Structural Studies on Aspartate Aminotransferase from Escherichia coli

THE JOURNAL OF’ BIOLOGICAL CHEMISTRY 0 1987 by The American Society of Biological Chemists, Inc.

Vol. 262, No. 18, Issue of June 25, pp. 8648-8659,1987 Prrnted m U. S. A.

Structural Studies on Aspartate Aminotransferase from Escherichia coli COVALENT STRUCTURE*

(Received for publication, October 17,1986)

Kiyoshi Kondo, Sadao WakabayashiS, and Hiroyuki KagamiyamaO From the Department of Medical Chemistry, Osaka Medical College, Takatsuki-shi, Osaka 569, Japan

The amino acid sequence of aspartate aminotransferase from Escherichia coli was established by sequence analysis and alignment of 39 tryptic peptides and 7 cyanogen bromide peptides. The total number of amino acid residues of the subunit was 396, and the molecular weight was calculated to be 43,573. A comparison of the primary structure of the E. coli enzyme with all known sequences of the two types of isoenzyme (mitochondrial and cytosolic enzymes) in vertebrates revealed that approximately 25% of all residues are invariant. The amino acid residues which were proposed from crystallographic studies on the vertebrate enzymes to be essential for the enzymic action are well conserved in the E. coli enzyme. The E. coli enzyme shows a similar degree of sequence homology to both the mitochondrial and cytosolic isoenzymes (close to 40%). The finding that the positions of deletions introduced into the sequence of E. coli enzyme to give the maximum homology agree well with those of the mitochondrial enzymes supports the endosymbiotic hypothesis of mitochondrial origin.

Aspartate aminotransferase (~-aspartate:2-oxoglutarate aminotransferase, EC 2.6.1.1) is widely distributed in animals, plants, and microorganisms. In vertebrates, two genetically distinct isoenzymes have been characterized (cytosolic aspartate aminotransferase and mitochondrial aspartate aminotransferase) (l), whereas in microorganisms only a single aspartate aminotransferase has been found (2-5). These enzymes are dimeric consisting of two identical subunits of approximately 400 amino acid residues. The complete amino acid sequences of the cytosolic aspartate aminotransferases from pig (6, 7) and chicken (8), and those of mitochondrial aspartate aminotransferases from pig (9, lo), rat (ll), chicken (12), and man (13) have been established. Comparative studies indicate that the degree of interspecies sequence identity of the homotopic isoenzymes is more than 80%, while that of the heterotopic isoenzymes in a single species is close to 50%. X-ray studies on the pig cytosolic (14), chicken cytosolic (15,

* This investigation has been supported in part by Grants-in-Aid 457073 for Scientific Research and 59106008 for Special Research Project from The Ministry of Education, Science and Culture of Japan and by a grant from the Naito Science Foundation. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “aduer- tisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

$Present address: Dept. of Biology, Faculty of Science, Osaka University, Toyonaka 560, Japan.

§ To whom correspondence should be addressed: Dept. of Medical Chemistry, Osaka Medical College, Takatsuki-shi, Osaka 569 Japan.

16), and chicken mitochondrial (17) enzymes revealed that their three-dimensional structures are virtually the same. These structural studies indicate that the cytosolic and mitochondrial isoenzymes of aspartate aminotransferase have evolved from a common ancestral form. In contrast to these extensive studies on the animal enzymes, little information on the structure of the procaryotic enzymes has been available.

To deduce the evolutionary history of the aspartate aminotransferase molecule, structural studies on the procaryotic aspartate aminotransferases are essential. Microbial aspartate aminotransferases have been obtained in crystalline form from Escherichia coli (18), Pseudomonas (2), and yeast (19) and characterized. The molecular weight, subunit number, absorption spectra, and Michaelis constants for the substrates of the microbial aspartate aminotransferases are similar to those of the animal isoenzymes (20). Once we had obtained enough pure enzyme from E. coli, the complete amino acid sequencing of this bacterial aspartate aminotransferase was undertaken.

A preliminary report in summary form of the present communication has been published (21) but it did not give details of experimental procedures. In this paper, therefore, we de- scribe the research leading to the completion of the primary structure of the E. coli aspartate aminotransferase. Attention is focused on its structural features, which are compared with those of the animal enzymes.

EXPERIMENTAL PROCEDURES AND RESULTS’

Complete Amino Acid Sequence of Aspartate Aminotrans- ferase of E. coli-The construction of the whole sequence of E. coli aspartate aminotransferase is summarized in Fig. 16. The sequence of the NH,-terminal 33 residues of the whole enzyme, provided previously by automated liquid-phase Se- quencer analysis of the whole protein (241, was identical with that of CN1, indicating that CN1 is from the NH, terminus of E . coli aspartate aminotransferase. The known sequence of T23 coincided with those of the 5 COOH-terminal residues of CN1 and 2 NH,-terminal residues of CN2, indicating that CN2 follows CN1. Since T26 has the Met-Lys sequence at

Portions of this paper (including “Experimental Procedures,” part of “Results,” Figs. 1-15, and Tables I-VII) are presented in miniprint at the end of this paper. The abbreviations used are: AST, aspartate aminotransferase; TPCK, ~-tosylamido-2-phenylethyl chloromethyl ketone; AEAP, 2-aminoethyl-3-aminopropyl; TFA, trifluoroacetic acid; PTH, phenylthiohydantoin. Miniprint is easily read with the aid of a standard magnifying glass. Full size photocopies are available from the Journal of Biological Chemistry, 9650 Rockville Pike, Be- tbesda, MD 20814. Request Document No. 86M-3605, cite the authors, and include a check or money order for $13.60 per set of photocopies. Full size photocopies are also included in the microfilm edition of the Journal that is available from Waverly Press.

8648

E. coli Aspartate Aminotransferase Covalent S t ruc ture 8649

the COOH terminus and no cyanogen bromide peptides other than CN3 were found to have an NHZ-terminal lysine, 1- residue overlap of T26 with CN3 was sufficient to indicate that CN3 follows CN2. T29 provided 10-residue overlap with CN3. Two cyanogen bromide peptides (CN4 and CN5) were found to have their NH,-terminal arginine to be overlapped with COOH-terminal arginine of T29. The useful peptide for determining the positions of CN4 and CN5 was T32 (Met- Arg). Since T32 was derived from a tryptic hydrolysate, it must be preceded by an arginine or lysine residue. Of the isolated cyanogen bromide peptides, only CN4 was found to have the Arg-Met sequence at its COOH terminus. Thus, 1- residue overlap of T32 with CN4 and CN5 was sufficient to align CN4 and CN5 in this order, leading to the conclusion that CN4 follows CN3. The amino acid composition of T36 coincided with the sum of those of the 4 COOH-terminal residues of CN5 and the first 8 residues of CN6, permitting the alignment of CN5 and CN6 in sequence. CN7 was assigned as the COOH-terminal peptide of the enzyme, since it was the only cyanogen bromide peptide without a homoserine residue. The sequence of the 11 NH2-terminal residues of T39 overlapped with the COOH-terminal sequence of CN6, and the sequence of the remaining COOH-terminal portion was identical with that of CN7, confirming that CN6 precedes CN7.

As summarized in Fig. 16, the alignment or seven cyanogen bromide peptides was thus established by the results of Se- quencer analysis of the whole protein and sequences of the tryptic peptides, leading to the reconstitution of the complete sequence.

The total number of residues is 396, and the molecular weight was calculated to be 43,573, which is very close to the value obtained by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (20). The amino acid composition calculated from the sequence agrees well with that obtained by hydrolysis of the protein (Table VIII).

DISCUSSION

Several lines of evidence described in this paper have established the complete amino acid sequence of E. coli aspartate aminotransferase (Fig. 16). This is the first report describing the complete primary structure of a microbial aspartate aminotransferase. The sequence was verified by recent reports (40, 41) on the nucleotide sequence of the aspC gene that encodes E. coli aspartate aminotransferase. The specific lysine residue that binds pyridoxal phosphate is located at position 246. The sum of the number of glutamic acid and aspartic acid residues is 7 residues more than that of arginine plus

S G L l K E ~ Y L R L R L ~ F G V Y A V A S G R ~ N V ~ G ~ T P Q N ~ A P l C E A l V l V l .I

4 C N O I C N ? ' 1 139

FIG. 16. Complete sequence of E. coli aspartate aminotransferase. -, Edman degradation of the whole protein (Ref. 24); CN, cyanogen bromide peptides; T, tryptic peptides.

TABLE VI11 Amino acid composition of E. coli aspartate aminotransferase

Values are given as mol of residue/mol of aspartate aminotransferase subunit.

Analysis" Sequence

Aspartic acid 42.6 20 Asparagine 23 Threonine 22.2b 22 Serine 18.5' 21 Glutamic acid 44.4 27 Glutamine 16 Proline 14.8 15 Glycine 29.7 30 Alanine 46.3 46 Cysteine 5.3' 5 Valine 25.4d 26 Methionine 7.9 8 Isoleucine 17.5d 17 Leucine 36.2 38 Tyrosine 11.3 11 Phenylalanine 20.3 20 Lysine 17.1 18 Histidine 5.6 6 Arginine 21.5 22 Tryptophan 4.5 5

Except where indicated, the values are averages of those obtained after different times of hydrolysis (24, 48, and 72 h).

* Values extrapolated to zero time of hydrolysis. e Determined as cysteic acid.

Values after 72-h hydrolysis.

lysine residues, which may contribute to the acidic isoelectric point (pH 4.2) of E. coli aspartate aminotransferase (20). In both the mitochondrial and the cytosolic isoenzymes from animal sources, the sum of the number of arginine plus lysine residues exceeds that of glutamic acid and aspartic acid residues. The E. coli enzyme contains 55.8% nonpolar and 44.2% polar residues, respectively. These values are in good agree- ment with those of the chicken and pig isoenzymes. There is no disulfide linkage in the E. coli aspartate aminotransferase molecule. Five thiol groups were titrated by 5,5'-dithiobis(2- nitrobenzoate) or 2,2'-dithiodipyridine in the presence of 6 M guanidine hydrochloride (data not shown). Preliminary ex- periments showed that in the native enzyme, one cysteine was readily modifiable with the thiol reagents without affecting the enzymatic activity, and the reactivity was increased by approximately three times in the presence of the substrate pairs (data not shown). A syncatalytic increase in the suscep- tibility of a particular suIfhydryl group toward thiol reagents by 1 or 2 orders of magnitude has been observed in the animal isoenzymes (42, 43). Since these syncatalytic reactivity changes are thought to reflect conformational adaptations of the enzyme-substrate complex in the catalytic reaction, iden- tification of the position of a cysteine residue with syncatal- ytically increased reactivity in the primary structure of E. coli aspartate aminotransferase will be important for understand- ing the integral feature of the mechanism of aspartate aminotransferase action.

Knowledge of the complete amino acid sequence of E. coli aspartate aminotransferase is of great importance in determining the structure-function relationships and in tracing the evolutionary history of the aspartate aminotransferase molecule. In Fig. 17, all known amino acid sequences of aspartate aminotransferases (mitochondrial enzymes from pig (9, IO), chicken (12), rat (111, and man (131, and cytosolic enzymes from pig (6, 7) and chicken (8) and that from E. coli (present report)) are compared. 410 and 412 amino acid residues con- stitute chicken and pig cytosolic enzymes, respectively, whereas polypeptide chains of the four mitochondrial enzymes

8650 E. coli Aspartate Amirwtral 10 20 ID *O 50 60 10

~~PSVfAiVPPA~PVLVFKLlADFREDPDPRKVNLOVMVRTDDCOPYVLPVVRKVi~RIANNSSLNHiY A A S I F A A V P R A P P V A V F K L l A D F R i D G D S R K V N L G ~ G A V R l D E G Q P Y V L P V V R K V E ~ ~ l A G N G S L M U i ~

S S Y Y A H V E M 6 P P D P I L G V T i A F K R D l N S K ~ N L G V G A V R D D N G K P ~ V L P S V R K A E A ~ l A A K N ~ L O K i V SSYVSHViMGPPOPILGVTEAFKRDlNSKKMNLGVGAVRDONGKP~VLNCVRKAEA~lAAKK-H~KiY SSYYT~Vi l (6GPDPIL6VTEAFKRDl ISKKMNLGVGAVRODNGKPVVLPSVRKAEAQlAGKN-LDKiY SSYYTWVIMCPPOPIL6VTiAFKRDlNSKKMNL6V6AVRDDNGK~VVLPSVRKAiA~lAAKN-LOKiY

M F E N l l A l P A D P 1 L G L A D L F R A D E R P 6 K 1 N L G I G V V K O f ~ G K l P V L l S V K K A ~ O Y L L i N ~ - T l K N V

" .*. . ........... . . . CPl9

* P i g cChlckcn

m t h l c k c n mRa t mMuman E.* . . . . . . . . ..........

220 cP iq V M K R R F L F P F F D S A V ~ G F A S G N L i K D A Y A l R V F V S E G F I l F C A ~ S F S K N F G L V ~ i R V G N L l V V A K E V O S l

230 240 250 260 210 280

m P l q L V K K N N L F A F F D M A l ~ 6 F A S G D G N K D ~ M A v R H F l E O G l N V C L C ~ S v A K N M G L Y G f R V G A F l V V C K D A i i A cChlcksn VMKRRCLFPFfDSAI~GFASGNLfKDAMAVRVFVSE6f IL fCAOSF5KNFGLVNERVGNLSV~GKDEDN~

m t h l c k e n V~KKRNLLAYFOUAVQ~FASGDlNROA~hLRH~lE~GlDVVLS~SIAXNU6L~GiRAG~FlVlCRDAiEA m R l t V V K K I I N L F A F ~ D U A V ~ G F A S G O G N K D A ~ A v R H F l i ~ G l I I V C L C a S V A K N M G L V G i R V G A F l V ~ C K O A i i A ~HUI I~ I I V V K K R ~ L F A F F D M A V ~ 6 F A S G D G N K D A Y A V R M F l i ~ G l I V C L C Q S V A K N H G L ~ G i R V G A F l ~ V C K D A D i ~ " i.coll L S v i K 6 ~ L P L F O F A ~ ~ 6 F A I G - L E i D A i G L R A F A A M M K E L l V A S S V S K N F G L Y M i R V G A C l L V A A O S E ~ V ............. ......... CPlO

" P l q CChlcken

In. t mChtctcn

.H"la" f.COll "

C P l O c c n l c k c n .PI9 m C l l c k c n .n. t .*".." ~ I . C O 1 I -~

. . . . . . . . . . . . . . . . . .

" . . . . . . . . . . . . . . . . . FIG. 17. Comparison of the sequences of aspartate amino-

transferases from different species. Deletions introduced into the sequences are indicated by dashes. Asterisks indicate residues identical in all aspartate aminotransferases. cpig, cytosolic aspartate aminotransferase from pig heart (6, 7); cChicken, cytosolic aspartate aminotransferase from chicken liver (8); mPig, mitochondrial aspartate aminotransferase from pig heart (9,lO); mchicken, mitochondrial aspartate aminotransferase from chicken liver (12); mRat, mitochondrial aspartate aminotransferase from rat liver (11); rntiuman, mitochondrial aspartate aminotransferase from human liver (13); E. coli, aspartate aminotransferase from E. coli (present report). In Ref. 11, glycine was assigned at position 140, but Wada, one ofthe authors, gave tryptophan in his recent personal communication.

contain 401 amino acid residues. The E. coli enzyme contains 5 residues less than the mitochondrial enzyme. Therefore, some deletions must be introduced to achieve maximum homology. In the following discussion, the amino acid residues are numbered according to the sequence of the cytosolic aspartate aminotransferase from pig to simplify the comparison. Fig. 17 shows that the E. coli aspartate aminotransferase and the animal isoenzymes of aspartate aminotransferase are homologous proteins. There are 108 amino acids (26% of the total) invariant in all aspartate aminotransferases compared. These residues, being invariant during the long evolution, may contribute to the formation of correct three-dimensional structure of aspartate aminotransferase or may be of func- tional importance. Based on the published results from the x- ray crystallographic analysis on the animal isoenzymes (14- 17, 44), we discuss some of the features of the amino acid sequence of the E. coli enzyme. All residues making the coenzyme-binding pocket are invariant with an exception of Ser-107 which is replaced by glycine in the E. coli enzyme: Tyr-70, Gly-108, Thr-109, Ser-255, Arg-266 (phosphate-binding site); Trp-140, His-143, Asn-194, Asp-222, Ala-224, Tyr- 225, Lys-258 (pyridoxal moiety-binding site). Residues making the substrate-binding site are also invariant. Arg-292 and -386 are responsible for binding the distal carboxylate and a- carboxylate groups of the substrate, respectively. In addition,

zsferase Covalent Structure

pockets for both carboxylates are provided by residues including Tyr-70, Trp-140, Asn-142, Asn-297, Ser-296, and Gly-38, Val-37 (isoleucine in the E. coli enzyme), Phe-360, respectively. Other invariable Asn-34, Leu-35, and Gly-36 may be involved in the active site. It is of interest that the E . coli enzyme, like the mitochondrial one, has Asp-15 (valine in the cytosolic enzyme) and lacks the Glu-141 present in the cytosolic enzymes. Arg-292 is proposed to interact with Asp-15 in the unliganded mitochondrial enzyme and with Glu-141 in the cytosolic one. This may be responsible for the small differences of the active sites of the E. coli and the mitochondrial enzymes from that of the cytosolic one. I t is noteworthy that 3 histidines, His-143, -189, and -193, are conserved. By interacting with Asp-222 and Thr(Ser)-139, they are pre- sumed to play an important role to facilitate the aspartate aminotransferase reaction. Each subunit of aspartate aminotransferase is divided into two major parts: a large domain constructed with residues 75-300 and a small domain com- posed of the carboxyl- (residues 301-358) and amino-terminal (residues 15-47) segments. According to Jansonius et al. (17) interdomain interface consists of residues 161-166, 192-199, and 228-231 in the large domain and 326-328 and 352-364 in the small domain. Since binding of the substrate induces the movement of the small domain over the large domain, as the essential process of the enzymic action, the structure of the interdomain interface must be important. Hence, highly conserved sequences in these regions are reasonable. Other highly conserved residues are: 133, 135, 185, 187,218, and 220 which contribute to hydrophobic core facing the coenzyme in the large domain; 33, 35, 337, 338, 341, 350, 353, 361, 363, 365, 370, 373, 377, 379, 381, 387, 389, 392, 397, and 399 which construct the nonpolar core of the small domain; and coenzyme-binding loop 255-273.

Amino acid sequences of the two isoenzymes are available in pairs for chicken and pig. Fig. 18 shows the degree of the sequence identity of the E. coli enzyme with the isoenzymes from chicken and pig as well as inter- and intraspecies identity of the animal isoenzymes. The enzyme shows a similar degree of identity (about 40%) with both the mitochondrial and cytosolic enzymes, without a species distinction, and the similarity is less than between the heterotopic isoenzymes of the same species. It has been proposed that cytosolic and mitochondrial aspartate aminotransferase evolved from a common ancestral form about one billion years ago (45), the time of the emergence of eukaryotic cells. Fig. 18 indicates that the ancestral form of procaryotic and eukaryotic aspartate aminotransferases is common and that the divergence to the original form of E. coli aspartate aminotransferase had occurred long before the divergence to the two types of the isoenzymes in eukaryotic cells. There has been 'an endosymbiotic hypothesis of mitochondrial origin that mitochondria evolved from ancient bacteria that had entered into a sym- biotic relationship with host cells (46). According to this hypothesis, the structure of the bacterial enzyme is expected to be more closely related to that of the mitochondrial than

pig Cytosolic AST

I 8 5 %

4 0 % 38 %

Chicken cytosolic AST

FIG. 18. Relationships among degree of sequence identity. AST;aspartate aminotransferase.

E. coli Aspartate Aminotransferase Covalent Structure 8651

that of the cytosolic isoenzymes. The present results, showing the same degree of sequence identity of the E. coli enzyme with both the mitochondrial and cytosolic enzymes, do not give support for this hypothesis. However, it is interesting to note that all but two positions of deletions introduced into the sequence of the E. coli enzyme are identical with those of the mitochondrial enzymes (Fig. 17). Besides NH2 terminus and COOH terminus, nine gaps were introduced internally (positions 65, 127, 128, 130, 131, 132, 153, 232, and 407) into the sequence of the E. coli enzyme and seven into that of the mitochondrial enzymes. All of the seven gaps in the sequences of the mitochondrial enzymes overlapped those in the E. coli enzyme. Since insertions or deletions are thought to be more infrequently accepted in the evolutionary course than amino acid substitution (47), occurrence of gaps of equal size at the same positions in the E. coli and the mitochondrial enzymes, and not in the cytosolic enzymes, would be an indication that the bacterial enzyme is more similar to the mitochondrial enzyme than to the cytosolic one.

The predicted secondary structure and hydropathy profile along the amino acid sequence of the E. coli enzyme are closely similar to those of the vertebrate enzymes (data not shown). Three-dimensional structures of the cytosolic and mitochondrial aspartate aminotransferases are known to be extremely similar (14), in spite of there being about 50% difference in the amino acid sequences. From these observa- tions the tertiary structure of the E. coli enzyme is expected to be similar to those of the vertebrate isoenzymes, allowing the functionally important residues to be located at the equiv- alent positions. However, there must be subtle but distinct differences in the active site structure between the E. coli enzyme and the vertebrate isoenzymes because the E. coli enzyme is known to catalyze transamination of aromatic amino acids at much higher rates ( 1 5 1 1 % those of the corresponding glutamate activity) than the vertebrate isoenzymes do (less than 1% those for glutamate) (20, 48).* The residues unique to the E. coli enzyme could be structural components responsible for such difference in the substrate specificity.

X-ray crystallographic studies of the E. coli enzyme are in progress (55) and will reveal detailed structure and provide a better knowledge of the evolutionary relationship between procaryotic and higher eukaryotic aspartate aminotransferases as well as of the catalytic functioning of the enzyme.

Acknowledgments-We wish to thank Dr. Toshiharu Yagi, Depart- ment of Agricultural Chemistry, Kochi University, for his valuable technical assistance in the earlier stages of this work. We also thank Dr. Hiroshi Matsubara, Faculty of Science, Osaka University, and Dr. Yoshimasa Morino, Department of Biochemistry, Kumamoto University Medical School, for their useful advice.

REFERENCES 1. Braunstein, A. E. (1973) in The Enzymes (Boyer, P. D., ed) 3rd

Ed., Vol. 9, pp. 379-481, Academic Press, Orlando, FL 2. Yagi, T., Toyosato, M. & Soda, K. (1976) FEBS Lett. 6 1 , 3 4 4 7 3. Chesne, S. & Pelmont, J. (1973) Biochimie 5 5 , 237-244 4. Powell, J. T. & Morrison, J. F. (1978) Eur. J. Biochem. 87, 391-

5. Mavrides, C . & Orr, W. (1975) J. Bwl. Chem. 2 5 0 , 4128-4133 6. Ovchinnikov, Y. A,, Egorov, T. A., Aldanova, N. A., Feigina, M.

Y., Lipkin, V. M., Abdulaev, N. G., Grishin, E. V., Kiselev, A.

* In earlier studies, mitochondrial aspartate aminotransferase had been reported to display considerable activity toward aromatic amino acids (49-52). However, we re-examined the activity of the aspartate aminotransferase isoenzymes toward aromatic amino acids and observed a relative activity of less than 1%. This value is comparable to those reported for the mitochondrial enzymes from rat (53) and chicken (54).

400

P., Modyanov, N. N., Braunstein, A. E., Polyanovsky, 0. L. & Nosikov, V. V. (1973) FEBS Lett. 29,31-34

7. Doonan, S., Doonan, H. J., Hanford, R., Vernon, C. A., Walker, J. M., Airoldi, L. P. daS., Bossa, F., Barra, D., Carloni, M., Fasella, P. & Riva, F. (1975) Biochem. J. 149,497-506

8. Shlyapnikov, S. V., Myasnikov, A. N., Severin, E. S., Myagkova, M. A., Torchinsky, Y. M. & Braunstein, A. E. (1979) FEBS

9. Kagamiyama, H., Sakakibara, R., Wada, H., Tanase, S., Morino, Y. & Wada, H. (1980) J. Biol. Chem. 256,6153-6159

10. Barra, D., Bossa, F., Doonan, S., Fahmy, H. M. A., Hughes, G. J., Martini, F., Petruzzelli, R. & Wittman-Liebold, B. (1980) Eur. J. Biochem. 108,405-414

11. Huynh, Q. K., Sakakibara, R., Watanabe, T. & Wada, H. (1980) Biochem. Bwphys. Res. Commun. 97,414-479

12. Graf-Hausner, U., Wilson, K. J. & Christen, P. (1983) J. Biol. Chem. 268,8813-8826

13. Martini, F., Angelaccio, S., Barra, D., Pascarella, S., Maras, B., Doonan, S. & Bossa, F. (1985) Biochim. Biophys. Acta 8 3 2 ,

14. Arnone, A., Rogers, P. H., Hyde, C. C., Briley, P. D., Metzler, C. M. & Metzler, D. E. (1985) in Transminases (Christen, P. & Metzler, D. E., eds) pp. 138-155, John Wiley & Sons, New York

15. Borisov, V. V., Borisova, S. N., Kachalova, G. S., Sosfenov, N. I. & Vainstein, B. K. (1985) in Transaminases (Christen, P. & Metzler, D. E., eds) pp. 155-164, John Wiley & Sons, New York

16. Harutyunyan, E. G., Malashkevick, V. N., Kochkina, V. M. & Torchinsky, Y. M. (1985) in Transaminases (Christen, P. & Metzler, D. E., eds) pp. 164-173, John Wiley & Sons, New York

17. Jansonius, J. N., Eichele, G., Ford, G. C., Picot, D., Thaller, C. & Vincent, M. G. (1985) in Transaminases (Christen, P. & Met- zler, D. E., eds) pp. 109-138, John Wiley & Sons, New York

18. Yagi, T., Kagamiyama, H., Motosugi, K., Nazaki, M. & Soda, K.

19. Yagi, T., Kagamiyama, H. & Nozaki, M. (1982) J. Biochem.

20. Yagi, T., Kagamiyama, H., Nozaki, M. & Soda, K. (1985) Methods

21. Kondo, K., Wakabayashi, S., Yagi, T. & Kagamiyama, H. (1984)

22. Tanase, S., Kojima, H. & Morino, Y. (1979) Biochemistry 18 ,

Lett. 106,385-388

46-51

(1979) FEBS Lett. 100,81-84

(Tokyo) 92,35-43

Enzymol. 113,83-89

Biochem. Biophys. Res. Commun. 122,62-67

3002-3007 23. Crestfield, A. M., Moore, S. & Stein, W. H. (1963) J. Bwl. Chem.

238,622-627 24. Kagamiyama, H. & Yagi, T. (1979) Biochem. Biophys. Res. Com-

25. Furlan, M. & Beck, E. A. (1974) J. Chromatogr. 101, 244-246 26. Kistler, W. S., Noyes, C., Hsu, R. & Heinrikson, R. L. (1975) J.

Biol. Chem. 250,1847-1853 27. Simpson, R. J., Neuberger, M. R. & Liu, T.-Y. (1976) J. Biol.

Chem. 251,1936-1940 28. Edman, P. & Henschen, A. (1975) in Protein Sequence Determi-

nation (Needleman, S. B., ed) 2nd Ed., pp. 232-262, Springer- Verlag, Heidelberg

mun. 89, 1347-1353

29. Laursen, R. A. (1977) Methods Enzymol. 47 , 277-288 30. Edman, P. & Begg, G. (1967) Eur. J. Biochem. 1,80-91 31. Bridgen, T., Graffeo, A. P., Karger, B. L. & Waterfield. M. D.

32. 33.

34.

35. 36.

37.

38.

39.

(1975) in Instrumentation in Amino Acid Sequence Analysis (Perham, R. N., ed) pp. 117-119, Academic Press, Orlando, FL

Inagami, T. (1973) Anal. Biochem. 62,318-321 Zimmerman, C. L., Apella, E. & Pisano, J. J. (1979) Anal.

Biochem. 77,569-573 Edman, P. (1970) in Protein Sequence Determination (Needle-

man, s. B., ed) 1st Ed., pp. 211-255, Springer-Verlag, Heidel- berg

Offord, R. E. (1977) Methods Enzymol. 47,51-69 Groskopf, W. R., Holleman, J. W., Margoliash, E. & Klotz, I. M.

(1966) Biochemistry ED, 3783-3795 Chang, J. Y., DeLange, R. J., Shaper, J. H. & Glazer, A. N. (1976)

J. Bwl. Chem. 251,695-700 Kaspar, C. B. (1970) in Protein Sequence Determination (Needle-

man, s. B., ed) 1st Ed., pp. 137-184, Springer-Verlag, Heidel- berg

Schroeder, W. A., Shelton, J. B. & Shelton, J. R. (1969) Arch. Biochem. Biophys. 130,551-556

8652 E. coli Aspartate Aminotransferase Covalent Structure

40. Kuramitsu, S., Okuno, S., Ogawa, T., Ogawa, H. & Kagamiyama,

41. Malcolm, B. A. & Kirsch, J. F. (1985) Biochem. Biophys. Res.

42. Birchmeier, W., Wilson, K. J. & Christen, P. (1973) J. Biol.

43. Gehring, H. & Christen, P. (1978) J. Biol. Chem. 263,3158-3163 44. Amone, A., Christen, P., Jansonius, J. N. & Metzler, D. E. (1985)

in Transaminases (Christen, P. & Metzler, D. E., eds) pp. 326- 362, John Wiley & Sons, New York

45. Christen, P., Graf-Hausner, U., Bossa, F. & Doonan, S. (1985) in Transaminases (Christen, P. & Metzler, D. E., eds) pp. 173- 185, John Wiley & Sons, New York

46. Margulis, L. (1970) in Origin of Eukaryotic Cells, Yale University Press, New Haven, CT

47. Dayhoff, M. 0. & Barker, W. C. (1972) in Atlas of Protein

H. (1985) J. Biochern. (Tokyo) 97,1259-1262

Commun. 132,915-921

Chem. 248,1751-1759

SUPPLEMENTARY KATERIAL

TO

STRUCTURAL STUDIES ON ASPARTATE RMINOTRANSFERRSE FROM E. coll. COVALENT STRUCTURE.

KiyoShl Rondo. Sadao Wakabayashl, and Hlroyukl Kagamlyama.

EXPERIMENTAL PROCEDURES

KATERIALS AST was purlfred from E. B essentially as drscrrbed prrv1Ou5Iy 1151. The following materials were purchased cammerclally: Sephadex G-50 from Pharmacla Flne Chemlcale Inc.. DEE-cellulose IDE521 from Whatman BlOChemlCal Ltd., trypsin (treated rlth L-tosylamldo-2-

CarboxvOeDtldase A from Worthlnaton Blochemlcal Cora.. phenylethyl chloromethyl ketone). o-chymotrypsxn, and

.. . StaDhvlococcus aureus VB protease from Mlles Lab.. and carbaxypept15aseFTFFom Orlental Yeast Co. Ltd.. O s a k a . S p e c i a l grade chemicals for sequence analyels were purchased from Wako Pure Chemlcais. Ltd.. Osaka.

- . .

Sodlum BOrOhydrlde Reduction and S-Carbaxymethylatlan o f the Holoenzyme. ]The holoenzyme I170 mgl w a s treated with sodium borol Hlhydrlde to convert the Ilnkaqe between

descrlbed previously 1221. Sodium baro13HI hydrlde 125mC~. pyridoxal 5"phosphate and the apoenzyme to a stable form a s

New England Nuclear1 was diluted wlLh 40 ma cold Sodrum borohydrlde in 0.3 mi cold water lust be€o;e use. T e

cpmirnol. The resultant prepdratran showed 2 absorption specrfic a c t l v l t y of Lhe reduced enzyme was 3.1 x 10%

bands a t 250 and 125 om. Carbuxymechylat~on of the borohydride-reduced hololST was performed a s descrrbed by Crestfleld & &. 1231 wlth some modlf1ratLOns 1241. Digestion W ~ t h Trypsln. The trlrium-labelled carboxymethylated AST Iapproxlmately 150 m q l was dlqested wlth TPCX-treated trypsin I2 mq) I" 50 m l of 0.1 M ammonium

Cleavage By CvdnOqen Bromide. The carboxpethylated AST blcarbonate a t 37OC for 6 hoors.

labout 100 mql was dissolved ~n 2 m l of 70% forrnlc acld and

The reaction mixture was then dllutcd vlrh 200 m i of reacted w l t h cyanaqen brornlde 181 mg) for 15 hours at 8OC.

Sequence and Structure (Dayhoff, M. O., ed) Vol. 5, pp. 41-45, National Biomedical Research Foundation, Wash. D. C.

48. Kagamiyama, H., Kondo, K. & Yagi, T. (1984) in Chemical and Biological Aspects of Vitamin B, Catalysis: Part B (Evangelo- poulos, A. E., ed) pp. 293-302, Alan R. Liss, Inc., New York

49. Novogrodosky, A. & Meister, A. (1964) Biochim. Biophys. Acta

50. Miller, J. E. & Litwack, G. (1971) J. Biol. Chem. 246,3234-3240 51. Scandurra, R. & Canella, C. (1972) Eur. J. Biochem. 26 , 196-

52. Shrawder, E. & Martinez-Carrion, M. (1972) J . Biol. Chem. 247,

53. King, S. & Phillips, A. (1978) J. Neurochem. 30, 1399-1407 54. Mavrides, C., & Christen, P. (1978) Biochem. Biophys. Res. Com-

55. Smith, D. L., Ringe, D., Finlayson, W. L. & Kirsch, J. F. (1986)

8 1,605-608

206

2486-2492

mun. 85, 769-773

J. Mol. Biol. 191 , 301-302

chpotrypsin~or V 8 protease were deslqnated a s C and 5 , respectlvely, followed by ~ r a b l c numerals I" accordance with the Order from the NH2-terminl of the parent peptides.

. .

RESULTS

Sequence Determlnallon of Tryptic Peptlde4. R e s u l t s of sequence determlnatmn of t h e tryptic peptldes are summa- rlred in Table 11. Manual Edman degradation establlshed the complete drnlno acld sequences Of 32 rryptlc peptlde~ IT1bT6. TB-TI?, T21-T31. and T37-T381. Tl, T l B , T19. T20, and T35


8654 E. coli Aspartate Aminotransferase Covalent Structure T a b l e I f c i l n f i n w d )

I TI4 Tll TI6 TI1 TIU TI9 T20 721 T22 T21 T11 T21 Ti6

0 .0111,

r a b l e 111. m l n o acld composition of cyanogen bromrde peptides from E. coll

Values arc taken from the sequence. aspartate amlnotransferase. Analyses were performed on 20 h hydrolysates

CN1 CN2 CN3 CN4 CN5 CN6 CN7

27.431271 2.691 31

14.911151 9.031 91 25.04(251 10.691111 21.141211 26.111261 11.951121

8 . 7 8 ( 91 24.811251 7.091 71 12.011121 12.051121

0.72111 4.30141

4.45151 1.88121

4.34141

2.07121 5.21151 3.38141 0.96f11 3.1613) 1.77121 1.81121 2.0412) 0.88111 1.9512) 0.7011l

5.17 (51 1 . 9 0 1 2 1 3.09(31 2.8313) 2.0512) 1.22111 8.02(81

2.52111 1.82121

3.4313) 0.90(11

0.90111 3.9314)

0.9011l 4.2714)

2.00121

0.91111 1.0611)

1.05121 1.72121

2.0712) 2.8713)

0.64111 2.0612)

1.05131 1.80121 2.6613) 4.23141 0.99fll 4.0314) 3.25131 4.53151

2.82131 0.82111 2.91131 1.0911l

2.7913) 1.76121

0.85111

1.02111 1.0411l

2.98131

0.62111 1.57121

1.91121

9 2.13(21

1.0411l

2.6513) 0.62111

."* I

1 .1311l 1.02111

0.85111 1.88121

4.101 41 9.871101 1.451 ? I 1.961 41 + I11

71.5 57.5 53.5 51.7 56.5 18.0 45.5

1-234 235.275 276-314 315-321 322-347 348-385 386-396

~

Cycle

39

Acld lnmol l Alal0.llJ AsplO.ll1 Gl"10.111 i i r q I O . 0 8 ) P r a 1 0 . 0 6 1 Gly10.081

PTH-Ammino

A'rg10.621 Alal0.711 AsplO.35l 61~10.421 RrqlO.251 rroIo.l91 Gly10.231

1 2 3

P h e l 5 . 9 0 1 Glull.61l 6 1 U l 4 . 3 4 1 Rsnil.081 Asn(2.681 lle10.47J

T h r Ile13.521 Thr

A l a 1 3 . 1 8 1 ~lai0.881

~la11.721 ProIO.641 11012.441 Ala10.781 Ala12.261 AsPl0.ll) Aspl1.781 PrOi0.281 Proll.261 lle10.24l

5 6

I l e l O . 0 6 1 A s n I O . 0 5 ) LeulO.051

Gly10.06) IielO.071

Vall0.051 Tyr10.041

Gly10.081

I l e I O . 2 0 1 Asnl0.191 Leul0.131 GlylO.22l Ile10.181 Glyl0.131 VallO.161 Tyr10.151

Rspl0.031

Thrltracel GlulO.02) AspI0.121

61u10.08J

E. coli Aspartate Aminotransferase Covalent Structure

.00111

T . > L l c V I . AutOmatCd solid-phase Edman deqradatlon of 59. 1 5 "mol Of s9 was used. The NH - t e r m m i residue wlis rdentliled by manual Edman aegradatlon. P'.H-amino acids were ldentifled by high-performance licluld chromatography ( s e e the t e x t ) .

Cycle PTH-Amino Acid Cyc~e PTH-Amino Acid Inmoll

2 3 G l y 10.491 1 Le"

("moll

2 Leu 13.51)

26 Ala 10.28) 4 Gly ( 2 . 7 2 1 I Phe 13.111

24 Thr 25 Gly 10.35)

5 6 Gly ( 2 . 6 1 )

27 Leu 10.341

1 ser 28 Acq (0.22)

8 Ala 12.38) 2 9 V a l 10.34) 30 Ala 10.26)

9 Leu 11.95) 10 Ile (2 .141

3 1 Ala 10.27)

11 A5n 12.341 32 ASP (0.18)

1 2 Asp ( 2 . 5 6 ) 3 3 Phe ( 0 . 2 6 )

1 3 34 Le" (0.18)

1 4 Arq ( 1 . 0 3 ) 35 Ala 10.191

1 5 Ala 11.351 36

1 6 Arg 10.851 37 A5n 10.101 38 Thr 39 ser 17 Thr

A l a 11.011 I i! V~l-(0.081 Gln 10.68) Thr A19 10.00

2 1 Pro (0.40) Val 10.081 22 Gly 10.531

o.8 t

FRACTION NUMBER Flg. 1 . Sephadex G-50 column Chromatography of the trypckc

2; aspartate aminotransferase. The absorbance at 280 digest of sodlum borohydrlde-reduced and 5-carbaxymethylated

nm l - I , and 325 nm I - - - ! and radloactlvlty l*....l of each f r d c t m n 110 m l l were measured, and approprlate f r d c r m n s were collected and lyophilized.

FRACTION NUMBER Flq. 2. DE-52 column chromatography of F r a c t l O n I 1 from Sephadex G-50 chromatography shown I" Flg. 1. The C o l u m n rids previously eqvillbrated vlth 0.0211 NH NCO . Peptldes were eluted with a linear gradient of 0 t8 1 . d n KC1 u s l n q two 500 m l chambers. FraCt~Ons of 5 m l Yere collected.

~~

8655

FRACTION NUMBER

8656 E. coli Aspartate Aminotransferase Covalent Structure

I I

T9

FRACTION NUMBER ~. -

F l q . 5. DE-52 column chrornatoqraphy of Flactlon V from Sephadex 6-50 chromatography shown I" F l q . 1. Candltrons a r e the same a5 described In the legend f o r Fig. 3 .

5 10 20

FRACTION NUMBER 50

TIME (mid

TIME (min) Flg. 9 . Chromatography of 0.1% trlfluoraacerrc a c l d - lnjoluble fractlon of F r a c t i o n I1 ~n Flg. 7 on a Cosmosll C i B column. See legend to F l g . 8 for detalls.

Fig. 7. Sephadex G-50 column chromatography of the soluble cyanogen bromlde peptldes from S-carboxymethylated E. toll aspartate arnlnatransferase. The peprrdes Were eluted wlth 0 . 0 5 M NHlON at a flow rate of 3 mllh. FraCtlOns 12.5 m1I were collected and measured far absorbance a t 280 nm.


FRACTION NUMBER Flq. 1 1 . Chromatoqraphy of the soluble par t of Staphylo- coccal protease peprldes from CN1 on Toyopearl X y - 5 0 I1 x 2 0 0 Cml . Peptldes were eluted with 0 . 1 " NHIHCOI, at a flow rate of 4 . 5 mllh, and 2.5 rnl fractlons were collected.

Y FRACTION NUMBER

F L ~ . 1 3 . Chromatoqraphy of a chymotrypsrn dlqest of S 9 on a Sephadex G-50 column. Flow rate was 3 . 2 m l / h and 2 . 2 mi fractions were collected.

I/. 1

FRACTION NUMBER 50 100 150

Big. 12. Chromatography of fractlons 1 ( A I . 2 101. and 3

Peprldes were eluted with a l m e a r gradient from 0.02 to I C 1 shown In Flq. 11 On a DE-52 column 11 .5 x 30 c m l .

0 . 5 M N H ~ R C O J 0 0 0 ml totall at a flow rate o f 9 mlih. Fractions Of 2.5 ml were collected.

CN2

CN3

CN4

CN5

CN6

CN7

Flg. 15. Sequence sfudles o n cyanogen bromlde peptides CN2-CN7. - and 7 rndrcare that the r e i r d u e above was ldentlfred by soild phase Edman deqradatlon and manual Edman deqradatlon. respectively. Numerals are ylelds of PTH-amlno acids lnmoll. F i f t e e n r#mal of CN2. CN3: ZOO nmOl of CN4: and 20 m o l of CN5, CN6, CN7 were used.

Structural Studies on Aspartate Aminotransferase from Escherichia coli

Documents