The Amino Acid Sequence of A5-3-Ketosteroid Isomerase of Pseudomonas testosteroni*

SUMMARY The primary structure of the identical subunits of crystalline A”-3-ketosteroid isomerase (EC 5.3.3.1) of Pseudomonas tesfosferoni has been determined by standard proce-dures to be the following:

The enzyme is of special interest from a mechanistic viewpoint (1). It has an extraordinarily high molecular turnover number of 17.6 X lo6 min-1 at' 25" and pH 7.0, with A5-androstene-3,17-dione as subst)rat,e, and is therefore probably the most active catalyst known.
The isomerase promotes a direct intramolecular cis cis, diaxial proton transfer from C-4/I to C-6& Ult,raviolet absorption and fluorescence spectroscopy, as well as isotope exchange experiments, favor the participation of an enolic intermediate in the enzymatic reaction (l-3). The * This study was supported by Grants AM 07322 and GM 1183 from the National Institutes of Health and by the Gustavus and Louise Pfeiffer Foundation of New York. I II oligomeric protein has a molecular weight of 40,800 and direct spectral and fluorescence-quenching studies indicate that there are three binding sites for steroidal inhibitors (3,4). This work completes the earlier report on the composition of some of the tryptic pept'ides (5).
Three times crystallized thermolysin was supplied by Calbiochem.
The analytical grade anion exchange resin, Dowex l-X2, was a product of Bio-Rat1 Corporation, Richmond, California.
Chromobeads A, :I specid form of Dorves 50, of fine and carefully graded partirle size, was obtained from Technicon, Ardsley, New York.
Hydrazine sulfate (J. T. Baker Chemical Company, Phillipsburg, New Jersey) and 95% hydrazinc (Eastman) were used for hydrazinolysis reactions. l-Dimethylaminonaphthalene-5-sulfonyl chloride was supplied by Pierce. Phenylisothiocyanate (Eastjman) was distilled before use. Pyridine (Fisher, infrared spectranalyzed grade) was distilled from ninhydrin. 25,1971 A. M. Benson, R. Jarabalc, and P. Talalay Amino Acid Analyses-Samples were hydrolyzed in vacua in glass-distilled constant boiling HCl at 105", usually for 20 hours. Analyses were performed on the Spinco model 12OC amino acid analyzer by an accelerst(ed version of the methods of Spackman, Stein, and Moore (7). Norleucine standard (Technicon) was added to the amino acid calibration mixture (Beckma,n). Paper Chromatography and Electrophoresis-Partition chromatography and high voltage elcct,rophoresis on either Whatman No. 1 or 3MM filter paper were used for examination of the purit,y of peptides and for peptide purification.
In analytical paper chromatography and electrophoresis runs the peptides were located by staining the entire paper with ninhydrin.
In preparative purification procedures, side bands were cut out and stained and the peptides were eluted from the unstained portion of the paper, except as indicat'ed.
Chymotryptic Digestion of Isomerase-CrysDalline isomerase (39.6 mg), having a specific activity of 54,600 unit#s per mg, was dissolved in 0.03 M potassium phosphate buffer, pII 7.0, dialyzed against distilled water, heated with stirring at 100" for 10 min, and then cooled in an ice bath.
To the resulting 11.7 ml of milky suspension were added 1.3 ml of 10% NHJIC03.
Initially and after 3: hours, 1.25 y0 (by weight') portions of chymotrypsin (dissolved in 0.001 M I-ICI) were added. The digest was magnetically stirred and was maintained at 37" for 44 hours, with the addition of 5 ml of 17; NH4HC03 at 3 hours. The digestion mixture was lyophilizcd rcpcatcdly from Hz0 to remove NHJIIC& Chromatographic Separation oj Chymotryptic Pep&es-The lyophilized chymotryptic digest was suspended in 0.3 M pyridine hydrochloride, pH 2.2, and centrifuged to remove insoluble material which was purified separately (see below).
The supernatant, fraction was subjected to chromatography in four portions on a column (0.9 x 29 cm) of the Beckman Spinco model 120C amino acid analyzer, which was packed t,o a height of 20 cm with Chromobeads A, maintained at 54.5", and equilibrated with starting buffer. The column was developed at a flow rate of 90 ml per hour with 495 ml of a 530.ml linear gradient from 0.2 M pyridine acetate, pIl 3.1, to 2.0 $1 pyridine acetate, pH 5.0, followed by 150 ml of the latter buffer. The efnuent stream was split, wit,h S.Sg, of the effluent automabicallp analyzed by t,he ninhydrin method and the remainder collected in 5.6-ml fractions.
Identical elution patterns of ninhydrin-positive materials were obtained in the four chrornntographies (Fig. I).

Further
Puri$cation of Chymotryptic Peptides--Fractions corresponding to peaks on the recorder rhart of the split stream column chromatography were examined for purity by partition chrolnntography on Whatman Ko. 3lKLI paper in Solvent 1. Those fractions which were foul~tl t,o contain multiple ninhydrinpositive components were purified by paper chromatography in the sanle solvent syst'em. The 24 chymotryptic peptides that were isolated are designated as C-l to C-24 according to the order of their occurrence in the final sequence.
The portion of t,he chyrnotryptic digest which was insoluble in 0. 3  1. Elution pattern of peptides from the chymotryptic digest of As-3-ketosteroid isomerase chromatographed on a column of Chromobeads A (0.9 X 20 cm) with pyridine acetate buffers.
protein and one important chymotryptic peptide (C-6). This residue was dissolved in 1.5 ml of 507, acetic acid and dialyzed against two 12.5-m] portions of the same solvent for a total of 20 hours. The outer fluid from t,hc dialysis was pooled and dried under vacuum.
The residue was dissolved in pyridine-collidineacetic acid buffer (10 ml of pyridine, 10 ml of collidine, and 1.1 ml of acetic a.cid per liter), pH 8.1, (8) and applied to a column (1.0 X 19 cm) of Dowel; AG l-X2 (100 to 200 mesh). The column was developed with 10 ml of each of the following solvents: pyridine-collidine-acetic acid buffer, pH 8.1; 0.2 M acetic acid; 0.4 M acetic acid; 1.0 M acetic acid; and, finally, 50% acetic acid. Fractions of 5 ml were collected, and an aliquot was taken from each fraction for hydrolysis and amino acid analysis.
Enzymatic Ilydrolysis of Peptides-Thermolysin digests were performed at 40" in 0.2 M ammonium acetate buffer, pH 8.2, with an enzyme to substrate ratio of 3 to 10% (by weight). Papain digests, in 0.2 M pyridine acetat'e buffer of pH 5.45 (0.01 M 2-mercaptoethanol), proceeded at 37" with 2 to 576 enzyme for 20 to 90 min. IIydrolysis with pepsin was carried out in 0.01 i+r HCI at 25" for 18 hours, with 5yG enzyme.
Tryptic and chymotryptic hydrolysis of peptides was performed in 0.02 M NH4HC03 at 40". A single elastaae digest was run at 43" in 0.2 M ammonium acetate (pH 8.2) with 10yO (by weight) elastase. All digests were carried out with magnetic stirring, in tubes which were flushed with nitrogen and stoppered.

Sequence
Xtudies-The subtrnct'ive method (9) of Edrnan degradation (10) was used most extensively in determination of amino-terminal residues and internal sequences of the peptides. Carboxyl-terminal residues were det,ermined by hydrazinolysis (11) and by the use of carboxypeptidases A and B. Hydrolysis with carboxypeptidase B (2 to 10GjO) was performed in 0.02 M NH,HCO, al 25" for 2 to 20 hours. Carbosypeptidase A digests were carried out in 0.2 M N-ethylmorpholine acetate, pH 7.6, at 40" for 4 to 24 hours, with enzyme to substrate ratios ranging from 3 to 10%. Carboxypeptidase digests were dried in a stream of nitrogen, dissolved in 0.01 M HCI, and applied directly to the columns of the amino acid analyzer.
In some cases, the use of carbosypeptidase A was preceded by dansylation (12) of the peptide.
In this procedure, residual peptides after carboxy-Primary Structure of A5-S-Ketosteroicl Isomerase Vol. 246, ATo. 24 peptidase digestion are not ninhydrin-positive and therefore Isolation of New Tryptic Pepfides T-6 and T-7 do not interfere with the analysis of the released amino acids.
The tryptic digest and the purification of all but two of the RESULTS AND DISCUSSION tryptic peptides have been previously described (5). These Carboxyl-terminal Residue of Isomerase peptides have been renumbered according to their sequence in the molecule, and will be designated by their new numbers in Hydrazinolysis of the intact protein yielded only alanine as this paper. Table II gives both old and new designations. Two the carboxyl-terminal residue.
The yield was 290%, based on a additional tryptic peptides (T-6 and T-7) have been isolated molecular weight of 40,800 (4).
Electrophoresis, pH 6.5: basic, therefore contains gIutamine. The values for arginine a-ere not determined in Steps 1 through 4, since this residue has been shown by carboxypeptidase treatment to occupy the carboxyl-terminal position of T-2. Hydrolysis of T-2 by chymotrypsin (3vo by weight) for 4 hours at 40°, followed by partition chromatography of the digest in Solvent 1, yielded only one fragment in pure form. 25,1971 A. M. Benson, R. Jarabak, and P. Talalay Electrophoresis, pH 6.5; neutral, therefore contains glutamic acid. Thermolysin (3% by weight) was used to hydrolyze T-2 in a 3+-hour reaction at 40". Ion exchange chromatography on a column of Chromobeads A yielded the elution pattern shown in Fig. 3A. Peptides Th-1 and Th-3 were resolved by gel filtration through a fine grade of Sephadex G-25. Th-2 was subjected to the same procedure followed by partition chromatography in Solvent 2. Electrophoresis, pH 6.5: neutral, therefore contains asparagine. FRACTION NUMBER FIG. 3. Elution patterns of thermolytic digests of tryptic peptides on a column of Chromobeads A (0.9 X 17 cm) with pyridine acetate buffers. Fraction size was 3.5 ml. The relative absorbance at 570 nm was determined after alkaline hydrolysis (14) and ninhydrin assay (13) Arginine, previously shown to be carboxyl-terminal in T-2, was not determined in Step 1. T-2-Th-4 was subjected to further hydrolysis with thermolysin (6.5% by weight) in a 16hour digest at 40". Two fragments were resolved on a column (0.9 x 17 cm) of Chromobeads A in 0.2 M pyridine acetate, pH 3.1.
T-6 was hydrolyzed with chymotrypsin (5oj, by weight) for 40 hours at 40". The products were separated on a column (0.9 x 17 cm) of Chromobeads A, in pyridine acetate buffers. Electrophoresis, pH 6.5: neutral, therefore one Glx residue is glutamic acid and the other is glutamine.
Electrophoresis, pH 6.5, of the residual peptide after Step 2: neutral, therefore the 3rd residue in T-6-C-2 is glutamic acid, and the 2nd residue is glutamine.
Since one step of Edman degradation, followed by direct amino acid analysis of the residue, yielded alanine, T-7-Th-2 must be a dipeptide. Electrophoresis, pH 6.5: basic, therefore contains glutamine. T-7 was hydrolyzed with elastase (lOyo by weight) for 19 hours at 43". The fragments were separated by high voltage eIect,rophoresis at pH 6.5 followed by partition chromatography in Solvent 2, detected by staining with 0.02% ninhydrin in acetone, and eluted with 0.02 M NH~HCOI. These values were obtained after 72 hours of hydrolysis. Evidence relative to the structure of T-S is summarized in Fig. 4. T-8 was hydrolyzed with thermolysin for 18 hours at 40". The products were separated by gradient elution from a column (0.9 x 17 cm) of Chromobeads A under conditions very similar to those described for the elution of the thermolytic fragments of T-7.  Arginine must be at the carboxyl terminus, in accordance with the specificity of trypsin. Step 1: Step 1: 2.0 1.8 1.0 Step 2: 2.0 1.7 0.9 Hydrazinolysis: threonine, 0.5.
Methionine was determined as the sulfone, after performic acid oxidation (15). The tryptic peptides are shown above the sequence and the chymotryptic peptides are shown below.