JouRNAL OF BACTERIOLOGY, OCt. 1975, p. 424-434 Copyright i 1975 American Society for Microbiology
Vol. 124, No. 1 Printed in U.S.A.
Effects of Specific Antibodies on the Catalytic Activity of L-Asparaginase from Serratia marcescens and Escherichia coli DONALD A. FERGUSON, JR., ARTHUR W. PHILLIPS,* AND A. ALVIN MARUCCI Department of Biology, Biological Research Laboratories, Syracuse University,* and the Department of Microbiology, Upstate Medical Center, State University of New York, Syracuse, New York 13210 Received for publication 9 June 1975
Rabbit antisera against highly purified L-asparaginase from Serratia and from Escherichia coli showed up to 60% inhibition of the catalytic amidohydrolysis of L-asparagine when combined with the homologous enzyme. This inhibition was diminished somewhat against the heterologous enzyme. Kinetic studies in the presence of these antisera showed an increased Km,&P for both homologous and heterologous enzymes using L-asparagine as substrate. In contrast, kinetic studies employing the poor substrate, L-glutamine, showed activation attributable to specific antibodies. This was seen in lower KmaPP values and up to twofold increases in the Vi.. over the normal rabbit serum controls. The high degree of cross-inhibition ('80%) and the low degree of cross-reactivity in the quantitative precipitin test (-34%) suggest that these two enzymes possess structural similarities located mainly in the regions of the catalytic sites. marcescens
Many enzymes retain some catalytic activity (This research has been submitted to Syrain the presence of specific antibody. A few cuse University by D.A.F. as part of a Ph.D. enzymes are completely inhibited; others are dissertation.) not at all affected by specific antibody (8). MXTERIALS AND METHODS Activation of enzyme by antibody has been Bacterial strain. S. marcescens ATCC no. 60 was reported in a few instances, especially with poor substrates (1, 8). The use of antibodies to cultivated in stock cultures on Trypticase soy agar C. enzymes as a tool in the study of structural slants at 28and harvesting of bacteria. An inoculum similarities between antigenic determinants was of Growth S. marcescens was grown in 100 ml of medium reviewed by Arnon (1). Boyd and Phillips (4, 5) containing 4% (wt/vol) autolyzed yeast (12, 13) in a have shown that complete regression of the 250-ml Erlenmeyer flask. This was incubated for 18 h Gardner lymphosarcoma in the C3H mouse at 34 C with shaking and then added to 10 liters of could be achieved with smaller amounts of sterile 3% yeast extract medium in a 15-liter fermenL-asparaginase from Serratia marcescens than tor (New Brunswick Scientific Co., New Brunswick, from Escherichia coli. This might be due to N.J.). Incubation was continued for 24 h at 34 C with some differential effect of antibody on the aeration and agitation. Antifoam A, silicone defoamer Corning), was added (100 ml) as a 10% solution enzyme from these two organisms. It is impor- (Dow 0 and 8 h of incubation. Cells were harvested in a tant, therefore, to investigate the characteris- at continuous flow centrifuge at 10 C and tics of inhibition of these enzymes by their Sharples washed twice with 0.05 N tris(hydroxymethyl)specific antisera. aminomethane (Tris)-hydrochloride buffer (pH 8.6). We have previously shown that there is im- About 20 g of packed cells per liter containing 100 to munologic cross-reactivity between the L- 150 IU of enzyme per g was obtained. Enzyme assays. L-Asparaginase assays were conasparaginases isolated from S. marcescens and E. coli (23). The question arises whether this ducted by nesslerization for routine work (19). The cross-reactivity, which has been measured by reaction mixture contained saturating substrate and enzyme in a total volume of 2.0 ml of 0.05 M complement fixation and precipitin tests, is also Tris-hydrochloride pH 7.4, and was incubated demonstrable by enzyme inhibition. The pres- at 37 C for specifiedbuffer, times. Turbidity due to serum ent work was undertaken to study (i) the proteins was avoided by adding 0.5 ml of 1.5 M triquantitative inhibition of enzyme by homolo- chloroacetic acid to stop the reaction. The mixture gous and heterologous antisera; and (ii) the was kept in an ice bath for 15 min and then ceneffects of antibody on enzyme kinetics using trifuged. The clear supernatant was mixed with 0.4 ml of 1.5 N KOH and assayed. To standardize protein L-asparagine and L-glutamine as substrates. 424
VOL. 124, 1975
L-ASPARAGINASE FROM S. MARCESCENS
activation or stabilization of L-asparaginase (14, 17), 0.5 mg of bovine serum albumin (BSA) was added to each assay tube prior to incubation. A different assay was employed in the kinetic studies. This was a continuous assay in which the asparaginase and glutamic dehydrogenase (GDH) reactions were coupled, and the stoichiometric oxidation of reduced nicotinamide adenine dinucleotide was measured at 340 nm. The reaction mixture contained 1 mg of BSA, 5 pmol of a-ketoglutarate, 0.025 Amol of reduced nicotinamide adenine dinucleotide, 0.333 mg of GDH, specified amounts of L-asparaginase, and antiserum, in a total volume of 1.0 ml of 0.05 M Tris-hydrochloride buffer, pH 8.6, at 22 C. The reaction was initiated by the addition of L-asparagine or L-glutamine. Details of this procedure have been described elsewhere (9). One international unit of L-asparaginase activity is defined as the amount of enzyme required to liberate 1 gmol of NH4+ from L-asparagine in 1 min under the assay conditions described. Protein determinations. Protein concentrations were estimated after Lowry et al. (18) with BSA as the standard. Dilute protein solutions (5 to 20 ug of protein/ml) were measured spectrophotometrically at 215 and 225 nm with BSA as standard (20). Chemicals and reagents. These materials were obtained from the following sources: L-asparagine monohydrate, A grade, from Calbiochem, Los Angeles; L-glutamine, crystalline, grade III, from Sigma, St. Louis (solutions were prepared daily); L-GDH, bovine liver, type II, in 50% glycerol, from Sigma (dialyzed before use); 1-nicotinamide adenine dinucleotide (reduced), disodium salt, grade III, low nicotinamide adenine dinucleotide content, from Sigma; a-ketoglutarate, monosodium salt, from Sigma; Whatman diethylaminoethyl cellulose from W and R Balston Ltd., London, U.K.; Bio Gel from Bio-Rad Laboratories, Richmond, Calif.; Tris from Sigma; acrylamide, bisacrylamide, N,N,N',N'-tetramethylethylenediamine, and glycine, from Eastman Organic Chemicals Inc., Rochester, N.Y.; phenol reagent (Folin-Ciocalteau) from Harleco, Philadelphia; BSA from Armour Co., Chicago, Ill.; potassium iodide, copper sulfate, and sodium carbonate from General Chemical Division of Allied Chemical and Dye Corp., New York, N.Y. All other chemicals were obtained from Mallinckrodt Chemical Works, St. Louis, Mo. L-Asparaginase from E. coli was a gift from Horace D. Brown, Merck Sharp and Dohme Research Laboratories, Rahway, N.J. Purification of L-asparaginase from S. marcescens. Unless stated otherwise, all procedures were conducted at 4 C. The packed cells were diluted 1:3 (wt/vol) with 0.05 M Tris-hydrochloride buffer, pH 8.6, and the enzyme was released in a sonic oscillator (Raytheon Manufacturing Co., Waltham, Mass.) at 1.5 A for 7 min. The preparation was centrifuged 30 min at 37,000 x g and the pellet was discarded. The enzyme was purified by a modification of the method described by Ho et al. (15) as follows. The supernatant was adjusted to pH 4.5 by dropwise addition of 2 N acetic acid, and the precipitate was removed by centrifugation for 20 min at 16,300 x g and washed twice with pH 4.5 buffer. The precipitate
425
was discarded. The supernatant and the washings were pooled and adjusted to pH 6.6 with 3 N KOH. The solution was concentrated to one-half its volume in an XM-50 Diaflo ultrafilter (Amicon Corp., Lexington, Mass.) under 53 lbs/in' of N,. The concentrate with about 10 mg of protein per ml was cooled to -15 C as 95% ethanol was added dropwise with stirring until a cloudiness appeared, when the mixture was kept at -15 C for 1 h without stirring. The precipitate was removed by centrifugation at 16,300 x g for 20 min at -15 C. In a typical purification successive precipitates of L-asparaginase occurred at 0.1, 0.2, 0.3, 0.5, 0.8, and 1.5 volumes of added ethanol; and each precipitate was treated as above. Fractionation with ethanol was repeated as above to obtain further purification and those fractions having similar specific activity were pooled after dissolving in a minimal volume of 0.02 M sodium phosphate buffer, pH 6.6. Each preparation was examined on analytical polyacrylamide electrophoresis as described previously (4). Fractions with high enzyme activity were loaded (5 ml) onto a Bio Gel P-150 column (46 by 2.5 cm) containing 200 ml packed bed volume with a void volume of 40 ml and a flow rate of 12 ml/h. Fractions (4 ml) were collected and assayed for protein and enzyme activity. Peak tubes were pooled and concentrated on an XM-50 Diaflo membrane under 20 lbs/in2 of N,. Concentrates containing 15 to 20 mg of protein per ml were made 20% (wt/vol) in sucrose and loaded on a polyacrylamide column (6.6% gel) in a preparative electrophoresis apparatus (Shandon Scientific Co. Ltd., London, U.K.). The buffer was 0.025 M Tris-glycine, pH 8.4, and the current was 50 mA. Fractions were collected, assayed, and concentrated as above. Enzyme yields were greatly improved by replacing the membrane and recovering trapped enzyme during the run. Homogeneous enzyme with a single band on gel electrophoresis and having a specific activity of about 200 IU/mg of protein was obtained (Fig. 1). The purification procedure is summarized in Table 1. Antisera. Anti-S. marcescens asparaginase sera were prepared in three New Zealand white does weighing 2.0 to 2.5 kg (rabbits 245, 246, 247). One milligram of S. marcescens enzyme (specific activity 200 IU) in complete Freund adjuvant was injected intramuscularly at multiple sites in each rabbit on days 1, 10, and 20. These animals were bled on days 27 and 28 (first course antisera). On day 37 each rabbit received S. marcescens antigen as above and second course antisera were obtained 11 days later. On day 77 S. marcescens antigen was again injected and third course sera were taken-11 days subsequently. Rabbit anti-E. coli L-asparaginase sera (223, 224) were prepared as above but with large doses of antigen since more of it was available. Rabbits received 5 mg of antigen on days 1 and 10 and 3 mg on day 25. Six days later first course sera were taken. On days 45 and 49 animals were given 3 mg of E. coli antigen intravenously, and second course sera were .taken about 6 days thereafter. Antisera are designated by rabbit number and course of injection, e.g., no. 223-2 is serum from rabbit no. 223, second course. Antibody inhibition of L-asparaginase activity.
FERGUSON, PHILLIPS, AND MARUCCI
J. BACTERIOL.
The effects of rabbit and anti-asparaginase sera on enzyme activity were investigated using the method described by Fuller and Marucci (11) for the horse liver alcohol dehydrogenase-antibody system. Various amounts (1 to 100 Mg) of antibody protein (determined by quantitative precipitin test) in 0.5 ml of buffer were added to L-asparaginase (1 ug) in 0.5 ml of 0.01 M sodium phosphate buffer, pH 6.9, containing 0.85% NaCl; no visible precipitation occurred. The enzyme-
antiserum mixtures were then incubated at 25 C for 15 min. Aliquots (0.4 ml) of enzyme-antibody mixtures were assayed by mixing with a solution of L-asparagine (10- M) in the above buffer, incubating at 37 C for 60 min, and measuring NH,+ by nesslerization. All determinations were done in duplicate and normal rabbit serum controls were run in each experi-
426
A
B
em
ment.
The effects of rabbit anti-asparaginase sera on asparaginase kinetics were measured with L-asparagine and with L-glutamine as substrates. A single ratio of antibody to enzyme protein of 25:1 (wt/wt) was employed since this ratio produced maximum inhibition of L-asparaginase activity by all antisera. Appropriate amounts of antigen and antisera were mixed in a total volume of 1.0 ml of 0.01 M sodium phosphate buffer, pH 6.9, containing 0.85% NaCl and incubated at 25 C for 15 min. The mixture was then diluted to 2.0 ml with 0.05 M Tris-hydrochloride, pH 8.6, and 0.1-ml aliquots were assayed by the GDH coupled assay method employing various concentrations of substrate (L-asparagine or L-glutamine). Initial velocities were determined for each substrate concentration (in duplicate) and appropriate controls with buffer and normal rabbit serum were conducted. Due to the low L-glutaminase activity of these enzymes (5%), fivefold more enzyme and antiserum were used with L-glutamine as substrate than with the L-asparagine experiments.
FIG. 1. Polyacrylamide gel electrophoresis pf Lasparaginase from S. marcescens (A) and E. coli (B). The gel columns (10 by 0.5 cm) contained 7.5% acrylamide and were loaded with 40 Mg of enzyme protein (specific activity 200 IU/mg of protein) in 0.04 ml of 0.005 M Tris-glycine buffer, pH 8.4. Each column was run at 2.5 mA. Gels were stained with Coomassie blue. The arrow points to the top of the column.
RESULTS Effects of specific antibody on i^asparaginase activity. The effects of specific antibodies on the catalytic activity of the enzymes towards L-asparagine were quantitatively investigated. The results showed partial inhibition of enzyme activity. The data are presented in Fig. 2. In the presence of S. marcescens antisera, S. marcescens L-asparaginase manifested about 60%, inhi-
TABLE 1. Purification of L-asparaginase from S. marcescens Purification step
PurificationsteP
Total
protein (g)
activity (IU)
Sp act (IU/mg of protein)
Enrichment
(-fold)
116
Total recovery
100
72,000
Intact cells Crude extract .................
Total enzyme
55,000
0.47
76
pH 4.5 supernatant ............
31.8
41,000
1.3
2.8
57
Concentrate ...................
26.4
40,000
1.5
3.2
55
Ethanol precipitates First ....................... Second ..................... Third .......................
8.1 2.6 0.74
35,000 18,300 10,600
4.3 7.0 14.3
9.2 15 30
49 25 15
P-150 gel ......................
0.45
10,000
22.2
47
14
Polyacrylamide gel ............
0.023
4,530
419
6
197
I
--
ANInM NTEINI
IFp
pg
FIG. 2. Inhibition of the catalytic activity of L-asparaginase on L-asparagine by specific rabbit antibodies. Different amounts of antibody protein (1 to 1,000 Ag) in 0.5 ml of buffer were added to I Ag of L-asparaginase in 0.5 ml of 0.01 M sodium phosphate buffer at pH 6.9 containing 0.85% NaCI. The mixtures were kept at 37 C for 15 min and aliquots were taken for enzyme assays by nesslerization. L-Asparaginase from S. marcescens (0; sim) and E. coli (0; ec) were selected, the former designated as S. marcescens and the latter as E. coli. The foUowing rabbit antisera were used: A, anti-S. marcescens serum no. 245-1; B, anti-S. marcescens serum no. 246-2; C, anti-S. marcescens serum no. 247-2; D, anti-E. coli serum no 223-2; E, anti-E. coli serum no. 224-1. The arrows represent the region of antibody-antigen equivalence for sm and ec. 427
428
FERGUSON, PHILLIPS, AND MARUCCI
bition, whereas E. coli L-asparaginase activity was less inhibited (Fig. 2A, B, and C). In all instances, however, it required a significantly greater antibody-antigen ratio to achieve maximum inhibition than to reach equivalence in the quantitative precipitin reactions (D. A. Ferguson, Jr., A. W. Phillips; and A. A. Marucci, manuscript in preparation). This was also true when E. coli antisera were employed where enzyme inhibition was observed albeit at a somewhat decreased level (Fig. 2D and E). Again the enzyme activity of the homologous antigen was significantly more inhibited than that of the heterologous one. The overall cross-inhibition, however, was high and comparable with both E. coli and S. marcescens antisera and complete inhibition of enzyme activity was never attained even in large excess of antibody. A control experiment was performed to show that the degree of inhibition was a function of the number of antibody molecules bound to each enzyme molecule and independent of concentration effects at a given ratio of antibody to enzyme (Fig. 3). These data show this to be true for both homologous and heterologous enzymes with all antisera tested. Our data on the crossinhibition of L-asparaginase activity and the immunologic cross-reactivity of antisera against these L-asparaginases are summarized in Table 2. Effects of specifc antibody on enzyme kinetics. The GDH coupled assay was employed in the determination of the effects of specific antibody on the kinetics of amidohydrolysis of L-asparagine and L-glutamine by L-asparaginase. In all of these experiments an antibody-antigen ratio of 25:1 (wt/wt) was employed which gave maximal inhibition in the studies described above. It was found that the presence of normal rabbit serum did not alter the kinetics in any of the reactions described. Under these conditions the K,,PP values for the L-asparagine reaction were 1.25 x 10- M and 1.2 x 10- M for S. marcescens (Fig. 4A) and E. coli (Fig. 4B) enzymes, respectively; and the values for L-glutamine were 7.4 x 10-' M for the S. marcescens enzyme (Fig. 5A, curve 3) and 2.5 x 10-' M for the E. coli enzyme (Fig. 8). The Vmax values for L-glutamine were considerably lower than the values for L-asparagine. These results agree well with measurements taken in the absence of normal serum (21, 29). Since normal rabbit serum had no effect on enzyme kinetics, any effects produced by the various antisera are attributable solely to their specific antibody content. With L-asparagine as substrate each of the S. marcescens antisera
J. BACTERIOL. I
I
I
-I
J
70
Go
so
A*
0
o-
0~~~~ A
a
2
a
a
p
__a
N
L
2X
11
ANTIIOI
L-ASPAUACINASE
41
FIG. 3. The effect on enzyme activity of varying the antibody and antigen concentration while maintaining their ratio constant. The antibody-antigen ratio is 25:1 in all samples. All original volumes of the antibody and antigen solutions were multiplied by the factors indicated on the abscissa, mixed, and assayed by nesslerization after incubation at 37 C in 0.01 M sodium phosphate buffer, 0.15 N NaCI (pH 6.9), with saturating L-asparagine. L-Asparaginase from S. marcescens (0) and E. coli (0) were employed. The following rabbit antisera were used: A, anti-S. marcescens serum no. 245-1; B, anti-S. marcescens serum no. 246-2; C, anti-S. marcescens serum no. 247-2; D, anti-E. coli serum no. 223-2; E, anti-E. coli serum no. 224-1.
decreased the V,,., of both S. marcescens and E. coli enzymes (Fig. 4). The S. marcescens enzyme (Fig. 4A) was inhibited to a greater extent (50 to 75%) than E. coli enzyme (Fig. 4B) by all S. marcescens antisera; and the Km"PP values were somewhat increased (mixed inhibition) except with serum no. 245-1, where little or no effect was observed with either enzyme (noncompetitive inhibition). The measurements taken on enzyme inhibition in the coupled assay were comparable to the values obtained by nesslerization. In contrast, with Lglutamine as substrate, an enhancement of
L-ASPARAGINASE FROM S. MARCESCENS
VOL. 124, 1975
enzyme activity by specific antibodies usually occurred. In these experiments the same three S. marcescen,s antisera described above were involved. Two antisera resulted in a decrease in KmaPP value and generally an increase in V.ax for both S. marcescens and E. coli enzymes (Fig. 5). The exception was antiserum no. 247-2 which inexplicably manifested little influence on either enzyme. The most pronounced effect was produced by antiserum no. 246-2 against the E. coli enzyme (Fig. 5). These results thus show that the binding of specific antibodies to the enzyme can alter the molecule in such a way as to significantly increase the activity of the catalytic site for L-glutamine. The above experiments were repeated with rabbit antisera against E. coli enzyme. When L-asparagine was the substrate there were significant decreases in the V.ax of E. coli and S. marcescens enzymes and an increase in the KmSPP values, indicating a mixed inhibition (Fig. 6). Antiserum no. 223-2 was considerably more inhibitory than antiserum no. 224-1, but neither of these sera manifested the degree of inhibition shown by S. marcescens antisera no. 245-1 and no. 246-2. Here again the greatest inhibition was shown to occur in the homologous system, and the inhibition values obtained by the GDH assay were comparable to those obtained by nesslerization. With L-glutamine as substrate in a reaction
429
mixture containing E. coli antiserum no. 224-1, there was a pronounced increase in the V.., and decrease in the Km.PP of the E. coli enzyme, indicating enhancement of enzyme activity (Fig. 7). However, there was little influence upon the S. marcescens enzyme by this antiserum (Fig. 7A). Also, E. coli antiserum no. 223-2 had little effect on the activity of E. coli and S. marcescens enzymes. The contrasting effects of E. coli antiserum no. 224-1 upon E. coli and S. marcescens enzymes with L-glutamine as substrate are the only instances of significant kinetic differentiation between the two enzymes by any of the antisera using either substrate. This may indicate a similarity in the structure of the active sites of these enzymes. However the exceptional behavior of anti-E. coli serum no. 224-1 does suggest that the active sites are not identical.
DISCUSSION Inhibition of L-asparaginase by specific antisera may be envisioned as occurring by (i) direct blocking of catalytic sites by antibody specific for these sites, (ii) blocking of the active centers by antibody which has combined with antigenic determinants in close proximity to these sites, and/or (iii) a conformational change in the enzyme molecule as a result of its combination with antibody. The heterogenous nature of antisera to various enzymes has been shown by fractionation of anti-enzyme sera into inhibitory, noninhibitory, TABLE 2. The relative cross-inhibition of and activating antibody preparations (10, 27, L-asparaginase activity between rabbit 28, 31). Arnon and Shapira (3) isolated antianti- L-asparaginase from S. marcescens and E. coli body specific for the active center of papain. and a comparison with the cross-reactivity as This antibody combines with and inhibits the measured by quantitative precipitin reactions catalytic activity of chymopapain. By use of specific antisera against highly Cross-Imu inhibiImmuL-asparaginase isolated from E. coli purified tion of ~~crosAntiserum nologic Antlserum S. we have made some postulaand marcescens, no. enzyme reacAntigen tions about the structural similarities of these activtionb %n itya enzyme preparations. The high degree of crossinhibition is consistent with structural similarin the region of the catalytic site. We cannot ity 31 223-2 78 E. coli L-asparadistinguish effects of antibody directed to the ginase catalytic sites and effects of antibody directed 82 23 224-1 E. coli L-asparato determinants in close proximity to these ginase 53 97 245-1 S. marcescens Lsites. However, it would seem that, in these asparaginase antisera, antibody to regions neighboring the 21 70 246-2 S. marcescens [_ catalytic site play the major role in inhibition. asparaginase This is so because antibodies directed to the 34 45 S. marcescens L247-2 catalytic site would be expected to result in asparaginase complete inhibition (1, 3, 7). We were never able to obtain complete inhibition of enzyme a The substrate was L-asparagine. I These data are from Ferguson, Phillips, and even in great antibody excess. The cross-inhibition of the enzyme was found Marucci, unpublished data. (
itsxlo, M
-
FIG. 4. The effects of rabbit antibodies against S. marcescens L-asparaginase on the kinetics of the amidohydrolysis of L-asparagine by L-asparaginase. Specified amounts of antisera and enzyme (antibodyantigen ratio of 25:1) were mixed in a total volume of 1.0 ml of 0.01 M sodium phosphate buffer, pH 6.9, containing 0.85% NaCI and incubated at 24 C for 15 min. The mixtures were then diluted to 2.0 ml with 0.05 M Tris-hydrochloride, pH 8.6, and 0.1-ml aliquots were assayed by the GDH coupled assay (see text). L-Asparaginases are from S. marcescens (0) and E. coli (0). (A) Reactions of S. marcescens with different anti-S. marcescens sera: curve 1, serum no. 245-1; curve 2, serum no. 246-2; curve 3, serum no. 247-2. The effect of normal rabbit serum is shown (0). (B) Reactions of E. coli with anti-S. marcescens sera: curve 1, serum no. 245-1; curve 2, serum no. 246-2; curve 3, serum no. 247-2. A normal rabbit serum control is shown (0). 430
VOL. 124, 1975
FiG. 5. The effects of rabbit antibodies against S. L-asparaginase on the kinetics of amidohydrolysis of L-glutamine by L-asparaginase. See legend to Fig. 4 for the reaction conditions, symbols, and abbreviations. (A) Reactions of S. marcescens with different anti-S. marcescens sera: curve 1, serum no. 246-2; curve 2, serum no. 245-1; curve 3, normal rabbit serum control (0). (B) Reactions of E. coli with anti-S. marcescens sera: curve 1, serum no. 246-2; curve 2, serum no. 245-1; curve 3, normal rabbit serum control (see Fig. 8). marcescens
to be about 80%. This is greater than the cross-reactivity as determined by quantitative precipitin analyses (ca. 34%) (D. A. Ferguson et
al., manuscript in preparation). From these data it might be postulated that regions of the enzyme not involved in catalysis are more structurally distinct than are the enzymic sites. On evolutionary grounds, this should be expected since only those regions necessary for catalytic activity would be essential and would be conserved. Furthermore, since S. marcescens antisera were more efficient in inhibiting the homologous enzyme, it could be inferred that the catalytic sites on this enzyme are more
L-ASPARAGINASE FROM S. MARCESCENS
431
susceptible to reaction with antibody than are similar sites on the E. coli enzyme. Experiments on kinetics of L-asparaginase hydrolysis in the presence and absence of antisera were done in an attempt to gain some idea about the mechanism of inhibition (16). In these studies most antisera showed a "mixed" pattern of inhibition (26); i.e., both KmaPP and Vmax vary in the presence of antiserum with both homologous and heterologous enzymes. This could be accounted for by either a conformational change in the enzyme which decreased both binding and catalysis or by a combination of conformational change and steric hindrance. Steric hindrance alone would not give this pattern of inhibition. Anti-S. marcescens serum no. 245-1 gave an unusual partial mixed type inhibition (Fig. 5A, curve 2) similar to the C5 system described by Segel (26). Future studies on the inhibitory effects of individual species of immunoglobulins from fractionated antisera, as well as the influence of inhibitor concentration on enzyme kinetics, should be useful aids in the elucidation of the inhibition mechanisms. The experiments with the poor substrate, L-glutamine (6, 21), further reinforce the view that antibody-induced conformational change can be a mechanism for altered activity of this enzyme. It was found that in the presence of antibody, the rate of catalysis of L-glutamine was significantly increased over normal serum controls. This is similar to the results of Pollock (24) with penicillinase, Okada et al. (22) with amylase, Rotman and Celada (25) with ,8-galactosidase, and Suzuki et al. (28) with ribonuclease. A probable mechanism for this effect is that constraint placed on the enzyme by bound antibody prevents conformative changes usually imposed by normal substrate (30, 31). Alternatively, the conformation of the enzyme may be so altered after reaction with antibody that the poor substrate is now more readily accepted. It would be of interest to make similar comparisons of L-asparaginases from other microbial species from the viewpoint of evolutionary relationships as well as for clinical applications to cancer chemotherapy. Concerning the latter, establishment of the mechanism of inhibition of L-asparaginase by human antibody should be a high priority item. What antigenic determinants are recognized on this enzyme by the human immunologic system? Can these be altered without affecting the enzymic activity? Drugs used in combined therapy with Lasparaginase ought to be examined for their effects on enzyme kinetics in the presence and in the absence of specific human antibodies.
432
FERGUSON, PHILLIPS, AND MARUCCI
J. BACTERIOL.
~~//
I A I-4-
I0-
A .*
,
It
6
4
2
0
2
ii511@3
4 N
6
FIG. 7. Effects of rabbit antibodies against E. coli L-asparaginase on the kinetics of L-glutamine and amidohydrolysis by L-asparaginase. See legend to Fig. 4 for reaction conditions, symbols, and abbreviations. The following mixtures were used: (A) serum no. 224-1 with S. marcescens; (B) serum no. 224-1 with E. coli.
8
6
4
2
2
lIIsIO
4
6
a
FIG. 6. Effects of rabbit antibodies against E. coli L-asparaginase on the kinetics of amidohydrolysis of L-asparagine by L-asparaginase. See legend to Fig. 4 for reaction conditions, symbols, and abbreviations. (A) Reactions of S. marcescens with different anti-E. coli sera: curve 1, serum no. 223-2; curve 2, serum no.
224-1; curve 3, normal rabbit serum control (0). (B) Reactions of E. coli with anti-E. coli sera: curve 1, serum no. 223-2; curve 2, serum no. 224-1; curve 3, normal rabbit serum control (0).
VOL. 124, 1975
L-ASPARAGINASE FROM S. MARCESCENS
28
24
20
12
~ ~ I I/1 1 6
4
2
0
, ~~I ~~~~~~ I 6
IISX
102
M
FIG. 8. Kinetics of amidohydrolysis of L-glutamine by E. coli L-asparaginase in the presence of normal rabbit serum. The reaction mixture consisted of 0.75 ml of 1:5 dilution of serum, 15 ,gg of enzyme, L-glutamine, 0.05 M Tris-hydrochloride buffer, pH 8.6, in a total volume of 2.0 ml at 25 C. Velocity is expressed as micromoles of NH4+ released per minute. This plot is curve 3 in Fig. 5B and the control plot in Fig. 7B (curve without symbols). ACKNOWLEDGMENTS This investigation was aided by a grant from the Hendricks Research Fund of Syracuse University to the State University of New York, Upstate Medical Center (A. A. Marucci and A. W. Phillips, co-investigators). We are grateful to H. Richard Levy, Robert W. Jackson, and Jack Bryan for helpful discussions, and we appreciate the skillful technical assistance of Kenneth Malament and Howard Telson. We thank the following persons for supplying us with materials: C. Gordon Zubrod and Samuel Poiley. National Cancer Institute, National Institutes of Health; Kenneth Price, John Godfrey, and Bernard Heineman, Bristol Laboratories, Inc., Syracuse, N.Y.; and Horace D. Brown, Merck Sharp and Dohme Research Laboratories, Rahway, N.J. We thank Georgia Ventura and Pam Dobbert for doing the typescript. LITERATURE CITED 1. Arnon, R. 1971. Antibodies to enzymes-a tool in the study of antigenic specificity determinants. Curr. Top. Microbiol. Immunol. 54:47-93. 2. Arnon, R.. and B. Schechter. 1966. Immunological studies on specific antibodies against trypsin. Immunochemistry 3:451-461. 3. Arnon, R., and E. Shapira. 1968. Comparison between the antigenic structure of mutually related enzymes. A study with papain and chymotrypsin. Biochemistry 7:4196-4200. 4. Boyd, J. W., and A. W. Phillips. 1971. Purification and
properties of L-asparaginase from Serratia J. Bacteriol. 106:578-587.
marcescens.
433
5. Boyd, J. W., and A. W. Phillips. 1971. Inhibition of lymphoma 6C3HED by L-asparaginase from Serratia marcescens. J. Natl. Cancer Inst. 46:1271-1276. 6. Campbell, H. A., and L. T. Mashburn. 1969. L-Asparaginase EC-2 from Escherichia coli. Some substrate specificity characteristics. Biochemistry 8:3768-3775. 7. Celada, F., and R. Strom. 1972. Antibody-induced conformational changes in proteins. Q. Rev. Biophys. 5:395-425. 8. Cinader, B. 1967. Antibodies to enzymes-a discussion of the mechanisms of inhibition and activation, p. 85-137. In B. Cinader (ed.), Antibodies to biologically active molecules. Pergamon Press, Oxford. 9. Ferguson, D. A., Jr., J. W. Boyd, and A. W. Phillips. 1974. Continuous assays of L-asparaginase by coupling with glutamic dehydrogenase and by cationic glass electrode. Anal. Biochem. 62:81-90. 10. Fuchs, S. P., P. Cuatrecasas, D. A. Ontjes, and C. B. Anfinsen. 1969. Correlation betwe.ewAOte antigenic and catalytic properties of staphylococc*1 nuclease. J. Biol. Chem. 244:943-950 11. Fuller, T. C., and A. A. Marucci. 1971. Immunoenzymology of liver alcohol dehydrogenase. I. Factors influencing the inhibition of horse liver alcohol dehydrogenase by rabbit and guinea pig antisera. J. Immunol. 106:110-119. 12. Heinemann, B., and A. J. Howard. 1969. Production of
tumor-inhibitory L-asparaginase by submerged growth of Serratia marcescens. Appl. Microbiol. 18:550-554. 13. Heinemann, B., A. J. Howard, and M. J. Palocz. 1970.
14.
15.
16.
17. 18.
19.
20. 21. 22.
23.
24.
Influence of dissolved oxygen levels on production of L-asparaginase and prodigiosin by Serratia marcescens. Appl. Microbiol. 19:800-804. Ho, P. P. K., and L. Jones. 1969. Non-specific protein stabilization of L-asparaginase. Biochim. Biophys. Acta 177:172-174. Ho, P. P. K., E. B. Milikan, J. L. Bobbitt, E. L. Grinnan, P. J. Burck, B. M. Frank, L. D. Boeck, and R. W. Squires. 1970. Crystalline L-asparaginase from Escherichia coli B. I. Purification and chemical characterization. J. Biol. Chem. 245:3708-3715. Kabat, E. A. 1968. Structural concepts in immunology and immunochemistry. Holt, Rinehart and Winston, Inc., New York. Lee, M. B., and J. M. Bridges. 1968. L-asparaginase activity in human and animal sera. Nature (London) 217:758-759. Lowry, 0. H., N. J. Rosebrough, A. L. Farr, and R. J. Randall. 1951. Protein measurement with the Folin phenol reagent. J. Biol. Chem. 193:265-275. Meister, A. 1955. Glutaminase, asparaginase and a-keto acid amidase, p. 380-385. In S. P. Colowick and N. 0. Kaplan (ed.), Methods in enzymology, vol. II. Academic Press Inc., New York. Murphy, J. B., and M. W. Kies. 1960. Note on spectrophotometric determinations of proteins in dilute solutions. Biochim. Biophys. Acta 45:382-384. Novak, E. K., and A. W. Phillips. 1974. L-Glutamine as a substrate for L-asparaginase from Serratia marcescens. J. Bacteriol. 117:593-600. Okada, Y., T. Ikenaka, T. Yagura, and Y. Yamamura. 1963. Immunological heterogeneity of rabbit antibody fragments against taka-amylase A. J. Biochem. (Tokvo) 54:101-102. Phillips, A. W., J. W. Boyd, D. A. Ferguson, Jr., and A. A. Marucci. 1971. Immunochemical comparison of L-asparaginase from Serratia marcescens and Escherichia coli. J. Bacteriol. 107:461-467. Pollock, M. R. 1964. Stimulating and inhibiting antibodies for bacterial penicillinase. Immunology
7:707-723. 25. Rotman, M. B., and F. Celada. 1968. Antibody mediated
434
FERGUSON, PHILLIPS, AND MARUCCI
J. BACTERIOL.
activation of a defective fl-D-galactosidase extracted from an Escherichia coli mutant. Proc. Natl. Acad. Sci. U.S.A. 60:660-667. 26. Segel, I. H. 1975. Enzyme kinetics, p. 189-192. WileyInterscience, New York. 27. Soru, E., and 0. Zaharia. 1974. Immunoenzymology of L-asparaginase from the BCG strain of Mycobacterium bovis. Immunochemistry 11:791-795. 28. Suzuki, T., H. Peliochova, and B. Cinader. 1969. Enzyme activation by antibody. I. Fractionation of immune sera in search for an enzyme activating antibody. J.
Immunol. 103:1366-1376. 29. Wriston, J. C., Jr., and T. C. Yellin. 1973. L-asparaginase: a review. Adv. Enzymol. 39:185-248. 30. Zyk, N., and N. Citri. 1968. The interaction of penicillinase with penicillins. VI. Comparison of free and antibody-bound enzyme. Biochim. Biophys. Acta 159:317-326. 31. Zyk, N., and N. Citri. 1968. The interaction of penicillinase with penicillins. VII. Effect of specific antibodies on conformative response. Biochim. Biophys. Acta 159:327-339.
Vol. 124, No. 1 Printed in U.S.A.
Effects of Specific Antibodies on the Catalytic Activity of L-Asparaginase from Serratia marcescens and Escherichia coli DONALD A. FERGUSON, JR., ARTHUR W. PHILLIPS,* AND A. ALVIN MARUCCI Department of Biology, Biological Research Laboratories, Syracuse University,* and the Department of Microbiology, Upstate Medical Center, State University of New York, Syracuse, New York 13210 Received for publication 9 June 1975
Rabbit antisera against highly purified L-asparaginase from Serratia and from Escherichia coli showed up to 60% inhibition of the catalytic amidohydrolysis of L-asparagine when combined with the homologous enzyme. This inhibition was diminished somewhat against the heterologous enzyme. Kinetic studies in the presence of these antisera showed an increased Km,&P for both homologous and heterologous enzymes using L-asparagine as substrate. In contrast, kinetic studies employing the poor substrate, L-glutamine, showed activation attributable to specific antibodies. This was seen in lower KmaPP values and up to twofold increases in the Vi.. over the normal rabbit serum controls. The high degree of cross-inhibition ('80%) and the low degree of cross-reactivity in the quantitative precipitin test (-34%) suggest that these two enzymes possess structural similarities located mainly in the regions of the catalytic sites. marcescens
Many enzymes retain some catalytic activity (This research has been submitted to Syrain the presence of specific antibody. A few cuse University by D.A.F. as part of a Ph.D. enzymes are completely inhibited; others are dissertation.) not at all affected by specific antibody (8). MXTERIALS AND METHODS Activation of enzyme by antibody has been Bacterial strain. S. marcescens ATCC no. 60 was reported in a few instances, especially with poor substrates (1, 8). The use of antibodies to cultivated in stock cultures on Trypticase soy agar C. enzymes as a tool in the study of structural slants at 28and harvesting of bacteria. An inoculum similarities between antigenic determinants was of Growth S. marcescens was grown in 100 ml of medium reviewed by Arnon (1). Boyd and Phillips (4, 5) containing 4% (wt/vol) autolyzed yeast (12, 13) in a have shown that complete regression of the 250-ml Erlenmeyer flask. This was incubated for 18 h Gardner lymphosarcoma in the C3H mouse at 34 C with shaking and then added to 10 liters of could be achieved with smaller amounts of sterile 3% yeast extract medium in a 15-liter fermenL-asparaginase from Serratia marcescens than tor (New Brunswick Scientific Co., New Brunswick, from Escherichia coli. This might be due to N.J.). Incubation was continued for 24 h at 34 C with some differential effect of antibody on the aeration and agitation. Antifoam A, silicone defoamer Corning), was added (100 ml) as a 10% solution enzyme from these two organisms. It is impor- (Dow 0 and 8 h of incubation. Cells were harvested in a tant, therefore, to investigate the characteris- at continuous flow centrifuge at 10 C and tics of inhibition of these enzymes by their Sharples washed twice with 0.05 N tris(hydroxymethyl)specific antisera. aminomethane (Tris)-hydrochloride buffer (pH 8.6). We have previously shown that there is im- About 20 g of packed cells per liter containing 100 to munologic cross-reactivity between the L- 150 IU of enzyme per g was obtained. Enzyme assays. L-Asparaginase assays were conasparaginases isolated from S. marcescens and E. coli (23). The question arises whether this ducted by nesslerization for routine work (19). The cross-reactivity, which has been measured by reaction mixture contained saturating substrate and enzyme in a total volume of 2.0 ml of 0.05 M complement fixation and precipitin tests, is also Tris-hydrochloride pH 7.4, and was incubated demonstrable by enzyme inhibition. The pres- at 37 C for specifiedbuffer, times. Turbidity due to serum ent work was undertaken to study (i) the proteins was avoided by adding 0.5 ml of 1.5 M triquantitative inhibition of enzyme by homolo- chloroacetic acid to stop the reaction. The mixture gous and heterologous antisera; and (ii) the was kept in an ice bath for 15 min and then ceneffects of antibody on enzyme kinetics using trifuged. The clear supernatant was mixed with 0.4 ml of 1.5 N KOH and assayed. To standardize protein L-asparagine and L-glutamine as substrates. 424
VOL. 124, 1975
L-ASPARAGINASE FROM S. MARCESCENS
activation or stabilization of L-asparaginase (14, 17), 0.5 mg of bovine serum albumin (BSA) was added to each assay tube prior to incubation. A different assay was employed in the kinetic studies. This was a continuous assay in which the asparaginase and glutamic dehydrogenase (GDH) reactions were coupled, and the stoichiometric oxidation of reduced nicotinamide adenine dinucleotide was measured at 340 nm. The reaction mixture contained 1 mg of BSA, 5 pmol of a-ketoglutarate, 0.025 Amol of reduced nicotinamide adenine dinucleotide, 0.333 mg of GDH, specified amounts of L-asparaginase, and antiserum, in a total volume of 1.0 ml of 0.05 M Tris-hydrochloride buffer, pH 8.6, at 22 C. The reaction was initiated by the addition of L-asparagine or L-glutamine. Details of this procedure have been described elsewhere (9). One international unit of L-asparaginase activity is defined as the amount of enzyme required to liberate 1 gmol of NH4+ from L-asparagine in 1 min under the assay conditions described. Protein determinations. Protein concentrations were estimated after Lowry et al. (18) with BSA as the standard. Dilute protein solutions (5 to 20 ug of protein/ml) were measured spectrophotometrically at 215 and 225 nm with BSA as standard (20). Chemicals and reagents. These materials were obtained from the following sources: L-asparagine monohydrate, A grade, from Calbiochem, Los Angeles; L-glutamine, crystalline, grade III, from Sigma, St. Louis (solutions were prepared daily); L-GDH, bovine liver, type II, in 50% glycerol, from Sigma (dialyzed before use); 1-nicotinamide adenine dinucleotide (reduced), disodium salt, grade III, low nicotinamide adenine dinucleotide content, from Sigma; a-ketoglutarate, monosodium salt, from Sigma; Whatman diethylaminoethyl cellulose from W and R Balston Ltd., London, U.K.; Bio Gel from Bio-Rad Laboratories, Richmond, Calif.; Tris from Sigma; acrylamide, bisacrylamide, N,N,N',N'-tetramethylethylenediamine, and glycine, from Eastman Organic Chemicals Inc., Rochester, N.Y.; phenol reagent (Folin-Ciocalteau) from Harleco, Philadelphia; BSA from Armour Co., Chicago, Ill.; potassium iodide, copper sulfate, and sodium carbonate from General Chemical Division of Allied Chemical and Dye Corp., New York, N.Y. All other chemicals were obtained from Mallinckrodt Chemical Works, St. Louis, Mo. L-Asparaginase from E. coli was a gift from Horace D. Brown, Merck Sharp and Dohme Research Laboratories, Rahway, N.J. Purification of L-asparaginase from S. marcescens. Unless stated otherwise, all procedures were conducted at 4 C. The packed cells were diluted 1:3 (wt/vol) with 0.05 M Tris-hydrochloride buffer, pH 8.6, and the enzyme was released in a sonic oscillator (Raytheon Manufacturing Co., Waltham, Mass.) at 1.5 A for 7 min. The preparation was centrifuged 30 min at 37,000 x g and the pellet was discarded. The enzyme was purified by a modification of the method described by Ho et al. (15) as follows. The supernatant was adjusted to pH 4.5 by dropwise addition of 2 N acetic acid, and the precipitate was removed by centrifugation for 20 min at 16,300 x g and washed twice with pH 4.5 buffer. The precipitate
425
was discarded. The supernatant and the washings were pooled and adjusted to pH 6.6 with 3 N KOH. The solution was concentrated to one-half its volume in an XM-50 Diaflo ultrafilter (Amicon Corp., Lexington, Mass.) under 53 lbs/in' of N,. The concentrate with about 10 mg of protein per ml was cooled to -15 C as 95% ethanol was added dropwise with stirring until a cloudiness appeared, when the mixture was kept at -15 C for 1 h without stirring. The precipitate was removed by centrifugation at 16,300 x g for 20 min at -15 C. In a typical purification successive precipitates of L-asparaginase occurred at 0.1, 0.2, 0.3, 0.5, 0.8, and 1.5 volumes of added ethanol; and each precipitate was treated as above. Fractionation with ethanol was repeated as above to obtain further purification and those fractions having similar specific activity were pooled after dissolving in a minimal volume of 0.02 M sodium phosphate buffer, pH 6.6. Each preparation was examined on analytical polyacrylamide electrophoresis as described previously (4). Fractions with high enzyme activity were loaded (5 ml) onto a Bio Gel P-150 column (46 by 2.5 cm) containing 200 ml packed bed volume with a void volume of 40 ml and a flow rate of 12 ml/h. Fractions (4 ml) were collected and assayed for protein and enzyme activity. Peak tubes were pooled and concentrated on an XM-50 Diaflo membrane under 20 lbs/in2 of N,. Concentrates containing 15 to 20 mg of protein per ml were made 20% (wt/vol) in sucrose and loaded on a polyacrylamide column (6.6% gel) in a preparative electrophoresis apparatus (Shandon Scientific Co. Ltd., London, U.K.). The buffer was 0.025 M Tris-glycine, pH 8.4, and the current was 50 mA. Fractions were collected, assayed, and concentrated as above. Enzyme yields were greatly improved by replacing the membrane and recovering trapped enzyme during the run. Homogeneous enzyme with a single band on gel electrophoresis and having a specific activity of about 200 IU/mg of protein was obtained (Fig. 1). The purification procedure is summarized in Table 1. Antisera. Anti-S. marcescens asparaginase sera were prepared in three New Zealand white does weighing 2.0 to 2.5 kg (rabbits 245, 246, 247). One milligram of S. marcescens enzyme (specific activity 200 IU) in complete Freund adjuvant was injected intramuscularly at multiple sites in each rabbit on days 1, 10, and 20. These animals were bled on days 27 and 28 (first course antisera). On day 37 each rabbit received S. marcescens antigen as above and second course antisera were obtained 11 days later. On day 77 S. marcescens antigen was again injected and third course sera were taken-11 days subsequently. Rabbit anti-E. coli L-asparaginase sera (223, 224) were prepared as above but with large doses of antigen since more of it was available. Rabbits received 5 mg of antigen on days 1 and 10 and 3 mg on day 25. Six days later first course sera were taken. On days 45 and 49 animals were given 3 mg of E. coli antigen intravenously, and second course sera were .taken about 6 days thereafter. Antisera are designated by rabbit number and course of injection, e.g., no. 223-2 is serum from rabbit no. 223, second course. Antibody inhibition of L-asparaginase activity.
FERGUSON, PHILLIPS, AND MARUCCI
J. BACTERIOL.
The effects of rabbit and anti-asparaginase sera on enzyme activity were investigated using the method described by Fuller and Marucci (11) for the horse liver alcohol dehydrogenase-antibody system. Various amounts (1 to 100 Mg) of antibody protein (determined by quantitative precipitin test) in 0.5 ml of buffer were added to L-asparaginase (1 ug) in 0.5 ml of 0.01 M sodium phosphate buffer, pH 6.9, containing 0.85% NaCl; no visible precipitation occurred. The enzyme-
antiserum mixtures were then incubated at 25 C for 15 min. Aliquots (0.4 ml) of enzyme-antibody mixtures were assayed by mixing with a solution of L-asparagine (10- M) in the above buffer, incubating at 37 C for 60 min, and measuring NH,+ by nesslerization. All determinations were done in duplicate and normal rabbit serum controls were run in each experi-
426
A
B
em
ment.
The effects of rabbit anti-asparaginase sera on asparaginase kinetics were measured with L-asparagine and with L-glutamine as substrates. A single ratio of antibody to enzyme protein of 25:1 (wt/wt) was employed since this ratio produced maximum inhibition of L-asparaginase activity by all antisera. Appropriate amounts of antigen and antisera were mixed in a total volume of 1.0 ml of 0.01 M sodium phosphate buffer, pH 6.9, containing 0.85% NaCl and incubated at 25 C for 15 min. The mixture was then diluted to 2.0 ml with 0.05 M Tris-hydrochloride, pH 8.6, and 0.1-ml aliquots were assayed by the GDH coupled assay method employing various concentrations of substrate (L-asparagine or L-glutamine). Initial velocities were determined for each substrate concentration (in duplicate) and appropriate controls with buffer and normal rabbit serum were conducted. Due to the low L-glutaminase activity of these enzymes (5%), fivefold more enzyme and antiserum were used with L-glutamine as substrate than with the L-asparagine experiments.
FIG. 1. Polyacrylamide gel electrophoresis pf Lasparaginase from S. marcescens (A) and E. coli (B). The gel columns (10 by 0.5 cm) contained 7.5% acrylamide and were loaded with 40 Mg of enzyme protein (specific activity 200 IU/mg of protein) in 0.04 ml of 0.005 M Tris-glycine buffer, pH 8.4. Each column was run at 2.5 mA. Gels were stained with Coomassie blue. The arrow points to the top of the column.
RESULTS Effects of specific antibody on i^asparaginase activity. The effects of specific antibodies on the catalytic activity of the enzymes towards L-asparagine were quantitatively investigated. The results showed partial inhibition of enzyme activity. The data are presented in Fig. 2. In the presence of S. marcescens antisera, S. marcescens L-asparaginase manifested about 60%, inhi-
TABLE 1. Purification of L-asparaginase from S. marcescens Purification step
PurificationsteP
Total
protein (g)
activity (IU)
Sp act (IU/mg of protein)
Enrichment
(-fold)
116
Total recovery
100
72,000
Intact cells Crude extract .................
Total enzyme
55,000
0.47
76
pH 4.5 supernatant ............
31.8
41,000
1.3
2.8
57
Concentrate ...................
26.4
40,000
1.5
3.2
55
Ethanol precipitates First ....................... Second ..................... Third .......................
8.1 2.6 0.74
35,000 18,300 10,600
4.3 7.0 14.3
9.2 15 30
49 25 15
P-150 gel ......................
0.45
10,000
22.2
47
14
Polyacrylamide gel ............
0.023
4,530
419
6
197
I
--
ANInM NTEINI
IFp
pg
FIG. 2. Inhibition of the catalytic activity of L-asparaginase on L-asparagine by specific rabbit antibodies. Different amounts of antibody protein (1 to 1,000 Ag) in 0.5 ml of buffer were added to I Ag of L-asparaginase in 0.5 ml of 0.01 M sodium phosphate buffer at pH 6.9 containing 0.85% NaCI. The mixtures were kept at 37 C for 15 min and aliquots were taken for enzyme assays by nesslerization. L-Asparaginase from S. marcescens (0; sim) and E. coli (0; ec) were selected, the former designated as S. marcescens and the latter as E. coli. The foUowing rabbit antisera were used: A, anti-S. marcescens serum no. 245-1; B, anti-S. marcescens serum no. 246-2; C, anti-S. marcescens serum no. 247-2; D, anti-E. coli serum no 223-2; E, anti-E. coli serum no. 224-1. The arrows represent the region of antibody-antigen equivalence for sm and ec. 427
428
FERGUSON, PHILLIPS, AND MARUCCI
bition, whereas E. coli L-asparaginase activity was less inhibited (Fig. 2A, B, and C). In all instances, however, it required a significantly greater antibody-antigen ratio to achieve maximum inhibition than to reach equivalence in the quantitative precipitin reactions (D. A. Ferguson, Jr., A. W. Phillips; and A. A. Marucci, manuscript in preparation). This was also true when E. coli antisera were employed where enzyme inhibition was observed albeit at a somewhat decreased level (Fig. 2D and E). Again the enzyme activity of the homologous antigen was significantly more inhibited than that of the heterologous one. The overall cross-inhibition, however, was high and comparable with both E. coli and S. marcescens antisera and complete inhibition of enzyme activity was never attained even in large excess of antibody. A control experiment was performed to show that the degree of inhibition was a function of the number of antibody molecules bound to each enzyme molecule and independent of concentration effects at a given ratio of antibody to enzyme (Fig. 3). These data show this to be true for both homologous and heterologous enzymes with all antisera tested. Our data on the crossinhibition of L-asparaginase activity and the immunologic cross-reactivity of antisera against these L-asparaginases are summarized in Table 2. Effects of specifc antibody on enzyme kinetics. The GDH coupled assay was employed in the determination of the effects of specific antibody on the kinetics of amidohydrolysis of L-asparagine and L-glutamine by L-asparaginase. In all of these experiments an antibody-antigen ratio of 25:1 (wt/wt) was employed which gave maximal inhibition in the studies described above. It was found that the presence of normal rabbit serum did not alter the kinetics in any of the reactions described. Under these conditions the K,,PP values for the L-asparagine reaction were 1.25 x 10- M and 1.2 x 10- M for S. marcescens (Fig. 4A) and E. coli (Fig. 4B) enzymes, respectively; and the values for L-glutamine were 7.4 x 10-' M for the S. marcescens enzyme (Fig. 5A, curve 3) and 2.5 x 10-' M for the E. coli enzyme (Fig. 8). The Vmax values for L-glutamine were considerably lower than the values for L-asparagine. These results agree well with measurements taken in the absence of normal serum (21, 29). Since normal rabbit serum had no effect on enzyme kinetics, any effects produced by the various antisera are attributable solely to their specific antibody content. With L-asparagine as substrate each of the S. marcescens antisera
J. BACTERIOL. I
I
I
-I
J
70
Go
so
A*
0
o-
0~~~~ A
a
2
a
a
p
__a
N
L
2X
11
ANTIIOI
L-ASPAUACINASE
41
FIG. 3. The effect on enzyme activity of varying the antibody and antigen concentration while maintaining their ratio constant. The antibody-antigen ratio is 25:1 in all samples. All original volumes of the antibody and antigen solutions were multiplied by the factors indicated on the abscissa, mixed, and assayed by nesslerization after incubation at 37 C in 0.01 M sodium phosphate buffer, 0.15 N NaCI (pH 6.9), with saturating L-asparagine. L-Asparaginase from S. marcescens (0) and E. coli (0) were employed. The following rabbit antisera were used: A, anti-S. marcescens serum no. 245-1; B, anti-S. marcescens serum no. 246-2; C, anti-S. marcescens serum no. 247-2; D, anti-E. coli serum no. 223-2; E, anti-E. coli serum no. 224-1.
decreased the V,,., of both S. marcescens and E. coli enzymes (Fig. 4). The S. marcescens enzyme (Fig. 4A) was inhibited to a greater extent (50 to 75%) than E. coli enzyme (Fig. 4B) by all S. marcescens antisera; and the Km"PP values were somewhat increased (mixed inhibition) except with serum no. 245-1, where little or no effect was observed with either enzyme (noncompetitive inhibition). The measurements taken on enzyme inhibition in the coupled assay were comparable to the values obtained by nesslerization. In contrast, with Lglutamine as substrate, an enhancement of
L-ASPARAGINASE FROM S. MARCESCENS
VOL. 124, 1975
enzyme activity by specific antibodies usually occurred. In these experiments the same three S. marcescen,s antisera described above were involved. Two antisera resulted in a decrease in KmaPP value and generally an increase in V.ax for both S. marcescens and E. coli enzymes (Fig. 5). The exception was antiserum no. 247-2 which inexplicably manifested little influence on either enzyme. The most pronounced effect was produced by antiserum no. 246-2 against the E. coli enzyme (Fig. 5). These results thus show that the binding of specific antibodies to the enzyme can alter the molecule in such a way as to significantly increase the activity of the catalytic site for L-glutamine. The above experiments were repeated with rabbit antisera against E. coli enzyme. When L-asparagine was the substrate there were significant decreases in the V.ax of E. coli and S. marcescens enzymes and an increase in the KmSPP values, indicating a mixed inhibition (Fig. 6). Antiserum no. 223-2 was considerably more inhibitory than antiserum no. 224-1, but neither of these sera manifested the degree of inhibition shown by S. marcescens antisera no. 245-1 and no. 246-2. Here again the greatest inhibition was shown to occur in the homologous system, and the inhibition values obtained by the GDH assay were comparable to those obtained by nesslerization. With L-glutamine as substrate in a reaction
429
mixture containing E. coli antiserum no. 224-1, there was a pronounced increase in the V.., and decrease in the Km.PP of the E. coli enzyme, indicating enhancement of enzyme activity (Fig. 7). However, there was little influence upon the S. marcescens enzyme by this antiserum (Fig. 7A). Also, E. coli antiserum no. 223-2 had little effect on the activity of E. coli and S. marcescens enzymes. The contrasting effects of E. coli antiserum no. 224-1 upon E. coli and S. marcescens enzymes with L-glutamine as substrate are the only instances of significant kinetic differentiation between the two enzymes by any of the antisera using either substrate. This may indicate a similarity in the structure of the active sites of these enzymes. However the exceptional behavior of anti-E. coli serum no. 224-1 does suggest that the active sites are not identical.
DISCUSSION Inhibition of L-asparaginase by specific antisera may be envisioned as occurring by (i) direct blocking of catalytic sites by antibody specific for these sites, (ii) blocking of the active centers by antibody which has combined with antigenic determinants in close proximity to these sites, and/or (iii) a conformational change in the enzyme molecule as a result of its combination with antibody. The heterogenous nature of antisera to various enzymes has been shown by fractionation of anti-enzyme sera into inhibitory, noninhibitory, TABLE 2. The relative cross-inhibition of and activating antibody preparations (10, 27, L-asparaginase activity between rabbit 28, 31). Arnon and Shapira (3) isolated antianti- L-asparaginase from S. marcescens and E. coli body specific for the active center of papain. and a comparison with the cross-reactivity as This antibody combines with and inhibits the measured by quantitative precipitin reactions catalytic activity of chymopapain. By use of specific antisera against highly Cross-Imu inhibiImmuL-asparaginase isolated from E. coli purified tion of ~~crosAntiserum nologic Antlserum S. we have made some postulaand marcescens, no. enzyme reacAntigen tions about the structural similarities of these activtionb %n itya enzyme preparations. The high degree of crossinhibition is consistent with structural similarin the region of the catalytic site. We cannot ity 31 223-2 78 E. coli L-asparadistinguish effects of antibody directed to the ginase catalytic sites and effects of antibody directed 82 23 224-1 E. coli L-asparato determinants in close proximity to these ginase 53 97 245-1 S. marcescens Lsites. However, it would seem that, in these asparaginase antisera, antibody to regions neighboring the 21 70 246-2 S. marcescens [_ catalytic site play the major role in inhibition. asparaginase This is so because antibodies directed to the 34 45 S. marcescens L247-2 catalytic site would be expected to result in asparaginase complete inhibition (1, 3, 7). We were never able to obtain complete inhibition of enzyme a The substrate was L-asparagine. I These data are from Ferguson, Phillips, and even in great antibody excess. The cross-inhibition of the enzyme was found Marucci, unpublished data. (
itsxlo, M
-
FIG. 4. The effects of rabbit antibodies against S. marcescens L-asparaginase on the kinetics of the amidohydrolysis of L-asparagine by L-asparaginase. Specified amounts of antisera and enzyme (antibodyantigen ratio of 25:1) were mixed in a total volume of 1.0 ml of 0.01 M sodium phosphate buffer, pH 6.9, containing 0.85% NaCI and incubated at 24 C for 15 min. The mixtures were then diluted to 2.0 ml with 0.05 M Tris-hydrochloride, pH 8.6, and 0.1-ml aliquots were assayed by the GDH coupled assay (see text). L-Asparaginases are from S. marcescens (0) and E. coli (0). (A) Reactions of S. marcescens with different anti-S. marcescens sera: curve 1, serum no. 245-1; curve 2, serum no. 246-2; curve 3, serum no. 247-2. The effect of normal rabbit serum is shown (0). (B) Reactions of E. coli with anti-S. marcescens sera: curve 1, serum no. 245-1; curve 2, serum no. 246-2; curve 3, serum no. 247-2. A normal rabbit serum control is shown (0). 430
VOL. 124, 1975
FiG. 5. The effects of rabbit antibodies against S. L-asparaginase on the kinetics of amidohydrolysis of L-glutamine by L-asparaginase. See legend to Fig. 4 for the reaction conditions, symbols, and abbreviations. (A) Reactions of S. marcescens with different anti-S. marcescens sera: curve 1, serum no. 246-2; curve 2, serum no. 245-1; curve 3, normal rabbit serum control (0). (B) Reactions of E. coli with anti-S. marcescens sera: curve 1, serum no. 246-2; curve 2, serum no. 245-1; curve 3, normal rabbit serum control (see Fig. 8). marcescens
to be about 80%. This is greater than the cross-reactivity as determined by quantitative precipitin analyses (ca. 34%) (D. A. Ferguson et
al., manuscript in preparation). From these data it might be postulated that regions of the enzyme not involved in catalysis are more structurally distinct than are the enzymic sites. On evolutionary grounds, this should be expected since only those regions necessary for catalytic activity would be essential and would be conserved. Furthermore, since S. marcescens antisera were more efficient in inhibiting the homologous enzyme, it could be inferred that the catalytic sites on this enzyme are more
L-ASPARAGINASE FROM S. MARCESCENS
431
susceptible to reaction with antibody than are similar sites on the E. coli enzyme. Experiments on kinetics of L-asparaginase hydrolysis in the presence and absence of antisera were done in an attempt to gain some idea about the mechanism of inhibition (16). In these studies most antisera showed a "mixed" pattern of inhibition (26); i.e., both KmaPP and Vmax vary in the presence of antiserum with both homologous and heterologous enzymes. This could be accounted for by either a conformational change in the enzyme which decreased both binding and catalysis or by a combination of conformational change and steric hindrance. Steric hindrance alone would not give this pattern of inhibition. Anti-S. marcescens serum no. 245-1 gave an unusual partial mixed type inhibition (Fig. 5A, curve 2) similar to the C5 system described by Segel (26). Future studies on the inhibitory effects of individual species of immunoglobulins from fractionated antisera, as well as the influence of inhibitor concentration on enzyme kinetics, should be useful aids in the elucidation of the inhibition mechanisms. The experiments with the poor substrate, L-glutamine (6, 21), further reinforce the view that antibody-induced conformational change can be a mechanism for altered activity of this enzyme. It was found that in the presence of antibody, the rate of catalysis of L-glutamine was significantly increased over normal serum controls. This is similar to the results of Pollock (24) with penicillinase, Okada et al. (22) with amylase, Rotman and Celada (25) with ,8-galactosidase, and Suzuki et al. (28) with ribonuclease. A probable mechanism for this effect is that constraint placed on the enzyme by bound antibody prevents conformative changes usually imposed by normal substrate (30, 31). Alternatively, the conformation of the enzyme may be so altered after reaction with antibody that the poor substrate is now more readily accepted. It would be of interest to make similar comparisons of L-asparaginases from other microbial species from the viewpoint of evolutionary relationships as well as for clinical applications to cancer chemotherapy. Concerning the latter, establishment of the mechanism of inhibition of L-asparaginase by human antibody should be a high priority item. What antigenic determinants are recognized on this enzyme by the human immunologic system? Can these be altered without affecting the enzymic activity? Drugs used in combined therapy with Lasparaginase ought to be examined for their effects on enzyme kinetics in the presence and in the absence of specific human antibodies.
432
FERGUSON, PHILLIPS, AND MARUCCI
J. BACTERIOL.
~~//
I A I-4-
I0-
A .*
,
It
6
4
2
0
2
ii511@3
4 N
6
FIG. 7. Effects of rabbit antibodies against E. coli L-asparaginase on the kinetics of L-glutamine and amidohydrolysis by L-asparaginase. See legend to Fig. 4 for reaction conditions, symbols, and abbreviations. The following mixtures were used: (A) serum no. 224-1 with S. marcescens; (B) serum no. 224-1 with E. coli.
8
6
4
2
2
lIIsIO
4
6
a
FIG. 6. Effects of rabbit antibodies against E. coli L-asparaginase on the kinetics of amidohydrolysis of L-asparagine by L-asparaginase. See legend to Fig. 4 for reaction conditions, symbols, and abbreviations. (A) Reactions of S. marcescens with different anti-E. coli sera: curve 1, serum no. 223-2; curve 2, serum no.
224-1; curve 3, normal rabbit serum control (0). (B) Reactions of E. coli with anti-E. coli sera: curve 1, serum no. 223-2; curve 2, serum no. 224-1; curve 3, normal rabbit serum control (0).
VOL. 124, 1975
L-ASPARAGINASE FROM S. MARCESCENS
28
24
20
12
~ ~ I I/1 1 6
4
2
0
, ~~I ~~~~~~ I 6
IISX
102
M
FIG. 8. Kinetics of amidohydrolysis of L-glutamine by E. coli L-asparaginase in the presence of normal rabbit serum. The reaction mixture consisted of 0.75 ml of 1:5 dilution of serum, 15 ,gg of enzyme, L-glutamine, 0.05 M Tris-hydrochloride buffer, pH 8.6, in a total volume of 2.0 ml at 25 C. Velocity is expressed as micromoles of NH4+ released per minute. This plot is curve 3 in Fig. 5B and the control plot in Fig. 7B (curve without symbols). ACKNOWLEDGMENTS This investigation was aided by a grant from the Hendricks Research Fund of Syracuse University to the State University of New York, Upstate Medical Center (A. A. Marucci and A. W. Phillips, co-investigators). We are grateful to H. Richard Levy, Robert W. Jackson, and Jack Bryan for helpful discussions, and we appreciate the skillful technical assistance of Kenneth Malament and Howard Telson. We thank the following persons for supplying us with materials: C. Gordon Zubrod and Samuel Poiley. National Cancer Institute, National Institutes of Health; Kenneth Price, John Godfrey, and Bernard Heineman, Bristol Laboratories, Inc., Syracuse, N.Y.; and Horace D. Brown, Merck Sharp and Dohme Research Laboratories, Rahway, N.J. We thank Georgia Ventura and Pam Dobbert for doing the typescript. LITERATURE CITED 1. Arnon, R. 1971. Antibodies to enzymes-a tool in the study of antigenic specificity determinants. Curr. Top. Microbiol. Immunol. 54:47-93. 2. Arnon, R.. and B. Schechter. 1966. Immunological studies on specific antibodies against trypsin. Immunochemistry 3:451-461. 3. Arnon, R., and E. Shapira. 1968. Comparison between the antigenic structure of mutually related enzymes. A study with papain and chymotrypsin. Biochemistry 7:4196-4200. 4. Boyd, J. W., and A. W. Phillips. 1971. Purification and
properties of L-asparaginase from Serratia J. Bacteriol. 106:578-587.
marcescens.
433
5. Boyd, J. W., and A. W. Phillips. 1971. Inhibition of lymphoma 6C3HED by L-asparaginase from Serratia marcescens. J. Natl. Cancer Inst. 46:1271-1276. 6. Campbell, H. A., and L. T. Mashburn. 1969. L-Asparaginase EC-2 from Escherichia coli. Some substrate specificity characteristics. Biochemistry 8:3768-3775. 7. Celada, F., and R. Strom. 1972. Antibody-induced conformational changes in proteins. Q. Rev. Biophys. 5:395-425. 8. Cinader, B. 1967. Antibodies to enzymes-a discussion of the mechanisms of inhibition and activation, p. 85-137. In B. Cinader (ed.), Antibodies to biologically active molecules. Pergamon Press, Oxford. 9. Ferguson, D. A., Jr., J. W. Boyd, and A. W. Phillips. 1974. Continuous assays of L-asparaginase by coupling with glutamic dehydrogenase and by cationic glass electrode. Anal. Biochem. 62:81-90. 10. Fuchs, S. P., P. Cuatrecasas, D. A. Ontjes, and C. B. Anfinsen. 1969. Correlation betwe.ewAOte antigenic and catalytic properties of staphylococc*1 nuclease. J. Biol. Chem. 244:943-950 11. Fuller, T. C., and A. A. Marucci. 1971. Immunoenzymology of liver alcohol dehydrogenase. I. Factors influencing the inhibition of horse liver alcohol dehydrogenase by rabbit and guinea pig antisera. J. Immunol. 106:110-119. 12. Heinemann, B., and A. J. Howard. 1969. Production of
tumor-inhibitory L-asparaginase by submerged growth of Serratia marcescens. Appl. Microbiol. 18:550-554. 13. Heinemann, B., A. J. Howard, and M. J. Palocz. 1970.
14.
15.
16.
17. 18.
19.
20. 21. 22.
23.
24.
Influence of dissolved oxygen levels on production of L-asparaginase and prodigiosin by Serratia marcescens. Appl. Microbiol. 19:800-804. Ho, P. P. K., and L. Jones. 1969. Non-specific protein stabilization of L-asparaginase. Biochim. Biophys. Acta 177:172-174. Ho, P. P. K., E. B. Milikan, J. L. Bobbitt, E. L. Grinnan, P. J. Burck, B. M. Frank, L. D. Boeck, and R. W. Squires. 1970. Crystalline L-asparaginase from Escherichia coli B. I. Purification and chemical characterization. J. Biol. Chem. 245:3708-3715. Kabat, E. A. 1968. Structural concepts in immunology and immunochemistry. Holt, Rinehart and Winston, Inc., New York. Lee, M. B., and J. M. Bridges. 1968. L-asparaginase activity in human and animal sera. Nature (London) 217:758-759. Lowry, 0. H., N. J. Rosebrough, A. L. Farr, and R. J. Randall. 1951. Protein measurement with the Folin phenol reagent. J. Biol. Chem. 193:265-275. Meister, A. 1955. Glutaminase, asparaginase and a-keto acid amidase, p. 380-385. In S. P. Colowick and N. 0. Kaplan (ed.), Methods in enzymology, vol. II. Academic Press Inc., New York. Murphy, J. B., and M. W. Kies. 1960. Note on spectrophotometric determinations of proteins in dilute solutions. Biochim. Biophys. Acta 45:382-384. Novak, E. K., and A. W. Phillips. 1974. L-Glutamine as a substrate for L-asparaginase from Serratia marcescens. J. Bacteriol. 117:593-600. Okada, Y., T. Ikenaka, T. Yagura, and Y. Yamamura. 1963. Immunological heterogeneity of rabbit antibody fragments against taka-amylase A. J. Biochem. (Tokvo) 54:101-102. Phillips, A. W., J. W. Boyd, D. A. Ferguson, Jr., and A. A. Marucci. 1971. Immunochemical comparison of L-asparaginase from Serratia marcescens and Escherichia coli. J. Bacteriol. 107:461-467. Pollock, M. R. 1964. Stimulating and inhibiting antibodies for bacterial penicillinase. Immunology
7:707-723. 25. Rotman, M. B., and F. Celada. 1968. Antibody mediated
434
FERGUSON, PHILLIPS, AND MARUCCI
J. BACTERIOL.
activation of a defective fl-D-galactosidase extracted from an Escherichia coli mutant. Proc. Natl. Acad. Sci. U.S.A. 60:660-667. 26. Segel, I. H. 1975. Enzyme kinetics, p. 189-192. WileyInterscience, New York. 27. Soru, E., and 0. Zaharia. 1974. Immunoenzymology of L-asparaginase from the BCG strain of Mycobacterium bovis. Immunochemistry 11:791-795. 28. Suzuki, T., H. Peliochova, and B. Cinader. 1969. Enzyme activation by antibody. I. Fractionation of immune sera in search for an enzyme activating antibody. J.
Immunol. 103:1366-1376. 29. Wriston, J. C., Jr., and T. C. Yellin. 1973. L-asparaginase: a review. Adv. Enzymol. 39:185-248. 30. Zyk, N., and N. Citri. 1968. The interaction of penicillinase with penicillins. VI. Comparison of free and antibody-bound enzyme. Biochim. Biophys. Acta 159:317-326. 31. Zyk, N., and N. Citri. 1968. The interaction of penicillinase with penicillins. VII. Effect of specific antibodies on conformative response. Biochim. Biophys. Acta 159:327-339.
