Báo cáo khoa học: Human anionic trypsinogen Properties of autocatalytic activation and degradation and implications in pancreatic diseases potx

Human anionic trypsinogen Properties of autocatalytic activation and degradation and implications in pancreatic diseases Zolta ´ n Kukor, Miklo ´ sTo ´ th* and Miklo ´ s Sahin-To ´ th Department of Molecular and Cell Biology, Goldman School of Dental Medicine, Boston University, Boston, USA Human pancreatic secretions contain two major trypsinogen isoforms, cationic and anionic trypsinogen, normally at a ratio of 2 : 1. Pancreatitis, pancreatic cancer and chronic alcoholism lead to a characteristic reversal of the isoform ratio, and anionic trypsinogen becomes the predominant zymogen secreted. To understand the biochemical conse- quences of these alterations, we recombinantly expressed and purified both human trypsinogens and documented characteristics of autoactivation, autocatalytic degradation and Ca 2+ -dependence. Even though the two trypsinogens are  90% identical in their primary structure, we found that human anionic trypsinogen and trypsin exhibited a signifi- cantly increased (10–20-fold) propensity for autocatalytic degradation, relative to cationic trypsinogen and trypsin. Furthermore, in contrast to the characteristic stimulation of the cationic proenzyme, acidic pH inhibited autoactivation of anionic trypsinogen. In mixtures of cationic and anionic trypsinogen, an increase in the proportion of the anionic proenzyme had no significant effect on the levels of trypsin generated by autoactivation or by enterokinase at pH 8.0 in 1m M Ca 2+ – conditions that were characteristic of the pancreatic juice. In contrast, rates of trypsinogen activation were markedly reduced with increasing ratios of anionic trypsinogen under conditions that were typical of potential sites of pathological intra-acinar trypsinogen activation. Thus, at low Ca 2+ concentrations at pH 8.0, selective degradation of anionic trypsinogen and trypsin caused diminished trypsin production; while at pH 5.0, inhibition of anionic trypsinogen activation resulted in lower trypsin yields. Taken together, the observations indicate that up-regulation of anionic trypsinogen in pancreatic diseases does not affect physiological trypsinogen activation, but significantly limits trypsin generation under potential pathological conditions. Keywords: anionic trypsin; cationic trypsin; autoactivation; autolysis; alcoholic pancreatitis. The human pancreas secretes three isoforms of trypsinogen, encoded by the protease, serine (PRSS)genes1,2and3. On the basis of their relative electrophoretic mobility, the three trypsinogen species are commonly referred to as cationic trypsinogen (product of PRSS1, OMIM 276000), anionic trypsinogen (product of PRSS2, MIM 601564), and mesotrypsinogen (product of PRSS3)(forareviewon human trypsinogen genes and proteins see [1] and references therein). While individual variations may be considerable, normally the cationic isoform constitutes about 2/3 of the total trypsinogen content, and anionic trypsinogen makes up approximately 1/3 [2–4]. Mesotrypsinogen is a minor species, accounting for less than 5% of trypsinogens in human pancreatic juice [5,6]. The evolutionary rationale for the existence of several isoforms has not been clarified yet, but it is believed that differences in inhibitor sensitivity may be advantageous in digestion of foods containing trypsin inhibitors. A characteristic feature of human pancreatic diseases as well as chronic alcoholism is the relatively selective up-regulation of anionic trypsinogen secretion [3,4]. In chronic pancreatitis, the total trypsinogen content of the pancreatic juice may be unchanged or decreased while in chronic alcoholism an increase in total trypsinogen secretion was demonstrated. In these conditions, the pro- portion of anionic and cationic isoforms becomes reversed, and anionic trypsinogen dominates pancreatic secretions. In acute pancreatitis, the ratio of trypsinogen isoforms in the pancreatic juice has not been investigated so far, but a preferential increase in immunoreactive anionic tryp- sin(ogen) in the serum was documented by several studies [7–10]. It is unclear whether or not elevated anionic trypsinogen secretion might cause or predispose for pan- creatitis. Alternatively, increased anionic trypsinogen secre- tion might be innocuous or even protective in pancreatic physiology. Recently, methodology has been developed for the recombinant expression, in vitro refolding and purification of human cationic trypsinogen [11–13]. This type of recombinant trypsinogen preparation has been used in a Correspondence to M. Sahin-To ´ th, Department of Molecular and Cell Biology, Goldman School of Dental Medicine, Boston University, 715 Albany Street, EVANS-4; Boston, MA 02118, USA. Fax: +1 617 414 1041; Tel.: +1 617 414 1070; E-mail: miklos@bu.edu Abbreviations: GPR-pNA, N-CBZ-Gly-Pro-Arg-p-nitroanilide; Hu1, human cationic trypsinogen; Hu2, human anionic trypsinogen. *Present address: Department of Medical Chemistry, Molecular Biology and Pathobiochemistry, Semmelweis University, Budapest, Puskin Street 9, Hungary, H-1088. (Received 13 January 2003, revised 23 February 2003, accepted 18 March 2003) Eur. J. Biochem. 270, 2047–2058 (2003) Ó FEBS 2003 doi:10.1046/j.1432-1033.2003.03581.x growing number of studies that investigated the effects of hereditary pancreatitis-associated mutations [11–17]. Here we report the successful production of recombinant human anionic trypsinogen in a pure and stable form. Activation and degradation characteristics of this recombinant prepar- ation was documented and compared to those of cationic trypsinogen. Furthermore, interactions between the two isoforms were studied and the results indicated that an increase in the proportion of the anionic proenzyme had no significant effect on physiological trypsinogen activation, but resulted in decreased trypsin generation under condi- tions that mimicked the potential milieu(s) of intracellular pathological trypsinogen activation. Experimental procedures Materials Reagent grade bovine serum albumin was purchased from Biocell Laboratories (Rancho Dominguez, CA, USA), N-CBZ-Gly-Pro-Arg-p-nitroanilide (GPR-pNA) was from Sigma, and bovine enterokinase was from Biozyme Labor- atories (San Diego, CA, USA). Plasmid construction The coding cDNA for human anionic trypsinogen was PCR-amplified from a commercial plasmid (pcDNA3.1/GS harboring GeneStorm clone no. H-M27602M, Invitrogen) and cloned in place of the cationic trypsinogen gene in the pTrap-T7/Hu1 expression vector using the flanking NcoI and SacI restriction sites (pTrap-T7/Hu2). The activation peptide sequence of recombinant anionic trypsinogen in pTrap-T7/Hu2 was Met-Ala-Pro-Phe-(Asp)4-Lys. One of the native EcoRI sites and the internal SacIsitewere removed by introducing silent mutations into the codons for Leu41 and Glu209 (numbering starts with Met1 of the native pretrypsinogen sequence). Mutation K23Q was introduced by linker mutagenesis. A synthetic oligonucleo- tide linker encoding the mutation was ligated between the NcoIandEcoRI sites of pTrapT7/Hu2. Construction of the pTrap-T7 expression plasmid harboring the wild-type cationic trypsinogen gene was described previously [11,14]. Expression and purification of trypsinogen Small scale expression and in vitro refolding of human trypsinogens was carried out as reported previously [11–13]. In a typical experiment, 200 mL cultures of Rosetta(DE3) (Novagen) cells harboring pTrap-T7/Hu1 or pTrap-T7/ Hu2 plasmid were grown in Luria–Bertani medium with 50 lgÆmL )1 carbenicillin and 34 lgÆmL )1 chloramphenicol to a D 600 nm of 0.5 1 , induced with 1 m M isopropyl thio-b- D - galactoside, and grown for an additional 5 h. Rosetta(DE3) host strains are BL21(DE3) derivatives designed to enhance the expression of eukaryotic proteins that contain codons rarely used in Escherichia coli. Cells were harvested by centrifugation, re-suspended in 0.1 M Tris/HCl (pH 8.0), 5m M K-EDTA, and disrupted by sonication. Inclusion bodies were pelleted by centrifugation 2 (5 min, 16 000 g)and washed twice with the same buffer. Solubilization of inclusion bodies and in vitro refolding of trypsinogen was performed as described previously [11–14], in 0.9 M guani- dine-HCl, 0.1 M Tris/HCl (pH 8.0), 2 m M K-EDTA con- taining 1 m ML -cystine and 1 m ML -cysteine. Refolded trypsinogens were purified to homogeneity by ecotin-affinity chromatography [18]. Both trypsinogens were stable when stored in 50 m M HCl on ice for several weeks. Concentra- tions of zymogen solutions were determined from their ultraviolet absorbance at 280 nm using calculated extinction coefficients of 36 160 M )1 Æcm )1 and 37 320 M )1 Æcm )1 for cationic and anionic trypsinogens, respectively. Autoactivation of trypsinogens Aliquots of trypsinogens (2 l M final concentrations) were incubated at 37 °Cin0.1 M Tris/HCl (pH 8.0) or 0.1 M Na-acetate buffer (pH 5.0), in the absence or presence of indicated concentrations of CaCl 2 in a final volume of 100 lL. Where indicated, 100 m M NaCl or 100 m M NaCl and 2 mgÆmL )1 BSA was included in the activation mixtures. At given times, 2.5 lL aliquots were removed for trypsin activity assays. Trypsin activity was determined using the synthetic chromogenic substrate, GPR-pNA (0.14 m M final concentration) in 200 lL final volume. Kinetics of the chromophore release was followed at 405 nm in 0.1 M Tris/HCl (pH 8.0), 1 m M CaCl 2 ,at22°C using a Spectramax Plus 384 microplate reader (Molecular Devices). Trypsin activity was expressed as percentage of the potential maximal activity, that was determined by entero- kinase activation (400 ngÆmL )1 final concentration) in 0.1 M Tris/HCl (pH 8.0), 10 m M CaCl 2 ,at22°Cfor60minon separate trypsinogen samples. Autolysis of trypsins Trypsinogens ( 10 l M final concentration) were activated with bovine enterokinase ( 1 lgÆmL )1 final concentration) in 0.1 M Tris/HCl (pH 8.0), 20 m M CaCl 2 ,at0°Cfor 120 min and loaded onto an ecotin column. Enterokinase, that does not bind to ecotin, was washed away with 20 m M Tris/HCl (pH 8.0), 0.2 M NaCl and trypsin was eluted with 50 m M HCl. Autocatalytic degradation of trypsin was followed by residual activity measurements at 37 °Cin 0.1 M Tris/HCl (pH 8.0) in the presence of the indicated concentrations of CaCl 2 . Where indicated, 100 m M NaCl was also included. At given times, 2.5 lL aliquots were removed and trypsin activity was determined using GPR- pNA (0.14 m M final concentration) in 200 lL final volume. Trypsin activity was expressed as percentage of the initial activity measured at the beginning of the incubation. SDS/PAGE analysis of trypsinogens Autoactivation and degradation of trypsinogens was also visualized by gel electrophoresis and staining. Typically, samples containing 2 l M trypsinogen in 100 lL volume were precipitated with trichloroacetic acid (10% final concentration), the precipitate was pelleted in an Eppendorf microcentrifuge, and solubilized in 20 lL2· Laemmli sample buffer. Trichloroacetic acid was neutralized with NaOH until the yellow color of the acidified Bromophenol Blue turned blue (1–2 lLof2 M NaOH), and dithiothreitol was added to a final concentration of 100 m M .Samples 2048 Z. Kukor et al. (Eur. J. Biochem. 270) Ó FEBS 2003 were heat-denatured at 95 °C for 5 min, and loaded onto 12% mini-gels. Gels were run at 30 mA, and stained for 30 min with a 0.5% Brilliant Blue R (Acros Organics, New Jersey, NJ, USA) solution containing 40% methanol and 10% acetic acid, followed by overnight de-staining with 30% methanol, 10% acetic acid. Where indicated, densito- metric quantitation of bands was also carried out. Gels were dried between two layers of cellophane according to the instructions of the Gel-Dry gel drying kit (Invitrogen). Dried gels were scanned at 600 d.p.i. resolution in gray-scale mode, and images were saved as TIFF files. Quantitation of gel bands was carried out with the IMAGEQUANT 5.2 (Molecular Dynamics) software. Rectangles were drawn around each band of interest, and an identical rectangle was used in each lane for background subtraction. Results Recombinant expression of human anionic trypsinogen The gene for human anionic trypsinogen was cloned under the control of the T7 promoter into the pTrap-T7 expression vector [11,14], that was developed originally for the expres- sion of human cationic trypsinogen. Over-expression of anionic trypsinogen was achieved in E. coli strains carrying an inducible T7 RNA polymerase gene, as described in Experimental procedures. Inclusion bodies containing denatured trypsinogen were isolated, solubilized with guani- dine-HCl and subjected to in vitro refolding [11–13]. To ensure that only trypsinogen that regained native confor- mation was used in the following experiments, the refolded material was purified to homogeneity via inhibitor-affinity chromatography on immobilized ecotin [18]. A single peak eluted from the ecotin column, suggesting that homogenous trypsinogen was obtained. Homogeneity was further con- firmed by anion-exchange chromatography (MonoQ) and size-exclusion chromatography (Superose 6), where the preparation also yielded single peaks (not shown). Further- more, analysis of the purified samples by native PAGE or SDS/PAGE revealed single bands, excluding the presence of multiple forms or oligomerization (not shown). Catalytic parameters of recombinant anionic trypsin (K M 11 ± 1 l M ; k cat 41 ± 1 s )1 ) were very similar to those of cationic trypsin (K m 15 ± 1 l M ; k cat 50 ± 1 s )1 ), as determined with the chromogenic peptide substrate GPR-pNA. The turnover number of anionic trypsin was also comparable to values reported previously for native trypsins on small synthetic substrates [13,19]. Finally, anionic trypsin was inhibited with a 1 : 1 stoichiometry by human pancreatic secretory trypsin inhibitor (not shown). Autoactivation of human trypsinogens at pH 8.0 Autoactivation was measured in 0.1 M Tris/HCl (pH 8.0), at 37 °C, both in the physiologically relevant Ca 2+ concentration range (0–1 m M , Fig. 1A), and in a higher, unphysiological Ca 2+ concentration range (1–20 m M , Fig. 1B), that is frequently used in biochemical assays. Human anionic trypsinogen exhibited minimal autoactiva- tion in the absence of Ca 2+ or at Ca 2+ concentrations up to 0.1 m M (Fig. 1A), and significant autoactivation was observed only at Ca 2+ concentrations of 0.5 m M and above. The rate of autoactivation increased up to 5 m M Ca 2+ , while higher Ca 2+ concentrations (10 m M and 20 m M ) slightly inhibited the activation rate, but still resulted in higher levels of trypsin (Fig. 1B). Analysis of anionic trypsinogen samples by SDS/PAGE revealed that the lack of autoactivation at 0.1 m M Ca 2+ and below was a consequence of massive zymogen degradation (Fig. 1C). Thus, in 50 l M Ca 2+ , the trypsinogen band disap- peared completely by 30 min, while a trypsin band was hardly visible. The rapid degradation at this low Ca 2+ Fig. 1. Autoactivation of human anionic trypsinogen. Approximately 2 l M trypsinogen (final concentration, in a final volume of 100 lL) was incubated at 37 °C, in 0.1 M Tris/HCl (pH 8.0) with the indicated concentrations of CaCl 2 . (A,B) Aliquots of 2.5 lLwerewithdrawn from reaction mixtures at indicated times and trypsin activity was determined with 0.14 m M (final concentration) GPR-pNA. Activity was expressed as percentage of the potential total activity, as deter- mined on similar zymogen samples activated with enterokinase at 22 °Cin0.1 M Tris/HCl (pH 8.0), 10 m M Ca 2+ .(C)Sampleswere precipitated with 10% trichloroacetic acid (final concentration), run on a 12% SDS/PAGE minigel under reducing conditions, and stained with Coomassie Blue. Ó FEBS 2003 Human anionic trypsinogen (Eur. J. Biochem. 270) 2049 concentration resulted only in faintly visible bands of larger peptide fragments, as the bulk of the protein was digested to small peptides. In contrast, in the presence of 5 m M Ca 2+ , the trypsinogen band was converted to trypsin and stable autolysis products were also detected. The appearance of larger peptide fragments at high Ca 2+ concentration was due to the significantly slower degradation rate and possibly the selective protection of certain cleavage sites by Ca 2+ . Human cationic trypsinogen exhibited characteristic differences from its anionic counterpart. In 0.1 M Tris/ HCl(pH8.0),at37°C, autoactivation was measurable even in the absence of added Ca 2+ , and it was significantly stimulated by Ca 2+ concentrations as low as 10 l M (Fig. 2A). Ca 2+ stimulated autoactivation in a concentra- tion-dependent manner up to 1 m M , while above this concentration autoactivation was progressively inhibited (Fig. 2B). As addition of 100 m M NaCl also significantly decreased the rate of autoactivation 3 or autolysis (see below), it appears that inhibition by the nonphysiologically high Ca 2+ concentrations was caused by ionic strength. Analysis of the Ca 2+ dependence of autoactivation revealed a biphasic activation curve (Fig. 2C); a typical saturation curve with an apparent EC 50 of  15 l M was followed by linear concentration dependence. The apparent half-maxi- mal stimulatory Ca 2+ concentration (15 l M ) was compar- able to the Ca 2+ concentration that stabilized cationic trypsin against autolysis half-maximally (20 l M ;seebelow).Tryp- sin stabilization by Ca 2+ is accomplished via binding to the high-affinity Ca 2+ binding site composed of five residues, between Glu75 and Glu85. Consequently, the observation that low concentrations of Ca 2+ stimulate autoactivation of cationic trypsinogen suggest that Ca 2+ exerts this effect through the same high affinity Ca 2+ binding site. Ca 2+ concentrations between 0.1 m M )1m M further stimulated autoactivation by binding to the low affinity site in the activation peptide. Determination of an EC 50 for the latter process was not feasible due to the inhibitory effect of Ca 2+ concentrations above 1 m M . SDS/PAGE analysis of autoactivation of human cationic trypsinogen at pH 8.0 was described in our previous studies (e.g. see Figs 3,4 in [11] and Fig. 1 in [16]). At pH 8.0, in the presence of 1 m M Ca 2+ , the typical banding pattern of autoactivated cationic trypsinogen is essentially identical to the picture shown below, which demonstrates autoactiva- tion of human cationic trypsinogen at pH 5.0. A notable feature of human cationic trypsin(ogen) is that it exists as an equilibrium mixture of single-chain and double-chain forms. The double-chain form, that in every functional aspect appears to be identical to the single-chain form, is generated by autocatalytic cleavage of the Arg122-Val123 peptide bond. The dynamic equilibrium between the two forms is maintained by continuous trypsin-dependent cleavage and resynthesis of the Arg122-Val123 bond. On reducing SDS/ PAGE gels, that dissociate the two chains, double-chain trypsinogen appears as a 15-kDa band, containing both the N- and C-terminal chains that are identical in size (band A). Activation of double-chain trypsinogen to double-chain trypsin results in the appearance of band B, which corresponds to the N-terminal chain of double-chain trypsin. For a more detailed description of the unique properties of double-chain trypsin(ogen) the reader is referred to our recent study [16]. Trypsinolytic degradation of human trypsinogens at pH 8.0 One of the striking observations from the comparative autoactivation studies of human trypsinogens at pH 8.0 was the marked susceptibility of anionic trypsinogen to auto- catalytic degradation. To get a more accurate comparison for the rates of zymogen degradation between the two trypsinogens, Lys23 in the activation peptide was replaced with Gln in human anionic trypsinogen. The resulting Fig. 2. Autoactivation of human cationic trypsinogen. Experimental conditions are given in Fig. 1. (A) Stimulation of autoactivation in the Ca 2+ concentration range 0.01 m M )1m M . (B) Inhibition of auto- activation by Ca 2+ in the concentration range 1 m M )20 m M .(C) Relative rates of autoactivation were plotted against the Ca 2+ con- centration between 0.01 m M )0.5 m M . The rate of autoactivation without any added Ca 2+ (0 m M inA)wasdesignatedas1. 2050 Z. Kukor et al. (Eur. J. Biochem. 270) Ó FEBS 2003 K23Q mutant trypsinogen is resistant to autoactivation, allowing selective examination of trypsinolytic zymogen degradation. Anionic K23Q-trypsinogen was purified to homogeneity and 5 l M zymogen was used as substrate in digestion experiments with 0.5 l M cationic trypsin as enzyme. Cationic trypsin was used, because it remained stable without significant loss of activity during the time course studied. Figure 3A demonstrates that in the absence of Ca 2+ (in 1 m M EDTA) cationic trypsin rapidly degraded anionic K23Q-trypsinogen, and densitometric quantitation indicated a half-life (t 1/2 ) of 2.25 min (Fig. 3C). Addition of 50 l M Ca 2+ (final concentration) afforded significant (fourfold) stabilization, and prolonged the t 1/2 to 10 min (Fig. 3B,C). Using the same strategy, in a recent study we determined the degradation of a K23Q-mutant of cationic trypsinogen by cationic trypsin [16]. At pH 8.0, in the absence of Ca 2+ (in 1 m M EDTA) 5 l M cationic K23Q- zymogen was degraded by 0.5 l M trypsin with a t 1/2 of 45 min. Thus, cationic trypsinogen is 20-fold more resistant to trypsinolytic degradation than anionic trypsinogen is. For comparison, densitometric quantitation data for K23Q cationic trypsinogen were also included in Fig. 3C. Autoactivation of human trypsinogens at pH 5.0 In contrast to the rapid autoactivation at pH 8.0, anionic trypsinogen autoactivated much slower at pH 5.0 (Fig. 4A), and Ca 2+ -stimulated autoactivation in a concentration dependent manner between 0.5 m M and 5 m M . Maximal levels of trypsin generation in 5 m M Ca 2+ did not exceed 30% of the total potential activity, indicating significant zymogen degradation. No further stimulation was apparent with 10 m M Ca 2+ , while 20 m M Ca 2+ inhibited the rate of autoactivation and yielded somewhat higher trypsin levels. SDS/PAGE analysis of anionic trypsinogen samples revealed that the lack of autoactivation at pH 5.0 in the absence of Ca 2+ was not due to rapid zymogen degrada- tion, as observed at pH 8.0 (see Fig. 1). Instead, zymogen activation was inhibited by the acidic conditions, and a stable trypsinogen band was observed over the 120 min Fig. 3. Degradation of K23Q-anionic trypsinogen (Hu2) by human cationic trypsin. Approximately 5 l M trypsinogen (final concentration, in a final volume of 100 lL) was digested with 0.5 l M cationic trypsin at 37 °C, in 0.1 M Tris/HCl (pH 8.0) in 1 m M EDTA (A) or in 50 l M Ca 2+ (B). Reactions were terminated at indicated times by trichloro- acetic acid precipitation, and analyzed by reducing SDS/PAGE and Coomassie Blue staining. In the 0 min samples, trichloroacetic acid was added before trypsin. (C) Densitometric quantitation of gels (n ¼ 3, error less than 15%). Also shown are data from ref [16], where trypsinolytic degradation of the K23Q mutant of human cationic trypsinogen (Hu1) was determined under identical conditions. Fig. 4. Autoactivation of human anionic trypsinogen at pH 5.0. Approximately 2 l M trypsinogen (final concentration, in a final vol- ume of 100 lL) was incubated at 37 °C, in 0.1 M Na-acetate buffer (pH 5.0) with the indicated concentrations of CaCl 2 .(A)Trypsin activity was determined and expressed as described in Fig. 1. (B) Samples (2 l M zymogen in 100 lL) were trichloroacetic acid-precipi- tated and analyzed by reducing SDS/PAGE (12%) and Coomassie Blue staining. Ó FEBS 2003 Human anionic trypsinogen (Eur. J. Biochem. 270) 2051 time-course studied (Fig. 4B). Addition of 5 m M Ca 2+ stimulated conversion of trypsinogen to trypsin, as a faint trypsin band became apparent at 60 min, and more significant trypsin generation was detectable by 120 min. In the absence of Ca 2+ , autoactivation of cationic trypsinogen was more rapid at pH 5.0 (Fig. 5A) than at pH 8.0 (Fig. 2). At pH 5.0, Ca 2+ caused a slight stimu- lation up to 1 m M , and inhibited autoactivation in a concentration-dependent manner between 2 m M and 20 m M (Fig. 5B). Comparing time-courses of autoactiva- tion at pH 5.0 on SDS/PAGE gels confirmed that in the absence of Ca 2+ anionic trypsinogen was not activated (see Fig. 4B), while cationic trypsinogen was fully activated over Fig. 5. Autoactivation of human cationic trypsinogen at pH 5.0. Approximately 2 l M trypsinogen (final concentration, in a final vol- ume of 100 lL) was incubated at 37 °C, in 0.1 M Na-acetate buffer (pH 5.0) with the indicated concentrations of CaCl 2 . (A,B) Trypsin activity was determined and expressed as described in Fig. 1. (A) Slight stimulation of autoactivation by Ca 2+ concentrations up to 1 m M .(B) Inhibition of autoactivation by Ca 2+ concentrations above 1 m M .(C) Samples (2 l M zymogen in 100 lL) were trichloroacetic acid-precipi- tated and analyzed by reducing SDS/PAGE (12%) and Coomassie Blue staining. Bands A and B correspond to the two chains of double- chain trypsin(ogen), see text for more explanation. Fig. 6. Autocatalytic degradation (autolysis) of human anionic trypsin. Trypsinogen was activated by enterokinase and purified on an ecotin- column, as described in Experimental procedures. Approximately 2 l M aliquots of anionic trypsin (final concentration) were incubated at 37 °Cin0.1 M Tris/HCl (pH 8.0) in the presence of the indicated concentrations of CaCl 2 . Aliquots of 2.5 lLwerewithdrawnfrom reaction mixtures at indicated times and trypsin activity was deter- minedwith0.14m M GPR-pNA (final concentration). Residual activities were expressed as percentage of trypsin activity measured at the beginning of the incubation. (B) Autolysis in the presence of 100 m M NaCl. (C) Effect of Ca 2+ ontherelativerateofautolysisinthe presence (s) or absence (d) of 100 m M NaCl. The rate determined in the absence of added Ca 2+ was designated as 1. 2052 Z. Kukor et al. (Eur. J. Biochem. 270) Ó FEBS 2003 the same time period (Fig. 5C). Conversion of cationic trypsinogen to trypsin was practically quantitative, with no significant zymogen degradation, and addition of 5 m M Ca 2+ had only a minor effect on the rate of autoactivation (Fig. 5C). Autolysis of human trypsins at pH 8.0 Previous experiments using purified native human trypsins indicated that human anionic trypsin was less stable and underwent faster autolysis than cationic trypsin [19,20]. To characterize the autolytic process of the recombinant trypsin preparations in more detail, we purified human anionic and cationic trypsin after enterokinase activation of the respective recombinant zymogens. In the absence of Ca 2+ , anionic trypsin suffered autolysis at a rapid rate (t 1/2 8min), and low concentrations of Ca 2+ stabilized the enzyme, with an IC 50 of 5 l M (Fig. 6A,C). Addition of 100 m M NaCl to anionic trypsin decreased the rate of autolysis threefold (in the absence of Ca 2+ t 1/2 was 24 min, Fig. 6B), and increased the IC 50 value for Ca 2+ stabilization sixfold (30 l M , Fig. 6C). Autolysis of cationic trypsin was signifi- cantly slower, and in the absence of Ca 2+ a t 1/2 of 90 min was measured (Fig. 7A). Thus, in the absence of Ca 2+ a >11-fold difference was apparent between the autolysis rates of the two trypsins (Figs 6A and 7A). Low concen- trations of Ca 2+ afforded stabilization with an IC 50 value of 20 l M (Fig. 7A and B). Surprisingly, addition of 100 m M NaCl diminished autolysis of cationic trypsin 14-fold, and even in the absence of Ca 2+ it took almost 21 h to observe a Fig. 7. Autocatalytic degradation (autolysis) of human cationic trypsin. (A) See Fig. 6 for experimental details. (B) Effect of Ca 2+ on the relative rate of autolysis. The rate determined in the absence of added Ca 2+ was designated as 1. Fig. 8. Autoactivation of physiological and pathological mixtures of human trypsinogens at pH 8.0, in 1 m M Ca 2+ . Autoactivation experi- ments were carried out as described in Fig. 1. Hu1 (h), human cationic trypsinogen (2 l M ); Hu2 (s), human anionic trypsinogen (2 l M ). Physiological mixtures (j) contained 1.33 l M (67%) Hu1 and 0.67 l M (33%) Hu2 trypsinogen. Pathological mixtures (d)contained0.67l M (33%) Hu1 and 1.33 l M (67%) Hu2 trypsinogen. Experiments were carried out under three conditions, in buffer only (A), in 100 m M NaCl (B) and in 100 m M NaCl with 2 mgÆmL )1 BSA (C). Ó FEBS 2003 Human anionic trypsinogen (Eur. J. Biochem. 270) 2053 50% loss of activity (not shown). Thus, there is a remarkable difference in salt sensitivity between the two human trypsins with respect to autolysis. Interactions between anionic and cationic trypsinogens and trypsins during trypsinogen activation The experiments presented in Figs 1–7 provided a detailed biochemical characterization of the autocatalytic activa- tion and degradation of the two major human trypsino- gens. The notably different behavior of the two zymogens suggested that changes in their ratio should have profound effects on the overall stability of the pancreatic trypsinogen pool and its susceptibility to autoactivation. To model these changes in vitro, we examined the effect of increasing anionic trypsinogen proportions in different mixtures of the two trypsinogens. In these experiments, the two human trypsinogens were mixed at two different ratios, 2 : 1 (physiological mixture; 67% cationic trypsinogen and 33% anionic trypsinogen) or 1 : 2 (pathological mixture, 33% cationic trypsinogen and 67% anionic trypsinogen). Autoactivation experiments were carried out at pH 8.0 andpH5.0.AtpH8.0,twodifferentCa 2+ concentrations were used, 1 m M or 50 l M .The1m M Ca 2+ concentration was selected to model the conditions in the pancreatic juice or in the duodenum, the physiological site of trypsinogen activation. The 50 l M Ca 2+ concentration modeled the intracellular conditions, where Ca 2+ concentrations are low. Although true cytoplasmic Ca 2+ concentrations are below micromolar levels, we chose to use 50 l M Ca 2+ because at this concentration autoactivation was somewhat faster and full time-courses could be analyzed within reasonable time limits. Qualitatively identical results were obtained when experiments were repeated at pH 8.0 without any added Ca 2+ . Finally, experiments at pH 5.0 modeled conditions in acidic intracellular vesicular compartments, that are known sites of pathological trypsinogen activation [21,22]. In addition, it was also important to demonstrate that any differences observed also existed in the presence of salts or other proteins, as the routinely used in vitro biochemical system obviously lacked the variety of salts and proteins present in the intra-acinar environment or in pancreatic secretions. Therefore, in addition to experiments performed in buffer only, autoactivation of mixtures was also compared in the presence of 100 m M NaCl or in the presence of 100 m M NaCl and 2 mgÆmL )1 BSA. AsshowninFig.8A,in0.1 M Tris/HCl (pH 8.0) and 1m M Ca 2+ , autoactivation of the two trypsinogens proceeded at comparable rates, but resulted in a twofold difference in final trypsin levels (Fig. 8A, white symbols). As demonstrated above (see Figs 1,2), this difference is due to the more rapid degradation of anionic trypsin(ogen) during autoactivation. Interestingly, when the two trypsino- gens were mixed either in a physiological or in a pathological mixture, rates of autoactivation did not change appreciably and final trypsin levels differed only by 20% (Fig. 8A, black symbols). Similarly, the physiolo- gical activator, enterokinase, generated approximately identical amounts of trypsin from both mixtures (not shown). Addition of 100 m M NaCl drastically reduced the rate of autoactivation by cationic trypsinogen, while anionic trypsinogen was much less affected (Fig. 8B, white symbols). Mixtures of the two trypsinogens, however, exhibited not too different activation rates and yielded essentially identical trypsin levels (Fig. 8B, black symbols). Finally, in the presence of 100 m M NaCl and 2 mgÆmL )1 BSA autoactivation of the two trypsinogen mixtures exhibited rates and final trypsin levels that showed a  20% difference only (Fig. 8C, black symbols). Interest- ingly, the BSA preparations used noticeably inhibited autoactivation of anionic trypsinogen (compare Figs 8B,C, white circles), while cationic trypsinogen was not affected. Although this problem was not investigated any further, this effect was in all likelihood due to the contaminating presence of a serum trypsin inhibitor in some of the commercial BSA preparations. A different picture emerged when experiments were performed in the presence of 0.1 M Tris/HCl (pH 8.0) with 50 l M Ca 2+ . Under all three conditions tested, cationic trypsinogen autoactivated to significant levels, while essen- tially no trypsin generation was detectable with anionic trypsinogen (Fig. 9A–C, white symbols), due to practically total degradation (see Fig. 1). Accordingly, autoactivation of mixtures of the two trypsinogens was proportional to the cationic trypsinogen content, and pathological mixtures consistently exhibited activation rates and final trypsin levels that were at least twofold lower compared to physiological mixtures (Fig. 9A–C, black symbols). Experiments at pH 5.0 showed similar differences in the autoactivation characteristics of the two types of trypsino- gen mixtures. Once again, rates of autoactivation seemed to reflect the cationic trypsinogen content and autoactiva- tion rates of pathological mixtures were markedly sup- pressed (Fig. 10A–C, black symbols). Clearly, this difference was caused by the inability of anionic trypsino- gen to autoactivate at this acidic pH (Fig. 10A–C, white circles; also see Fig. 4). Due to the extended time-courses, final trypsin levels were not determined accurately, but it appeared that pathological mixtures should yield at least twofold less trypsin than physiological mixtures (Fig. 10A). The anomalous and distinct migration of the two human trypsinogens on SDS/PAGE gels allowed the visualization of both species present in the mixtures (Fig. 11). In 0.1 M Tris/HCl (pH 8.0) with 50 l M Ca 2+ , both mixtures contained only active cationic trypsin by the end of the 60 min incubation (Fig. 11A). Both the single-chain form and the double-chain form (denoted by bands A and B in Fig. 11) were observed. In agreement with the activity assays, the stronger intensity of the cationic trypsin band in the physiological mixture was noticeable. Furthermore, rapid disappearance of the anionic trypsinogen band without the appearance of a clearly detectable anionic trypsin band was also evident in both mixtures. Taken together, the observations confirmed that under these conditions (pH 8.0, 50 l M Ca 2+ , Fig. 11A) selective degra- dation of anionic trypsinogen resulted in lower trypsin generation in pathological mixtures of the two human trypsinogens. Finally, at pH 5.0 cationic trypsinogen was completely activated to trypsin in the physiological mix- ture, while significant amounts of anionic trypsinogen remained unactivated in the pathological mixture, reflect- ing the resistance of this trypsinogen species to activation at acidic pH (Fig. 11B,C). 2054 Z. Kukor et al. (Eur. J. Biochem. 270) Ó FEBS 2003 Discussion How do the two major trypsinogen isoforms of the human pancreas interact with respect to autocatalytic activation and degradation? What are the biochemical consequences of the up-regulation of anionic trypsinogen in pancreatic secretions of patients with pancreatic diseases or chronic alcoholism? To address these questions, we recombinantly produced human anionic trypsinogen and purified it in a stable form. Although recombinant expression of anionic trypsin activity per se was reported in a few studies [6,13], pure and stable zymogen preparations were difficult to achieve, due to the notoriously unstable nature of this trypsinogen isoform. In this respect, methodology devel- oped earlier for the recombinant production of human Fig. 10. Autoactivation of physiological and pathological mixtures of human trypsinogens at pH 5.0. Autoactivation experiments were car- ried out in 0.1 M Na-acetate buffer (pH 5.0). See Fig. 8 for other experimental details. Fig. 9. Autoactivation of physiological and pathological mixtures of human trypsinogens at pH 8.0, in 50 l M Ca 2+ . See Fig. 8 for experi- mental details. Ó FEBS 2003 Human anionic trypsinogen (Eur. J. Biochem. 270) 2055 cationic trypsinogen was critical [11–13], including the use of immobilized ecotin for the final purification step [18]. To understand the behavior of trypsinogens in more complex mixtures, first we documented their properties individually, under the typical experimental conditions used in recent literature. At least four major differences were observed. (a) Trypsinolytic degradation of anionic trypsi- nogen or trypsin was 10 to 20-fold faster. As a consequence of their highly different stability, the two trypsinogens exhibited distinct autoactivation profiles. Thus, essentially no trypsin activity was detectable during autoactivation of anionic trypsinogen at pH 8.0 in 0.1 m M Ca 2+ or lower. At these Ca 2+ concentrations autoactivation was relatively slow, and could not keep up with pace of trypsin(ogen) degradation. Only in millimolar Ca 2+ concentrations was significant autoactivation detected, when the rate of auto- activation exceeded the rate of degradation. In contrast, because degradation of cationic trypsin(ogen) was much slower, autoactivation resulted in the development of significant trypsin activity even in the absence of added Ca 2+ . (b) Acidic pH stimulated autoactivation of cationic trypsinogen, but inhibited activation of anionic trypsinogen. (c) Anionic trypsin bound Ca 2+ fourfold stronger than cationic trypsin, as judged by the stabilizing effect of Ca 2+ on autolysis. Binding of Ca 2+ to the high-affinity site also stimulated autoactivation of cationic trypsinogen, while this effect was either absent in anionic trypsinogen or it was masked by the rapid degradation. Interestingly, Ca 2+ in the concentration range between 1 m M and 10 m M stimulated autoactivation of anionic trypsinogen, almost in a manner that was observed for bovine trypsinogen [23] or rat anionic trypsinogen [24]. In contrast, autoactivation of cationic trypsinogen was progressively inhibited by Ca 2+ concen- trations between 1 m M )20 m M . Although this observation is important for the correct interpretation of autoactivation assays performed under a variety of Ca 2+ concentrations in the literature; the (patho)physiological significance of such a Ca 2+ -mediated inhibition mechanism is questionable. (d) Autoactivation of cationic trypsinogen and autolysis of cationic trypsin were markedly inhibited by 100 m M NaCl, while anionic trypsin(ogen) was significantly less sensitive to this salt effect. In physiological terms, this observation would suggest that anionic trypsinogen can autoactivate much faster than cationic trypsinogen under conditions prevailing in the pancreatic juice. Trypsinogen autoactivation and degradation were studied previously with purified native trypsinogen preparations [19,20,25–27]. Although experimental conditions (pH, tem- perature, salt and buffer concentrations) were frequently varied in these studies, some of the results regarding the characteristic differences between the two human tryp- sin(ogen)s were similar to our findings. Thus, compared to the other isoform, anionic trypsin exhibited much more rapid autolysis, and cationic trypsinogen autoactivated more prominently at acidic pH. On the other hand, our observa- tions disagree with previous results in some detail. In our study, cationic trypsin was more stable in the absence of Ca 2+ than reported before [19]. Relative to cationic trypsi- nogen, anionic trypsinogen autoactivated faster at pH 8.0 in 20 m M Ca 2+ whereas the opposite relationship was described previously [25,26]. Finally, in our experiments, high Ca 2+ concentrations consistently inhibited autoactiva- tion of cationic trypsinogen, while both stimulation and inhibition was found in early studies [25–27]. Characterization of the individual trypsinogens set the stage for the analysis of their mixtures. These experiments sought to answer one question: what happens to trypsino- gen activation and degradation when the normal ratio of cationic and anionic trypsinogen is reversed, as seen in pancreatic diseases or chronic alcoholism? The results indicated that trypsin generation by autoactivation or enterokinase activation was not affected significantly by the ratio of the two isoforms, under conditions that were typical of the pancreatic juice. This observation suggests that the primary trypsin functions, i.e. activation of other zymogens and digestion of ingested proteins; are unaffected by up-regulation of anionic trypsinogen. In contrast, trypsinogen activation was markedly diminished by an increased ratio of anionic trypsinogen under conditions that mimicked potential intracellular sites of pathological tryp- sinogen activation, such as the cytoplasm or acidic vesicles. Increasing the ratio of anionic trypsinogen resulted in decreased overall trypsin generation at pH 8.0 in the presence of low Ca 2+ concentrations, due to the selective degradation of anionic trypsin(ogen). Similarly, total trypsin formation was suppressed at pH 5.0, where the acidic pH selectively inhibited activation of anionic trypsinogen. Under both conditions, the concentration of cationic trypsinogen seemed to determine the rate of autoactivation and the final levels of trypsin generated. Consequently, an Fig. 11. Autoactivation of physiological (67% Hu1–33% Hu2) and pathological (33% Hu1–67% Hu2) mixtures of human trypsinogens at pH 8.0 in 50 l M Ca 2+ (A) and pH 5.0 (B and C). Autoactivation experiments were carried out asin Fig. 9A (A) and 10(B and C). Samples (2 l M total zymogen in 100 lL) were trichloroacetic acid-precipitated and analyzed by reducing SDS/PAGE (12%) and Coomassie Blue staining. Panel C is an enlargement from the 120 min lanes of panel B, demonstrating the resolution of the four human trypsin(ogen) species in the gel. Tg, trypsinogen; bands A and B correspond to the two chains of double-chain trypsin(ogen), see text for more explanation. 2056 Z. Kukor et al. (Eur. J. Biochem. 270) Ó FEBS 2003 [...]... role in pancreatic physiology As a defensive mechanism, acinar cells increase secretion of the anionic isoform in pancreatic diseases or toxic conditions, thereby decreasing the chance for premature trypsinogen activation inside the pancreas, while maintaining acceptable trypsin function in the duodenum In other words, increased anionic trypsinogen levels do not cause or predispose to pancreatitis; instead,... immunoreactive anionic and cationic trypsinogen in urine and serum in human acute pancreatitis Int J Pancreatol 25, 165–170 ´ 11 Sahin-Toth, M (2000) Human cationic trypsinogen Role of Asn-21 in zymogen activation and implications in hereditary pancreatitis J Biol Chem 275, 22750–22755 ´ 12 Sahin-Toth, M (2001) The pathobiochemistry of hereditary pancreatitis: studies on recombinant human cationic trypsinogen Pancreatology... increase in the proportion of the anionic trypsinogen resulted in a Õloss of trypsinogen functionÕ, as the ratio of cationic trypsinogen was decreased Although this study clarified the biochemical consequences of increased anionic trypsinogen secretion, the (patho)physiological sequelae have remained contentious The most straightforward interpretation of the results is to suggest that anionic trypsinogen. .. Guy, O (1979) The two human trypsinogens Evidence of complex formation with basic pancreatic trypsin inhibitor – proteolytic activity Biochim Biophys Acta 570, 397–405 27 Figarella, C., Miszczuk-Jamska, B & Barrett, A.J (1988) Possible lysosomal activation of pancreatic zymogens Activation of both human trypsinogens by cathepsin B and spontaneous acid activation of human trypsinogen 1 Biol Chem Hoppe-Seyler... latter conditions the increased trypsinogen synthesis may render the pancreas more susceptible to inappropriate zymogen activation, despite the protective effects of anionic trypsinogen In contrast to a possible safeguard role, chronically increased anionic trypsinogen levels and ensuing lower intrapancreatic trypsin concentrations may also be regarded as a disease-causing factor This interpretation relies... trypsin(ogen) and possibly other proteases thus preventing the escalation of intra-acinar digestive enzyme activation [28] More recently, a genetically engineered mouse deficient in one of the zymogen granule membrane proteins (integral membrane-associated protein-1, Itmap-1) was shown to develop more severe secretagogue- and dietinduced experimental pancreatitis, but with diminished intrapancreatic trypsinogen. .. effect of calcium and other ions on the autocatalytic formation of trypsin from trypsinogen J General Physiol 25, 53–73 ´ ´ 24 Sahin-Toth, M & Toth, M (2000) High-affinity Ca2+ binding inhibits autoactivation of rat trypsinogen Biochem Biophys Res Comm 275, 668–671 25 Colomb, E & Figarella, C (1979) Comparative studies on the mechanism of activation of the two human trypsinogens Biochim Biophys Acta 571, 343–351... for trypsinogen- 1 and 2 in serum reveal preferential elevation of trypsinogen- 2 in pancreatitis J Laboratory Clin Med 115, 712–718 9 Borgstrom, A & Andren-Sandberg, A (1995) Elevated serum levels of immunoreactive anionic trypsin (but not cationic trypsin) signals pancreatic disease Int J Pancreatol 18, 221–225 10 Petersson, U., Appelros, S & Borgstrom, A (1999) Different patterns in immunoreactive anionic. .. Juhasz, G & ´ Graf, L (2001) Comparative in vitro studies on native and recombinant human cationic trypsins Cathepsin B is a possible pathological activator of trypsinogen in pancreatitis J Biol Chem 276, 24574–24580 ´ ´ 14 Sahin-Toth, M & Toth, M (2000) Gain -of- function mutations associated with hereditary pancreatitis enhance autoactivation of human cationic trypsinogen Biochem Biophys Res Commun 278,... Pal, G & Sahin-Toth, M (1998) Affinity purification of recombinant trypsinogen using immobilized ecotin Protein Expr Purif 12, 291–294 19 Colomb, E., Guy, O., Deprez, P., Michel, R & Figarella, C (1978) The two human trypsinogens: catalytic properties of the corresponding trypsins Biochim Biophys Acta 525, 186–193 2058 Z Kukor et al (Eur J Biochem 270) 20 Mallory, P.A & Travis, J (1973) Human pancreatic . acidic pH inhibited autoactivation of anionic trypsinogen. In mixtures of cationic and anionic trypsinogen, an increase in the proportion of the anionic proenzyme. Human anionic trypsinogen Properties of autocatalytic activation and degradation and implications in pancreatic diseases Zolta ´ n Kukor,