THE JOURNAL OF BIOLOGICAL CHEMISTRY 0 1985 by The American Society of Biological Chemists, Inc

Vol. 260, No. 14, Issue of July 15, pp. 8320-8324, 1985 Printed in U.S.A.

Amino Acid Sequence of the Active Site Peptide of Bovine Intestinal 5’-Nucleotide Phosphodiesterase and Identification of the Active Site Residue as Threonine*

(Received for publication, July 26, 1984)

Jeffrey S. Culp, Harold J. Blytt, Mark Hermodson, and Larry G . Butler From the Department of Biochemistry, Purdue University, West Lafayette, Indium 47907

We report here the identification of the amino acid residue which forms the covalent intermediate in the catalytic mechanism of bovine intestinal 5”nucleotide phosphodiesterase and the sequence of the neighboring amino acids. The active site of 5‘-nucleotide phospho- diesterase was labeled using thymidine 5’-[a-s2P]tri- phosphate as substrate. A single labeled cyanogen bro- mide peptide was isolated using reversed-phase high performance liquid chromatography. After subdiges- tion with endoproteinase Lys-C and chymotrypsin, the entire amino acid sequence of the 60-residue active site peptide was obtained using automated Edman degra- dation. All of the radioactivity of the active site peptide was localized to a hexapeptide with sequence Thr-Phe- Pro-Asn-His-Tyr. Phosphoamino acid analysis of this peptide indicated that the labeled residue was threo- nine. We are not aware of any other enzymes in which threonine is phosphorylated as a covalent intermediate in the catalytic mechanism.

The catalytic mechanism of 5’-nucleotide phosphodiester- ase (EC 3.1.4.1) from bovine intestine includes formation of a covalent phosphoryl enzyme intermediate (Kelly and Butler, 1977). By utilizing radioisotope-labeled substrate such as [“C] CAMP (Landt and Butler, 1978) or [3H]dTTP1 and quenching the reaction with phenol or with acid, a stable covalent linkage between denatured enzyme and a substituted phosphoryl group of the substrate has been demonstrated. Incorporation of radioactivity from substrate into the enzyme is inhibited by competitive inhibitors and enhanced by the uncompetitive inhibitor imidazole (Landt and Butler, 1978). The radioiso- tope-labeled enzyme-bound intermediate is stable in mild acid and at neutral pH. It is labile in alkali and is also cleaved from the protease-digested protein by active phosphodiester- ase (Landt and Butler, 1978). These properties suggest that the intermediate is a hydroxylic amino acid in phosphodiester linkage with the substituted phosphoryl group of the sub- strate.

Snake venom phosphodiesterase, which has catalytic and structural properties (Pollack et al., 1983; Dolapchiev et al., 1980) similar to the bovine intestinal enzyme (Kelly et al., 1975), has been similarly shown to be covalently labeled by

* This work was supported in part by National Institutes of Health Grant R01 GM-23028. This is Journal Paper 9962 from the Purdue University Agricultural Experiment Station. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Blytt, H. J., Brotherton, J. E., and Butler, L. G. (1985) Anal. Biochem., in press.

radioisotope-labeled substrate (Rugevics and Witzel, 1982). These authors suggested that the phosphorylated amino acid of snake venom phosphodiesterase is histidine.

We have labeled bovine intestinal phosphodiesterase using [32P]dTTP and isolated and sequenced the radioactive cyan- ogen bromide active site peptide. The active site amino acid residue which forms the covalent intermediate is threonine.

MATERIALS AND METHODS

Labeling of the Phosphodiesterase Active Site-The active site of phosphodiesterase was labeled by reaction with thymidine 5’-[a-32P] triphosphate ([3’P]dTTP); the conditions were somewhat different from the procedure of Landt and Butler (1978). The reaction mixture (1.0 ml) contained 0.65 M imidazole formate (pH 4.0), 10 mM ZnCI2, 3 m y [32P]dTTP (2 mCi, Amersham Corp.), and (added last) 9 mg of bovine intestinal phosphodiesterase purified to homogeneity by the method of Landt and Butler (1978). Nonradioactive dTTP carrier (Sigma) was purified on HPLC.’.’ Assuming one active site/enzyme with M, = 107,000 (Kellyet al., 1975), the mixture originally contained 35 molecules of dTTP/enzyme active site.

After incubation for 10 s at room temperature, 88% formic acid was added to a final concentration of 12%. Four samples of phospho- diesterase were subjected to the labeling procedure, and the combined mixtures were applied to a Sephadex G-50 column (1.5 X 29 cm) equilibrated with 9% formic acid in order to separate labeled protein from residual dTTP and product dTMP. The labeled protein was lyophilized before reduction, alkylation, and digestion with CNBr.

Reduction, Alkylation, and Cyanogen Bromide Cleavage-Lyophi- lized labeled enzyme was dissolved in a minimum volume of 0.13 M Tris-HC1 (pH 7.6), containing 6 M guanidine HCl and 1 mg/ml EDTA, and was reduced and alkylated by a modification of the method of Hermodson et al. (1973). Dithiothreitol (ICN) was added in a 30-fold molar excess over the concentration of protein cysteine residues, and the solution was incubated a minimum of 4 h at room temperature. Redistilled 4-vinylpyridine was added at a ratio of 3 mol of 4- vinylpyridine/mol of dithiothreitol, and the reaction continued for 1.5 h in a sealed container in the dark at room temperature. The protein was separated from the reagents on a Sephadex G-50 column (1.5 X 35 cm) equilibrated with 9% formic acid and lyophilized.

Reduced and alkylated labeled phosphodiesterase was digested with CNBr in a minimum volume of 70% formic acid. One large crystal of CNBr (Kodak) was added, and the solution was kept for 24 h in a glass-stoppered container in the dark at room temperature. The mixture was diluted with distilled H20, lyophilized, dissolved in 9% formic acid, and applied to a Sephadex G-75 column (2.5 X 100 cm) equilibrated with 9% formic acid. The eluant was monitored at 280

and purified by HPLC. nm. Fractions containing labeled peptide were pooled, lyophilized,

Endoproteinase Lys-C Digestion-Peptides were digested using en- doproteinase Lys-C (Boehringer Mannheim) which cleaves polypep- tides primarily on the carboxyl side of lysine residues (Steffens et al., 1982). Endoproteinase Lys-C (2% (w/w)) was added to peptides dissolved in 3.5 M urea, 25 mM Tris-HC1, 1 mM EDTA, pH 7.7, and the mixture was incubated for 4 h at 37 “C. The reaction was stopped by the addition of one-quarter volume of 9% formic acid, and the

The abbreviation used is: HPLC, high performance liquid chro- matography.

8320

Threonine, Active Site Residue of Phosphodiesterase

peptides were purified on reversed-phase HPLC. Chymotrypsin Digestion-Peptides were digested using No-p-tosyl-

L-lysine chloromethyl ketone-chymotrypsin (Sigma) under conditions identical to those described for endoproteinase Lys-C digestion except that peptides were dissolved in 100 mM ammonium bicarbonate, pH 9. The peptides were purified on reversed-phase HPLC.

Clostripain Digest-Peptides were digested by a modification of the method of Mitchell (1977). Clostripain (239 units/mg of protein, Sigma) was added (1% (w/w)) to peptides dissolved in 1.7 ml of 0.6 M sodium phosphate, pH 7.7, 3.2 M urea and the digestion proceeded at 25 'C for 5 h. Then an additional aliquot of clostripain (1% (w/w)) was added, and the reaction continued for 4 h. Peptides were purified on reversed-phase HPLC.

Reversed-phase HPLC-Peptides were separated and purified us- ing a Varian 5000 Liquid Chromatograph equipped with a Varian UV-50 detector on a Brownlee Aquapore RP-300 10-rm C-18 column (4.6 mm, inner diameter, X 250 mm) using 0.1% trifluoroacetic acid (Pierce Chemical Co.) as the mobile phase and 0.1% trifluoroacetic acid in 2-propanol (Mallinckrodt Chemical Works) as the organic modifier (Mahoney and Hermodson, 1980; Pearson et al., 1981). All solvents were HPLC-grade.

Amino Acid Analysis and Sequence Determinution-Peptides were hydrolyzed in 6 N HCl in acid-washed, evacuated, and sealed tubes for 24 h at 110 "C. Acid was removed under a stream of nitrogen. All analyses were performed on a Durrum D-500 amino acid analyzer in the laboratory of Dr. Michael Laskowski, Jr., Chemistry Department, Purdue University. The amino acid sequences of peptides were deter- mined using automated Edman degradation (Mahoney et al., 1981).

Phosphoamino Acid Analysis-Peptides and phosphoamino acid standards (Sigma) were hydrolyzed in 5.7 N HCI under a vacuum at 110 "C for 2 h. After concentration by lyophilization, samples were spotted on K2-cellulose plates (20 X 20 cm, 250-pm layer thickness, Whatman) and analyzed using one-dimensional electrophoresis (Manai and Cozzone, 1982). Reference compounds were detected with ninhydrin. Radioactivity was detected by exposing the plate to Kodak X-Omat AR film for 65 h followed by film development according to the manufacturer's recommendations.

Quantitation of Radioactivity and Protein-Cerenkov radiation (Dyer, 1980) was used to quantitate 32P radioactivity. This procedure allowed for 100% sample recovery because no scintillation fluid is used. Counting efficiency was approximately 14%. Protein concentra- tion was determined from absorbance at 280 nm and the mass extinction coefficient of 0.88 A mg-I ml" for phosphodiesterase.'

RESULTS

Labeling and Sequence Determination of the Phosphodies- terase Active Site-A kinetically competent, covalent inter- mediate of bovine intestinal 5'-nucleotide phosphodiesterase was isolated by a modification of a previously reported pro- cedure (Landt and Butler, 1978). Under the conditions em- ployed for labeling with [32P]dTTP, 23% (82 nmol) of the 350 nmol of phosphodiesterase were labeled based on the specific activity of the [32P]dTTP.

Cyanogen bromide peptides from the labeled enzyme were separated on Sephadex G-75 (Fig. 1). The fourth peak de- tected at 280 nm contained the labeled active site peptide, as was found in several previous preparations using ['*C]cAMP as substrate. This peptide contained 78% of the trichloroacetic acid-precipitable radioactivity and 99% of the phenol-extract- able radioactivity in the labeling mixture. The remaining radioactive peak corresponds to unhydrolyzed [32P]dTTP as identified by reversed-phase HPLC.'

The peak from Sephadex G-75 containing the active site peptide was pooled, lyophilized, and chromatographed on HPLC (Fig. 2 A ) . The second of the two peaks observed by monitoring at 230 nm contained all the radioactivity. After lyophilization of this peptide, automated Edman degradation identified the majority of the first 38 amino acid residues (Fig. 3).

A separate portion of the CNBr active site peptide (Fig. 2 A ) was digested using endoproteinase Lys-C (Fig. 2B). The amino acid composition of the HPLC-purified peptides (Table

1.2 -

1.0 - - I 0.8 - - N 8 a 0.6 -

0.4 -

I i h l

I n q 3

832 1

'20 40 60 80 1 0 0 120 140 0

Fraction Number

FIG. 1. Elution profile of radioactive phosphodiesterase CNBr peptides chromatographed on Sephadex G-75. Lyophi- lized radioactive phosphodiesterase CNBr digestion products were dissolved in 9% formic acid and applied to a Sephadex G-75 column (2.5 X 100 cm) equilibrated with 9% formic acid. The flow rate was 0.3 ml/min. Fractions (5 ml) were collected directly into scintillation vials and analyzed for Cerenkov radiation (- - -) and absorbance at 280 nm (-). Fractions 78-88 were pooled and lyophilized.

I I I I

B

4

ELUTION TIME (rnin) FIG. 2. HPLC of radioactive phosphodiesterase CNBr pep-

tide and endoproteinase Lys-C digest of radioactive phospho- diesterase CNBr peptide. Fractions were collected directly into scintillation vials and analyzed for Cerenkov radiation. Peaks con- taining the majority of the radioactivity were pooled, lyophilized, and subjected to automated Edman degradation. A, pooled peak 4 from Sephadex G-75 (Fig. 1) containing radioactive CNBr peptide; B, endoproteinase Lys-C digest of radioactive CNBr peptide (A,peak 2). - , absorbance at 230 nm; - - -, per cent 2-propanol.

8322 Threonine, Active Site Residue of Phosphodiesterase

45 50 55 60 P-N-H-Y~-I-v-T-G-L-Y-P-E-s-H-G-I-I-D-N-(M)

L y s I1

Chy I 1 chy I1 I FIG. 3. Phosphodiesterase active site peptide sequence and

peptide map. Carboxyl-terminal Met was detected as homoserine. The residue prior to Asp 1 is presumed to be Met from the specificity of the CNBr digest. Since Ala is the amino-terminal residue of the protein (J. S. Culp, H. J. Blytt, M. Hermodson, and L. G. Butler, unpublished results), this must be an internal cyanogen bromide fragment. Parentheses in peptide map indicate tentative identifica- tions in the sequencing run which were confirmed by amino acid analysis or a second sequenator analysis. Arrows indicate peptide has been sequenced. Lys IV, Chy I, and Chy I1 were subjected to amino acid analyses only. *, identified as the active site residue.

TABLE I Amino acid composition of peptides

Peptides were hydrolyzed in sealed, evacuated tubes in 6 N HCl at 110 "C for 24 h. Numbers in parentheses are values obtained from the sequence. Glx values are elevated in peptides containing homo- serine. Methionine was detected by the presence of homoserine lac- tone.

Amino acid

Asx Thr Ser Glx Pro

Ala Val Ile Leu TY r Phe His LY s Arg Met CYS Tn,

G ~ Y

CNBr

7.4 (6) 6.7 (7) 3.0 (3) 5.1 (4) 5.5 (5) 4.2 (4) 2.3 (2) 3.1 (4) 3.0 (4) 5.5 (5) 5.4 (5) 2.1 (2) 2.8 (2) 3.0 (3) 2.3 (2) -a (1)

NDb (1) - (1)

Lys IV Lys I

2.7 (3) 2.0 (2) 1.6 (2) 1.3 (1) 1.1 (1) 2.3 (2) 1.1 (1) 1.2 (1) 2.2 (2) 1.3 (1) 1.1 (1) 1.1 (1) 1.0 (1) 1.0 (1) 0.9 (1) 1.0 (1) 1.9 (2) 1.0 (1) 0.8 (1) 1.5 (2) 0.9 (1)

1.1 (1) 1.0 (1) 1.0 (1) 0.9 (1)

ND (1) ND - (1)

Lys I1

3.7 (3) 2.3 (3) 1.3 (1) 1.6 (1) 1.7 (2) 2.4 (2)

0.9 (1) 1.8 (3) 1.1 (1) 1.5 (2) 0.8 (1) 1.6 (2)

- (1)

ND

ChyI Chy I1

1.3 (1) 3.1 (2) 0.7 (1) 2.0 (2)

1.1 (1) 1.4 (1)

1.0 (1) 1.1 (1) 2.1 (2)

0.9 (1) 2.2 (3) 1.0 (1)

0.9 (1) 0.9 (1) 1.0 (1) 1.0 (1) 1.0 (1)

ND ND ~~ ~

-, identified at low levels. ND, not determined.

I) indicated that the first peak (Lys I) corresponded to resi- dues 23-38, peak 2 (Lys 11) contained all the radioactivity and was the carboxyl-terminal fragment, and peaks 3 and 4 were two forms of the same fragment (Lys IV) containing residues 1-20 (Fig. 3). (The numbering of phosphodiesterase CNBr peptide began at the amino-terminal residue of the peptide. The position of this peptide in the overall enzyme sequence is unknown.) The dipeptide containing residues 21-22 was

not recovered by chromatography. The amino acid sequence of the phosphodiesterase CNBr

active site peptide was completed by automated Edman deg- radation of Lys I and Lys I1 (Fig. 3). Leu-14 was identified from the amino acid composition of Lys IV. The overlap at positions 38 and 39 was obtained by degradation of a mixture of peptides from a clostripain digest of the CNBr fragment containing residues 21-61, 33-61, and 1-20 in a ratio of 3:1:1, respectively (Fig. 4, peak 9). The overlap at 38-39 was ob- tained on both the 21-61 and 33-61 fragments, and the residues observed in all cycles were consistent with those obtained on the pure fragments from which the sequence was derived.

The active site residue was not identified using the sequen- ator because the [32P]dTMP-amino acid derivative was too hydrophilic to be released from the spinning cup during sequencing. A portion of the [32P]dTMP Lys-C peptide (Fig. 2B, peak 2) was further digested with chymotrypsin, and the three peptides from the digest were purified by HPLC (Fig. 5). The first peptide to elute consisted of the 6 amino-terminal residues and contained the radioactive label (Fig. 3). The second peptide, which was not radioactive, contained the remaining 17 residues, and the third peptide was undigested material.

Because the radioactive amino acid could not be identified using the sequenator, an alternative method was utilized. The bond between the nucleotide and the active site residue was known to be acid-stable and alkali-labile, and the nucleotide moiety could be removed by active phosphodiesterase (Landt and Butler, 1978). This strongly suggested that the covalent intermediate consists of a nucleotidyl moiety linked through a phosphodiester bond to serine, threonine, or tyrosine (Landt and Butler, 1978). The amino acid containing the label had to be either threonine or tyrosine because serine is not present

0.8 -

0.6 - - I N N

dN 0.4 -

0.2 -

/

Elution Time ( m i n ) FIG. 4. HPLC of clostripain digest of radioactive phospho-

diesterase CNBr peptide. Peaks were analyzed as described for Fig. 2. -, absorbance at 222 nm; - - -, per cent 2-propanol.

Threonine, Active Site Residue of Phosphodiesterase

A Pi ”

8323

8

0 us 0 4 8 12 16 20 0

Elution Time (min) FIG. 5 . HPLC of chymotrypsin digest of endoproteinase

Lys-C radioactive peptide. Endoproteinase Lys-C radioactive pep- tide isolated on HPLC (Fig. 2 H ) was t.reated with N”-p-tosyl-L-lysine chloromethyl ketone-chymotrypsin and chromatographed on HPLC as descrihed under “Materials and Methods.” Fractions were collected and analyzed as before (Fig. 2) . Peaks 1 and 3 contained radioactivity and were subjected to amino acid analysis and phosphoamino analy- sis. -, absorbance at 222 nm; - - -, per cent ?-propanol.

in the first chymotrypsin peptide. Phosphoserine, phospho- threonine, and phosphotyrosine can be identified electropho- retically after acid hydrolysis of phosphoproteins (Manai and Cozzone, 1982). Under the very strong acid conditions used for amino acid hydrolysis, the nucleotidyl amino acid deriva- tive of phosphodiesterase was hydrolyzed to phosphoamino acid.

Acid-hydrolyzed chymotrypsin peptide 1 (Fig. 5 and Table I) was analyzed for [“’P]phosphothreonine or [“’Plphospho- tyrosine as described under “Materials and Methods.” Follow- ing electrophoresis at pH 3.5, three well-resolved spots cor- responding to the phosphoamino acid standards (Fig. 6A) were produced by ninhydrin staining. The radioactivity of the acid-hydrolyzed peptide co-migrated with phosphothreonine, strongly suggesting that threonine (residue 39 in the CNBr peptide) is the active site residue of phosphodiesterase. The same result was obtained with chymotrypsin peptide 3.

Following electrophoresis at pH 1.9 (Fig. 6B) , the three phosphoamino acid standards migrated at two distinct spots. One spot was identified as phosphoserine, and the other spot contained co-migrating phosphothreonine and phosphotyro- sine. One radioactive compound migrated in the position of phosphothreonine, confirming that threonine is the active site residue.

Other radioactive compounds were detected using autora- diography. The compound with the greatest mobility detected after electrophoresis at pH 1.9 is “‘Pi as reported previously (Manai and Cozzone, 1982). The radioactive compounds mi- grating slower than phosphotyrosine may be partially hydro- lyzed ‘“P-peptides because they migrate slower than the phos- phoamino acid standards (Werth and Pastan, 1984). Their radioactivity is stable to phosphodiesterase treatment and labile to alkaline phosphatase treatment,3 so they are phos- phomonoester products of acid hydrolysis.

Partial Hydro1

Origin.

Partially Hydrolyzed

Origin- FIG. 6. Autoradiography of the electrophoretogram of hy-

drolyzed radioactive phosphodiesterase peptide. Chymotrypsin radioactive phosphodiesterase peptide (Fig. 5 , peak I ) was analyzed for phosphoamino acids as descrihed under “Materials and Methods.” Phosphoamino acids were detected using ninhydrin (standards mi- grated as indicated). Radioactivity was detected by autoradiography. A, electrophoretic separation at pH 3.5; H, electrophoretic separation at pH 1.9.

DISCUSSION

Radioactivity was incorporated into the active site of phos- phodiesterase through the formation of a [“‘PjdTMP-enzyme kinetically competent covalent intermediate. The purified CNBr peptide contained 78% of the acid-precipitable radio- activity and 99% of the phenol-extractable radioactivity in the labeling mixture. Incorporation was inhibited by compet- itive inhibitors of phosphodiesterase and by addition of acid to the incubation mixture before addition of active enzyme.’ Active phosphodiesterase removed the incorporated radioac- tivity from the labeled chymotrypsin peptides, and the label was stable at pH 1.0 (Culp, 1983). These properties are identical to the int,ermediate reported previously for phospho- diesterase (Landt and Butler, 1978), strongly suggesting that the int,ermediates are hydroxylic amino acid-[””PIdTMP in- termediates with phosphodiester linkages.

Threonine was identified as the hydroxylic amino acid that forms the covalent intermediate of bovine intestinal phospho- diesterase. Threonine was also identified as the active site residue of 5’-nucleotide phosphodiesterase from two different snake venoms that were labeled and analyzed using proce- dures identical to those described here.:’ Although threonine has been suggested as a metal ligand in alkaline phosphatase (Wyckoff et al., 1983) and phosphothreonine has been previ- ously reported as a regulatory site in ATP-citrate lyase (Pucci et al., 1983), we are not aware of any previous reports of an active site threonine forming a covalent intermediate for any enzyme.

The chemical reactivities of serine and threonine are quite similar (Means and Feeny, 1971), but the extra methyl group of threonine may have a significant steric effect. The phos- phorylated serine intermediate of alkaline phosphatase, a phosphohydrolase very similar catalytically to phosphodies- terase (Culp, 1983), readily transfers its phosphoryl group to :’.J. S. Culp and L. G. Butler, manuscript in preparation.

8324 Threonine, Active Site Residue of Phosphodiesterase

acceptors other than water (Morton, 1965). Attempts to dem- onstrate transfer of the nucleotidyl group of phosphodiester- ase to an acceptor other than water were not successful (Kelly, 1974). The methyl group of threonine, as well as the bulky nucleoside group esterified to the phosphorus atom, may play a very important role in limiting the access of nucleotidyl acceptors to the covalent intermediate of phosphodiesterase.

Acknowledgments-We wish to thank Dr. Robert L. Geahlen and Dr. Marietta L. Harrison for their assistance with the phosphoamino acid analyses.

REFERENCES

Culp, J. S. (1983) Ph.D. thesis, Purdue University Dolapchiev, L. B., Vassileva, R. A., and Koumanov, K. S. (1980)

Dyer, A. (1980) Liquid Scintillation Counting Practice, pp. 77-79,

Hermodson, M. A., Ericsson, L. H., Neurath, H., and Walsh, K. A.

Kelly, S. J. (1974) Ph.D. thesis, Purdue University Kelly, S. J., and Butler, L. G. (1977) Biochemistry 16 , 1102-1104

Biochim. Biophys. Acta 622,331-336

Heyden & Sons, London

(1973) Biochemistry 12 , 3146-3153

Kelly, S. J., Dardinger, D. E., and Butler, L. G. (1975) Biochemistry

Landt, M., and Butler, L. G. (1978) Biochemistry 17, 4130-4135 Mahoney, W. C., and Hermodson, M. A. (1980) J. Biol. Chem. 255 ,

Mahoney, W. C., Hogg, R. W., and Hermodson, M. A. (1981) J. Biol.

Manai, M., and Cozzone, A. J. (1982) Anal. Biochem. 124 , 12-18 Means, G. E., and Feeney, R. E. (1971) Chemical Modification of

Mitchell, W. M. (1977) Methods Enzymol. 47, 165-170 Morton, R. K. (1965) Comp. Biochem. 16,55-84 Pearson, J. D., Mahoney, W. C., Hermodson, M. A., and Regnier, F.

Pucci, D. L., Ramakrishna, S., and Benjamin, W. B. (1983) J. Biol.

Pollack, S. E., Uchida, T., and Auld, D. S. (1983) J. Protein Chem.

Rugevics, C . U., and Witzel, H. (1982) Biochem. Biophys. Res. Com-

Steffens, G. J., Gunzler, W. A., Otting, F., Frankus, E., and Flohb, L.

Werth, D. K., and Pastan, I. (1984) J. Biol. Chem. 259,5264-5270 Wyckoff, H. W., Handschumacher, Murthy, K., and Sowadski, J. M.

14,4983-4988

11199-11203

Chem. 256,4350-4356

Proteins, p. 57, Holden-Day, Inc., San Francisco

E. (1981) J. Chronatogr. 207,325-332

Chem. 258,12907-12911

2 , l - 1 2

mun. 106,72-78

(1982) Hoppe-Seyler’s 2. Physiol. Chem. 363 , 1043-1058

(1983) Adu. Enzymol. 55, 453-480