Intramolecular Activation of Porcine Pepsinogen*

Abstract Conversion of pepsinogen to pepsin at acid pH involves an intramolecular reaction in which the unproteolyzed zymogen cleaves itself. This conclusion is based upon experiments in which pepsinogen, at low concentrations, was activated in the presence of substrate, hemoglobin. Under these conditions, the activation of pepsinogen is independent of pepsinogen concentration, and addition of pepsin does not enhance the rate of autoactivation. The activity of Sepharose-bound pepsinogen was independent of the average amount of pepsinogen bound per g of Sepharose, suggesting that the activation process is not caused by neighboring molecules mutually activating. Apparently, in more concentrated solutions that were previously employed for autoactivation, the concentration of pepsinogen was such that the reaction could be catalyzed by one of the products (an autocatalytic reaction). When the activation process is performed at low pepsinogen concentration in the presence of substrate, or in columns of Sepharose-bound pepsinogen, the results indicate that reaction between a trace of pepsin and pepsinogen or between molecules of pepsinogen is not a prerequisite for activation. Amino-terminal studies on Sepharose pepsinogen yielded leucine as the amino-terminal acid. Exposure of the Sepharose pepsinogen to acid yielded leucine and isoleucine as amino-terminal amino acids. Treatment of base-denatured pepsinogen with acid failed to yield isoleucine as the new amino-terminal. These results suggest that cleavage of the peptide bond which results in exposure of a new amino terminus involves an intramolecular enzymatic process. Precipitin tests with antipepsinogen and antipepsin sera which had been passed through Sepharose-pepsinogen columns exposed to various conditions indicated that Sepharose-bound pepsinogen retained its antigenic determinants upon conversion to pepsin.


Induction
of biological activity in the precursors of proteases may require proteolysis by previously formed enzyme, proteolysis by a different enzyme, or reduction of a particular sulfhydryl * This work was supported by Grant 06-010 with the National Institutes of Health, United States Public Health Service, Bethesda, Maryland.
$ Recipient of the American Cancer Society Postdoctoral Fellowship PF-473. group (reviewed in References l-4).
We have investigated the possibility that the conversion of pepsinogen to pepsin involves an intramolecular reaction in which the unproteolyzed zymogen cleaves itself.
Herriott (5) showed in 1938 that the conversion of pepsinogen to pepsin is an autocatalytic process in which the product of the reaction (pepsin) accelerates the formation of itself. During the activation of the zymogen to the enzyme, seven to nine peptide bonds are hydrolyzed, and a peptide segment originating from the amino-terminal part of pepsinogen is released (6,7). The release of the peptide segment (representing about 12% of the zymogen) yields the enzyme whose conformation and physical properties differ markedly from those of the zymogen (8). The mechanism of activation of pepsinogen to pepsin is not well understood (9). Rajagopalan, Moore, and Stein (10) have shown that the homogeneity of a pepsin preparation is markedly dependent on the pH at which pepsinogen is activated. Ong and Perlmann (9) suggested that a conformational change in pepsinogen at the pH of activation may expose the active site of pepsin and that the zymogen itself may exhibit catalytic activity. This postulate implies that for the autocatalytic conversion of pepsinogen to pepsin, previously formed pepsin is not necessary. In addition, it raises the question whether during the activation process 2 molecules of unfolded, active, unproteolyzed zymogen mutually activate, or whether the conversion of pepsinogen to pepsin involves an intramolecular reaction.
To gain an insight into the activation process we have examined the appearance of peptic activity in pepsinogen in the presence of an excess of hemoglobin substrate, which could competitively inhibit proteolysis of the pepsinogen.
In addition, we have covalently bound pepsinogen to Sepharose beads, thereby reducing the possibility of a bimolecular reaction occurring during the activation of pepsinogen.
The extinction coefficients at 278 rnp for pepsin and pepsinogen were those used by Rimon (12). The hemoglobin was first converted to the ferric state with potassium ferricyanide, according to the method of Austin and Drabkin (13). After extraction of the heme with methyl ethyl ketone, the globin was exhaustively dialyzed against water. The precipitate which formed at pH 7 was separated by centrifugation and then dissolved in 0.1 N HCl. The absorbance at 280 rnp of the globin solution was 20. All spectrophotometric measurements were performed in cuvettes of l-cm path length.
Coupling of Protein to Sepharose-Pepsinogen or pepsin was coupled to cyanogen bromide activated Sepharose beads by a slight modification (14) of the method of Porath, Axen, and Ernback (15). Sepharose 4B was washed with distilled water on sintered glass funnels until the absorbance at 280 mp of the filtrate was below 0.02. Five grams of the moist Sepharose (partially dried on the funnels) were suspended at room temperature in 15 ml of distilled water and the pH adjusted to 10.5 to 11 with 2 N NaOH. To the Sepharose suspension 0.5 g of solid cyanogen bromide (Eastman) was added (in some cases the Sepharose was suspended in 10 ml of water and the cyanogen bromide was initially dissolved in 5 ml of water). The suspension was constantly stirred and the pH kept between 10.5 and 11 for 15 min. The activated Sepharose was washed with chilled (4") water followed by 0.1 M sodium bicarbonate, pH 8.8. The .washed, activated Sepharose was added to protein solutions of various concentrations.
The protein concentration varied from 1 to 7 mg per ml and the buffer was either 0.1 M sodium phosphate, pH 6.7, or 0.1 M sodium bicarbonate, pH 8.8. The protein-Sepharose suspensions were shaken overnight at 4' and then washed (sintered glass funnel) with 0.1 M sodium bicarbonate until the absorbance at 280 rnp was below 0.02. The amount of protein bound was estimated from the absorbance at 280 rnp recovered in the filtrate. The exact amount bound was determined by amino acid analyses (16). Denaturation of Pepsinogen-This was accomplished by incubation at pH 11.5 at 23" for 2 hours (6,17).
Assay Procedures-Proteolytic activity in solution was measured with the hemoglobin assay developed by Anson (18), with .dialyzed hemoglobin solutions as substrate (17). The reaction was terminated by the addition of equal volume of 10% trichloracetic acid. The mixture was allowed to stand at room temperature for 10 to 30 min, centrifuged, and the absorbance at 280 rnp of the supernatant measured.

Preliminary
Digestion of Hemoglobin-To a 2.1% solution of hemoglobin at pH 1.8 and 37" pepsinogen was added to a final concentration of 1 pg per ml. The digests were incubated at 37" for 10 min. The peptic activity was destroyed by bringing the digest to pH 11.5 and keeping it at this pH and 23' for 2 hours. Prior to the addition of fresh pepsinogen the hemoglobin digest was acidi6ed to pH 1 (to dissolve the heme and the globin) and the pH of the digest was adjusted to the desired value with 1 N NaOH.

Activity Measurements on Sepharose-bound
Pepsinogen-Pepsinogen Sepharose was diluted to the desired concentration of pepsinogen by adding the appropriate amounts of unactivated, washed Sepharose. The mixture of pepsinogen Sepharose and Sepharose was then added to columns (8.5 x 0.5 cm). The total amount of Sepharose added to each column was approximately 0.6 g. The total amount of pepsinogen per column rvked from 5 ho 100 ,q.
To mca~urc the ra.&tc of g-lo& digcstkq ~0.5 ml of globin solution was added to the column at room temperature (20-21"). After all the globin was adsorbed on the Sepharose, 0.01 N HCl was added to wash off the globin. The time between the adsorption of the 1st drop of the slightly colored globin solution onto the Sepharose and the emergence of 1.5 ml was recorded. To compare the digestion of globin by Sepharose-bound pepsinogen and activated pepsinogen in solution, the time between the adsorption of the 1st drop of globin onto the Sepharose and the emergence of the 1st slightly colored drop was considered to be the average time that the globin was in contact with the pepsinogen Sepharose. The amount of digestion was measured by the appearance of trichloracetic acid-soluble peptides.
To check whether any activity was eluted from the column at pH 2, columns containing 1.5 to 2.0 mg of Sepharose-bound pepsinogen were prepared. The columns were first washed with 0.02 M sodium phosphate, pH 7.4, 0.15 M NaCl, and then with water until the absorbance of the eluate at 230 rnp was less than 0.02. The columns were brought to acid pH with 0.015 N HCl and the eluate was collected. Activity measurements on the eluate were done with the hemoglobin assay. Immunization Procedure-Rabbits were immunized by an injection of 4 mg of either pepsin or pepsinogen in complete Freund's adjuvant at multiple intradermal sites. Ten days after the injection a booster of 2 mg of protein in complete Freund's adjuvant was given (1 mg was injected at multiple intradermal sites and 1 mg was injected intramuscularly).
Starting 8 days after the booster, blood was collected weekly from the marginal ear vein. After the antibody titer of the various individual sera was checked by the precipitin test, the antipepsin sera and the antipepsinogen sera were separately pooled.

Quantitative Precipitin
Test-Precipitin reactions were performed with whole antisera. Increasing amounts of the antigen were added to a constant amount of the antibody.
The precipitates formed after 1 hour at 37" and overnight at 4" were were washed, dissolved in 0.1 N NaOH, and quantitatively determined by measuring the absorbance at 280 mp within 10 min.
Amino Acid Analyses and Determination of Amino-termina,l Amino Acids-Acid hydrolysis of proteins and Sepharose-coupled protein and determination of amino acid composition was carried out as described by Moore and Stein (16). The l-dimethylaminonaphthalene-5-sulfonyl method of Gray (19) was slightly modified (20). The dansyll derivatives of the amino acids were extracted with an acetone-acetic acid mixture (3:2) and identified by thin layer chromatography (21) as described previously (20). Control experiments with Sepharose-bound alanine were performed.

Activation of Pepsinogen in Presence of Substrate-Herriott
(5) has shown that the conversion of pepsinogen to pepsin is accelerated by addition of enzyme and that the process is dependent on the pepsinogen concentration (5). We have undertaken a study of the activation of pepsinogen in the presence of hemoglobin at low (0.5 to 5 pg per ml) concentrations of pepsinogen. In initial tests, in which pepsinogen was added to hemoglobin solutions (at pH 1.8), an increase in pepsinogen concentration resulted in an apparent increase in the observed specific activity as measured by the appearance of trichloracetic acid-soluble Issue of February 10, 1971 M. Bustin and A. Conway-Jacobs 617 peptides. There was a lag period in the appearance of these peptides and the plot of activity versus time was concave. These results could be caused by activation of pepsinogen by traces of pepsin and subsequent interaction between catalytically active molecules. It is equally probable, however, that the dependence of the rate of hemoglobin digestion on the concentration of pepsinogen resulted from a change in the susceptibility of the substrate (hemoglobin) to digestion during the course of the reaction. Indeed, when the hemoglobin was initially digested (see "Experimental Procedure") prior to the addition of pepsinogen, the plot of activity versus time was linear and the observed specific activity was independent of pepsinogen concentration, as shown in Fig. 1.
In a further test hemoglobin was digested by a mixture containing 90% pepsinogen and 10% pepsin (the zymogen and the enzyme were separately added to the substrate). The results suggested that at pH 1.8 pepsin did not enhance the conversion of pepsinogen to pepsin. Actually, it can be seen from Fig. 2 that under these conditions the proteolytic activity is identical, irrespective of whether pepsinogen or pepsin was added to the hemoglobin.
It is known, however, that at pH 1.8 the conversion of pepsinogen to pepsin is very rapid and while the experiments described above minimize the possibility that the proteolytic activity observed is a result of activation of pepsinogen molecules by traces of pepsin, it is still possible that the activation process is so fast that it could not be detected.
Therefore are presented in Fig. 3. When pepsinogen is added to hemoglobin solutions at pH 3 and 23' there is a distinct lag period in the appearance of proteolytic activity.
This lag period could represent the removal of the pepsin inhibitor peptide (5-7) from pepsin. At pH 3, the amount of trichloracetic acid-soluble peptides resulting from 20-min digestion of hemoglobin by pepsin was approximately 4 times the amount obtained when pepsinogen was added to the same amount of substrate. When hemoglobin was incubated with a mixture of pepsin and pepsinogen (separately added to the substrate), the rate of digestion equaled the sum of the rates observed when hemoglobin was incubated with each component separately. It can be seen that the rate of hemoglobin digestion by the mixture is significantly lower than the rate obtained by an equivalent amount of pepsin. At pH 3 and 23", just as is found at pH 1.8 and 37", the specific activity is independent of pepsinogen concentration. These results support the notion that the activity displayed by the pepsinogen solution did not result from activation by pepsin or by the interaction of several molecules. Apparently, when activation is performed at low pepsinogen concentration in the presence of hemoglobin, the substrate may act as a competitive inhibitor and the activation process ceases to be autocatalytic.  2 g of Sepharose were added to 1 ml of a solution of 1.0 mg per ml of pepsinogen, 85% of the pepsinogen added bound to Sepharose. In contrast, when the concentration of Sepharose was 0.25 g per ml only 21% of the pepsinogen added bound to Sepharose. When dissolved cyanogen bromide was used, the percentage of pepsinogen that bound to Sepharose was significantly higher than when solid cyanogen bromide was used. The amino acid composition of the Sepharose-pepsinogen preparations indicated that the entire molecule was bound to the resin.
The activation process was studied only with preparations from which no detectable pepsin activity was eluted upon exposure of Sepharose-bound pepsinogen at pH 2. At pH 6.7 pepsinogen was bound to Sepharose through relatively few bonds (22) and upon exposure of this preparation to pH 2, 50% of the peptic activity bound to Sepharose was released. When the zymogen was bound to Sepharose at pH 8.8, the amount of activity eluted was dependent upon the amount of pepsinogen bound to Sepharose.
Detectable activity was eluted when more than 2.5 mg of pepsinogen was bound to 1 g of Sepharose.
When there were less than 2.5 mg of pepsinogen per g of Sepharose, no detectable activity was eluted. The amino acid composition of these Sepharose-pepsinogen preparations was not altered by exposure to pH 2. Apparently all the peptide fragments resulting from exposure of the Sepharose pepsinogen to acid are covalently bound to the Sepharose. The Sepharose-bound pepsinogen was poured into columns and the proteolytic activity of the columns was measured with globin (hemoglobin was not suitable for this purpose because the heme precipitated on top of the column). The proteolytic activity of Sepharose-bound pepsinogen columns was about 30% of the activity of pepsinogen in solution.
As shown in Fig. 4, the amount of globin digested was proportional to the amount of pepsinogen in a column. The specific activity of Sepharosebound pepsinogen preparations in which the amount of pepsinogen was less than 2.5 mg per g of Sepharose was independent of the ratio of pepsinogen to Sepharose.
For example, preparations originally containing 2 and 0.45 mg of pepsinogen per g of Sepharose were mixed with unactivated Sepharose to give 10 pg of pepsinogen per column. The specific activity (absorbance at 280 rnp x min-l X mg-I) of the pepsinogen in these columns was 2.6 and 2.3, respectively.
Apparently the activation process is independent of the average distance between pepsinogen molecules and interaction between neighboring molecules can probably be excluded as a major factor in the activation process. It has been previously shown (6,17)  Prior to acid exposure, the only spot observed was that corresponding to dansyl-leucine; after acid exposure, both dansyl-leucine and dansyl-isoleucine were detected.
Dansylation of base-denatured pepsinogen which was exposed to pH 2 yielded only leucine as the amino-terminal amino acid.
These experiments suggest that cleavage of the peptide bond which results in the exposure of isoleucine involves an enzymatic process. Studies-We were also interested to see whether the cleavage of the peptide bond (possibly the glutamyl-isoleucine bond (9)) in Sepharose-bound pepsinogen gave rise to a marked conformational change in the protein.
From Fig. 5, D and E it can be seen that a large amount of highly cross-reacting antibodies were obtained when rabbits were intradermally injected with pepsin or pepsinogen.
To find out whether Sepharosebound pepsinogen removes preferentially antipepsinogen or antipepsin antibodies, sera were passed through columns containing limiting amounts of pepsinogen. When 5.8 ml of antipepsinogen was passed through a column containing 150 c(g of pepsinogen, 21'% of the antipepsin and 50% of the antipepsinogen activity were removed (see Fig. 5A). In contrast to the pepsinogen columns which preferentially bound antipepsinogen activity, Sepharose-pepsin columns did not show preference for either the antipepsin or antipepsinogen activities of the antipepsinogen sera (Fig. 5C). Exposure of the Sepharosepepsinogen column to pH 2 did not alter the amount of antipepsinogen and antipepsin activity which bound at pH 7.4 (compare Fig. 5, A with B). From these experiments it was concluded that there was no detectable loss of pepsinogenspecific antigenic determinants, nor appearance of new pepsinspecific antigenic determinants upon activation of the Sepharosebound pepsinogen. DISCUSSION We conclude from the experiments presented in the previous sections that the conversion of pepsinogen to pepsin at acid pH involves an intramolecular reaction in which the unproteolyzed zymogen cleaves itself.
Kinetic studies of the activation of pepsinogen in the presence of the substrate, hemoglobin, showed that the rate of activation was independent of pepsinogen concentration and was not enhanced by the addition of pepsin. These results, together with the finding that the activity of Sepharose-bound pepsinogen was independent of the average amount of pepsinogen bound per g of Sepharose (i.e. the average distance between adjacent molecules), showed that a bimolecular reaction between traces of pepsin and pepsinogen or between molecules of pepsinogen is not a necessary step in the activation process of pepsinogen.
Apparently, in more concentrated solutions that were previously employed for autoactivation, the concentration of pepsinogen was such that the reaction could be catalyzed by one of the products, i.e. the reaction was autocatalytic (5). When the activation process is performed at low pepsinogen concentrations in the presence of hemoglobin, the substrate may act as a competitive inhibitor and the reaction ceases to be autocatalytic.
The appearance of the pepsin amino-terminal acid (isoleucine, see Reference 9) upon exposure of Sepharose-bound pepsinogep to acid suggests that the unproteolyzed pepsinogen molecule may be capable of splitting peptide bonds. At least, it can cleave itself in an intramolecular reaction. Indeed, this cleavage appears to be enzymatic, requiring a particular alignment of catalytic residues, as indicated by our findings that acid exposure of base-denatured pepsinogen fails to give rise to a new ammo terminus.
Possibly, as suggested by Ong and Perlmann (9), exposure of the precursor to an acid pH brings about a conformational change which ultimately results in self-proteolysis.
It is not clear, however, whether the unproteolyzed zymogen is also capable of hydrolyzing other substrates (i.e. hemoglobin) or whether self-proteolysis is a prerequisite for appearance of proteolytic activity. Although other studies have shown that the conformation of pepsin differs markedly from that of pepsinogen (6, S), our studies indicate that Sepharose-bound pepsinogen retains most of its antigenic determinants after exposure to acid. Possibly, the binding of the pepsin portion of pepsinogen to Sepharose stabilizes the original conformation.
The presence of activation peptides on the Sepharose may further stabilize the native conformation of pepsinogen. The zymogen-enzyme transformation of pepsinogen resembles that of streptococcal proteinase (23) and that of prochymosin (prorennin) (24). In all these cases the autocatalytic activations do not require the presence of active proteolyzed species; the unproteolyzed zymogens have the ability to cleave themselves. As suggested before (23,25), it is possible that other enzymes, for example, papain (3), are derived from precursors in a similar manner.