Vol. January

166,

No. 30,

2, 1990

BIOCHEMICAL

AND

BIOPHYSICAL

RESEARCH

COMMUNICATIONS

Pages

1990

A SINGLE

AMINO

DEHYDROGENASE

ACID

SUBSTITUTION

IMPROVES

WITH

IN LACTATE

THE CATALYTIC

AN ALTERNATIVE

667-672

EFFICIENCY

COENZYME

Raymond Feeney, Anthony R. Clarke and J. John Holbrook

Molecular Recognition Centre, Department of Biochemistry, University of Bristol, School of Medical Sciences, Bristol BS8 lTD, U.K. Received

December

5,

1989

Using site-directed mutagenesis, the NADH-linked lactate dehydrogenase from Bacillus stearothemophilus has been specifically altered at a single residue to shift the coenzyme specificity towards NADPH. The single change is at position 53 in the amino acid sequence where a conserved aspartate has been replaced by a seiine. This substitution was made to reduce steric hindrance on binding of the extra phosphate group of NADPH and to remove the negative charge of the aspartate group. The resultant mutant enzyme is 20 times more catalytically efficient than the wild-type enzyme with NADPH. 0 mo Academic press, W.

The majority of redox enzymes involved in catabolic processes use the coenzyme NADH whilst those of anabolic processes use NADPH; despite the structural similarity between these coenzymes, only occasionally can either be used by the same enzyme, i.e. glutamate dehydrogenase (1). The NADPH/NADH specificity of dehydrogenases arises through their ability to distinguish between the presence and absence of a phosphate group on carbon-2 of the adenine ribose ring. In this study we use the NADH-specific lactate dehydrogenase (LDH) from B.stearothermophi1u.s to evaluate the possibility of using protein engineering to convert it to an NADPH-specific enzyme. LDH catalyses the following reaction: Pyruvate + NADH

+ H + + Lactate + NAD +.

As shown in table 1, the K, for NADH is small ( 15pM) and the enzyme has a rapid turnover number (220 s-l). However, with NADPH as the coenzyme, the Km Abbreviations

FBP, Fructose-1,6-bisphosphate; LDH, Lactate dehydrogenase; NADH, Nicotinamide-adenine dinucleotide; NADPH, Nicotinamide-adenine dinucleotide phosphate; Pyr, Pyruvate; TEA, Triethanolamine hydrochloride. 0006-291X/90 667

$1.50

Copyright 0 1990 by Academic Press, Inc. All rights of reproduction in any form reserved.

Vol. 166, No. 2, 1990

BIOCHEMICAL

AND BIOPHYSICAL

RESEARCH COMMUNICATIONS

is increased to 60kM and the turnover number diminished to 2d1, thus the bacterial enzyme is similar to the mammalian enzymes in selecting NADH over NADPH (2). In an attempt to enhance the catalytic efficiency of the enzyme with NADPH, a single mutation was made to replace aspartate 53 by serine. In the wild-type enzyme the aspartate residue forms a hydrogen bond with the 2’ hydroxyl of the adenine ribose of NADH as shown in fig.1 (3). To produce a site which will accept the 2’ phosphate group of the NADPH ribose ring more readily, aspartate was replaced by the smaller, uncharged serine. This mutation tests the hypothesis that this conserved aspartate is a major structural element in conferring NADH specificity in LDHs (4).

Materials

and Methods

A single point mutation was introduced into the gene coding for B. stearothermophilus LDH at position 53 in the amino acid sequence using a 22 mer synthetic oligonucleotide ( 5’ TITCATTCGCACI’GATGAGCACG 3’ ) substituting a wildtype aspartate for a serine (5). The mutant gene was then overexpressed in the E. cofi high expression vector pKK 223-3 in TG2 cells. Approximately 30% of total soluble cell protein was produced as mutant protein as is usual with this system (6 & 7). The muta@ protein was purified by ion exchange chromatography on Q- Sepharose Fast Flow in 1OOmMTriethanolamine hydrochloride (TEA) pH 7.5 and eluted on a NaCl gradient from 0 to 0.35 M with the protein eluting at approximately 0.2 M. The resultant protein sample was applied to an oxamateSepharose affinity column in 50mM TEA pH 6.0,20mM NADH and 12mM FBP and the column washed in 50mM TEA pH 6.0,lmM NADH and 5mM FBP. LDH was eluted with 50mM TEA pH 9.0 no NADH or FBP. The enzyme was precipitated by adding 43Og/l (NH4)2S04 and stored at 4’C. Prior to use the protein was treated with charcoal to remove any coenzyme present (8). All steady-state kinetic parameters were measured in 1OOmM TEA pH 6.0 buffer at 25’C by monitoring the optical density change at 340nm due to NADH consumption and fitted using the ENZFHTER non-linear regression data analysis programe from Elsevier-BIOSOFT (9).

Results The steady-state catalytic properties of both the wild-type and mutant (asp53ser) with NADH and NADPH as coenzymes are shown in Table 1. Taking k&Km values (where kcat refers to the turnover rate at saturating concentrations of both coenzyme and pyruvate and Km is the Michaelis constant for the coenzyme) as a measure of catalytic efficiency, the mutation reduces efficiency with NADH by a fat tor of about 3 but has little effect with NADPH . This is true whether or not the enzyme has been activated by fructose-1,6-bisphosphate (FBP). Although this represents a shift in specificity toward NADPH, the mutation has more effect on dis criminating against NADH than selecting for NADPH. However, in evaluating the relative efficiency of an enzyme with alternative coenzymes, the substrate Km must be taken into account, i.e. the ability of the enzyme-coenzyme to bind the substrate must be considered. In the absence of activator but with NADH as coenzyme pyruvate Km is increased by a factor of 2. A realistic measure of overall catalytic efficiency must take account of the maximal turnover rate 668

Vol.

BIOCHEMICAL

166, No. 2, 1990

AND BIOPHYSICAL

RESEARCH COMMUNICATIONS

Table 1

asp-53-ser

Wild-type -FBP Coenzyme kca,(il) &II

Coenzyme (FM) kadh x lo%r’i’

KmPyr(mM) kc&m x l&M-‘s-l

NADH

-FBP

+FBP

NADPH

NADH

NADPH

NADH

+ FBP

NADPH

NADH

NADPH

220

2

170

30

200

5

145

55

15

60

1.5

60

40

200

40

100

133

0.033

11.3

0.5

5

0.025

3.625

0.55

9

0.1

0.4

4

2

0.1

0.5

1.7

0.075

0.05

0.0025

1.45

0.11

2 0.11

0.002

Fructose-l,&bispbospbate (FBP), where used,waspresent at a concentration of SmM.

and of Km values for both substrate and coenzyme. We define this “overall catalytic efficiency” as: kit (s-I) Ldpyr)(M)xIL(coenzyme)(M) The units of this measure of overall efficiency are Mm2.se1and values are listed in Table 2. In the absence of activator the mutation leads to a 3.5-fold improvement in efficiency (as measured above) with NADPH and a 6- fold deterioration with NADH. Whilst the mutant is still 100 times more efficient with NADH as coenzyme, the ratio of efficiencies (i.e. the specificity of the enzyme) is shifted 20- fold towards NADPH. However in the presence of FBP the effect is much less marked with almost no change in efficiency with NADPH and a 3 fold deterioration in efficiency with NADH. Although the mutant is 33 times more efficient with NADH than NADPH,

Table 2

Wild-type -FBP kcaJ(Km(pyr).K&NADH)) k&(K&pyr).&(NADPH)) See text for definition

x106 M-*it x106 M*s-’

-FBP

+FBP

7333

113333

1250

36250

3.7

1250

12.5

1100

of “overall catalytic efficiency”. 669

asp- 53-ser

+FBP

Vol.

166, No. 2, 1990

BIOCHEMICAL

AND BIOPHYSICAL RESEARCH COMMUNICATIONS

there is only a 3 fold shift towards NADPH by FBP.

efficiency in the mutant when activated

Discussion

Figure 1 shows the structure of NADH bound to B. steurothemophilus LDH. The 2’ hydroxyl group is within hydrogen bonding distance of aspartate 53 and it also forms a hydrogen bond to the 3’ hydroxyl of the NADH adenine ribose. To represent bound NADPH a phosphate group has been modeled on to the 2’ ribose of NADH (Figure 2a). Charge repulsion between the phosphate group and the aspartate would occur due to their close proximity. Figure 2b indicates that with NADPH bound, a serine residue at position 53 could retain some hydrogen bonding ability to the 3’ hydroxyl but would also remove negative charge repulsion and reduce steric hindrance on the 2’ phosphate of the NADPH. This suggestion is supported by the results of Table 3: the mutant enzyme shows a loss of binding energy of about 1.8 kcal which is consistent with the loss of one hydrogen bond (10). Although this mutant has not reduced Km for NADPH it has increased the kcat and weakened NADH binding; in doing so it has produced a “transhydrogenase” which is less specific for NADH. This observation suggests aspartate 53 may not be the only residue responsible for selecting against NADPH in this class enzymes. Nevertheless the mutation has already produced a dehydrogenase with dual coenzyme specificity which is suitable for regenerating the expensive NADPH coenzyrnes in enzyme reactors. References

(1) Smith, E.L., Austen, B.M., Blumenthal, K.M. & NYC, J.F. (1975) ‘The Enzymes” (Boyer, P.D., ed), 3rd Edn., Vol lla Academic Press, New York, pp 294-367.

Figure 1.

The view shownis from the quaternary form crystal structure of wild-type B. (Ref. 3) with the a-carbon backbone represented in ribbon form and the NADH and the aspartate 53 in ball and stick with the carboxylate oxygens hatched (generated by the INSIGHT molecular graphics programme from Biosym Technologies Inc.). The secondarystructure nomenclature is from Ref. 4. stearothermophilus

670

Vol.

166,

No.

2, 1990

BIOCHEMICAL

AND

BIOPHYSICAL

RESEARCH

COMMUNICATIONS

b

a

Figure 2a.

A phosphategroup hasbeenmodeledon to the 2’ hydroxyl of the adenine ribosering showingthe closepositioningof the highly chargedphosphategroup and the aspartatein wild- type LDH with NADPH asa cofactor. Figure 2b.

Thisshowsamodeledstructureof the serine53mutationwithboundNADPH. The serinecanform ahydrogenbondwith the 3’ hydroxylgroupof the adenineribose and removesthe negativechargerepulsionfrom the aspartateactingon the phosphategroup.

(2) Fawcett, C.P. & Kaplan (1962) Journal of Biological Chemistry 237,1709-1715. (3) Wigley, D.B., Muirhead, H. and Holbrook, J.J. Personal Communication. (4) Holbrook, J.J., Liljas, A., Steindel, S.J. and Rossmann, M.G. (1975) in “The Enzymes” (Boyer, P.D., ed), 3rd Edn., Vol.lla Academic Press, New York, pp 191-292.

Table 3

1.Wild-type -FBP

kc;,l/[Kn,(PYR).K,(NADPII)]

= 0.0005

k,t/[Km(PYR).K,(NADH)I

2. asp53-ser-FBP

3. Wild-type + FBP

kc,l/[K,(PYR).K,(NADPH)I

= ” o1oo

k,t/[K,(PYR).K,(NADH)]



k&[Krn(PYR).K,(NADPH)]

= 0.0110

kot/[K,(PYR).K,(NADH)l

4. asp53-ser+ FBP

kca@m( PYR).Km(NADPH)]

k,t/[K,(PYR).K,(NADH)]

= ” o3“”



The differencebetweenthe free energiesfor the pseudoequilibrium constantof 2 minus1is 1.8kcal andfor 4 minus3 is0.59kcal. 671

Vol.

166, No. 2, 1990

BIOCHEMICAL

AND BIOPHYSICAL

RESEARCH COMMUNICATIONS

(5) Winter, G., Fersht, A.R., Wilkinson, A.J., Zoller, M. and Smith, M. (1982) Nature 299,756-758. (6) Barstow, D.A., Clarke, A.R., Chia, W.N., Wigley, D.B., Sharman, A.S., Holbrook, J.J., Atkinson, T. and Minton, N.T. (1986) Gene 46,47-55. (7) Clarke, A.R., Wigley, D.B., Chia, W.N., Barstow, D.A., Atkinson, T. and Holbrook, J.J. (1986) Nature 324,699-702. (8) Hart, K.W., (1989) Ph.D Thesis, Bristol University. (9) Leatherbarrow, R.J., (1987) ENZFITIER. A Non- linear Regression Data Analysis Program for the IBM-PC. Elsever- BIOSOFT, Cambridge, England. (10) Leatherbarrow, R.J. and Fersht, A.R. (1986) Protein Engineering I, 7-16.

672

A single amino acid substitution in lactate dehydrogenase improves the catalytic efficiency with an alternative coenzyme.

Using site-directed mutagenesis, the NADH-linked lactate dehydrogenase from Bacillus stearothermophilus has been specifically altered at a single resi...
352KB Sizes 0 Downloads 0 Views