ARCHIVES OF BIOCHEMISTRY AND BIOPHYSICS
Vol. 340, No. 2, April 15, pp. 355–358, 1997 Article No. BB979925
Engineered Porcine Pepsinogen Exhibits Dominant Unimolecular Activation Takuji Tanaka and Rickey Y. Yada1 Department of Food Science, University of Guelph, Guelph, Ontario, N1G 2W1, Canada
Received September 4, 1996, and in revised form January 21, 1997
An engineered pepsinogen, which was a fusion protein of thioredoxin and pepsinogen, exhibited dominant self-activation (unimolecular reaction; intramolecular activation) in contrast to recombinant pepsinogen which exhibited both unimolecular and bimolecular reactions (intermolecular activation mediated by pepsin released during activation). At pH values of 1.1, 2.0, and 3.0, activation curves for the engineered pepsinogen were hyperbolic rather than sigmoidal, indicating that self-activation was the dominant activation mechanism in comparison to the slower bimolecular activation. To confirm which activation mechanism was dominant, an equal mole of pepsin was added to accelerate the bimolecular reaction during activation. The addition of exogenous pepsin did not affect the activation rate of the engineered pepsinogen but accelerated pepsinogen activation through the bimolecular reaction. The above results indicated that the engineered pepsinogen exhibited, primarily, a self-activation mechanism and that bimolecular activation was negligible. q 1997 Academic Press Key Words: intermolecular activation; intramolecular activation; fusion protein; pepsin; aspartic protease.
Pepsinogen, the zymogen of pepsin, is hydrolyzed at the Leu44p-Ile1 bond resulting in pepsin. This activation which occurs at acidic pH values changes the static charges of residues and is followed by the destabilization of the prosegment (1). Examination of the crystal structures of pepsin and pepsinogen indicates that some drastic conformation changes occur during activation (1–3). The activation mechanism of pepsinogen into pepsin can occur via two different mechanisms. One is a bimolecular reaction (an intermolecular reaction), in which a pepsin molecule converts pepsinogen into pepsin, and 1
To whom correspondence should be addressed. Fax: 519-8240847. E-mail:
[email protected].
the other is an unimolecular reaction (self-activation; an intramolecular reaction), in which a pepsinogen molecule cleaves itself to yield a pepsin molecule (4– 7). Al-Janabi et al., in a study of pepsinogen activation, observed that below pH 4.0, in the pH range which activation occurs, both unimolecular and bimolecular activations were observed. At pH 3.0, the bimolecular activation was 6.5 times faster than self-activation, while at pH 2.0 self-activation was 2 times faster (6). To determine the contribution of the unimolecular and bimolecular mechanisms to activation, the authors calculated a rate constant for only the unimolecular reaction under conditions which prevented bimolecular activation, e.g., activation in the presence of large amounts of substrate. The rate constants for bimolecular activation were calculated from the overall activation rate which was measured in the absence of large amounts of substrate (6, 8). It was assumed that the rate constants for both the unimolecular and bimolecular reactions did not change under these experimental conditions. In a recent study, we reported the construction of a fusion protein to express cloned pepsinogen as a soluble protein (9). This fusion protein can be directly processed into pepsin. In the present study, activation kinetics of the fused pepsinogen were compared to recombinant pepsinogen. The activation kinetics indicated that the engineered pepsinogen showed a different behavior from the native porcine pepsinogen used in the Al-Janabi et al. study (6). Results of the study may aid in the investigation of mutations in the prosegment region. MATERIALS AND METHODS Recombinant pepsinogen fused with thioredoxin (Trx-PG)2 was expressed in Escherichia coli as previously described (9). The expressed
2 Abbreviations used: Trx-PG, Escherichia coli thioredoxin–porcine pepsinogen A fusion protein; r-PG, recombinant porcine pepsinogen A; r-pepsin, recombinant porcine pepsin A.
355
0003-9861/97 $25.00 Copyright q 1997 by Academic Press All rights of reproduction in any form reserved.
AID
ABB 9925
/
6b32$$$$81
03-12-97 06:56:03
arca
356
TANAKA AND YADA
fusion protein was extracted by sonicating the cells which were recovered from 1.5 L culture. The fusion protein was precipitated from the crude extract using 40% saturated ammonium sulfate. After removing the salt by dialysis, the fusion protein was applied on a DEAE Sephacel column (5 1 25 cm, Pharmacia Biotech AB, Uppsala, Sweden). The column was washed with 0.2 M sodium chloride/1 M urea/20 mM Tris–HCl (pH 8.0) and then eluted with a linear gradient to 0.35 M sodium chloride/1 M urea/20 mM Tris–HCl (pH 8.0). The fractions containing the fusion protein were pooled. The pH was then decreased to 6.5 with 8 M phosphoric acid after which the pooled fractions were applied onto a DEAE Sepharose CL-6B column (2.5 1 25 cm, Pharmacia Biotech AB). The fusion protein was recovered with 450 mM phosphate buffer (pH 6.5) following washing with 300 mM phosphate buffer (pH 6.5). The ammonium sulfate concentration was then increased to 1 M and the fractions were applied on a PhenylSepharose Fast Flow column (2.5 1 20 cm, Pharmacia Biotech AB). The fusion protein was eluted with 20 mM Tris–HCl (pH 8.0) after washing the column with a linear gradient from 1 M ammonium sulfate to 0.2 M. Recombinant pepsin (r-pepsin) was obtained from Trx-PG as described previously (9). Recombinant pepsinogen (r-PG) was prepared from the fusion protein by tryptic digestion, and therefore, contained an equal molar concentration of thioredoxin. All chemicals were of the highest grade available. Activity measurements. The activities of samples were determined using the milk-clotting assay. A 1-mL aliquot of 0.32% (w/v) powdered skim milk solution was placed in an optical cuvette. After incubation at room temperature for at least 1 min to confirm a stable absorbance, 200 mL of enzyme solution was added. Time to reach A505 nm Å 0.4 was recorded. The inverse of the time in seconds was taken as the activity. pH dependency of activation. Trx-PG and r-PG (50 mL, 14.4 nM) were treated with 50 mL of 100 mM sodium citrate–HCl (pH 1.1, 1.5, 2.0, 2.5, 3.0, 3.5, 4.0, 4.5, 5.0, 5.5, and 6.0) for 1 h at room temperature. Activation was terminated by the addition of 50 mL of 1 M sodium acetate (pH 5.3). Activation kinetics. Two hundred microliters of Trx-PG (14.4 nM) or r-PG preparation (14.4 nM) solution were placed in a microtube. Activation was initiated by the addition of 400 mL of 48 mM sodium citrate–HCl (pH 1.1, 2.0, and 3.0). At 15-s intervals, 50 mL of the mixture were withdrawn and neutralized with 15 mL of 1 M Tris– HCl (pH 8.0). Samples were then left at 47C for 12 h in order to inactivate pepsin which had been activated. The residual fusion protein was activated by adding 3 mL of 6 N HCl. After a 30-min activation period at room temperature, activation was terminated with 100 mL of 1 M sodium acetate (pH 5.3). The activity of these later samples was taken as the amount of residual fused pepsinogen during the course of activation. To estimate the effects of pepsin on the activation, r-pepsin (a 1:1 molar ratio to Trx-PG or r-PG) was added simultaneously at the time of acidification. Each assay was conducted a minimum of three times. Calculation of activation rate constants. The rate constants for activation were determined using a Guggenheim plot (10).
RESULTS AND DISCUSSION
Both r-PG and Trx-PG were activated at low pH values (Fig. 1). The plots indicated that both zymogens were similar upon activation and exhibited near 100% activation below pH 3.0. To estimate the effect of pH on activation kinetics, pH values of 1.1, 2.0, and 3.0 were used for further experiments. Activation kinetics of r-PG were plotted in Fig. 2. rPG at pH 2.0 and 3.0 (solid triangles and solid squares,
AID
ABB 9925
/
6b32$$$$81
03-12-97 06:56:03
FIG. 1. pH dependency of activation of recombinant pepsinogen (rPG) and fusion pepsinogen (Trx-PG). Activity of the samples was measured after 1 h activation. Each sample was plotted relative to the value at pH 1.1. Open circles represent r-pepsinogen samples, while solid circles represent Trx-PG samples. Each data point represents the mean of three determinations.
respectively, in Fig. 2A) exhibited an initial lag phase after which the rate of activation accelerated. This observation would indicate that the pepsin molecule which is initially activated from r-PG by a unimolecular reaction began to hydrolyze other r-PG molecules (bimolecular activation) and that bimolecular activation became the dominant reaction (due to increasing amounts of released pepsin) compared to self-activation. At pH 1.1, the activation curve for r-PG was hyperbolic (solid circles in Fig. 2A). These results would indicate that either the unimolecular activation was dominant or the lag phase of a sigmoidal curve was too short to be detected within the 15-s sampling intervals used in the present study. Results for recombinant pepsinogen were consistent with those previously reported for native pepsinogen (4–7). Activation experiments in the presence of exogenous r-pepsin (molar ratio to r-PG was 1:1) were used to confirm bimolecular activation. If the dominant activation mechanism was bimolecular, accelerated activation would be observed upon addition of exogenous pepsin. Activation of r-PG was noticeably accelerated even at pH 1.1 where a hyperbolic curve was observed in the absence of exogenous pepsin; therefore, in the presence of large amounts of exogenous pepsin, the much faster bimolecular activation is the predominant reaction such that unimolecular activation is not observed during the course of activation. When both bimolecular and unimolecular activations occur, first-order and second-order rate constants should be determined in order to determine the activation rate constants (6). The equation used to fit the data is 0D[r-PG]/Dt Å k1[r-PG] / k2[r-PG][r-Pepsin]
arca
DOMINANT SELF-ACTIVATION IN ENGINEERED PEPSINOGEN
357
and would describe a sigmoidal activation curve as long as k1 and k2 exist. No sigmoidal curve was observed at any of the pH values examined for the Trx-PGs. In addition, the fusion protein exhibited no difference (within error) in activation in the absence or presence of a 1:1 molar ratio of pepsin molecules (Fig. 2B), whereas r-PG activation was accelerated in the presence of exogenous pepsin (Fig. 2A). Again, if activation of Trx-PG by pepsin is much faster than self-activation, as was observed in r-PG, a faster activation would be expected with exogenous pepsin. No such effect was observed at pH 1.1 and 2.0. At pH 3.0, activation was slightly accelerated. The above results do not discount the existence of bimolecular activation of Trx-PG, but would suggest
FIG. 3. Three-dimensional structure of pepsinogen and thioredoxin and schematic scheme of activation of Trx-PG. (a) Three-dimensional structures of pepsinogen and thioredoxin (Protein Data Bank 3PSG and 2TRX, respectively) are shown in the same scale. The prosegment of pepsinogen is shown in ball-and-stick models. The Nterminal of pepsinogen and C-terminal of thioredoxin are indicated with arrows. Thioredoxin consists of 108 amino acid residues, while pepsinogen is 371 amino acid residues long. In Trx-PG, both protein are connected by a 20-amino-acid residue linker. (b) Proposed scheme of how thioredoxin prevents pepsin from cleaving the fusion protein. In both the fusion protein (Trx-PG) and pepsinogen, two aspartic active sites (two ‘‘D’’s in the figure) are covered with a prosegment (the thick lines in the figure). Pepsinogen can be activated into pepsin through either self-activation or bimolecular activation. Trx-PG has a large independent domain, i.e., thioredoxin portion, at the amino terminal of the prosegment. This large domain would prevent pepsin from approaching the susceptible site on the prosegment, and therefore, the bimolecular reaction could not occur. FIG. 2. Activation time course of recombinant pepsinogen (r-PG, A) and fusion pepsinogen (Trx-PG, B). The ratio of activated zymogens was calculated from the activity measurements. Nonactivated samples are defined as 0% and the plateau of the activation curves is defined as 100%. Open symbols represent pepsinogen activated in the presence of exogenous pepsin, while solid symbols represent pepsinogen activated in the absence of exogenous pepsin. Circles, triangles, and squares represent pH 1.1, 2.0, and 3.0, respectively. Each data point represents the mean of a minimum of three determinations.
AID
ABB 9925
/
6b32$$$$81
03-12-97 06:56:03
that bimolecular activation was extremely slow in comparison to unimolecular activation. Based on the above results, the activation of Trx-PG followed 0D[Trx-PG]/Dt Å k[Trx-PG] and can be analyzed by a conventional Guggenheim
arca
358
TANAKA AND YADA
plot (10) in order to calculate the rate constant for unimolecular activation for Trx-PG. The rate constants for Trx-PG were 0.0276 { 0.0004, 0.0120 { 0.0010, and 0.0099 { 0.0001 s01 at pH 1.1, 2.0, and 3.0, respectively. The present study showed that the unimolecular activation of the fusion protein was dominant and different from that of r-PG. The difference in the activation mechanism was definitely caused by thioredoxin. The presence of thioredoxin itself, however, did not alter the r-PG activation mechanism since the sigmoidal curve of r-PG activation indicated that the unimolecular activation was extremely slow even in the presence of thioredoxin molecules, which existed in the r-PG solution as the cleaved prosegment. Only when thioredoxin covalently bonded to pepsinogen was the unimolecular activation dominant. The thioredoxin portion is approximately one-quarter of the entire fusion molecule (Fig. 3a). The bulky prosegment of the fusion protein would stabilize the catalytic intermediate of unimolecular activation and could also retard bimolecular activation. The bulkiness of the prosegment of the fusion protein could prevent pepsin molecules from approaching the cleavage site, Leu44p-Ile1 (Fig. 3b). In conclusion, this engineered pepsinogen exhibited a dominant unimolecular activation. The fused thioredoxin could prevent pepsin from digesting pepsinogen. The behavior was different from the recombinant pepsinogen which exhibited the same activation mecha-
AID
ABB 9925
/
6b32$$$$81
03-12-97 06:56:03
nism as observed by Al-Janabi et al. (6). The nature of the fusion protein could provide an effective means to estimate the effect of mutations on the activation of pepsinogen, e.g., mutations in the prosegment. ACKNOWLEDGMENTS The authors appreciate the assistance of Dr. Alejandro G. Marangoni for his help with the reaction model and Kyra M. Lamb for her assistance on the kinetic measurements. This work was supported by a grant from the Natural Sciences and Engineering Research Council of Canada.
REFERENCES 1. James, M. N. G., and Sielecki, A. R. (1986) Nature 319, 33–38. 2. Andreeva, N. S., Zdanov, A. S., Gustchina, A. E., and Fedorov, A. A. (1984) J. Biol. Chem. 259, 11353–11365. 3. Abad-Zapatero, C., Rydel, T. J., and Erickson, J. (1990) Proteins 8, 62–81. 4. Herriott, R. M. (1939) J. Gen. Physiol. 22, 65–78. 5. Bustin, M., and Conway-Jacobs, A. (1971) J. Biol. Chem. 246, 615–620. 6. Al-Janabi, J., Hartsuck, J. A., and Tang, J. (1972) J. Biol. Chem. 247, 4628–4632. 7. McPhie, P. (1974) Biochem. Biophys. Res. Commun. 56, 789– 792. 8. Kageyama, T., and Takahashi, K. (1987) Eur. J. Biochem. 165, 483–490. 9. Tanaka, T., and Yada, R. Y. (1996) Biochem. J. 315, 443–446. 10. Guggenheim, E. A. (1926) Phil. Mag. 2, 538–543.
arca