory gastrins, progastrin also release other fragments. The N-terminal fragments have for instance been identified in extracts from gastrinoma tissue [8– 10], and in normal human antral tissue [11]. Hence, in addition to the large amidated gastrin-71, fragments 1 –35, 6– 35, and 20 –35 from the N-terminal part of progastrin were also present in G* Corresponding author. Tel.: +45 3545 4640; fax: +45 3545 8323. E-mail address: [email protected] (J.P. Goetze). 0167-0115/$ - see front matter D 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.regpep.2005.09.016
cells with fragment 1 – 35 being the most abundant after purification [11]. So far, the biology of N-terminal progastrin fragments has not been examined. But recently, we reported that the Nterminal fragment 1– 35 is not required for cellular sorting to the secretory pathway and subsequent regulated secretion of gastrins [12]. Other reports have suggested that the intact progastrin molecule in itself can stimulate colonic cell growth [13 – 17]. However, the active site within progastrin has not always been defined, which hampers a molecular interpretation of the observed effects. Furthermore, the normal and pathophysiological occurrence of the N-terminal progastrin derivatives is virtually unknown. In order to examine the secretion and metabolism of N-terminal progastrin fragments, we therefore developed a sensitive and specific radioimmunoassay for the N-terminus of human progastrin. Our results reveal a surprisingly different cellular and circulatory profile compared to gastrin.
48
J.P. Goetze et al. / Regulatory Peptides 133 (2006) 47 – 53
2. Methods and materials 2.1. Peptides Human progastrin fragments 1– 35, 1 –19 and 6 – 35 were custom synthesized by Cambridge Research Biochemicals Ltd (Cheshire, UK). Also, progastrin 1 –10 extended C-terminally with a tyrosyl residue and fragment 1 – 10 extended Cterminally with cysteine were synthesized for tracer and standard, and for immunization, respectively. The structure and identities of all synthetic peptides were verified by reversed-phase HPLC, amino acid analysis and mass spectrometry before use. 2.2. Radioimmunoassay A radioimmunoassay directed against sequence 1 –10 of human progastrin was developed using 10 mg of the 1 –10 fragment extended C-terminally with cysteine and coupled to 20 mg bovine serum albumin using the m-maleimidobenzoylN-hydroxysuccinimide ester conjugation method [18]. The coupled product was dissolved in 15 ml of distilled water (conjugate solution). The antigen solution (2 ml) was mixed with 2.5 ml isotonic saline and emulsified with an equal volume of complete Freund’s adjuvant (The Danish Serum Institute, Copenhagen, Denmark) and used for the first immunization. For booster injections, 1 ml of the antigen solution was mixed with 4 ml of saline and an equal volume of incomplete Freund’s adjuvant. Eight random-bred white Danish rabbits were immunized subcutaneously over the lower back at 8-week intervals. Twenty millimeter of blood was collected from an ear vein 14 days after each immunization, and the serum was stored at 20 -C for evaluation. For tracer use, the tyrosine-extended 1 – 10 fragment (4.5 nmol) was monoiodinated using the chloramine-T method as previously described [19] and purified on reversed-phase HPLC (Pierce C8 column, 4.6 220 mm) eluted by a linear ethanol gradient (5 – 25%) in 1% trifluoroacetic acid. The gradient was selected to ensure separation of the non-labeled peptide from the iodinated tracer. The specific tracer radioactivity was determined by self-displacement [20]. For measurement of gastrins, we used a radioimmunoassay based on antiserum no. 2604, which is specific for the C-terminal octapeptide sequence of the aamidated gastrins [21]. This assay measures all amidated gastrins with equimolar potency and the reactivity with the related hormone cholecystokinin is negligible. 2.3. Human antral tissue Tissue biopsies of human antral mucosa (n = 5) were obtained from patients undergoing surgery for pancreatic cancer (Whipple’s operation). The use of human tissue was approved by the local ethics committee (KF01-352/96) and informed consent was obtained from all patients. The tissue biopsies were immediately frozen in liquid nitrogen and stored at 80 -C until extraction. While frozen, the tissues were
sliced into small pieces and immersed in boiling water (10 ml/g tissue) for 20 min, homogenized and centrifuged at 10000 g for 30 min. The supernatant was stored and the pellet extracted in 0.5 mol/l acetic acid (10 ml/g tissue), homogenized and centrifuged as above. 2.4. Gel filtration chromatography The molecular composition of N-terminal progastrin fragments in tissue was examined by gel chromatography (Sephadex G-50 Superfine, 10 1000 mm, Pharmacia, Sweden). Columns were calibrated with synthetic progastrin 1– 35 and 1 –19. A barbital buffer containing 0.1% bovine albumin was used as eluent at a flow rate of 4 ml/h (4 -C). The buffer was supplemented with 0.05 M NaCl to avoid protein binding. Void and total volumes were determined by eluting 125Ialbumin and 22NaCl, respectively. 2.5. Plasma Blood samples were collected from 26 healthy and fasting volunteers. Six of the volunteers were later served at proteinrich meal, and blood samples were obtained from a cubital vein at set time intervals ( 35, 5, 0, 10, 20, 30, 40, 50, 70, 95 min). Plasma was stored at 20 -C until analysis. In addition, plasma from 48 patients with hypergastrinemia due to gastrinoma or achlorhydria was collected from the local gastrin analysis facility in our department. 2.6. Elimination of progastrin 1– 35 in humans Synthetic progastrin 1 – 35 (100 pmol/kg) was injected intravenously as a bolus into fasting, healthy subjects (n = 6). Blood samples were collected from the opposite arm (0, 2, 4, 6, 8, 10, 15, 20, 25, 30, 40, 50 and 60 min after injection, and then every 15 min until 180 min from the time of bolus injection). After plasma measurement, the pharmacokinetics was analyzed according to a two-compartment open model: C t = Ae at + Be bt , where A and B is the zero intercepts with the ordinate of the individual exponential terms and a and b are the slopes of the curves. 2.7. Organ extraction of progastrin 1 –35 in pigs To further elucidate the metabolic fate of progastrin 1– 35, we used a previously described animal model [22]. Eight pigs (Danish landrace – Yorkshire breed, 30 –40 kg) were anesthetized and intubated. Isotonic saline was infused at 10 ml/kg/ h during the whole experiment, and donor blood from siblings was additionally infused in volumes equal to that removed during sampling. Cardiac output was measured regularly with a Swan – Ganz thermodilution catheter (Swan – Ganz pediatric 5F; Baxter Health Care, CA, USA). Polyethylene catheters for blood sampling were placed in the thoracic aorta via the left carotid artery and cephalically in the left internal jugular vein. Via the external jugular veins, two angiography catheters were introduced into a major hepatic vein and the left renal vein.
J.P. Goetze et al. / Regulatory Peptides 133 (2006) 47 – 53
Blood from a peripheral lung artery was drawn from the Swan – Ganz catheter. The positions of all catheters were verified on fluoroscopy. Of note, neither the hepatic nor the pulmonary catheter was wedged to avoid mixing with portal and pulmonary venous blood when collecting blood samples. After a midline laparotomy, a catheter was placed at the entrance of the portal vein via the splenic vein and another catheter was positioned in one of the major tributaries of the mesenteric vein draining the small intestine. Blood from a femoral vein was drawn from an inserted catheter after exposure of the vessels in the groin. Progastrin 1– 35 and indicator substance were infused into catheters with their tips placed in the right atrium. After completion of the surgical procedure, a 1 h stabilization period was allowed before starting the experiments. Bloodflow in the renal, femoral and portal vein was measured with an electromagnetic flowmeter (Nycotron, Oslo, Norway) and pulmonary flow was regarded as equivalent to cardiac output. Blood flow was converted to plasma flow based on the hematocrit. Hepatic plasma flow was measured by continuous infusion of indocyanine green (Cardiogreen, Becton and Dickinson, MD, USA) at 0.130 Amol/min. After calibration, simultaneous samples from the aorta and the hepatic vein were drawn at set time intervals. The concentration of indocyanine in plasma was measured spectrophotometrically, and hepatic plasma flow was calculated according to Fick’s principle. The elimination of progastrin 1 – 35 was determined as the concentration gradient across the vascular beds during infusion of 100 pmol/kg/h of the peptide. Arterial samples were taken from the aorta every 15 min during infusion. After 90 min, venous samples were drawn every 15 min for 1 h from a lung artery and the femoral, renal, portal, hepatic, internal jugular, and cranial mesenteric veins. The fractional extraction was calculated as: E R = (C i C o) / C i, where C i and C o are the inflow and outflow concentrations in plasma, respectively. 2.8. Elimination of progastrin 1 –35 in pigs When the infusion of progastrin 1 –35 was terminated after 170 min, aortic blood was sampled at regular intervals for 90 min to determine the half-life in circulation. A two-compartment model was used to determine the half-lives as above, but as data were obtained after a continuous infusion, the values of A and B were calculated as: A ¼ RðaT Þ= 1 eaT ; B ¼ S ðbT Þ= 1 ebT where R and S represent the intercepts with the ordinate at the start of the post-infusional period, and T is the period of infusion [23].
49
samples. Correlations were performed on logarithmically transformed data prior to regression analysis. A two-tailed level for significance was set a 0.05. 3. Results 3.1. Radioiodination The incorporation of 125I in the tyrosine-extended progastrin 1– 10 varied between 60 –95%. The non-specific binding was <5% and labeled tracer was completely separated from the non-labeled peptide. Dilution curves for labeled and unlabeled antigen were parallel. The specific radioactivity of the iodinated tracer peptide was 1.08 Ci/Amol. 3.2. Antiserum evaluation The characteristics of the three best antisera are shown in Table 1. Antiserum no. 94023 demonstrated the highest avidity expressed by the effective equilibrium constant, K 0eff = 0.39 1012 l/mol [24]. This antiserum was therefore chosen for further characterization and measurements of gel filtration fractions, tissue extracts and plasma samples. The index of heterogeneity according to Sips was 1.02, indicating that the ligand binding is highly homogenous and that the diluted antiserum acts like a solution of monoclonal antibodies [25]. A high sensitivity, expressed as the analytical detection limit, was achieved for several antisera (Table 1). Specificity expressed as the ID50 of the N-terminal truncated progastrin 6 –35 peptide compared to the ID50 of progastrin 1– 10 Tyr was found to be < 0.0001. In contrast, the binding of both progastrin 1 –19 and 1 –35 did not differ from the binding of progastrin 1 – 10 Tyr. Antiserum no. 94023 was also tested for reactivity with N-terminal progastrin fragments in porcine antral extracts. Notably, the N-terminus of porcine progastrin is homologous to the human sequence at the first 4 N-terminal amino acid residues but with substitutions in position 5 and 6 (Arg–Ser to Gly– Phe, Fig. 1). No binding to porcine fragments could be demonstrated. The FASTA computer program was then used to search available databases for amino acid sequences resembling the N-terminal sequence 1 –10 of human progastrin. No relevant sequences other than the N-terminal progastrin regions from other mammals were found. Finally, dilution curves for tissue extracts (antral mucosa) were parallel with the calibrator Table 1 Antiserum characteristics Antiserum no. (rabbit and bleeding)
Titera (103)
K 0effb 1012 (l/mol)
ac
Detection limit (pmol/l)
2.9. Statistical analysis
94023/4 94024/4 94027/4
570 450 160
0.39 1.32 2.40
1.02 0.80 0.83
0.07 0.04 0.04
Results are expressed as means T SE, except for the measurements in hypergastrinemic plasma with a non-Gaussian distribution. Comparative analyses were performed either by a paired Student’s t-test or the Wilcoxon’s test for paired
a: Antiserum dilution at which 35% of the tracer is bound at equilibrium. b: Effective equilibrium constant according to Ekins and Newman [24]. c: Index of heterogeneity according to Sips [25]. d: Detection limit with ten replicates of 0-calibrator, one-sided 95% confidence interval.
50
J.P. Goetze et al. / Regulatory Peptides 133 (2006) 47 – 53
Fig. 1. Amino acid sequence of N-terminal progastrin 1 – 10 in mammals. Note the homology of the first N-terminal residues, and the basic arginyl residue in position 5.
curves (data not shown). The recovery of progastrin 1– 35 in plasma kept at room temperature was unaffected after 8 h but reduced to 27 T 3% after 24 h. 3.3. N-terminal progastrin fragments in human antrum The total concentration of N-terminal progastrin immunoreactivity in antral mucosa was lower than that of the amidated gastrins with only ¨60% N-terminal progastrin fragments compared to the gastrin contents (Table 2). Gel chromatography revealed distinct elution peaks corresponding to progastrin fragments 1 – 35 and 1– 19 (Fig. 2). Also, a larger molecular form was detected, which may correspond to intact progastrin.
Fig. 2. Gel chromatography of human antral tissue. The positions of the calibrator peptides, progastrin fragments 1 – 35 and 1 – 19, were determined in separate experiments. The first immunoreactive peak (K D ¨ 0.25) presumably corresponds to intact progastrin.
correlation between the N-terminal progastrin and the gastrin concentrations.
3.4. N-terminal progastrin fragments in normal plasma
3.6. Elimination of progastrin fragment 1 – 35 in man
The basal concentration of N-terminal progastrin immunoreactivity in healthy individuals was 285 T 20 pmol/l, which was almost 30-fold higher than the concentration of amidated gastrins (9.8 T 0.8 pmol/l, P < 0.0001). Meal stimulation increased both N-terminal progastrin fragments and amidated gastrins (r = 0.97, P = 0.005) with gastrin concentrations reaching a plateau 30 min after stimulation. The concentration of Nterminal progastrin fragments remained elevated throughout the entire time period studied (90 min) after stimulation (Fig. 3). Gel chromatography of pooled plasma 30 min after meal stimulation revealed only one peak corresponding to the calibration position of progastrin fragment 1 – 35 (data not shown).
The elimination of progastrin 1 – 35 after a bolus injection (112 T 12 pmol/kg) was biexponential with half-lives: T 1 / 2a = 7.0 T 1.2 and T 1 / 2b = 30.5 T 8.2 min (Fig. 5). Clearance and V dss (apparent volume of distribution at steady state) were
3.5. N-terminal progastrin fragments in hypergastrinemic plasma The N-terminal progastrin concentration in hypergastrinemic plasma was 2.7-fold higher than the gastrin concentration (median 198 pmol/l (range 18 –37000) vs. 540 pmol/l (130 – 2070), P = 0.02, Fig. 4). Of note, there was no significant Table 2 N-terminal progastrin and amidated gastrin in human antral tissue extracts Antrum no. N-terminal progastrin Amidated gastrin [N-terminal progastrin] (pmol/g tissue) (pmol/g tissue) [gastrin] 1 2 3 4 5 Mean
5718 7061 14411 3871 2630 6738
9813 10948 28890 6324 4320 12059
0.58 0.64 0.50 0.61 0.61 0.59
Fig. 3. N-terminal progastrin fragments (upper panel) and gastrin concentrations (lower panel) in normal subjects (n = 26) during a protein-rich meal. The results are shown as means T SEM.
J.P. Goetze et al. / Regulatory Peptides 133 (2006) 47 – 53
51
Table 3 Fractional extraction of progastrin fragment 1 – 35
Kidney Liver Gut Head Hindlimb Lungs
Progastrin 1 – 35
Gastrin-17
0.16 T 0.03* 0.04 T 0.04 0.03 T 0.01 0.11 T 0.02* 0.02 T 0.01 0.03 T 0.02
0.40 T 0.04* 0.07 T 0.03 0.13 T 0.03* 0.32 T 0.04* 0.42 T 0.04* 0.01 T 0.03
Values are means T SEM; n = 8, *P < 0.05. The gastrin-17 results have been published previously [22].
3.7. Organ extraction of progastrin fragment 1– 35 in pigs
Fig. 4. N-terminal progastrin fragments and amidated gastrin concentrations in hypergastrinemic plasma. The concentrations were transformed (Log10) prior to plotting and statistical analysis. The horizontal lines represent median concentrations.
calculated to 6.8 T 0.6 ml/kg per min and 126.0 T 1.3 ml/kg. Analysis according to a two-compartment model revealed that the fractions of elimination during the two phases, f a and f b , A=a B=b expressed as: fa ¼ AU C and fb ¼ AU C , were 0.29 and 0.71, indicating that most of the peptide was eliminated in the hphase.
The infused dose rate of human progastrin fragment 1– 35 was 52 T 6 pmol/kg/h. Extraction of the peptide was recorded over the kidney (metabolic clearance rate, MCR = 0.9 T 0.2 ml/ kg/min) and head (MCR = 1.1 T 0.1 ml/kg/min), whereas no significant extraction was observed across the lungs, liver, gut or hind limb (Table 3). The elimination kinetics of progastrin 1– 35 in pigs was then calculated from post-infusion data with biexponential half-lives: T 1 / 2a = 2.7 T 0.7 and T 1 / 2b = 35.1 T1.9 min. The total MCR and the V dss were 3.7 T 0.2 ml/kg/min and 130.8 T 12.7 ml/kg, respectively. The fractions of elimination were f a = 0.37 and f b = 0.63, implying that total progastrin 1– 35 elimination in pig occurred mainly during the h-phase. Gel chromatography on pooled blood after the infusion period (t 135) showed identical elution profiles from all organ veins, and only immunoreactivity eluting in a position corresponding to progastrin 1– 35 was detected (data not shown). 4. Discussion
Fig. 5. Elimination of progastrin fragment 1 – 35 in humans. A bolus dose (100 pmol/kg) was intravenously injected into healthy subjects (n = 6). Plasma concentrations are shown after subtraction of baseline values. Rate constants were: k 12 = 0.032 T 0.008, k 21 = 0.066 T 0.090 and k 10 = 0.057 T 0.003.
This study describes the development of an assay for measurement of human N-terminal progastrin fragments. Such assay has not been reported previously. In human antral tissue, the mono-specific immunoassay revealed large quantities of progastrin 1 – 35 and a smaller fragment corresponding to progastrin 1 – 19. Nevertheless, the antral contents of Nterminal progastrin fragments were only half of that of the amidated gastrins. In circulation, the opposite situation was observed with N-terminal progastrin fragments present in almost 30-fold higher concentrations. Examination of both stimulated secretion and metabolic fate of progastrin fragment 1– 35 did not explain the discrepancies. The elaborate post-translational processing of progastrin in G-cells includes several endoproteolytic cleavages at monoand bibasic sites. Two major cleavage sites are located immediately N-terminal of gastrin-34 and gastrin-17, which in turn releases the two dominant gastrins (Fig. 6). Besides the gastrins, progastrin processing also gives rise to complementary N-terminal flanking fragments, some of which have already been identified [11]. In this early report, however, the fragments were identified after multistep purification, which may have led to loss of peptides during the procedures. Thus, the tissue concentrations of the different fragments were not established. The present study therefore adds to the prevailing
52
J.P. Goetze et al. / Regulatory Peptides 133 (2006) 47 – 53
Fig. 6. Schematic figure of human progastrin (top), the N-terminal fragments (black bars) and the gastrins (white bars). The evolutionary preserved progastrin 1 – 5 sequence (lower left) and the pentapeptide-sequence required for gastrin bioactivity (lower right) are also shown.
knowledge in several ways. First, a progastrin fragment 1 –19 also seems to be present in normal antral tissue (Figs. 2 and 6). This fragment represents the complementary N-terminal flanking fragment to gastrin-52 after cleavage at the arginyl residue in position 19 of progastrin. Moreover, the dominant fragment in antral G-cells was progastrin 1 –35, which accordingly also may be the dominant form released to circulation. Finally, we did not detect any larger N-terminal fragments, as for instance a complementary N-terminal fragment to gastrin-17, i.e., progastrin 1 – 52. Although gel filtration of antral tissue extracts did reveal a larger molecular form, this chromatographic peak is more likely to represent intact progastrin (Fig. 2). As neither intact progastrin or progastrin 1 –52 was available for column calibration, it was not possible to determine the molecular identity of this larger form. The immunoassay also allowed us to compare the molar relationship between the N-terminal progastrin fragments and the gastrins. In normal antral tissue extracts, there were consistently lower contents of N-terminal progastrin fragments (Table 2). This may partly be due to endoproteolytic cleavage at the arginyl residue in position 5 (Fig. 1). In fact, some evidence corroborates this possibility as large quantities of a truncated progastrin fragment 6 –35 also have been identified [11]. Another plausible mechanism for the low cellular contents could be that the N-terminal fragments are secreted in a way different from the gastrins. Meal-stimulated G-cell secretion, however, disclosed almost parallel increases in Nterminal progastrin fragments and gastrins (Fig. 3). The Nterminal progastrin fragments are therefore, at least to some extent, co-released during meal-stimulated secretion. The high basal concentration of N-terminal progastrin fragments compared to amidated gastrin in normal plasma is intriguing. A feasible explanation would be major differences in the metabolic handling. However, the half-life of progastrin fragment 1 – 35 of ¨30 min in humans (Fig. 5) is similar to the half-life of the dominant gastrin-34 form in circulation [6,26]. Further studies in pigs consolidated the half-life of progastrin 1 – 35, which was found to be ¨35 min after infusion. Consequently, pharmacokinetics does not explain the high basal plasma concentrations. In pigs, progastrin 1– 35 was extracted across the kidney and in the vasculature to the head
(Table 3). Interestingly, this extraction pattern closely resembles that of gastrin-52 [22]. Although this type of study does not reveal the underlying molecular mechanism of peptide degradation, it is nevertheless captivating to speculate that a common sequence in progastrin 1 – 35 and gastrin-52, i.e., amino acids 20– 35, may target both peptides for degradation in the kidneys and vasculature to the head. In some support of this, it has been shown that recombinant progastrin 6 –80 is stable in vitro but degraded in vivo [27]. Thus, it may be worthwhile to specifically examine the organ extraction of progastrin fragment 20 –35. A more feasible explanation may be that the vascular endothelium in the kidneys and to the head expresses an aminopeptidase responsible for cleavage at the N-terminus. In turn, such N-terminal truncation would lead to diminished antibody binding and measurement of both peptides. In hypergastrinemic plasma, the concentration of N-terminal progastrin fragments was only 2.7-fold increased compared to gastrin (Fig. 4). Moreover, there was no correlation between the peptide concentrations. This shift in the ratio of N-terminal progastrin fragments and gastrins are likely to be caused by changes in the secretion from the affected cells. As a result, Gcells overexpressing the gastrin gene may change the otherwise regulated release into a more constitutive-like secretion. An alternative mechanism may be that N-terminal progastrin fragments are increasingly processed at the monobasic arginyl residue in position 5 (Fig. 1). Either way, measurement of Nterminal progastrin fragments in hypergastrinemic plasma does not merely reflect increased gastrin concentrations. Rather, it may hold a new type of information in patients suspected for a gastrin-producing malignancy. A promising perspective could be that measurement of both N-terminal progastrin fragments and gastrins in plasma may improve the diagnostic separation between conditions with increased antral G-cell secretion and uncontrolled secretion from malignant gastrinoma cells. Precursor-derived peptides have gained interest in recent years. Rather than being considered as a mere byproduct from post-translational precursor processing, N-terminal fragments have been demonstrated to be intricately involved in both intracellular and endocrine mechanisms. For instance, Nterminal fragments from some prohormones contain important regulatory motifs for the biosynthetic process including intracellular sorting and transportation [28,29]. From a diagnostic point of view, N-terminal precursor fragments have also proven useful as plasma markers of disease. In this respect, a timely example of such function seems to be cardiac-derived pro-B-type natriuretic peptide fragments, which are likely to become the most frequently measured peptide in modern medicine [30]. In contrast, elucidation of the N-terminal progastrin fragments has only just begun. As the fragments do not seem to be required for intracellular processing and sorting of gastrin [12], it may be worthwhile to look for biological functions outside the G-cells. Some recent reports have in fact suggested that progastrin can impose gastrinindependent stimulatory effects on colonic cell growth [13 – 17]. Notably, increased colonic proliferation has been demonstrated in mice overexpressing glycine-extended gastrin [31].
J.P. Goetze et al. / Regulatory Peptides 133 (2006) 47 – 53
However, the precise stimulatory region within progastrin is not always defined, which hampers a molecular interpretation of some of the observed effects. Furthermore, the basic physiology behind the experimental findings, i.e., the normal and pathophysiological occurrence of the stimulatory peptides, is often not known. Whether some of the effects could be attributable to N-terminal progastrin fragments remains to be determined. Acknowledgements The expert technical assistance of Lone Olsen, Letty Klarskov and Mette Olesen is gratefully acknowledged. The study was supported by grants from the Danish Cancer Union and the P. A. Messerschmidt Foundation. References [1] Rehfeld JF. The new biology of gastrointestinal hormones. Physiol Rev 1998;78:1087 – 108. [2] Gregory RA, Tracy HJ. The constitution and properties of two gastrins extracted from hog antral mucosa. Gut 1964;46:103 – 14. [3] Yalow RS, Berson SA. Size and charge distinctions between endogenous human plasma gastrin in peripheral blood and heptadecapeptide gastrins. Gastroenterology 1970;58:609 – 15. [4] Gregory RA, Tracy HJ. Isolation of two ‘‘big gastrins’’ from Zollinger – Ellison tumour tissue. Lancet 1972;2:797 – 9. [5] Rehfeld JF, Stadil F, Vikelsoe J. Immunoreactive gastrin components in human serum. Gut 1974;15:102 – 11. [6] Walsh JH, Isenberg JI, Ansfield J, Maxwell V. Clearance and acidstimulating action of human big and little gastrins in duodenal ulcer subjects. J Clin Invest 1976;57:1125 – 31. [7] Rehfeld JF, Bundgaard JR, Goetze JP, Friis-Hansen L, Hilsted L, Johnsen AH. Naming progastrin-derived peptides. Regul Pept 2004;120:177 – 83. [8] Reeve Jr JR, Walsh JH, Tompkins RK, Hawke D, Shively JE. Amino terminal fragments of human progastrin from gastrinoma. Biochem Biophys Res Commun 1984;123:404 – 9. [9] Desmond H, Pauwels S, Varro A, Gregory H, Young J, Dockray GJ. Isolation and characterization of the intact gastrin precursor from a gastrinoma. FEBS Lett 1987;210:85 – 188. [10] Huebner VD, Jiang RL, Lee TD, Legesse K, Walsh JH, Shively JE, Chew P, Azumi T, Reeve Jr JR. Purification and structural characterization of progastrin-derived peptides from a human gastrinoma. J Biol Chem 1991;266:12223 – 7. [11] Rehfeld JF, Johnsen AH. Identification of gastrin component I as gastrin71. The largest possible bioactive progastrin product. Eur J Biochem 1994;223:765 – 73. [12] Bundgaard JR, Birkedal H, Rehfeld JF. Progastrin is directed to the regulated secretory pathway by synergistically acting basic and acidic motifs. J Biol Chem 2004;279:5488 – 93.
53
[13] Singh P, Velasco M, Given R, Varro A, Wang TC. Progastrin expression predisposes mice to colon carcinomas and adenomas in response to a chemical carcinogen. Gastroenterology 2000;119:162 – 71. [14] Baldwin GS, Hollande F, Yang Z, Karelina Y, Paterson A, Strang R, Fourmy D, Neumann G, Shulkes A. Biologically active recombinant human progastrin(6 – 80) contains a tightly bound calcium ion. J Biol Chem 2001;276:7791 – 6. [15] McQueen K, Kovac S, Ho PK, Rorison K, Pannequin J, Neumann G, Shulkes A, Baldwin GS. Preparation of biologically active recombinant human progastrin(1 – 80). J Protein Chem 2002;21:465 – 71. [16] Wu H, Owlia A, Singh P. Precursor peptide progastrin(1 – 80) reduces apoptosis of intestinal epithelial cells and upregulates cytochrome c oxidase Vb levels and synthesis of ATP. Am J Physiol Gastrointest Liver Physiol 2003;285:G1097 – 110. [17] Ottewell PD, Watson AJ, Wang TC, Varro A, Dockray GJ, Pritchard DM. Progastrin stimulates murine colonic epithelial mitosis after DNA damage. Gastroenterology 2003;124:1348 – 57. [18] Liu FT, Zinnecker M, Hamaoka T, Katz DH. New procedures for preparation and isolation of conjugates of proteins and a synthetic copolymer of D-amino acids and immunochemical characterization of such conjugates. Biochemistry 1979;18:690 – 3. [19] Stadil F, Rehfeld JF. Preparation of 125 I-labelled synthetic human gastrin I for radioimmunoanalysis. Scand J Clin Lab Invest 1972;30:361 – 8. [20] Morris BJ. Specific radioactivity of radioimmunoassay tracer determined by self-displacement: a re-evaluation. Clin Chim Acta 1976;73:213 – 6. [21] Rehfeld JF, Stadil F, Rubin B. Production and evaluation of antibodies for the radioimmunoassay of gastrin. Scand J Clin Lab Invest 1972;30: 221 – 32. [22] Hansen CP, Goetze JP, Stadil F, Rehfeld JF. The metabolism of gastrin-52 and gastrin-6 in pigs. Am J Physiol Gastrointest Liver Physiol 2000;279:G552 – 60. [23] Loo JC, Riegelman S. Assessment of pharmacokinetic constants from postinfusion blood curves obtained after I. V. infusion. J Pharm Sci 1970;59:53 – 5. [24] Ekins R, Newman B. Theoretical aspects of saturation analysis. Acta Endocrinol Suppl (Copenh) 1970;147:11 – 36. [25] Sips R. On the structure of a catalyst surface. J Chem Phys 1948;16: 490 – 5. [26] Walsh JH, Debas HT, Grossman MI. Pure human big gastrin. Immunochemical properties, disappearance half time, and acid-stimulating action in dogs. J Clin Invest 1974;54:477 – 85. [27] Paterson AC, Baldwin GS, Shulkes A. Metabolism of recombinant progastrin in sheep. Am J Physiol 2002;283:E449 – 56. [28] Cool DR, Fenger M, Snell CR, Loh YP. Identification of the sorting signal motif within pro-opiomelanocortin for the regulated secretory pathway. J Biol Chem 1995;270:8723 – 9. [29] Baertschi AJ, Monnier D, Schmidt U, Levitan ES, Fakan S, Roatti A. Acid prohormone sequence determines size, shape, and docking of secretory vesicles in atrial myocytes. Circ Res 2001;89:E23 – 9. [30] Goetze JP. Biochemistry of pro-B-type natriuretic peptide-derived peptides: the endocrine heart revisited. Clin Chem 2004;50:1503 – 10. [31] Koh TJ, Dockray GJ, Varro A, Cahill RJ, Dangler CA, Fox JG, Wang TC. Overexpression of glycine-extended gastrin in transgenic mice results in increased colonic proliferation. J Clin Invest 1999;103:1119 – 26.