WO1997044483A1

WO1997044483A1 - Enzyme-electron acceptor assembly

Info

Publication number: WO1997044483A1
Application number: PCT/GB1997/001376
Authority: WO
Inventors: Nigel Shaun Scrutton
Original assignee: University Of Leicester
Priority date: 1996-05-17
Filing date: 1997-05-19
Publication date: 1997-11-27
Also published as: AU2907397A; GB9610376D0

Abstract

The present invention concerns an enzyme-electron acceptor assembly comprising TMADH or a partially modified form thereof or a phenotypic mutant thereof and an electron acceptor, the electron acceptor being chemically bonded to the enzyme and the assembly being arranged such that upon catalysis of a reaction by the enzyme, electron transfer occurs from the enzyme to the electron acceptor.

Description

Enzyme-Electron Acceptor Assembly

The present invention concerns an enzyme-electron acceptor assembly.

Enzymes can be extremely accurate and sensitive detectors of the presence of particular molecules, the molecules affecting the rate of a reaction catalysed by the enzyme. This is particularly the case when the molecule is a substrate for the enzyme. However, quantifying the rate of catalysis by the enzyme can be difficult, inconsistent and inconvenient. One particularly useful enzyme is trimethylamine dehydrogenase (TMADH) (EC 1.5.99.7) from the methylotrophic bacterium Methylophilus methylotrophus (s.p. W3A1 ) which forms a physiological electron transfer complex with the electron transferring flavoprotein (ETF), ETF being an electron acceptor. TMADH acts to catalyse the conversion of trimethylamine to dimethylamine and formaldehyde, during which catalysis electrons flow (are transferred) from TMADH to ETF. Despite the fact that TMADH and ETF are only transiently associated and thus not susceptive to conventional analysis methods, e.g. x-ray crystallography, the present inventor has succeeded in identifying the point in TMADH from which electrons are transferred (or at least are preferentially transferred) to ETF.

To date the crystal structure of TMADH has been solved (Lim, J.W. et al., 1986, J. Biol Chem., 26J.: 15140 - 15146), the genes encoding TMADH (Boyd, G. et al., 1992. FEBS Lett.. 3J&: 271-276) and ETF (Chem, D.W. and Swenson. R.P.. 1994. J Biol Chem., 269: 32120 - 32128) cloned, a large area (1200 A^:) on the surface of TMADH has been postulated as the interaction site for ETF (Wilson, E.K. et al., 1995, Biochemistry. 3_4: 2584-2591 ) and comparisons made to other enzymes (Yang. C-C. et al., 1995, Eur. J. Biochem., 212: 264-271 ).

The present invention allows the creation of novel enzyme electron acceptor assemblies which may, for example, allow the rapid, simple and accurate quantification of the catalysis of a reaction by the enzyme.

According to the present invention there is provided an enzyme-electron acceptor assembly comprising TMADH or a partially modified form thereof or a phenotypic mutant thereof and an electron acceptor, the electron acceptor being chemically bonded to the enzyme and the assembly being arranged such that upon catalysis of a reaction by the enzyme, electron transfer occurs from the enzyme to the electron acceptor.

As described in the "Experimental" section below, novel TMADH-electron acceptor assemblies have been created. TMADH may, of course, be modified in order to change its substrate specificity or in other ways which do not substantially affect its properties or characteristics as an enzyme which transfers electrons to an electron acceptor upon catalysis of a reaction. As such, "TMADH" encompasses partially modified forms and phenotypic mutants of TMADH, partivularly with regard to its catalytic domain.

The electron acceptor may be chemically bonded to. or near to, the preferential exit point (i.e. point of transfer) for electrons from the enzyme. It may be bonded to residue 442 of TMADH which may be cysteine. The bonding may be a chemical bond, i.e. a non-transient bond. The bonding of TMADH may be to another molecule or to a metal and may occur via, for example, a carbon or sulphur atom. The bonding may, for example, be covalent bonding or it may be co-ordinate bonding. By binding the electron acceptor to the enzyme, electron transfer from the enzyme to the electron acceptor may be ensured.

The electron acceptor may be other than ETF.

Electron transfer to the electron acceptor may preferentially occur from residue 442 of the enzyme (the preferential exit point for electrons). Residue 442 of TMADH is Tyrosine and it may, of course, be modified. In modified forms of the TMADH enzyme which have, for example, additions, deletions or substitutions, electron transfer may occur from a residue other than residue 442. Substitution may for example be of Tyr-442 to Cys-442. An alignment plot (for example using the PILE-UP program on SEQNET at the Daresbury Laboratories. UK) of the modified form of TMADH against TMADH may identify a residue equivalent to residue 442 of TMADH. For example, electrons may preferentially leave a modified form of TMADH from a residue neighbouring residue 442 or from a residue near to residue 442.

The electron acceptor may be a redox acceptor.

The electron acceptor may be a flavin, for example, 8-C1-FAD. For example, simple reactions may be used to covalently bond a flavin to TMADH, electron transfer occurring from a cysteine residue substituted for tyrosine at residue 442 to which the electron acceptor is bound. The covalent bonding of a flavoprotein to an enzyme may be readily achieved, for example, as described by Moore, E.G. et al. (1978, Journ. Biol. Chem., 253: 6413-6422). Such a flavin may produce hydrogen peroxide upon electron transfer, which hydrogen peroxide may be detected by the use of a colorimetric hydrogen peroxidase assay.

Alternatively, simple modification of TMADH may allow it to be attached to an electrode (for example, a gold electrode by co-ordinate bonding) and electron transfer to the electrode directly quantified using standard apparatus.

The invention will be further apparent from the following description, with reference to the several accompanying figures. Of the figures:

Figure 1 shows kinetic transients observed for the wild-type (panel A) and Y442G mutant (panel B) complexes. Absorbance changes were recorded at 370 nm and reaction components were contained in 50 mM potassium phosphate buffer, pH 7. Concentration of TMADH was 3.8 μM (wild-type) and 3.8 μM (Y442G) and concentrations of ETF were 36 μM (panel A) and 20 μM (panel B). The solid line represents the non-linear least squares fit to the experimental data using equation 1 ;

Figure 2 shows molecular graphics of one subunit of TMADH showing the position of Tyr-442 at the centre of the putative ETF docking site. All side chains are omitted with the exception of Tyr-442. The FMN and 4Fe-4S center are depicted. The representation is generated from the refined crystallographic coordinates of TMADH. Arrows indicate approximate position of two- fold axis of symmetry; Figure 3 shows (Panel A) plots of observed rate constant against ETF concentration at 5, 15, 25 and 35 °C for the wild-type TMADH:ETF electron transfer complex. Data were fitted to equation 2. Limiting rate constants and kinetically-determined dissociation constants are; k_lm (s^"1) 73 ± 5, 143 ± 3. 187 ± 18 and 310 ± 34 at 5, 15, 25 and 35 °C respectively; K_d (μM) 24.6 ± 4.5, 15.2 ± 1.1 , 15.8 ± 5.4 and 21.2 ± 6 at 5, 15, 25 and 35 °C respectively. Panel B shows dependence of k_Um on temperature for the wild-type TMADH:ETF electron transfer complex. The solid line represents the fits of the data to equations 3 and 4;

Figure 4 shows mid-point potential determination of the FAD in ETF. Oxidised ETF (35 μM) contained in 50 mM potassium phosphate buffer, pH 7 was mixed with 35 μM toluyene blue, 2 μM methyl viologen, 250 μM xanthine and 20 nM xanthine oxidase under anaerobic conditions. Scans were recorded every 30 seconds (not all data shown). Inset: plot of log [ETF]_ox/[ETF]_red against log [toluyene blue]_ox/[toluyene blue]_red. Midpoint redox potential is 0.141 V;

Figure 5 shows plots of observed rate constant against ETF concentration at 25 °C for the wild-type, Y442F, Y442L and Y442G TMADH:ETF electron transfer complexes. Data were fitted to equation 2. Limiting rate constants and kinetically-determined dissociation constants are: k_Um (s^"1) 186 ± 18, 129 ± 6.8, 85.1 ± 8 and 6.1 ± 2 and K_d (μM) 15.8 ± 5.4, 20.6 ± 3, 37.8 ± 7.7 and 179 ± 70 for the wild-type, Y442F, Y442L and Y442G complexes, respectively; and

Figure 6 shows plots of observed rate constant against ETF concentration at 5, 15, 25 and 35 °C for the Y442G TMADH:ETF electron transfer complex. Data were fitted to equation 2. Limiting rate constants and kinetically-determined dissociation constants are: k_Um (s ¹) 0.5 ± 0.005, 1.3 ± 0.24, 6.1 ± 2 and 12.2 ± 6 at 5. 15, 25 and 35 °C, respectively; K_d (μM) 28.2 ± 7, 50.8 ± 14, 179 ± 70 and 21 1 ± 126 at 5, 15. 25 and 35 °C respectively.

Experimental

The kinetics of electron transfer from TMADH to ETF have been studied in wild-type and mutant TMADH-ETF electron transfer complexes in order to identify surface residues of TMADH that interact with ETF and potentiate electron transfer. The results of the study indicate that residue Tyrosine 442 of TMADH is the major (i.e. preferential) gateway for electron transfer to ETF.

Methods

Complex bacteriological media were from Difco Laboratoties and all media were prepared as described by Sambrook et al. (1989, Molecular Cloning: a laboratory manual, 2nd ed., Cold Spring Harbor Laboratory, Cold Spring Harbor NY). Ultrapure agarose and cesium chloride were from Life Technologies Inc.. ethidium bromide was from Bachem and Timentin from Beecham Research Laboratories. Sodium dithionite was purchased from Virginia Chemicals, phenvlhvdrazine from Eastman Kodak and toluylene blue from Sigma and Aldrich library of rare chemicals. Ferricenium hexafluorophosphate was synthesised as described by Lehman et al. (1990, Anal. Biochem., 186: 280-284). All other chemicals were analytical grade wherever possible and glass-distilled water was used throughout. Restriction enzymes Eco RI and Hin dill were purchased from Pharmacia. Calf intestinal alkaline phosphatase was from Boehringer Mannheim. T4 DNA ligase and T4 polynucleotide kinase were from Amersham International.

Mutagenesis. plasmid construction and DNA sequencing: Bacteria were cultured in 2YT media supplemented where appropriate with Timentin. Plasmid DNA and bacteriophage RF DNA were prepared by cesium chloride density centrifugation and general cloning methods were adopted from Sambrook et al. ( 1989, supra). Site-directed mutagenesis was performed on a derivative of Ml 3 containing the coding strand of the tmd gene as described previously (Scrutton et al., 1994, J. Biol. Chem., 2_6_9_: 13942-13950). The mutagenic oligonucleotides:

5'-CACGATAATC GCGGTGACCG CTCCACTCAC-3' (Y442G) (SEQ ID NO: 1; 5'-AATCGCGGTG GACCGCTCCA CTCAC-3' (Y422F) (SEQ ID NO: 2); and 5'-CACGATAATC GCGGTGCAGG CTCCACTCAC-3' (Y442L) (SEQ ID NO: 3) were used to isolate bacteriophage constructs containing the desired mutations. Putative mutants were screened directly by dideoxynucleotide sequencing (Sanger et al., 1977, Proc. Natl. Acad. Sci. USA., 74: 5463-5467) using the T7 system supplied by Pharmacia. Each mutant gene was re-sequenced in its entirety to ensure that spurious changes did not arise during the mutagenesis procedure. Each mutant gene was subcloned as an Eco RI/Hin dill fragment into the expression construct pSV2tmdveg (Scrutton et al., 1994, supra) where it replaced the analogous wild-type Eco RI/Hin dill fragment.

Purification of wild-type and mutant proteins:

Recombinant forms of TMADH were prepared from cultures of E. coli strain TGI transformed with the appropriate plasmid expression vector as described previously (Scrutton et al., 1994, supra). TMADH was purified from Methylophilus methylotrophus (W3A1) (Steenkamp, D.J. and Mallinson, J., 1976, Biochim. Biophys. Acta, 429: 705- 719) incorporating the modifications of Wilson et al. ( 1995. supra). The flavin content of mutant enzymes was determined spectrophotometrically (Scrutton et al., 1994. supra). The concentrations of mutant and wild-type TMADH enzymes were determined at 280 nm (e_;80 = 201,610 M^"1 cm^"1). TMADH was modified with phenylhydrazine as described by Nagy, J. et al. ( 1979, J. Biol. Chem., 254: 2684-2688) and Kasprazak, AA. et al. (1983, Biochem J., 211: 535-541). ETF was purified from Methylophilus methylotrophus (W3A1) as previously described (Steenkamp. D.J. and Gallup. M., 1978, J. Biol. Chem., 253: 4086-4089), except that the final gel filtration step was performed using Sephacryl 200-HR. Complete oxidation of ETF was effected by treatment with potassium ferricyanide and desalted by chromatography using Sephadex G25. The concentration of oxidised ETF was determined spectrophotometrically at 438 nm [e₄₃₈ = 1 1, 300 M^'1 cm^"1 (Steenkamp & Gallup, 1978, supra).

Potentiometry:

Potentiometric titrations of the FAD contained within ETF were conducted using the xanthine oxidase method developed by Massey, V. (1990, in Flavins and Flavoproteins (ed. Curti, B., Ronchi. S. & Zanetti, G.) Walter de Gruyter, Berlin, pp 59-66). ETF was contained in 50 mM potassium phosphate buffer, pH 7. Oxidized ETF (35 μM) was made anaerobic in a side-arm cuvette along with 250 μM xanthine. 2 μM methyl viologen and 35 μM toluylene blue, total volume 1 ml; xanthine oxidase (8 μl of a 2.5 μM stock) was placed in the side-arm of the apparatus. Following the achievement of anaerobosis by repeated evacuation and flushing with O₂-free argon, the UV visible spectrum of the mix was recorded using a Hewlett Packard 8452a diode array spectrophotometer. Reduction of the dye was initiated by tipping xanthine oxidase from the side-arm into the mix. Spectra were recorded at 30 second intervals. The extent of reduction of ETF and the dye was determined from absorbance changes at 420 nm and 646 nm, respectively. The mid-point potential of ETF was determined from a plot of log [ETF]_0X/[ETF]_red against log [toluylene blue]_ox/[toluylene blue] J_rr_ιed- Steady-state and stopped-flow kinetic analyses:

Steady-state kinetic parameters were determined at 25 ^CC in reaction mixtures containing 50 mM potassium phosphate, pH 7 and 50 mM trimethylamine. The concentration of TMADH was fixed and that of ETF varied. Reactions were initiated by the addition of trimethylamine. Reaction rates were calculated using a difference molar extinction coefficient between the oxidized and semiquinone forms of ETF (e₄₃₈ = 7830 M^{" 1} cm'¹). Data were fitted to the Michaelis-Menton equation using the fitting program Kaleidograph (Abelbeck software, C A). Stopped-flow experiments were performed using a Kinetic Instruments Inc. stopped-flow apparatus equipped with an On-Line Instrument Systems (OLIS) model 3920Z data collection system. Phenylhydrazine-treated TMADH was placed in a tonometer equipped with a ground glass joint (for the dithionite-titration syringe), a side-arm cuvette and a three-way stopcock valve with a male Luer connector. The solution was made anaerobic by alternately evacuating and flushing with oxygen-free argon, and the iron-sulfur centre of TMADH was reduced (when appropriate) by titration with sodium dithionite. In stopped-flow experiments, the concentration of ETF was at least five-fold greater than that of TMADH. thereby ensuring pseudo first order conditions. Kinetic transients were monitored as transmittance voltage and collected by a high speed A/D converter, and converted into absorbance changes using OLIS software. Reactions were conducted in 50 mM potassium phosphate buffer, pH 7 and monitored at 370 nm. All data were best described by a single exponential (Figure 1 ) and rate constants were obtained by non-linear least squares fitting to kinetic transients using the following relationship (equation 1):

AA =AA e ^'b where AA ₀ represents the total absorbance change, and AA , the observed absorbance change at time /. Where saturation kinetic behaviour was observed, plots of the observed rate constant against [ETF] were constructed and values for the limiting rate constant (k_Um) and the apparent dissociation constant K_d were calculated using Equation 2 (Strickland, S. et al., 1975. J. Biol. Chem., 25.0: 4048-4052):

_k - ^kJjETF] " [ETF] +K_d

Results

Results are shown in Figures 1 -6 and Table 1. The shortest distance from the 4Fe-4S center of wild-type TMADH to the surface of the protein is 1 1.6 A leading to the hydroxyl of Y442 (Figure 2). Tyr-442 is therefore an attractive candidate residue for mediating electron transfer from TMADH to ETF. For this reason, three mutant forms of TMADH were constructed in which the tyrosine residue was exchanged for a phenylalanine (as found in the related dimethylamine dehydrogenase). leucine or glycine residue, and the detailed consequences of these alterations on the electron transfer kinetics to ETF investigated .

In the steady-state, plots of reaction rate against ETF concentration revealed that the reaction of wild-type and mutant forms of TMADH are described by simple Michaelis-Menton kinetics. Steady-state parameters are displayed in Table 1. The determined Michaelis constants are in the range 14.3 μM to 26.2 μM. The values of k^ were found to decrease moderately for the Y442L mutant and substantially for the Y442G mutant. The implication is that removal of the aromatic side-chain at residue 442 has compromised electron transfer from TMADH to ETF. The rate-limiting step for the wild-type enzyme at low substrate concentrations is associated with product release in the reductive-half reaction. For the Y442G mutant enzyme, it appears from the steady-state data that the oxidative half- reaction, i.e. electron transfer to ETF, has become rate limiting.

Electron transfer from TMADH to ETF can be conveniently monitored in the stopped-flow apparatus following modification of the C4a atom of the enzyme-bound flavin with phenylhydrazine (Nagy et al., 1979, supra; Kasprzak et al., 1983, supra). Modification renders the flavin redox-inert, and the 4Fe-4S center can therefore be reduced selectively to the level of one electron by titration with dithionite (Huang, L., et al., 1995, J. Biol. Chem., 22J2: 23958-23965). This procedure simplifies the analysis of the oxidative half-reaction considerably, and has been used to follow the transfer of the electron from reduced wild-type TMADH to ETF (Huang et al., 1995, supra). To extend this work, the temperature dependence of the rate of electron transfer from the reduced 4Fe-4S center to ETF in the wild-type complex was investigated. Pseudo-first order reactions were performed at 5, 15, 25 and 35 °C using a fixed concentration (3.8 μM) of phenylhydrazine-treated TMADH and various ETF concentrations ranging from 18 to 65 μM. The reaction at all temperatures was monophasic and exhibited saturation behaviour with respect to the ETF concentration (Figure 3). The kinetically-determined dissociation constant was similar at each temperature (range 15 to 25 μM), illustrating that increases in temperature over the selected range did not significantly affect complex assembly. The limiting values for the electron transfer at each temperature are given in the legend of Figure 3. An Arrhenius plot of the klim values was linear and yielded an activation energy of 32.6 kJ mol '. From the temperature perturbation data, it is possible to calculate the physical parameters governing the rate of electron transfer and also the tunnelling pathway distance between the 4Fe-4S center and the FAD of ETF. The factors controlling the rate of electron transfer are calculated from Marcus theory (Marcus. R.A. and Sutin, N., 1985, Biochim. Biophys. Acta, Mi: 265-322) which relates the electron transfer rate (k_E7) as a function of free energy change (ΔG°), temperature (T), the electronic coupling matrix element (H_AB), and the sum of the inner sphere (higher vibrational frequency of immediate ligands of the redox centres) and outer sphere (lower vibrational frequency of solvent and protein atoms) reorganisational energies (λ). The equation relating the nuclear and electronic factors to the rate of electron transfer is: (Equation 3)

, - ( Δ G ' + λ )² / 4 λΛ r

and comprises a classical component associated with nuclear motion and a quantum mechanical component (Η^: _ΛB) associated with electron tunnelling. R is the gas constant, h is Planck's constant and T the absolute temperature. Analysis of the temperature dependence of die electron transfer rate using Equation 3 requires knowledge of ΔG° for the reaction. The driving force for the reaction was calculated as -3.66 kJ mol^"1 from the known mid-point potential of the 4Fe-4S center (0.102 V; Barber, M.J. et al.. 1988, Biochem. J., 256: 657-659) and the measured mid-point potential of ETF (0.141V; Figure 4). Analysis of the temperature dependence of electron transfer using Equation 3 (Figure 3) yielded values for the electronic coupling matrix element. H² _AB, of 0.82 cm^"' ± 0.58 and the reorganisational energy, λ, of 136 kJ mol^'1 ± 14 (1.4eV). The tunnelling pathway distance , r, was calculated from Equation 4: ET σ

where k₀ is the characteristic frequency of the nuclei and is assigned a value of IO¹³ s^'1 (Marcus and Sutin, 1985, supra; Rees, D.C. and Farrelly, D., 1990, The Enzymes, Vol. 19, pp 37-96, Academic Press, Inc.. NY), and r₀ represents the van der Waals distance (3 A), β, the electronic decay factor, is a coefficient that relates the decay of the electronic coupling matrix element as a function of distance, r. For homogeneous bridging material, the decay of the electronic coupling matrix element is given by Gamow's tunnelling equation (Equation 5):

H r² = H o² e - ^pr

where H₀ is the tunnelling matrix element at van der Waals separation. In proteins, the precise value of β changes throughout the tunnelling pathway, and depends on various factors, for example whether the bridging structure is made up of covalent, hydrogen bond or 'through space' connectivities (Beratan, D.N. et al., 1991, Science, 252: 1285- 1288; Beratan. D.N. et al, 1992, J. Phys. Chem., 9_6_: 2852-2855). Since in most proteins tunnelling pathways involve all three types of connectivity, the value of β is averaged and approximated to 1.4 A^"1 (Moser et al., 1992, Nature, 3_5_5: 796-802) - a value that is intermediate of that for covalently linked pathways (β = 0.7 A^'1) and for pathways that pass through vacuum (β = 2.8 A^"1). Using a β value of 1.4 A^"1, a tunnelling pathway distance of 1 1.3 A ± 1 was calculated from Equation 4 for the electron transfer event from TMADH to ETF. The kinetically-determined pathway distance (11.3 A) is similar to the distance (1 1.6 A) from the 4Fe-4S center to the hydroxyl of Tyr-442 determined from the crystallographic coordinates of TMADH. The kinetic data collected for the wild-type complex is, therefore, consistent with Tyr-442 being located in the electron tunnelling pathway assuming the FAD of ETF to be in very close proximity (perhaps van der Waals contact) with Tyr-442 in the productive electron transfer complex. Naturally, the length of the tunnelling pathway is under the control of the electronic decay factor β and therefore may change slightly if the properties of the bridging protein are not accommodated in the average value of 1.4 A^"'. In this analysis, it is also important to note that the driving force ΔG° may fluctuate with temperature due to conformational flexibility of the proteins at higher temperatures. However, given the relatively large reorganisational energy calculated for the reaction and also the nature of Equations 3 and 4, any deviations in driving force would have to be substantial to compromise the analysis presented here.

That Tyr-442 resides in the electron tunnelling pathway to ETF was substantiated further by studies of the electron transfer kinetics for mutant TMADH enzymes. As with native enzyme, each mutant enzyme was modified with phenylhydrazine prior to reduction with sodium dithionite and studies of electron transfer rate as a function of [ETF] were conducted at 25 °C (Figure 5). The behaviour of the Y442F and Y442L mutants were found to be similar to the behaviour of the wild-type enzyme (k_hm 1.4 and 2.2 fold lower than wild-type), but the limiting electron transfer rate for the Y442G mutant was found to be substantially lower than that for the wild-type (30 fold reduction in k_Um). The kinetically-determined dissociation constant for the Y442G mutant TMADH was also found to be about one order of magnitude higher than the same parameter for the wild-type, Y442F and Y442L mutant enzymes.

The substantial reduction in electron transfer rate for the Y442G mutant enzyme and the higher dissociation constant measured at 25 °C was subsequently followed by a temperature perturbation analysis of electron transfer for this enzyme. As for the wild-type complex, stopped-flow experiments were conducted at 5, 15, 25 and 35 °C. Over this temperature range, k_im rose from 0.5 s^"1 to 12.2 s^'1, but perhaps more importantly the dissociation constant for the complex was raised by one order of magnitude (Figure 6). At 5 °C, the dissociation constant for the Y442G-ETF complex is similar to the dissociation constant for the wild-type TMADH-ETF complex at all temperatures studied and the Y442F-ETF and Y442L-ETF complexes studied at 25 °C. The effect of mutating Tyr-442 to a glycine residue is, therefore, to impair complex assembly (ΔG_Y442G - ΔG_WT ⁼ 5.88 kJ mol^{" 1} at 35 °C) in addition to compromising the electron transfer rate to ETF. Values for r, H_AB and λ for the Y442G mutant enzyme could not be calculated because, unlike for the wild-type complex, the dissociation constant for the Y442G-ETF complex was temperature dependent. Consequently, the reorganisational energy will not be constant over the temperature range studied. A change in the structure of the productive electron transfer complex as a function of temperature may also alter the tunnelling pathway distance. For these reasons, it was judged inappropriate to analyse the data in the manner described for the wild-type complex.

Discussion

Hence residue Y442 of wild-type TMADH is the preferential point from which electrons are transferred to the electron acceptor ETF. Modification of TMADH may allow the coupling (bonding) of an electron acceptor to TMADH or may be used to optimise the TMADH-electron acceptor interactions. Table 1 - Steady-state kinetic parameters of wild-type and mutant forms of TMADH. The kinetic values are apparent, and measured at a trimethylamine concentration of 50 μM

Enzyme K_m (ETF) (μM) £_cat(s^'')

Wild-type 17.5 ±2 16.5 ±0.8

Y442F 14.3 ±3 10.3 ±0.9

Y442L 19.6 ±4 3.8 ±0.3

Y442G 26.2 ± 5 0.25 ± 0.02

SEQUENCE LISTING

(1) GENERAL INFORMATION (l) APPLICANT-

(A) NAME: The University of Leicester

(B) STREET. University Road

(C) CITY. Leicester

(E) COUNTRY UK

(F) POSTAL CODE (ZIP) : LEI 7RH

(ii) TITLE OF INVENTION- Enzyme-Electron Acceptor Assembly

(ill) NUMBER OF SEQUENCES- 3

(iv) COMPUTER READABLE FORM:

(A) MEDIUM TYPE. Floppy disk

(B) COMPUTER IBM PC compatible

(C) OPERATING SYSTEM. PC-DOS/MS-DOS

(D) SOFTWARE Patentln Release #1 0, Version #1 30 (EPO)

(vi) PRIOR APPLICATION DATA.

(A) APPLICATION NUMBER GB 9610376.7

(B) FILING DATE: 17-MAY-1996

(2) INFORMATION FOR SEQ ID NO: 1:

(l) SEQUENCE CHARACTERISTICS:

(A) LENGTH. 30 base pairs

(B) TYPE- nucleic acid

(C) STRANDEDNESS: single

(D) TOPOLOGY: unknown (xi) SEQUENCE DESCRIPTION- SEQ ID NO: 1-

CACGATAATC GCGGTGACCG CTCCACTCAC 30

(2)- INFORMATION FOR SEQ ID NO- 2.

(l) SEQUENCE CHARACTERISTICS.

(A) LENGTH 25 base pairs

(B) TYPE, nucleic acid

(C) STRANDEDNESS single

(D) TOPOLOGY unknown

(xi) SEQUENCE DESCRIPTION SEQ ID NO: 2:

AATCGCGGTG GACCGCTCCA CTCAC 25

(2) INFORMATION FOR SEQ ID NO. 3.

(l) SEQUENCE CHARACTERISTICS.

(A) LENGTH 30 base pairs

(B) TYPE: nucleic acid

(C) STRANDEDNESS single

(D) TOPOLOGY unknown

(xi) SEQUENCE DESCRIPTION SEQ ID NO- 3-

CACGATAATC GCGGTGCAGG CTCCACTCAC 30

Claims

1. An enzyme-electron acceptor assembly comprising TMADH or a partially modified form thereof or a phenotypic mutant thereof and an electron acceptor, the electron acceptor being chemically bonded to the enzyme and the assembly being arranged such that upon catalysis of a reaction by the enzyme, electron transfer occurs from the enzyme to the electron acceptor.

2. An enzyme-electron acceptor assembly according to claim 1 wherein the electron acceptor is covalently bonded to the enzyme.

3. An enzyme-electron acceptor assembly according to any one of the preceding claims wherein the electron acceptor is other than ETF.

4. An enzyme-electron acceptor assembly according to any one of the preceding claims wherein the electron transfer to the electron acceptor occurs from residue 442 of the enzyme.

5. An enzyme- electron acceptor assembly according to claim 4 wherein residue 442 is cysteine.

6. An enzyme-electron acceptor assembly according to any one of the preceding claims wherein die electron acceptor is a redox acceptor.

7. An enzyme-electron acceptor assembly according to any one of the preceding claims wherein the electron acceptor is a flavin.

8. An enzyme-electron acceptor assembly according to claim 7 wherein the electron acceptor is 8-C 1-FAD.