(enzymology/evolution/differentiation). CHARLES H. WILLIAMS, JR.*, L. DAVID ARSCOTT*, AND GEORG E. SCHULZt. *Veterans Administration Medical Center ...
Proc. NatLi Acad. Sci. USA Vol. 79, pp. 2199-2201, April 1982
Biochemistry
Amino acid sequence homology between pig heart lipoamide dehydrogenase and human erythrocyte glutathione reductase (enzymology/evolution/differentiation)
CHARLES H. WILLIAMS, JR.*, L. DAVID ARSCOTT*,
AND
GEORG E. SCHULZt
*Veterans Administration Medical Center and Department of Biological Chemistry, University of Michigan, Ann Arbor, Michigan 48105; and tMax-Planck-Institut fur Medizinische Forschung, 6900 Heidelberg, Federal Republic of Germany
Communicated by J. L. Oncley, December 28, 1981 Extensive amino acid sequence homology has been found between nine tryptic peptides of pig heart lipoamide dehydrogenase [NADH:lipoamide oxidoreductase, EC 1.6.4.3] and the sequence of human erythrocyte glutathione reductase [NAD(P)H:glutathione oxidoreductase, EC 1.6.4.2]. The average homology is 40%. Six lipoamide dehydrogenase peptides are homologous with segments of the two parts of the FAD domain of glutathione reductase, one with the NADPH domain, and two with the interface domain. Thus, the homology extends throughout the molecule. ABSTRACT
Lipoamide dehydrogenase [NADH-lipoamide oxidoreductase, EC 1.6.4.3] and glutathione reductase [NAD(P)H:glutathione oxidoreductase, EC 1.6.4.2] have both mechanistic and structural properties in common (1). Both flavoproteins contain a cystine residue which undergoes reduction-oxidation during catalysis. In catalysis, both cycle between the oxidized and twoelectron-reduced forms. The two-electron-reduced form appears to be a charge transfer complex in which a thiolate anion is the donor and FAD is the acceptor (2, 3). The amino acid sequences around the oxidation-reduction active cystine residue are highly homologous (4-8). Reaction of either enzyme with iodoacetamide at the twoelectron-reduced level yields a monoalkylated species in which the nascent thiol nearer the amino terminus is modified. Based on the properties ofthese monoalkylated derivatives, roles have been assigned to the nascent thiols ofthe two-electron-reduced form. Thus, the thiol nearer the amino terminus interacts with the dithiol (disulfide) substrate whereas the thiol nearer the carboxyl terminus interacts with the FAD (9-11). The same roles have been assigned on the basis of x-ray diffraction analysis and chemical sequence determination of human erythrocyte glutathione reductase (8, 12, 13). The structural and mechanistic equivalence of the nascent thiols at the two-electron-reduced level in these enzymes together with the extensive homology in the region of the cystine residue suggests that homology may exist at other points in the lipoamide dehydrogenase and glutathione reductase molecules. Therefore, each tryptic peptide of known sequence from pig heart lipoamide dehydrogenase has been tested for homology with the sequence of human erythrocyte glutathione reductase which has recently been completed (14-16). The comparison is made more objective by use of a computer program that incorporates the empirical amino acid exchange frequencies (17) based on comparisons of closely related sequences from 34 superfamilies. The same program using a matrix based on apparent similarities of amino acids has been practical in locating the sequences of peptides of glutathione reductase in a sequence
of that enzyme derived exclusively from the electron density map (18-20). The utility of such programs lies in their reliance on similarity between amino acid residues at a given position rather than on identity. MATERIALS AND METHODS Lipoamide dehydrogenase from pig heart was purchased from Miles Laboratories (Seravac Division) and was purified by passage over calcium phosphate gel. The tryptic digest of alkylated lipoamide dehydrogenase was prepared as described (21). The peptides were purified by using high-performance liquid chromatography on a Waters Associates C18 ,uBondapak column (0.4 x 30 cm) with a linear gradient from 0.1% phosphoric acid to 50% acetonitrile in 0.1% phosphoric acid over 70 min at a flow rate of 1.9 ml/min. Peptide mixtures derived from up to 30 nmol of protein (FAD) could be separated. Peptides in individual peaks were further purified in the same system but at pH 7.2. Amino acid sequences were determined by the automated Edman degradation procedure in a Beckman 890C Sequenator with a modified 0.1 M Quadrol buffer program and 1.5 mg of Polybrene as a carrier. The phenylthiazolinone amino acid derivatives were converted to the phenylthiohydantoin amino acids in a Sequamat Autoconverter. The identification system was similar to that reported by Tarr (22). The computer program used to test for homology compares the peptide sequence with every possible position in the protein sequence in three modes. The first mode is a simple comparison; the second allows for the deletion of a single amino acid residue at each position; the third allows for an insertion at each position. A score is assigned for each position in a given comparison based on the matrix of natural substitution frequencies in homologous proteins (17); the scores are summed. The distribution of summed scores for all the comparisons is approximately gaussian. As a consequence, it is meaningful to calculate the SD of the scores from their mean value and to express each score by its distance from the mean as measured in SD units. The score with the highest distance from the mean in the positive wing of the gaussian represents the most likely position of the peptide sequence in the total sequence.
RESULTS AND DISCUSSION The amino acid sequence ofpig heart lipoamide dehydrogenase is being determined in the Ann Arbor laboratory. Thus far, approximately 40 tryptic peptides have been purified and 20 have been subjected to sequence analysis. The sequence of each tryptic peptide has been compared with the sequence of human erythrocyte glutathione reductase by using a computer program developed in the Heidelberg laboratory. With this program, two different sequences that nevertheless contain many fre-
The publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement" in accordance with 18 U. S. C. §1734 solely to indicate this fact.
2199
2200
Proc. Natl. Acad. Sci. USA 79 (1982)
Biochemistry: Williams et al.
quently encountered changes can still be recognized as related. In contrast, sequences with a low level of identity relative to their chain length and a large number ofunlikely changes would be designated nonhomologous (18-20). The sequences of the tryptic peptides oflipoamide dehydrogenase are aligned with the homologous segments of glutathione reductase in Table 1. The score, as expressed in SD units, and the fraction of identical residues are indicators of the quality of fit. Peptide a represents the amino terminus of lipoamide dehydrogenase as determined by sequence determination on the whole protein. Glutathione reductase has an amino-terminal extension of 14 residues beyond the terminus of lipoamide dehydrogenase.The amino termini of four flavoproteins can now be aligned and it is seen that p-hydroxybenzoate hydroxylase, lipoamide dehydrogenase, and glutathione reductase extend beyond the sequence of D-amino acid oxidase by 2, 6, and 20 residues, respectively (23). This region is crucial in the binding of FAD and has been highly conserved in these enzymes which represent three of the major classes of flavoenzymes: transhydrogenases (class 1), oxidases (class 2), and oxygenases (class 3) (24). The second peptide, b, is only seven residues long. Properly aligned, it is identical in four of the seven residues and yields a score of 4.3 SD from the mean. At a length of only seven residues, however, such scores are not significant and we would not include this peptide in Table 1 on its own ground. However, because peptide b is exactly placed between peptides a and c, it can be inserted with greater confidence. Peptide c contains the highly conserved oxidation-reduction active disulfide and, as would be expected, has the highest sequence identity and the highest score distance from the mean. The final peptide, i, in Table 1 extends to within 16 residues of the carboxyl terminus of glutathione reductase. Of the 20 tryptic peptides analyzed thus far, 9 have been
aligned. The remaining peptides are all short, 11 residues or less. At about 40% sequence identity such short peptides in general can not be placed with certainty. However, like peptide b, their positions presumably can be determined as soon as the adjacent peptides are located. The first four peptides in Table 1 are homologous with segments of the FAD-binding domain of glutathione reductase; peptide e is homologous with a segment of the pyridine nucleotide-binding domain; peptides fand g are homologous with segments of the second part of the FAD domain; and the last two peptides are homologous with segments of the interface domain. Thus, the homology between lipoamide dehydrogenase and glutathione reductase extends throughout the molecule as illustrated in Fig. 1. The average fraction of identical amino acids in the aligned peptides is 40% (65 of 162 residues). The aligned residues comprise about one-third of the chain. We think that these findings offer strong evidence for an evolutionary relationship between all segments of these two enzymes which have previously been shown to be mechanistically similar. It should be noted that in lysozyme and lactalbumin, the classical pair ofhomologous differentiated proteins, the proportion of identical amino acid residues is also approximately 40% (25, 26). Both lipoamide dehydrogenase and glutathione reductase are found in prokaryotes and eukaryotes (1). The sequences around the oxidation-reduction cystine residue have been determined for lipoamide dehydrogenase of Escherichia coli and pig heart and for glutathione reductase from human erythrocytes, yeast, and E. coli (Table 2). It will be of interest to determine whether the total sequences of the prokaryote species are also homologous. It is not possible to say if lipoamide dehydrogenase predates glutathione reductase, but lipoamide dehydrogenase is found in the anaerobe Peptococcus glycinophilus (27, 28) which is known from 16S RNA sequences to be an ancient species (29). Although this does not establish the antiquity of lipoamide de-
Table 1. Sequences of tryptic peptides from pig heart lipoamide dehydrogenase and homologous regions of human erythrocyte glutathione reductase Score distance Identity, from mean, SD % Sequences Peptide
ADQPIDADVTVIGSGPGGYVAAIK
5.1 38 AGAVA SYDYLVI GGGSGGLASARR 4.3 57 b AAQLGFK AAE LGAR 52 7.6 c TVC I EKNETLGGTCLNVGC IPSK AAVVE S HK-LGGTCVNVGCVPKK 5.0 29 L LNNGH A * H* AH * K d I IR-GHAAFTSDPK 4.1 36 e SEEQL KE EGI E-YK CT EEL ENAGVEVLK 5.1 44 f PF TQN L GLE ELG I ELR PNTKD L S LNKLG I QTD 3.7 43 g I PNIAA IGDVVAGP VKG I Y A VGDVC - GK 6.1 35 h CVPSV I YTHPEVAWVGK NIP TVV FSHPP I G TVGL i VLGAHIIGPGAGEMINEAALALEY VVG I HMQGLGCDEMLQGFAVAVKM 6.1 36 GASC E DIAR GATK A DFDN The lipoamide dehydrogenase sequence is on the top in each comparison. *, Residue at that position could not be identified in the lipoamide dehydrogenase sequence (for the purpose of the computer comparison, an alanine residue has been inserted); -, deletion from that sequence. The score distance from the mean is the "quality of fit" as defined in the text. The one-letter code for amino acid residues has been used: A, alanine; C, cysteine; D, aspartate; E, glutamate; F, phenylalanine; G, glycine; H, histidine; I, isoleucine; K, lysine; L, leucine, M, methionine; N, asparagine; P, proline; Q, glutamine; R, arginine; S, serine; T, threonine; V, valine; W, tryptophan; Y, tyrosine. a
a
b
Is I
I
Biochemistry:
Williams et aL
c
d
e
FAD-i ', -1 FAD N.iNj I
O I
FAD - 2
1
Proc. Natl. Acad. Sci. USA 79 (1982)
INTERFACE
NADPH NADPH
|
lipoamide dehydrogenase glutathione reductase
FIG. 1. Homology between glutathione reductase and lipoamide dehydrogenase. The'thick bar represents the amino acid sequence of human erythrocyte glutathione reductase with the boundaries of the amino-terminal extension (N), part 1 of the FAD domain (FAD-1), the NADPH domain (NADPH), part 2 of the FAD domain (FAD-2), and the interface domain (INTERFACE) indicated by vertical dashedlines. The small bars above the glutathione reductase sequence represent the tryptic peptides from pig heart lipoamide dehydrogenase listed in Table 1. They are aligned with the glutathione reductase sequence.
hydrogenase, because the information for the synthesis of an enzyme may be widely circulated in aplasmid, this seems highly unlikely because lipoamide dehydrogenase plays a crucial role in the metabolism of glycine, the sole carbon source for this organism. Glutathione reductase is hypothesized to have arisen in response to the advent of oxygen (30). It is possible that the amino acid sequences of lipoamide dehydrogenase from P. glycinophilus, E. coli, and pig heart will clarify the ill-defined divergence during the oxygen build-up era and that the sequences of E. coli and human erythrocyte glutathione reductase will clarify the point of divergence of that enzyme from lipoamide dehydrogenase. In contrast with lysozyme and lactalbumin, lipoamide dehydrogenase and glutathione reductase have evolved much more slowly. Both pairs contain approximately 40% sequence identity, but lysozyme and lactalbumin differentiated comparatively recently, with the emergence of mammals, whereas lipoamide dehydrogenase and glutathione reductase are both present in E. coli, indicating that they diverged rather early. Although the mechanism of lipoamide dehydrogenase is essentially preserved in glutathione reductase, both substrate specificities have changed. One important difference between the two mechanisms is that whereas in two-electron-reduced lipoamide dehydrogenase the presence of NAD+ changes the oxidation-reduction potential of the FAD relative to the dithiol so that electrons pass to the FAD, no such change is observed in glutathione reductase (11, 31). This evolution allows glutathione reductase to function physiologically in the opposite chemical direction. The structural basis of the difference is not known and it is hoped that relevant information will emerge as the amino acid sequence of pig heart lipoamide dehydrogenase is fitted to the electron.density map of human erythrocyte glutathione reductase. Table 2. Sequences around the oxidation-reduction active cystine residue in lipoamide dehydrogenase (LipDHase) and glutathione reductase (GRase) isolated from several species TLGGTC LNVGC I P S K PigheartLipDHase T LGGVC LNV GC I P S K E.coliLipDHase KLGGTCVNVGCVPKK Human erythrocyte GRase ALGGTCVNVGCVPK YeastGRase QLGGTCVNVGCVPK E.coliGRase The sequence of E. coli glutathione reductase peptide has not been published previously. Reference to other sequences will be found in the Introduction. The one-letter code is used for the amino acid residues as given in the footnote to Table 1.
2201
We thank Drs. R. H. Schirmer, R. L. Krauth-Siegel, M. Saleh, and R. Untucht-Grau for making available to us sequence data on glutathione reductase prior to publication. This research was supported by the Medical Research Service of the Veterans Administration and in part by Grant GM-21444 from the National Institute of General Medical Sciences. 1. Williams, C. H., Jr. (1976) in The Enzymes, ed. Boyer, P. D. (Academic, New York), Vol. 13, pp. 89-173. 2. Kosower, E. M. (1966) in Flavins and Flavoproteins, ed. Slater, E. C. (Elsevier, Amsterdam), pp. 1-14. 3. Massey. V. & Ghisla, S. (1974) Ann. N.Y. Acad. Sci. 227, 446-465. 4. Burleigh, B. D., Jr. & Williams, C. H., Jr. (1972)J. Biol Chem. 247, 2077-2082. 5. Brown, J. P. and Perham, R. N. (1972) FEBS Lett. 26, 221-224. 6. Matthews, R. M., Arscott, L. D. & Williams, C. H., Jr. (1974) Biochim. Biophys. Acta 370, 26-38. 7. Jones, E. T. & Williams, C. H., Jr. (1975) J. Biol Chem. 250, 3779-3784. 8. Krohne-Erich, G., Schirmer, R. H. & Untucht-Grau, R. (1977) Eur. J. Biochem. 80, 65-71. 9. Williams, C. H., Jr., Thorpe, C. & Arscott, L. D. (1978) in Mechanisms of Oxidizing Enzymes, eds. Singer, T. P. & Ondarza, R. N. (Elsevier/North-Holland, New York), pp. 2-6. 10. Thorpe, C. & Williams, C. H., Jr. (1981) Biochemistry 20, 1507-1513. 11. Arscott, L. D., Thorpe, C. & Williams, C. H., Jr. (1981) Biochemistry 20, 1513-1520. 12. Pai, E. F., Schirmer, R. H. & Schulz, G. E. (1978) in Mechanisms of Oxidizing Enzymes, eds. Singer, T. P. & Ondarza, R. N. (Elsevier/North-Holland, New York), pp. 17-22. 13. Schulz, G. E., Schirmer, R. H., Sachsenheimer, W. & Pai, E. F. (1978) Nature (London) 273, 120-124. 14. Schiltz, E., Blatterspiel, R. & Untucht-Grau, R. (1979) Eur. J. Biochem. 102, 269-278. 15. Untucht-Grau, R., Schirmer, R. H., Schirmer, I. & Krauth-Siegel, L. (1981) Eur. J. Biochem. 120, 407-419. 16. Krauth-Siegel, R. L., Saleh, M., Blatterspiel, R., Schiltz, E., Schirmer, R. H. & Untucht-Grau, R. (1982) Eur. J. Biochem. 121, 259-267. 17. Dayhoff, M. 0. (1978) Atlas of Protein Sequence and Structure (Natl. Biomed. Res. Found., Washington, D.C.), Vol. 5, Suppl. 3. 18. Untucht-Grau, R., Schulz, G. E. & Schirmer, R. H. (1979) FEBS Lett. 105, 244-248. 19. Schirmer, R. H., Pai, E., Schiltz, E., Schulz, G. E. & UntuchtGrau, R. (1980) in Methods of Peptide and Protein Sequence Analysis, ed. Birr, C. (Elsevier/North-Holland, Amsterdam), pp. 267-283. 20. Schulz, G. E. & Schirmer, R. H. (1979) Principles of Protein Structure (Springer-Verlag, New York), chapt. 9, pp. 166-205. 21. Thorpe, C. & Williams, C. H., Jr. (1976) J. Biol Chem. 251, 3553-3557. 22. Tarr, G. E. (1981) AnaL Biochem. 111, 27-32. 23. Hofsteenge, J., Vereijken, J. M., Weijer, W. J., Beintema, J. J., Wierenga, R. K. & Drenth, J. (1980) Eur. J. Biochem. 113, 141-150. 24. Massey, V. & Hemmerich, P. (1980) Biochem. Soc. Trans. 8, 246-257. 25. Brew, K., Vanaman, T. C. & Hill, R. L. (1967) J. Biol Chem. 242, 3747-3749. 26. Brew, K., Castellino, F. J., Vanaman, T. C. & Hill, R. L. (1970) J. Biol. Chem. 245, 4570-4582. 27. Baginsky, M. L. & Huennekens, F. M. (1966) Biochem. Biophys. Res. Commun. 23, 600-605. 28. Klein, S. M. & Sagers, R. D. (1967)J. Biol Chem. 242, 297-300. 29. Fox, G. E., Stackebrandt, E., Hespell, R. B., Gibson, J., Maniloff, J., Dyer, T. A., Wolfe, R. S., Balch, W. E., Tanner, R. S., Magrum, L. J., Zablen, L. B., Blavkemore, R., Gupta, R., Bonen, L., Lewis, B. J., Stahl, D. A., Luegrsen, K. R., Chen, K. N. & Woese, C. R. (1980) Science 209, 457-463. 30. Fahey, R. C. (1977) Adv. Exp. Med. BioL 86A, 1-30. 31. Matthews, R. G., Ballou, D. P. & Williams, C. H., Jr. (1979)J. BiOL Chem. 254, 4974-4981.
