THE JOURSAL OF Elrom~rca~ CHEMIIISTRY Vol. 24i, No. 16, Issue of August 25, pp. 50-H-5047, 1972 Printedin G.S.A. The Developmental Biochemistry of Cottonseed Embryogenesis and Germination II. CATALYTIC PROPERTIES OF THE COTTON CARBOXYPEPTIDASE” (Received for publication, December 13, 1971) JAMES N. IHLE~ AKD LEON S. DURE, III$ From the Depadment of Biochemistry, University of Georgia, 4 thens, Georgia 30601 SUMMARY involve a serine-mediated hydrolysis, which is unusual for The cotton carboxypeptidase hydrolyzes several ester carboxypeptidases, generally considered to be metal ion-mediated substrates, small polypeptides, and denatured proteins. It hydrolases (2, 3). Third, the “serine protease” type of mecha- does not hydrolyze dipeptides and hydrolyzes tripeptides very nism has not been definitively demonstrated for any plant slowly. It appears to hydrolyze any amino acid from the protease, most of which have been found to utilize the sulfhydryl- COOH-terminal of polypeptides, including proline. Its mediated type of mechanism (4). Fourth, the cotton car- activity against esters and peptides is not inhibited by boxypeptidase appears to have active hydrolytic sites on two of sulfhydryl antagonists, and no requirement for metal ions its three subunits. Evidence for these properties is presented in has been demonstrated. Its esterase and carboxypeptidase this paper. activities are inhibited by diisopropylfluorophosphate, and phosphoserine can be isolated from the diisopropylfluorophos- EXPERINENTAL PROCEDURE phate-inhibited enzyme after mild acid hydrolysis, indicating that the enzyme employs a “serine protease” type of hy- Materials drolytic mechanism. Enzyme, Subsfmtes, and inhibitors-The cotton carboxypep- Two molecules of diisopropylfluorophosphate must react tidase was purified to homogeneity by the procedure presented in with a molecule of native enzyme to completely inhibit its the preceding paper (1). Yeast enolase and sea pansy luciferase activity. When the enzyme is inhibited with radioactive were obtained from Drs. J. M. Rrewer and XI. J. Cormier, re- diisopropylfluorophosphate, the two large polypeptide chains spectively, and bovine chymotrypsin (tosyllysine chloromethyl of the enzyme incorporate radioactivity. Thus, two active ketone-treated) from Dr. J. Travis, all of the Biochemistry De- sites, each on a separate chain, are indicated for the enzyme. partment, University of Georgia. Ilovine trypsin and soybean Chymotrypsin digestion of the enzyme inhibited with radio- trypsin inhibitor were obtained from Worthington; subtilisin and active diisopropylfluorophosphate yields a single radioactive diisopropylfluorophosphate from Sigma. UAEE,’ A’I’EE, tosyl- peptide, indicating that the sequence of amino acids around L-arginine methyl ester, phenylmethylsulfonylfluoride, and the active site serine of each subunit is identical. glucagon were obtained from Calbiochem. Angiotensin II monoacetate hydrate (Lot No. 69620), all dipeptides, and leucyl- glycylglycine were obtained from Miles Labs. p-Nitrophenyl lnu- rate, p-nitrophenyl acetate, and N-ethylmaleimide were obtained from Pierce Chemical Co. Hippuryllysine, hippurylphenylala- The cntalvtic capabilities and the mechanism of catalysis of nine, and the A1 chain of bovine insulin (S-aminoethylated) were the cotton carboxypeptidase, whose purification is presented in purchased from Mann. Cbz-arginine methyl ester, Cbz-alanine the preceding paper (l), appear to be unusual in several aspects. p-nitrophenyl ester, Cbz-leucylglycylglycine methyl ester, and First, t,he enzyme exhibits esterase activity against a broad range 1-chloro-3-tosylamido-7-amino-2-heptanone (tosyllysine chloro- of endoprotease substrates, as well as carboxypeptidase activity, methyl ketone) were obtained from Cycle Chemical Co., ‘I’riace- which combination is not generally found in a single protease. tin was obtained from Schwarz BioResenrch; 1, IO-phenanthro- Second, the mechanism of action for both activities appears to line from Fisher; and cottonseed oil from Will. Radiochemicals-[l”C]Diisopropylfluorophosphate was obtained from ?;ew England Nuclear with a specific activity of 111 PC1 per from the National Commission * This investigation Contract was supported SRO-AT-(3%1).638 in part and by Atomic Grant Energy pmole, and [~2P]diisopropylfluorophosphate from Amersham- Searle with an initial specific activity of 62.3 &i per pmole. Present $ United tory, Oak Ridge, address, States Science Foundation. GB-16632 Biology Public Division, Health Service Oak Ridge Predoctoral National Labora- Fellow. ester; 1 The ATEE, abbreviations acetyl-L-tyrosine used are: BAEE, ethyl ester; benzoyl-L-arginine Cbz-, benzyloxycar- ethyl velopment $ Recipient Award of United Tennessee 37830. GM 35309-05. States Public Health Service Career De- boxyl-; DIP, diisopropylphosphate. 5041 Downloaded from http://www.jbc.org/ by guest on June 17, 20155042 TABLE I 1 Rate of hydrolysis of ester substrates by cotton carboxypeptidase NH2ASP-ARG-VAL-TYR-ILE-HIS-PRO-PHECOOH Esterase activities were determined by pH-stat titration as described under “Methods” with the use of 0.05 mg of enzyme in 25 ml of solution containing the indicated concentration of sub- strate. Micromoles Substrate Concentration hydrolyzed/ min/mg protein N-Benzoyl- L - arginine ethyl ester . . . . . . . . . . . . . . . . 10-s 25.5 N-Acetyl-L-tyrosine ethyl es- ter......................... 10-a 23.6 Cbz-L-alanine p-nitrophenyl ester. ..____. _.._.__., 10-a 23.2 Cbz-L-arginine methyl ester. 10-s 8.2 Tosyl-L-arginine methyl ester. 10-a 3.2 Cbz-leucylglycylglycine methyl ester. 10-s 2.2 Triacetin . 1% emulsion (v/v) 0 p-Nitrophenyl acetate. . 10-S 0 p-Nitrophenyl laurate. 10-a 0 Cottonseed Oil. 4% emulsion (v/v) 0 Methods Measurements of E’sterase Activity-E&erase activities of the carboxypeptidase against the substrates listed in Table I were determined by pa-stat titration with the use of a Radiometer auto-burette, model ABU lb/SBR, B/TTT 1, and 0.01 JI NaOH as titrant to maintain pH 6.6. For the determination of Michaelis constants, the hydrolysis of benzoyl-L-arginine ethyl ester (BAEE) and N-acetyl-L-tyrosine ethyl ester (ATEE) was measured by the spectrophotometric assay of Schwert and Takenaka (5) and N-benzoyloxycarbonyl-L-alanine p-nitrophenyl ester (Cbz-alanine p-nitrophenyl ester) hydrolysis by the spec- trophotometric method of Janoff (6), all in 0.1 L{ sodium phos- phate, pH 6.6. In all other instances, e&erase activity was determined by the spectrophotometric assay of BAEE hydrolysis (5). All spectrophotometric assays were performed with a Cary model 15 spectrophotometer. Xeasurements of Peptidase Activity-The hydrolysis of casein and bovine serum albumin were assayed by the method of Anson (7), in which 0.1 mg of enzyme was incubated with 1 ml of a 1% solution of substrate in 0.01 M sodium phosphate, pH 6.6, at 25” for 15, 30, and 45 min, after which the reaction was stopped by the addition of 3 ml of 5y0 trichloroacetic acid. Enzyme ac- tivities against benzoyl-L-arginine p-nitroanilide, hippuryl-n- phenylalanine, and hippuryl-L-lysine were assayed spectropho- tometrically in 0.01 M sodium phosphate, pH 6.6, by the methods of Erlanger et al. (8), Folk and Schirmer (9), and Wolff et al. (lo), respectively. Measurements of Carboxypeptidase Activity-The rate of release of amino acids from Angiotensin II, insulin, glucagon, luciferase, enolase, and leucylglycylglycine was followed by analysis of free amino acids with a Beckman/Spinco amino acid analyzer, model 120C. Details are given in the legend to Fig. 1 and in Tables II and III. The rates were calculated from a summation of the micromoles of the individual amino acids released after a single interval of incubation time, except in the case of Angio- / PHE FIG. II by the cotton 1. Kinetics of the release of amino acids from Angiotensin enzyme. Reaction mixtures containing 70 pg of enzyme phosphate and 1 mg of Angiotensin II in 0.2 ml of 0.001 M sodium 30,60, and 120 min by the addition were incubated at 25” and the reactions stopped at 0, pH of 1.8 ml of 0.2 M sodium citrate, 2.2. Each incubation mixture was then subjected to amino acid analysis. The amino acid sequence of Angiotensin II is shown at the top of the figure. tensin II in which amino acid release was measured at several time points. Dipeptide hydrolysis was assayed by the ninhydrin method of Matheson and Tattrie (11) in which 0.1 mg of carboxypeptidase was incubated with 1 pmole of dipeptide in 1 ml of 0.01 M sodium phosphate, pH 6.6, at 25”. Measurements were made after 1 and 2 hours of incubation. To determine the effect of pH and ionic strength on carbox- ypeptidase activity, the rate of hydrolysis of the A chain of insulin was obtained by the ninhydrin assay. To determine the effect of pH, 0.08 mg of enzyme was incubated with 1 mg of substrate at 25” in 1 ml of 0.01 112 sodium acetate or sodium phosphate at various pH values, aliquot samples of 0.25 ml were removed and assayed after 15, 30, and 45 min, and initial re- action velocities determined. To determine the effect of ionic strength, the same concentrations of enzyme and substrate as above were incubated at 25” in a solution containing 0.001 >f sodium phosphate, pH 5.8, and various concentrations of NaCl. Aliquot samples were removed and assayed as above. Other LVethods-The incorporation of radioactivity from [r4C]diisopropylfluorophosphate into the carboxypeptidase was measured by applying aliquot samples of an enzyme-[r4C]- diisopropylfluorophosphate incubation mixture to Whatman No. 3MM filter paper disks which were washed four times with 500 ml of 10% trichloroacetic acid for 30 min at 5”. The disks were dried, placed in scintillator vials containing 5 ml of scin- tillation fluid, and the radioactivity determined in a Packard Tri-Carb scintillation counter. Sodium dodecyl sulfate-poly- acrylamide gel electrophoresis was performed by the method of Weber and Osborn (12). ‘Radioactivity profiles of the poly- acrylamide gels were obtained by slicing the gel into 1.5-mm sections and treating each section in a scintillator vial with 0.5 ml of 1 M NaOH for 16 hours at 25” to hydrolyze and solubilize the 14C-labeled isopropyl groups from the diisopropylphosphoserine- protein. The solutions were neutralized with 0.5 ml of 1 M acetic acid and 10 ml of aqueous scintillator solution added for radioactivity determination. The carboxypeptidase was oxi- dized with performic acid by the method of Hirs (13). Downloaded from http://www.jbc.org/ by guest on June 17, 2015TABLE II Hydrolysis of peptide and protein substrates by cotton cnrboxypeptidase Hydrolyses were carried out as indicated below and the reac- tions stopped by the addition of 0.2 M sodium citrate buffer, pH 2.2. The concentrations of liberated amino acids were determined by amino acid analysis of the reaction mixture. Substrate released Amino acid -/- Insulina Glucag&’ Luciferas& EllOhS& pmJles Tryptophan -e 0.127 0.004 0.015 Lysine - 0.032 0.089 0.054 Histidine. - - 0.015 0.028 Arginine . . - 0.152 0.036 0.030 Aspartic acid. . - 0.144 0.050 0.043 Threonine - + 0.034 0.018 Glutamine.. S.f + + + Asparagine + + + + Serine.................. + 0.091 + + Glutamic acid. 0.148 - 0.072 0.044 Proline. - - 0.037 0.014 Glycine 0.047 - 0.041 0.056 Alanine 0.150 0.075 0.053 0.074 Cysteine - - + - Valine . 0.128 0.102 0.041 0.047 Methionine. - 0.127 0.014 0.006 Isoleucine . . . . - - 0.028 0.017 Leucine . . 0.289 0.201 0.055 0.080 Tyrosine 0.290 0.063 0.016 0.009 Phenylalanine - 0.105 0.024 0.029 Aminoethylcysteine.. 0.160 - - - a One milligram (0.41 pmole) of the A chain of insulin was re- acted with 0.146 mg of cotton carboxypeptidase for 90 min at 25” in 0.2 ml of 0.001 M sodium phosphate, pH 6.6. * One milligram (0.287 pmole) of glucagon was reacted with 0.149 mg of cotton carboxypeptidase for 90 min at 25” in 1 ml of 0.001 M sodium phosphate, pH 6.6. c Heat-denatured sea pansy luciferase (1.7 mg) was reacted with 0.046 mg of cotton carboxypeptidase for 5 hours at 37” in 1 ml of 0.005 M sodium phosphate, pH 6.2. d Acid-denatured yeast enolase (8 mg) was reacted with 0.015 mg of cotton carboxypeptidase for 20 hours at 37” in 1 ml of 0.05 M sodium phosphate, pH 7.2. 0 Indicates the amino acid was not present. / Indicates that the amino acid was present but not quantita- tively determined. RESULTS E&erase Activity The range of esterase activity of the purified enzyme was examined utilizing the pH-stat assay. Table I lists the esterase substrates assayed and the rates of their hydrolysis where ap- plicable. This table shows that in addition to BAEE, a trypsin substrate, the cotton enzyme hydrolyzed ATEE, a chymotryptic substrate (5), and Cbz-n-alanine p-nitrophenyl ester, an elastase substrate (6), all at approximately equivalent rates when assayed at 10 mM concentration. The Michaelis constants calculated for these three substrates were 4.17 x low4 M for BAEE, 1.34 X 10W2 M for ATEE, and 2.38 X 10m4 M for the Cbz-r-alanine p-nitrophenyl ester. It should be noted that the rate of hy- 5043 TABLE III Rate of hydrolysis of peptide and protein substrates by cotton carboxypeptidase Amino acids Substrate Molarity’- released/ min/mg peptidase carboxy- pm&s Glucagon*...................... 3 x 10-d 0.121 A chain of insulin*. 10-s 0.150 Angiotensin IIc. 5 x 10-s 0.186 Denatured sea pansy luciferase”. 8 X 1O-6 0.052 Denatured yeast enolaseb. 1 x 10-d 0.035 Leucylglycylglycined. 1 x 10-s 0.007 Alanylalaninee. 1 x 10-Z 0 Valylglutamatee. 1 x 10-Z 0 Tryptophanylalaninee. 1 x 10-S 0 Histidylalaninee 1 x 10-a 0 Glycylaspartatee. 1 x 10-a 0 Glutamylvaline6. 1 x 10-s 0 D Molarity based on the number of subunits where applicable. * Determined by amino acid analysis of the reaction mixtures given in Table II. c Determined from the data presented in Fig. 1. d Determined by amino acid analysis of a l-ml reaction mixture containing 1 pmole of substrate, 8Opg of enzyme in 0.001 M sodium phosphate, pH 6.6, incubated for 180 min at 25”. e Determined by the ninhydrin assay described under “Meth- ods.” drolysis of ATEE was obtained at an ATEE concentration that is much lower than the Michaelis constant for this substrate. This was unavoidable because of the low solubility of ATEE at pH 6.6. Presumably the rate of hydrolysis of ATEE would be much greater than those of BAEE or Cbz-L-alanine p-nitro- phenyl ester at saturating concentrations. The ability to hydrolyze these three substrates is unusual for a single protease and could indicate contaminating enzyme activities. However, as shown in the preceding paper (l), the cotton carboxypeptidase appears homogeneous by several criteria, and, second, the ratios of these esterase activities to each other and to carboxypeptidase activity are constant from one preparation of the purified enzyme to another. Table I also shows that the cotton enzyme did not hydrolyze lipase ester substrates such as triacetin, p-nitrophenyl acetate, p-nitrophenyl laurate, or an emulsion of cottonseed oil. Thus, as shown by Table I, the enzyme’s esterase activities resemble those of endoproteolytic enzymes more closely than those of exoproteases. Peptidase Activity No proteolytic activity could be detected when the enzyme was incubated with casein or bovine serum albumin as described under “Methods.” Furthermore, the enzyme did not hydrolyze either benzoyl-L-arginine p-nitroanilide, a trypsin substrate requiring amide bond hydrolysis, or the carboxypeptidase A and B substrates, hippurylphenylalanine and hippuryllysine, re- spectively. The only proteolytic activity we have observed for the enzyme is carboxypeptidase activity against peptides and denatured proteins. This type of activity is demonstrated by the hydrolysis of the octapeptide hormone, Angiotensin II. When the hydrolysis of Angiotensin II by the enzyme was Downloaded from http://www.jbc.org/ by guest on June 17, 20155044 followed by analyses of the free amino acid liberated, the results showu in Fig. 1 were obtained. In the figure, the micromoles of the individual amino acids released are plotted with respect to incubation time. This plot shows a sequential release of amino acids from the COOKterminal of the peptide whicll would be expected from a nonspecific carboxypeptidase. Thus, approxi- mately 40 7O of the COOKterminal phenylalanine was released from the peptide after 120 min, followed by about 35% of the proline, 28% of the histidine, 2576 of the isoleucine, and 22$& of TABLI; IV Effect of inhibitors on e&erase activity Estcrase activity against RAISE u-as assayed spectrophotomet- rically as described under “Methods.” Enzyme (50 pg) s-as in- cubated with the inhibitor at 25” in 1 ml of 0.1 N sodium phos- phate, pH 0.0, for the time indicated prior to assay, escept, for EDTA and l,lO-phenanthroline where the inhibition TVas deter- mined after dialysis of an enzyme solution against nn aqueous solution of the inhibitor for 12 hollrs at 4”. the tyrosine. This pattern of release reflects the sequence of amino acids through the first 5 residues from the COOH-terminal. The results in Fig. 1 also demonstrate that the hydrolysis by the cotton carboxypeptidase was not stopped by COOI-I-terminal proline, as is the case with carboxypept.idase A and B (2, 3). The release of valine was very slow, and arginine and aspartic acid were not released. This presumably indicates that the rate of hydrolysis decreases at the tripeptide level, and that the enzyme cannot hydrolyze dipeptides. These conclusions are further substantiated by experiments described below. The apparent lack of specificity of the enzyme for the COOH- terminal amino acid was further examined by analyzing the release of amino acids from large polypeptides (glucagon and the A chain of insulin) and from two denatured enzymes (sea pansy luciferase and yeast enolase). The results are given in Table II, whicll slmws that after a SO-min incubation of each of the peptides with the enzyme, the release of most of t,he different amino acids colltnilled ill them could be demonstrated. When the two denatured enzymes were incubated with the carboxypeptidase for extended ljeriods, the release of all 20 amino acids was demon- strable. Thece data. suggest that no amino acid in the COOH- terminal l)osition is resistant to hydrolysis by the cotton car- bosypeptida,~e. The rate of amino atid release from several peptides and denatured enzymes catalyzed by the enzyme is presented in Table ITT. It appears from the data in this table that the rate of hydrolysis was much slower with large poly- l>eptides (represented by the denatured enzymes) than with illtermetliate hize poly~~epticles (represented by glucagon, the A\\ chain of insulin, and Angiotensin 11). However, the molar concentrations of substrates used were 5- to lo-fold lower in the case of the denatured proteins, and, hence, probably represented subsaturatina levels of polypeptide chains. The table does show that t,he rate of hydrolysis of the tripeptide, leucylglycylglycine, IT-as 20-fold lower than that of the A chain of insulin at identical concentration.<, and shows further that the dipept’ides were not hydrolyzed. Comparison oj Esterase and Carboxypeptidase Activities The effect5 of pH and ionic strength on these two activities were compared by measuring the rate of hydrolysis of BbEE ,sl)ectrophotometrically as indicative of esterase activity and the rate of release of amino acids from the A chain of insulin by the ninhydrin assay as indicative of carboxypeptidase activity. LI single concentration of the two substrates, which was subsatu- rating in both cases, was used in both assay systems. Thus, the results may have only a practical significance. Under these conditions carboxypeptidase activity was maximal at pH 5.8, TT-hereas esterase activity was maximal at pH 6.6. Esterase ac- tivity did not vary when assayed in 0.001, 0.01, or 0.1 M NaCl and 0.001 M sodium phosphate. On the other hand, carboxy- peptidase activity in 0.1 JZ NaCl-0.001 M sodium phosphate was only 21% of that found in 0.001 M sodium phosphate alone. Inhibitor JIolarit> Percentage inhibition of Diisopropylfluorophosphate 10-a 15 min 100 Phenylmethylsulfonylfluo- ride. 10-a GO min 100 Tosyllysine chloromethyl ~ ketone. i 10-a 240 min I~--Ethylmaleimide. 10-a 240 min p-Nitrophenyl acetate. .~ 10-Z 240 min Soybean trypsin inhibitor. 1 mg/ml 240 min l,lO-I’henanthroline. .j 10-Z 12 hrs EDT-4 . . . / 10-Z 12 hrs (These parameters have proven important when this enzyme has been used as a reagent to digest peptides.) Xeclzanis?n of Action Inhibitors Xfudies-The mechanism of action of t,hr elqme was suggested by the data in Table IV which summarize the ef- fects of various inhibitors on its esterase activity determined b! the hydrolysis of B,UZE. Of the \\-nrious inhibitors tested only diisopropylfiuorophoaphate and phriir-lmethyls~~lfo~~~lfluoride inhibited est,erase a&vi@. These results suggested that the active site of the enzyme employ;; the serine-hi&dine tyl)e of catalytic mechanism as do trypsin, chymotrypsin, and many other entloproteases (14). However, unlike trypsin, the enzyme was not inhibited by tosyllysine chloromethyl ketone which specifically inhibits trypsin by reacting with its active site his- tidiue (15). However, failure to inhibit enzyme act,ivity with histidine-binding reagents does not necessarily rule out the in- volvement of histidine at the active site, since elastase, another protease that is known to have a serine-histidine active site, is not inhibited by chloromethyl ketone derivatives of the elast,ase substrates (16). Table IV shows that a number of other pro- teolytic inhibitors did not affect enzyme activity. In particular, the enzyme was not inhibited by N-ethylmaleimide which rapidly inhibits plant sulfhydryl proteases such as papain and ficin (4), nor was it inhibited by dialysis against EDTA or l,lO-phenan- throline. The latter observation suggests that no metal ion is required for activity, which is in contrast to CarboxTpeptidase tissues, both of which require Zn*++ at the active site (2, 3). It should be noted that the esterase activities against BAEE, ATEE, and Cbz-L-alanine p-nitrophenyl est’er are all inhibited by diisopropylfluorophosphate at the same rate, further t,hat a single enzyme is responsible for these The effects of some of the inhibitors presented in Table IV on carboxypeptidase activity were also determined. When the ac- of the cotton enzyme against BAEE was completely in- with diisopropylfluorophosphate, the carboxypeptidase against glucagon was also completely inhibited. Con- Downloaded from http://www.jbc.org/ by guest on June 17, 2015A and B of animal indicating activities. tivity hibited activity 5043 - -- A -c--T?- B L‘ 6 A 25 I Ni?&ION 75 100 ‘I 10 20 s~&o~” 50 60 “1. FIG. 2 (left). Paper electrophoresis of peptides from 32P- Aliquot samples (2.5 ~1) were removed with time for measurements labeled, acid-hydrolyzed carboxypeptidase. Enzyme activity of radioact,ivity incorporated into the enzyme as described under was completely inhibited with [32P]diisopropylfluorophosphate, “Methods” and 10-J aliquot samples were removed wit,h time to excess [32P]dii sopropylfluorophosphate removed by dialysis, and determine residual esterase activity, measured by the L~AEE as- the enzyme lyophilized. The lyophilized enzyme was hydrolyzed say. Plotted are the moles of [l%]DIP incorporated per mole of in 1.0 ml of 6 x HCl for 1 hour at 110” and dried in a rotary evapo- enzyme ?IBTSUS the percentage of esterase activity inhibited. rator. The dried hydrolysate was resuspended in pyridine-glacial FIG. 4 (right). Sodium dodecyl sulfate-polyacrylnmide gel ncet,ic acid-water (v/v, 1:10:289), pH 3.6, and subjected to elec- electrophoresis of [14C]diisopropylfluorophosphate inhibited car- trophorcsis on Whatman No. 3MM filter paper (20 X 60 cm) at boxypeptidase. Enzyme activity was inhibited with [‘4C]diiso- pH 3.6, for FO min at 2000 volts. Aliquots of 0.1 pmole of phospho- propylfluorophosphate, the enzyme dialyzed to remove excess threonine (A) and phosphoserine (B) were run as references. [‘%]diisopropylfluorophosphate, lyophilized, and redissolved in Phosphoserine and phosphothreonine were detected with ninhy- 0.01 hr sodium phosphate buffer, pH 7.0, containing lyb sodium drin and radioactivity was visualized with x-ray film. dodecyl sulfate and 1% mercaptoethanol. Enzyme (10 fig) was FIG. 3 (center). The stoichiometry of the inhibition of enzyme subject,ed to electrophoresis on a loo/ polyacrylamide gel for 5 by incubated diisopropylfluorophosphate. with 5.5 X Enzyme (3.2 X 10m3 pmoles) was hours at 8 m per tube. The gel was sltbseqllently sectioned and 1OV pmoles the radioactivity profile determined as described under “Meth- phate in 0.5 ml of 0.1 M sodium phosphate of [‘4C]diisopropylfluorophos- buffer, pH 6.6, at 2.5”. ods.” Tersely, dialysis of the enzyme against 1, IO-phenanthroline, reacted with [14C]diisopropylfluorophosphate, dissociated into which did not inhibit esterase activity, did not inhibit carboxy- separate chains by incubation with sodium dodecyl sulfate and peptidsse activity against glucagon. mercaptoethanol (I) and the chains were subjected to electro- Formation of diisopropylphosphoserine-To confirm the pres- phoresis on a polyacrylamide gel. The gel was subsequently ence of serine at the active site, we allowed the enzyme to react sectioned and the location of radioactivity on the gel determined with [Vjdiisopropylfluorophosphate, performed a limited acid as shown in Fig. 4. The two peaks of radioactivity correspond hydrolysis of the product, and subjected the hydrolysate to to the 33,000 and 31,000 molecular weight chains of t,he enzyme electrophoresis with phosphoserine and phosphothreonine as (I), indicating the formation of DIP-serine on each of the two reference standards. Fig. 2 is a diagram of the resulting elec- large chains. The small chain (24,000 molecular weight) on tropherogram, and it shows that a radioactive component was the other hand did not contain any radioactivity. present in the hydrolysis products that reacted to electrophoresis We further compared the two active sites by digesting the identically with phosphoserine. This indicated the formation of [l*C]diisopropylfluorophosphate-labeled, performic acid-oxidized covalently linked diisopropylphosphoserine during the inhibition enzyme with specific endoproteases. The peptides formed by of enzyme activity by diisopropylfluorophosphate which was the enzymatic hydrolysis were separated by electrophoresis and hydrolyzed to phosphoserine during the acid hydrolysis. the number of radioactive peptides (containing [14C]DIP-serine) Evidence for Two Active Sites-The existence of two active was determined. When the radioactive enzyme was hydrolyzed sites per molecule of native enzyme was first indicated by experi- with trypsin and the hydrolysate subjected to electrophoresis at ments with radioactive diisopropylfluorophosphate. The en- pH 2.0, 3.6, or 6.7, all radioactivity remained at the origin. zyme was incubated with [l*C]diisopropylfluorophosphate, ali- This fact presumably indicates the large size of the radioactive quot samples were removed with time and assayed for esterase peptide produced by trypsin. In contrast, hydrolysis of the activity, and the amount of radioactivity incorporated into the [~*C]diisopropylfluorophosphate-labeled enzyme with chymo- enzyme as DIP. The disappearance of enzyme activity was trypsin produced a single radioactive peptide that moved as an linear with the formation of radioactive DIP-enzyme, and 2 anion when subjected to electrophoresis at pH 3.6 as shown by moles of DIP bound per mole of enzyme were required to give Fig. 5. The fact that only one radioactive peptide could be complete inhibition of activity as shown in Fig. 3. An enzyme demonstrated by this technique indicated that the DIP-serine molecular weight of 84,500 (1) was assumed in calculating in containing peptides from t’he two chains must have the same or molarity of the enzyme. very similar amino acid sequences. It is not unusual for the The presence of two active sites was further demonstrated by sequence around the serine active site of different, proteolytic an analysis of the incorporation of radioactive diisopropylfluoro- enzymes to be similar (17). To determine the existence of any phosphate into the enzyme polypeptide chains. Enzyme was sequence homology between the active site serine peptide of t’he Downloaded from http://www.jbc.org/ by guest on June 17, 20155046 cotton enzyme and that of bovine trypsin or subtilisin, we com- pared electrophoretically the radioactive peptides produced by + mild acid hydrolysis of these three enzymes after they had been inactivated with [32P]diisopropylfluorophosphate. Fig. 6 shows YJ an autoradiograph of the peptides subjected to electrophoresis, ? and it is apparent that no sequence homology exists between x bovine trypsin and the cotton enzyme, but it is shown that some c, h degree of similarity exists between the radioactive peptides of subtilisin and the cotton enzyme. DISCUSSION Other plant carboxypeptidases appear to have some of the 20 0 properties of the cotton enzyme. Wells (18) has reported the MIGRj?OTION (CM) purification of an enzyme from French bean leaves that hydro- FIG. lyzes benzyloxycarbonyldipeptides including those that have electrophoresis 5. Radioactivity profile obtained by high voltage paper COOH-terminal proline. This apparent carboxypeptidase is not isopropylfluorophosphate-labeled of chymotryptic peptides obtained from [ikC]di- inhibited by EDTA or 1, lo-phenanthroline, which indicates inactivated move excess diisopropylfluorophosphate, with [W]diisopropylfluorophosphate, enzyme. Enzyme (1 mg) was lyophilized, dialyzed and to re- per- that, like the cotton carboxypeptidase, it is not a metallo-car- formic acid-oxidized as described under “Methods.” The ox- boxypeptidase. Visuri et al. (19) have described a carboxypep- idized tidase from germinating barley that also hydrolyzes a wide va- buffer, pH 8.5, and incubated enzyme was dissolved in 1 ml of 0.2 M with 50 pg of chymotrypsin N-ethylmorpholine for 2; riety of benzyloxycarbonyldipeptides, but not those containing hours and lyophilized. COOH-terminal proline. Like the cotton carboxypeptidase, 0.1 ml of pyridine-acetic The hydrolyzed acid buffer, pH 3.6, and aliquot enzyme was dissolved samples in the barley enzyme possesses no dipeptidase activity and is in- were subjected to electrophoresis 3 hours at 2000 volts on What- hibited by diisopropylfluorophosphate. This enzyme has a mo- man No. 3MM obtained with a Packard filter paper radiochromatogram strips. The radioactivity scanner. profile was lecular weight of 90,000 and a pH optimum of 5.2, properties also similar to those of the cotton enzyme. Zuber (20) has purified a carboxypeptidase from citrus fruit that also hydrolyzes benzyloxycarbonyldipeptides that have COOH-terminal proline. This enzyme is also inhibited by diisopropylfluorophosphate. Its pH optimum is approximately 5.5, like that of the cotton enzyme, but its reported molecular weight is 148,500. A fur- ther characterization of the bean, barley, and citrus enzymes with respect to esterase activity, type of catalytic mechanism, subunit structure, and number of active sites may bring to light more properties common to these plant carboxypeptidases. From the properties described for the cotton carboxypeptidase and the other plant carboxypeptidases mentioned above, it appears that they may represent a distinct class of carboxypep- tidase enzymes. They differ from the carboxypeptidases A and B in two basic characteristics: (a) they appear to utilize the “serine protease” mechanism of hydrolysis rather than the metal ion mechanism; and (b) with the possible exception of the barley enzyme, they appear to have little or no restrictions re- garding the nature of the COOH-terminal amino acid. Because of these distinctive properties, we feel that the term, carboxy- peptidase C, originally suggested by Zuber (20), should be used to designate this class of hydrolases that is characterized by a nonspecific carboxypeptidase activity and the “serine protease” mechanism of catalysis as typified by the cotton and citrus en- zymes. A third distinction which would be of interest from a phylogenetic point of view may emerge as more carboxypepti- FIG. 6. Paper electrophoresis of peptides from 32P-labeled, dases from different organisms are characterized. Namely, all acid-hydrolyzed bovine trypsin, subtilisin, and cotton carboxy- the presumptive carboxybeptidase C enzymes have been found peptidase. only in higher plants, whereas carboxypeptidases that utilize the and cotton One milligram carboxypeptidase of bovine (C) was reacted trypsin (A), subtilisin with [32P]diiso- (B), metal ion or sulfhydryl mechanisms of catalysis have been found propylfluorophosphate, pH 6.6, for 3 hours. each in 5 ml of 0.01 M Excess diisopropylfluorophosphate sodium phosphate, was in several animal phyla and in fungi (2, 3, 21). Thus it may be removed by dialysis, the [32P]D1P-enzymes lyophilized, dissolved that carboxypeptidase C enzymes are confined to higher plants. in 1 ml of 6 N drolyzed enzymes were dried HCl, and hydrolyzed on a rotary evaporator, at 110” for 20 min. dissolved The hy- in pyridine-glacial acetic acid-water (v/v, 1:10:289), pH 3.6, and 3MM aliquot samples subjected to electrophoresis on Whatman No. Radioactive filter paper peptides (20 X 60 cm) at pH 3.6 at 2000 volts for 80 min. were visualized with x-ray film. Downloaded from http://www.jbc.org/ by guest on June 17, 20155047 It is curious that these plant carboxypeptidases utilize the “serine protease” mechanism which heretofore has been demonstrated solely for endoproteases (14), and yet the known plant endo- proteases utilize the sulfhydryl mechanism of catalysis (4). The apparent existence of three nonidentical polypeptide chains, two of which contain an active site serine, makes the cotton carboxypeptidase unique among proteolytic enzymes to date. Although the nonidentity of the two large chains requires more corroborative evidence, the evidence for an active site on Two molecules of two of the three chains is more substantial. diisopropylfluorophosphate must react with each enzyme mole- cule to totally inhibit its esterase activity (Fig. 3) and both of some means. This could be achieved by its compartmentaliza- tion in some fashion that allows for aleurone protein degradation alone. On the other hand, its inability to degrade native pro- teins may render it harmless to the other enzymes of this tissue. This possibility is supported by the fact that the carboxypepti- dase does not appear to hydrolyze itself on prolonged storage at 5”. We would like to point out the value of this enzyme in de- termining the amino acid sequence of small peptides which stems from its ability to hydrolyze any amino acid from the COOH- terminal of peptides. Other investigators have used this en- zyme effectively for this purpose with the use of the technique the large chains appear to bind equal amounts of the inhibitor (Fig. 4). It would be of interest to know if the two active sites have distinct catalytic properties. We attempted to dis- tinguish between the activities of the two sites by measuring the rat.es at which they incorporate [14C]diisopropylfluorophos- phate. This was done by the sodium dodecyl sulfate electro- phoresis of the chains. However, both chains incorporated the inhibitor at the same rate. Another similarity between the two active sites is the apparent identity of the amino acid sequence of the peptide from each chain containing the diisopropylfluoro- phosphate-reactive serine that is produced by chymotrypsin digestion (Fig. 5). The sequence of amino acids around the diisopropylfluorophosphate-reactive serine is apparently more closely related to the subtilisin family sequence that has as its core Gly-Thr-Ser-Met than to the trypsin family sequence that has a Gly-Asp-Ser-Gly core (Fig. 6). The functioning of the cotton carboxypeptidase in viva can be surmised from the fact that it appears at the time in the de- velopment of the cotyledon when the stored aleurone protein is degraded to support the growth of the emerging tissues of the seedling. The synthesis of the carboxypeptidase stops as the rate of disappearance of the aleurone protein reaches a peal~,2 and the activity of the enzyme itself decays after this point. Although the sequence of catalytic events that bring about the mobilization of amino acids from aleurone protein is not known, the initial hydrolytic event appears to be an endoproteolytic cleavage of both of the aleurone protein subunit.s that produces half-sized polypeptides.2 Endoproteolytic activity that could account for this observation has been detected in isolated cotton aleurone granules by Yatsu and Jacks (22). The carboxypep- tidase could presumably participate in the hydrolysis of these endopeptidase products. Such an in viva role for the carboxy- peptidase seems feasible in view of its inability to hydrolyze na- tive proteins. Presumably a dipeptidase(s) would be required to complete the hydrolysis. Since the carboxypeptidase exists inside cells, the functioning enzyme protein of the cell must be protected from its action by 2 J. N. Ihle and L. S. Dure III, unpublished observations. we have described in Fig.1 for the hydrolysis of Angiotension II. REFERENCES 1. IEILE, J. N., AND DURE, L. S. (1972) J. Biol. Chem. 247, 5034- 5040 2. HARTSUCK, (Editor), J. A., AND LIPSCOMB, W. N. (1971) in I?. D. BOYER The enzymes, Vol. III, p. 1, Academic Press, New York 3. FOLK, J. E. (1971) in P. D. BOYER (Editor), The enzymes, vol. III, n. 57, Academic Press, New York 4. GLSZ;;, A..N., AND SMITH, k. L. (1971) in P. D. BOYER (Ed- itor), The enzymes, Vol. III, p. 501, Academic Press, New York 5. SCHWERT, G. W., AND TUXNAIU, Y. (1955) Biochem. Biophys. Acta, 16, 570-575 6. JANOFF, A. (1969) Biochem. J. 114, 157-159 7. ANSON, B. L. (1938) J. Gen. Physiol. 22, 79 8. ERLANGER, B. F., KOKOTVSI~Y, N., AXD COHEN, W. (1961) Arch. Biochem. Biophvs. 95, 271 9. FOLK, J. E., AND SCH~RI&R, k. W. (1963) J. Biol. Chem. 238, 3884-3894 10. WOLFF, E. C., SCHIRMER, E. W., AKD FOLK, J. E. (1962) J. Biol. Chem. 237, 3094-3099 11. MATHESON, A. T., AND TATTRIE, B. L. (1964) Can. J. Biochem. 42, 95 12. WEBER, K., AND OSBORN, M. (1969) J. Biol. Chem. 244, 4406- 4412 13. HIRS, C. H. W. (1956) J. Biol. Chem. 219, 611 14. SHOTTON, D. M., AND HARTLEY, B. S. (1970) Nature 226, 80% 806 15. SHAW, E., AND SPRINGHORN, S. (1967) Biochem. Biophys. Res. Commun. 27, 391 16. KAPLAN, H., SYMONDS, V. B., DUG~S, H., AND WHITAKER, D. R. (1970) Can. J. Biochem. 48, 649 17. ~~~~~~~~~ ‘M. O., AND ECK, R. ?. (1968) Atlas of protein se- quence and structure 1967-1968, p. 57, The National Biomedi- 18. cal Research Foundation, Silver Spring, Maryland WELLS, J. R. E. (1965) Biochem. J. 97, 228-235 19. VISURI, K., MIICOLA, J., AND ENARI, T. M. (1969) Eur. J. Biochem. 7, 193 20. Zun~n, V. H. (1968) Hoppe-Seyler’s 2. Physiol. Chem. 349, 1337 21. MATSUBARB, H., AND FEDER, J. (1971) in P. D. BOYER (Ed- itor), The enzymes, Vol. III, p. 721, Academic Press, New York 22. Y~TSU, L., AND JACICS, J. (1968) Arch. Biochem. Biophys. 124, 446 Downloaded from http://www.jbc.org/ by guest on June 17, 2015ARTICLE:
The Developmental Biochemistry ofCottonseed Embryogenesis and Germination: II. CATALYTICPROPERTIES OF THE COTTON CARBOXYPEPTIDASE
James N. Ihle and Leon S. Dure III
J. Biol. Chem. 1972, 247:5041-5047.
Access the most updated version of this article at http://www.jbc.org/content/247/16/5041
Find articles, minireviews, Reflections and Classics on similar topics on the JBC Affinity Sites.
Alerts: • When this article is cited
• When a correction for this article is posted
Click here to choose from all of JBC's e-mail alerts
This article cites 0 references, 0 of which can be accessed free athttp://www.jbc.org/content/247/16/5041.full.html#ref-list-1
Downloaded from http://www.jbc.org/ by guest on June 17, 2015