WO2006114616A1 - Engineered peroxidases with veratryl alcohol oxidase activity - Google Patents

Abstract

Peroxidases engineered to replace substitute an amino acid residue corresponding to lignin peroxidase tryptophan 171 with a tryptophan, and wherein at least two acid residues are in close proximity, are capable of oxidising lignin and can be used in the paper industry and the treatment of pollutants containing PCBs, for example.

Description

ENGINEERED PEROXIDASES WITH VERATRYL ALCOHOL OXIDASE ACTIVITY

The present invention relates to peroxidases and their uses.

Peroxidases are a widespread group of enzymes that are present in many organisms and have a wide variety of physiological roles. Most peroxidases are metalloenzymes and a large proportion of these contain haem (Smith & Veitch, 1998). They are used in immunoassay systems as a reporter enzyme and have been extensively patented for biobleaching, chemical polymerisation and catalysis of 'difficult' oxidative chemistry. Lignin peroxidase is notable for being able to oxidise highly electropositive substrates with half potentials >1.4V (Pasti-Grigsby et al, 1992). It is one of only three enzymes implicated in the destruction of lignin by wood rot fungi and can oxidise even recalcitrant environmental pollutants such as PCBs.

To fully realise the use of renewable materials such as cellulose and lignin as sources of fuels and chemicals, clean technologies for lignin bleaching are required. Current technologies for paper bleaching utilise hypochlorite, and result in the production of polychlorobenzenzenes (PCB's) that have extensively contaminated the environment in several countries. The use of so-called 'green technologies', utilising enzymes to bleach both lignin and other environmental pollutants including waste dyes from the dyeing industry, are urgently required.

Although lignin peroxidase enzymes from both brown and white wood rot fungi have been proposed for this purpose, they are difficult to produce in quantity on any kind of commercial scale, and are relatively sensitive to inactivation, unfolding and proteolysis during production or under application-based conditions. Coprinus cinereus peroxidase is very similar in protein fold to lignin peroxidase, but does not have the ability to oxidise very electropositive substrates, such as veratryl alcohol. The gene for this enzyme has been cloned by Novozymes (formerly, Novo Nordisk) and successfully over-expressed in a heavily protected Aspergillus orsaryze expression system at g/litre levels. PCT/DK93/00189 discloses cloning the enzyme, although the only commercial use has been in the context of denim bleaching and has not found widespread application in the market place.

The general reaction catalysed by haem peroxidases involves a 2e~ oxidation of the peroxidase to yield Compound I. Compound I then undergoes two Ie" reactions with a reducing substrate. The first of these reactions produces a second intermediate, Compound II. A second reaction with a reducing substrate returns the peroxidase to its resting state (Smith & Veitch 1998).

Haem peroxidases are divided into two broad superfamilies; mammalian peroxidases such as prostaglandin H synthase, and plant peroxidases, which include those from plants, fungi, and bacteria. There is also a third, less defined, group that includes chloroperoxidase (Smith & Veitch, 1998). Plant peroxidases can be divided into three classes: Class I contains yeast cytochrome c peroxidase and gene duplicated bacterial peroxidases. Class II includes peroxidases from fungi, while those from higher plants fall into Class III (Welinder, 1992). Coprinus cinereus peroxidase (CIP) is a Class II haem peroxidase produced by a black ink cap mushroom from the Basidiomycete family. CIP forms conventional Compounds I and II and follows the general model of peroxidase action as shown in Figure 1 {supra, Dunford, 1999).

CIP is characterised by high stability and turnover with traditional peroxidase substrates and its general kinetic characteristics are similar to those of horseradish peroxidase. CIP is the only known peroxidase produced commercially from a recombinant system and is closely related evolutionarily to another peroxidase, lignin peroxidase (LiP) which is one of three enzymes known to have a role in the break down of lignin.

Fungal peroxidases have been the focus of significant interest in the last two decades, mainly due to their possible industrial and environmental applications. LiP is a constituent of the ligninolytic cycle that involves an assortment of enzymes including manganese peroxidase (MnP), laccase and H₂O₂ producing enzymes. Enzymes of the ligninolytic cycle are produced by many wood-degrading fungi, the best known of which being Phanaerochaete chrysosporium the white wood rot degrading fungus. The ligninolytic cycle is responsible for the degradation of the complex biopolymer lignin, which is found as a component of plant cell walls.

Lignin is a renewable non-phenolic aromatic polymer, consisting of phenol propanoid units linked by a variety of non-hydrolysable C-C and C-O bonds. Lignin is, therefore, an abundant source of carbon and plays a vital role in the carbon cycle, its degradation being the rate limiting step (Piontek et al, 2001). The structure of lignin implies that LiP and other enzymes of the ligninolytic cycle are able to breakdown substrates with high redox potentials and it is now known that LiP is able to oxidise aromatic compounds with redox potentials in excess of 1.4 V (Pasti-Grigsby et al, 1992). Therefore, LiP may also be able to oxidise a whole host of environmental recalcitrant pollutants including DDT and other pesticides, dyes, munitions, cyanides, cross-linked acrylic polymers and polycyclic and chlorinated aromatic compounds.

Given the large size of lignin and its potential oxidation by LiP, it becomes apparent that the active-site channel of LiP is not of sufficient size to allow the large polymer lignin direct access to the haem. This has been studied using a substrate of LiP, the small molecule VA (veratryl alcohol or 3,4-dimethyloxybenzyl alcohol) which is a secondary metabolite of P. chrysosporium and other wood-degrading fungi, produced at the same time as LiP (Piontek et al, 2001). The direct oxidation of VA by LiP involves a one- electron oxidation to the VA cation radical (VA⁺) and results finally in the formation of veratryl aldehyde among other products (Dunford 1999). It has been proposed that the VA^' acts as a diffusible oxidant, oxidising other substrates and in turn regenerating VA (Harvey et al 1986). More recent work has also suggested the formation of a LiP- VA⁺ complex (Khindaria et al. 1996). VA is known also to act as a mediator for the oxidation of compounds that would normally not be substrates of LiP, such as chlorpromazine, guaicol, 4-methoxymandelic acid and dimethoxylated aromatics (Goodwin 1995).

It has recently been found that Wl 71 is hydroxylated at it Cβ atom in the fungal LiP X-ray structure (Choinowski et al. 1999). This post-translational modification has been shown to be a result of auto-catalysis which occurs under turnover conditions and highlights this residue as important (Blodig et al. 1998). The role of this tryptophan was examined by site-directed mutagenesis and it was found that substitution by a serine or phenylalanine residue led to a complete loss of activity with respect to VA, but not with the two dye substrates, ABTS (2,2'-azinobis-(3-ethylbenzthiazoline-6-sulphonic acid)) or DFAD (4- [(3,5-difluor-4-hydroxyphenyl)azo]benzene-sulphonate) (Doyle et al, 1998). A charge neutralisation mutation in the "classical haem edge", active site channel of LiP, however, did not lead to significant loss of activity with respect to VA.

These experiments offered the first direct evidence for the presence of two substrate interaction sites in LiP: - one at the haem-edge for small dye substrates e.g. DFAD and the other site positioned around Wl 71 for substrates such as VA. Comparison of the crystal structures of the pristine (non-hydroxylated) form, the Cβ-hyrdroxylated form and of the W171F mutant of LiP revealed inconsequential structural differences between the enzymes (Blodig et ah, 2001). From these findings it has been concluded that the absence in VA oxidation activity of the Wl 7 IF/ Wl 71 S mutants is due to the lack of the redox active indole side-chain. Experiments have also shown that the fast conversion of Compound I to Compound II observed in wild-type LiP was absent in the Wl 71 mutants (Doyle et al, 1998). It was, therefore, suggested that the enzyme may contain an electron-transfer pathway from a surface site in close proximity to Wl 71 to the haem. It has since been shown, using spin-trapping experiments, that Wl 71 is redox active and forms an indole radical by transfer of an electron to the haem of Compound I and/or II (Blodig et al, 1999). The straightest path from a carbon atom of the haem to an indole carbon of tryptophan 171 is l lA and from other systems it is known that biological electron transfer over this distance is feasible (Blodig et al, 1999).

Apart from LiP no other examples have been found in which a tryptophan cation radical is directly involved in the oxidation/activation of a substrate (note, a Plerotus eryngii peroxidase containing a tryptophan residue at a homologous position to tryptophan 171 appears to be both a lignin and manganese peroxidase (Perez-Boada et al. 2002)). Tryptophan 171 in LiP is surrounded by four surface negatively charged amino acid residues, D165, E168, E250, D264. Mutation of E168, E250 and D264 suggest that these negative charges may help stabilise the VA⁺, as the K_m for VA is significantly increased over that of the wild-type, in site-directed mutants at these positions and electron transfer rates (k_cat) are decreased, particularly if two charges are removed.

Surprisingly, we have now found that substituting the aspartate at position 179 of Coprinus cinereus peroxidase (CIP) with tryptophan, together with an increase in the acidity of the local environment of this residue, allows the enzyme to oxidise veratryl alcohol (VA), in a manner similar to that of lignin peroxidase (LiP). Thus, it is possible to engineer an enzyme, producible on an industrial scale to provide lignin peroxidase activity. . Coprinus cinereus peroxidase would not normally be regarded has having a sufficient oxidising potential to achieve this kind of chenmistry.

Accordingly, in a first aspect, the present invention provides a peroxidase capable of oxidising veratryl alcohol, wherein said peroxidase is a peroxidase not normally capable of oxidising veratryl alcohol, and wherein a residue equivalent to tryptophan 171 in LiP, but which is not tryptophan, is substituted by tryptophan, and, if necessary, providing one or more acidic amino acid residues such that there are at least two acidic amino acid residues in sufficient proximity to the indole ring of the tryptophan to be able to enhance the stability of any charge on the indole ring and/ or substrate or intermediate formed therewith.

In the case of CIP, the tryptophan residue was inserted at position 179 in the CIP protein sequence, which is equivalent to the tryptophan at position 171 in LiP, this being the redox active residue associated with VA oxidation.

Thus, we are able to provide lignin peroxidase activity in a commercially valuable enzyme that was previously incapable of catalysing this chemistry.

The original peroxidase will not normally be capable of oxidising VA to any significant degree, by which is meant that such activity is not measurable, or is less than 1% of LiP. For the avoidance of doubt, CIP is a peroxidase not normally capable of oxidising veratryl alcohol.

The peroxidase is preferably a plant peroxidase, and is preferably a Class II peroxidase. The most preferred peroxidase is from Coprinus cinereus, and is referred to herein as CIP.

In preferred peroxidases, a glutamate residue is already present in close proximity to the position selected for change to the redox active tryptophan. For example, in CIP, glutamate is present at position 176, which is three residues removed from the aspartate present at position 179 that is changed to a tryptophan.

It is further preferred in CIP to replace one or both arginine residues at positions 258 and 272 with an acidic amino acid residue. The residue may be either glutamate or aspartate, and may be the same or different for each position. It is most preferred, in CIP₅ that position 258 is substituted with a glutamate, while position 272 is substituted with an aspartate, these residues corresponding to those in LiP.

It is particularly preferred that the arginine at position 258 be replaced by glutamate, as there is evidence that glutamate interacts, via H-bonding, with the tryptophan indole.

It is further preferred that serine at position 173 of CIP is substituted by aspartate. In LiP₃ four acidic amino acid residues surround the redox active tryptophan, and such a substitution corresponds to the forth acidic amino acid present in LiP.

It is preferred that there are at least 3, and preferably 4, acid residues in proximity to the tryptophan residue.

Thus, in a preferred aspect, the present invention provides CIP, or a modified version thereof, substituted at position 179 with tryptophan and at one or more of positions 173, 258 and 272 with an acidic amino acid residue. Preferably, position 258 is substituted with a glutamate, and positions 173 and 272 are substituted each with an aspartate.

CIP, with the preferred mutations at positions 179, 258 and 272 is illustrated herein as SEQ ID NO: 1 (coding sequence) and SEQ ID NO: 2 (protein sequence). Note that the synthetic sequence is six residues shorter in numbering compared to the mature fungal sequence. AU residue positions (e.g. 179, 258 and 272) referred to herein are reported using the mature sequence numbers. By "modified version" is meant CIP which has been modified either at the peptide level or nucleic acid level to facilitate handling or expression, for example. Such modifications are within the skill of those in the art, and are routinely used to facilitate cloning or protein expression, for example. It will also be understood that peroxidases having at least 80% sequence identity with that of SEQ ID NO: 1 are encompassed by the present invention, provided that the necessary peroxidase activity is present. Sequence identities in excess of 85%, 90% and 95% are preferred.

It will be appreciated that the sequence illustrated herein is characteristic of one embodiment of the invention. The invention further provides variants and mutations of naturally occurring peroxidases mutated to provide veratryl alcohol oxidising ability. Mutations may be any suitable mutations, and include deletions, insertions, inversions and substitutions, always provided that the resulting enzyme has the required peroxidase activity. Variants are naturally occurring sequences which vary from one another but still have the necessary peroxidase activity.

The necessary activity is generally a measurable veratryl alcohol oxidising activity, with preferred levels being at least 10% that of lignin peroxidase. A preferred measuring technique is as provided in the accompanying Examples. Such levels are readily reached in CIP when positions 179, 258 and 272 are substituted. Higher levels may also be reached when position 173 is substituted.

It will be appreciated that residues corresponding to positions 173, 176, 179, 258 and 272 are readily discernable in other plant peroxidases by protein sequence comparison. For example serine 173 of CIP corresponds to position 165 of LiP.

In addition, at the veratryl alcohol interaction site, it is preferred that positions 174, 175 and 275 of CIP, and the corresponding positions of other peroxidases, are substituted with glutamate, leucine and phenylalanine, respectively. Also, it is preferred that a further residue(s) is(are) inserted between residues 171 and 172 of CIP, and the equivalent position in other peroxidases.

Peroxidases of the present invention are generally producible in high quantities, especially where an existing commercially produced peroxidase can be modified in accordance with the present invention. Thus, the present invention further provides DNA encoding a peroxidase as defined above. Further provided is a vector comprising said DNA, especially where said vector is an expression vector. Hosts comprising said vectors are especially useful, and cultivation of said hosts to provide expression of peroxidases of the invention is a preferred method for obtaining the peroxidases. Subsequent isolation and/or purification of the peroxidases is desirable, but not always necessary, prior to use. In addition, any X-ray crystal structure determined from a peroxidase as defined above is provided.

The peroxidases of the present invention may be used in various areas, but are preferably used in areas such as in the oxidation of environmental pollutants, especially PCB 's, in biobleaching, such as in paper manufacture, and in treating waste dyes from the dyeing industry.

In the above uses, the peroxidases may be formulated in any suitable manner. Although the peroxidase may be provided and stored in the form of a solution, it is generally preferred to store the peroxidase in the form of a dry powder, optionally together with one or more stabilisers, for example.

Comparison of the crystal structures of the pristine (non-hydroxylated) form, the Cβ- hydroxylated form and of the W171F mutant of LiP isoenzyme H8 revealed inconsequential structural difference between the structures. From these findings it has been concluded that the absence in VA oxidation activity of the W171F/ W171S mutant is due to the lack of the redox active indole side-chain (Blodig et al, 2001). Experiments have also shown that the fast conversion of Compound I to Compound II observed in wild-type LiP was absent in the tryptophan 171 mutants (Doyle et al, 1998). It was, therefore, suggested that the enzyme may contain an electron-transfer pathway from a surface site in close proximity to tryptophan 171 to the haem (Doyle et al, 1998). It has since been shown, using spin-trapping experiments, that tryptophan 171 is redox active and forms an indole radical by transfer of an electron to the haem of Compound I and/or II (Blodig et al, 1999). The straightest path from a carbon atom of the haem to an indole carbon of tryptophan 171 is 11 A and from other systems it is known that biological electron transfer over this distance is feasible (Blodig et al, 1999). Apart from LiP no other examples have been found in which tryptophan^"1" is directly involved in the oxidation/activation of a substrate (a Plerotus eryngii peroxidase containing a tryptophan residue at a homologous position to tryptophan 171 appears to be both a lignin and manganese peroxidase (Perez-Boada et al 2002)). Tryptophan 171 in LiP is surrounded by four surface negatively charged amino acid residues, D165, E168, E250, D264. Mutation of E 168 and D264 suggest that these negative charges may help stabilise the VA⁺, as the K_m for VA is significantly increased over that of wild-type and electron transfer rates (k_cat) are decreased.

Another enzyme that is similar to LiP, but which is mainly a manganese peroxidase, has been shown to have weak VA activity. This so-called "versatile peroxidase" possesses a tryptophan residue equivalent to Wl 71 that gives the enzyme its weak VA activity (Ruiz-Duenas et. al. , 2001).

Figure 2 shows that the tryptophan of LiP fits well into the proposed mutation site in CIP, ^■ D 176. This fit is further enhanced by replacing the two sterically hindering, surface arginine residues (CIP R258 and R272) with the appropriate negatively charged residues, E and D, required for optimal VA oxidation in LiP.

Also provided according to the present invention is a method of oxidising veratryl alcohol, comprising the steps of:

(i) contacting said veratryl alcohol with a peroxidase according to the present invention; and

(ii) allowing said peroxidase to catalyse oxidation of said veratryl alcohol.

Preferably, the veratryl alcohol is contacted with the peroxidase under suitable reaction conditions for the peroxidase. Appropriate reaction conditions are detailed further herein and will be readily apparent to one of ordinary skill in the art.

Also provided according to the present invention is a method of determining the effect of at least one substance upon the catalytic activity of a^' peroxidase according to the present invention, comprising the steps of: (i) contacting a veratryl alcohol with said peroxidase and said at least one substance; (ii) measuring the catalytic activity of said peroxidase in the oxidation of said veratryl alcohol; and

(iii) comparing the results of step (ii) with control results in order to determining the effect of said least one substance upon said catalytic activity of said peroxidase.

Preferably, the control results are obtained by measuring the catalytic activity of said peroxidase in the oxidation of said veratryl alcohol in the absence of said at least one substance.

The at least one substance can be an agonist or an antagonist of the catalytic activity of the peroxidase. Alternatively, it can have no effect upon the catalytic activity of the peroxidase. The reaction conditions can also be varied, for example the temperature, pH or other conditions can be varied and the method can determine the effect of the at least one substance upon catalytic activity of the peroxidase under those conditions. For example, the at least one substance might enhance the pH tolerance of the peroxidase or extend its operating temperature, for example by stabilising it at high temperatures.

The measuring of the catalytic activity of the peroxidase can, of course, be done simply by measuring oxidation of the veratryl alcohol.

The invention will be further apparent from the following description with reference to the figures of the accompanying drawings which show, by way of example only, forms of proteases. Of the Figures: Figure 1 shows a general peroxidases reaction scheme. Compund I (C-I) is top-right, Compound II (C-II) is bottom, and the native state (N S) is shown top-left;

Figure 2 shows a superposition of LiP and CiP in the region of LiP W171.

Structural superposition was on the basis of matched residues from the amino acid sequence alignment;

Figure 3 shows a WPAM site-directed mutagenesis scheme. Starting with the

6.2kb pFLAGl + CIP at "A", Primers 1 and 2 are used to generate a linear PCR product ("B"), which is then ligated ("C") to generate a new circular structure ("D") with a new restriction site (10) created by mutagenesis;

Figure 4 shows VA assays with varying hydrogen peroxide concentrations in a time course graph. X-axis shows time (seconds). Y-axis (left) shows Abs. Y-axis (right) shows H₂O₂ concentration (μM);

Figure 5 shows VA assays with varying hydrogen peroxide concentrations in a time course graph. X-axis shows time (seconds). Y-axis (left) shows Abs. Y-axis (right) shows VA concentration (mM);

Figure 6 shows a comparison of enzyme effectiveness with veratryl alcohol as substrate, Wild type LiP₅ D179W and D179W:R258E:R272D

CIP. On the plot, Y-axis shows turnover (s^"1) and X-axis shows VA (mM). On the bar chart, Y-axis shows turnover (s^"1); and

Figure 7 shows the crystal structure of recombinant LiP H8 (ref [ I]) showing the acidic environment of Trp- 171 and residues chosen for mutagenesis.

EXAMPLES

Materials and Methods

The CiP synthetic gene

The synthesis of the CiP synthetic gene is known in the art. Briefly, the mature sequence was taken from the database and poorly translated codons modified to standard E. coli I yeast codon usage. The gene was constructed from over 20 overlapping oligonucleotides by recursive PCR. The DNA sequence was checked and then 6 independent errors were repaired by repetitive site-directed mutagenesis to restore the wild-type sequence. The first 50 codons were optimised for E. coli expression and the first 5 residues of the N-terminus were removed as these were disordered in the structure and had a high GC content when encoded at the DNA level. Two GC rich islands at the N-terminus were subsequently removed in order to optimise the gene for E. coli expression. After final confirmation of DNA sequence the gene was cloned in the E. coli expression vector pFLAGl .

Site directed mutagenesis

The wild-type CIP gene was cloned within the pFLAGl plasmid, which was used as a template for first round site-directed mutagenesis that would introduce a tryptophan residue at position 179 (equivalent to the tryptophan at position 171 in LiP). A new recognition sequence for the restriction endonuclease BspEI was inserted to facilitate selection of mutant genes. Due to problems finding unique restriction enzyme sites for cassette based mutagenesis, the whole plasmid amplification method (described in Doyle et al. , 1998) (WPAM) was used. WPAM is illustrated in Fig 3. Primers were designed so that one primer encoded half the BspEI site and the other being encoded on the second primer. This resulted in a linear PCR product that when correctly self ligated should cut once with BspEI. Pfu polymerase was employed, as it produces fewer errors than taq polymerase as it has proofreading capabilities that taq polymerase does not, which reduces the error rate compared to taq by a factor of 5-12 fold. Two identical PCR reactions were performed, each tube containing the following:

lμl Template DNA: pFLAGl CIP

5μl Each Primer: RTCIP3 and RTCIP4

5μl dNTPs: From a mixed ImM stock

5μl Buffer: Pfu polymerase buffer

28.5μl ddH^O: To make up a final volume of 50μl

The two tubes were placed in the PCR machine and heated to 95°C for 5 minutes, and the machine was then paused to allow the addition of:

0.5μl Pfu polymerase

The reaction then continued for 25 cycles consisting of:

95°C for 1 minute: Allows denaturation of template DNA

60⁰C for 1 minute: Allows primers to anneal to template DNA

72°C for 16 minutes: Allows extension of new DNA by Pfu polymerase.

Upon completion of the 25 cycles, the reaction was held at 72°C for ten minutes to allow any proof reading reactions to complete, after which it was cooled to 4°C prior to analysis or storage at -20°C.

Down Stream Processing of PCR Product

Samples of the reaction were run on an agarose gel. l μl of Ficoll gel loading buffer was added to 5μl of each of the two PCR mixtures. These were loaded onto a 1 % agarose gel and run alongside a lkb ladder marker. The gel was then stained with ethidium bromide and photographed in the presence of UV light. To process the PCR product l μl of Dpnl was added to each tube to destroy wild-type (methylated) copies of the gene. This was incubated at 37°C for 1 hour. The contents of both tubes were then transferred to a 1.5ml microfuge tube and the volume was made up to 150μl with ddH2θ. I SOμl of Phenol:Chloroform:Isoamyl alcohol (25:24: 1) was added to form an emulsion. This was then put in the microfuge at 13000 rpm for 2 minutes. The resulting top, aqueous, layer was transferred to a fresh tube and mixed with an equal volume of chloroform in order to remove traces of phenol. The mixture was again spun at 13000 rpm for 2 minutes and the aqueous layer collected. To adjust the concentration of monovalent cations to 0.3M Na, 3M Na acetate was added as 1/9 of the sample volume i.e. 16.7μl of Na acetate in 150μl sample. The DNA could then be recovered using ethanol precipitation.

300μl of ice cold ethanol was added and the mixture left on ice for 15 minutes. Following this it was centrifuged at 13000 rpm for 10 minutes. The supernatant was removed and 500μl of 70% ethanol was added. This was then centrifuged at 13000 rpm for 5 minutes and the supernatant discarded. The open tube was dried in the laminar flow hood and the pellet re-suspended in l Oμl of ddE^O.

For the kinase step, the following mixture was incubated at 37°C for an hour:

8μl extracted DNA

l μl 1 OX ligase buffer, containing ATP

l μl T4 PNK

The linear DNA was then recircularised during the ligase step. For this the following were incubated overnight at 4°C:

l Oμl kinase step mixture

l μl l OX ligase buffer

l μl T4 DNA ligase

8μl (IdH₂O

Sequences

In the above WPAM site-directed mutagenesis, using WT CIP as a template and changing D 179 to W179 (equivalent to W171 in LiP), the section of the pFLAGl plasmid to be mutated is given in SEQ ID NO : 3. C24T and T27G mutations do not alter the sequence of the translation product, but do result in a BspEl restriction site (TCCGGA). G40T, A41 G and C42G mutations transform an encoded Asp (D) residue (D 179) to a Trp (W) residue (W 179). BspEl cuts between C26 and G27.

RTCIP3 (D 179W, 27mer, Tm = 66 ⁰C) has the sequence of SEQ ID NO: 5

RTCIP4 (reference, 21mer, Tm = 68 ⁰C) has the sequence of SEQ ID NO: 6

BspEl does not cut in the CIP gene but does cut pFLAGl once. Therefore, if the new BspEl site is present then a 765bp size would be expected.

Transformation of DH5oc Competent cells

lOμl of DNA was added to l OOμl of DH5oc competent Escherichia coli cells that were then left on ice for 10 minutes. The cells were then heat shocked at 42°C for 1 minute then returned to the ice for 5 minutes. I mI of L-broth was added to the mixture, which was then left to shake at 37°C for 1 hour, to allow expression of the ampicillin resistance gene on the pFLAGl plasmid. The cells were then centrifuged at 6000 rpm for 2 minutes and most of the supernatant discarded. The remaining supernatant, containing the DH5oc cells, was plated out onto ampicillin containing L-agar. The plate was then incubated overnight at 37⁰C.

Purification of plasmid DNA

12 individual colonies from the plate were selected and placed in separate tubes each containing 5ml of L-broth and 5μl of 100mg/ml ampicillin stock. These tubes were grown up overnight in the shaker at 37°C. The plasmid DNA from each of these 12 colonies was then purified using the QIAprep miniprep DNA purification system.

Digestion with Bsp EI

For each of the 12 DNA samples the following restriction digests were prepared:

7μl DNA

l μl 1 OX BSA

l μl 1 OX NEB buffer 3

I μl Bsp EI

Also, for each sample, an undigested control was included.

This contained the same ingredients as above except the Bsp EI, which was replaced with ddH^O. All the samples were then incubated in a water bath at 37°C for 1 hour. 2μl of Ficoll loading buffer was added to all 24 tubes, which were then run on a 1 % agarose gel against a 1 Kb ladder marker DNA. The gel was then stained and photographed as before (results not shown).

Purification of Plasmid DNA for sequencing

Two colonies that had successfully been cut were chosen for scaled up DNA preparation by the midi prep procedure. Cells from the small overnight bottles were transferred to larger bottles containing 50ml of L- broth and 50μl of 100mg/ml Ampicillin. These were grown overnight in the shaker at 37⁰C. The DNA was then purified using the Promega wizard plus midi prep system.

Calculation of DNA concentration

The absorbance of each DNA sample was measured with a spectrophotometer, using Tris buffer, pH8.0, to set the baseline. The absorbance of l Oμl of DNA in 490μl of Tris was measured at 260 and 280 nm wavelengths. These were used to calculate the DNA concentration.

Preparation of DNA for sequencing

One of the previously purified DNA samples was selected for DNA sequencing, owing to the size of the CIP gene, two samples were sent so that sequencing could be performed from both the 5' and 3' ends using universal and forward and reverse primers to obtain the full sequence. To each of two tubes the following was added:

8μl DNA

42μl dd H₂O

12.5 μl 1 OM ammonium acetate

125μl ice cold ethanol

The mixtures were then left on ice for 15 minutes. Next, the tubes were centrifuged for 10 minutes at 13000 rpm and the supernatant discarded. 500μl of 70% ethanol were added to each tube and they were centrifuged at 13000 rpm for 5 minutes. The supernatant was then discarded and the pellet dried out in the laminar flow hood. The tubes were then labelled and sent for sequencing.

Transformation of W1130 competent cells Wl 130 cells were transformed with mutant plasmid DNA, using the same protocol as for DH5oc cells, to verify that the protein could still be expressed. A second sample of Wl 130 cells was transformed with DNA for wild type horseradish peroxidase (HRP), which encodes a protein of an equivalent size to CIP. These cells then served as a positive control as it was known that they would express protein if the experiment were performed correctly. Three 5ml flasks of L-broth were prepared with 5 μl of 100mg/ml ampicillin stock. Cells transformed with mutant CIP DNA were added to two flasks and the other contained cells transformed with wild type HRP. The wild type and one mutant-containing flask were induced with 20μl of a 25OmM IPTG stock. The other mutant was not induced and so served as a negative control.

Preparation of a protein extracted from E. coli inclusion bodies

1.5ml of each of the three culture samples were centrifuged at 6000 rpm for 2 minutes. The cells were then resuspended in 500μl of 2OmM Tris, pH 8.0,

ImM EDTA, I mM DTT and then sonicated. Each extract was centrifuged at

13000 rpm for 15 minutes. The supernatant, containing any soluble proteins was removed and frozen for later use. Another 500μl of the above buffer was added to the pellet, which was then sonicated and left at room temperature for 10 minutes. The extracts were then centrifuged at 13000 rpm for 15 minutes and the supernatant discarded.

The pellet was resuspended in 30μl 5OmM Tris, pH 8.0, I mM EDTA, 3OmM DTT, 1 OM urea and centrifuged at 13000 rpm for 10 minutes to remove the insoluble membrane proteins. The solubilised ligninase containing supernatant was then removed and the pellet discarded. The three samples, together with the three earlier samples were then analysed on a 12% SDS- PAGE gel (results not shown).

PCR based second round mutagenesis to introduce R258E-.R272D into the D179W background The mutant plasmid DNA made in the first round was used as a template for a second round of site directed mutagenesis. The reaction was performed as before except that instead of primers RTCIP3 and RTCIP4, the primers RTCIP5 and RTCIP6 were used. Due to the high annealing temperature predicted for the primers, annealing and extension were performed in one 17 minute step at 72⁰C. The PCR product was then processed and used to transform DH5cc cells as described above.

Twelve mini preps were set up using the previously described protocol, and were run on an agarose gel. Two of the colonies that produced successful minipreps were then set up as midipreps. One sample was then prepared for sequencing as before. The triple CIP mutant DNA was then used to transform Wl 130 cells, ready for protein expression.

Sequences

Using the WPAM method and D179W CIP (above) as a template, R258E and R272D (equivalent to W171 and surrounding acidic environment in LiP) are changed.

The relevant section of the D 179W CIP sequence is given in SEQ ID NO: 7. The primers used to achieve the desired changes are RTCIP5 and RTCIP6 (below). The R258E mutation is achieved by the following mutations: C 13G, GHA and T15G. C30T and C33T mutations do not alter the sequence of the translation product, but do result in a Spel restriction site (ACTAGT). C55G and G56A mutations transform an encoded Arg (R) residue (R272) to an Asp (D) residue (D272). Spel cuts between T30 and A31.

RTCIP5 (R272D, 36mer, Tm = 100 ⁰C): SEQ ID NO: 9

RTCIP6 (29mer, Tm = 80 ⁰C): SEQ ID NO: 10

Small Scale Protein Expression Two A small scale protein expression experiment was performed on the triple mutant. This confirmed that the gene still be expressed well.

Flask scale protein expression

Eight 2 litre flasks containing 500ml each of terrific broth were prepared and autoclaved. Four flasks were used to grow the single mutant and four contained the triple mutant. When the optical density, at 500nm, of the broth reached around 1.6, IPTG was added to each flask at a concentration of 0.5mM. This induced the cells to start expressing the mutant CIP gene.

The flasks were than incubated for a further three hours. The contents of each flask were centrifuged at 4000 rpm for 30 minutes and the supernatant discarded.

Lysis and washing of cells

All four pellets of each mutant were resuspended in 100ml of 5OmM Tris, pH 8.0, 1 OmM EDTA, 5mM DTT. Lysozyme was added at a concentration of 2mg/ml, mixed well and left for 30 minutes. The lysozyme broke down the E. coli cells and the mixture became viscous and stringy due to the release of DNA from the cells. The mixture was sonicated in 50ml aliquots to break up the DNA and centrifuged at 15000 rpm for 30 minutes. The supernatant was discarded.

The eight pellets were each resuspended in 20ml of 2OmM Tris, pH 8.0, ImM EDTA, 5mM DTT using a homogeniser. To each tube, 10 ml of the above solution, with the addition of 1 % w/v Triton x- 100 was added. The tubes were then left at room temperature for 15 minutes, centrifuged at 15000 rpm for 30 minutes and the supernatant discarded. The above step was then repeated, with another Triton wash, followed by a Triton free wash.

The Folding Procedure

Because CIP is expressed in E. coli inclusion bodies, the protein needs to be folded in vitro. Each of the eight pellets were resuspended in 5OmM Tris, pH 8.0, 6M urea, ImM EDTA, I mM DTT₅ using a homogeniser. The tubes were then left at room temperature for 15 minutes, then centrifuged at 15000 rpm for 30 minutes.

The four pellets from each mutant were mixed together and the protein concentration measured using Biorad reagent. The protein concentration of each of the two mutants was adjusted to approximately 2mg/ml by diluting with more of the above buffer.

Each mutant protein preparation was diluted into three 400ml folding reactions containing:

20ml 10OmM CaCIa

5.6ml 5OmM GSSG

10OmL 5OmM Tris, pH 9.5, 6M Urea

230ml 5OmM Tris, pH 9.5

4ml ImM Haemin

40ml mutant CIP enzyme ^f

The six folding mixtures were left at room temperature and in the dark overnight. Haem could also be added as haemoglobin at 20μM final concentration. Use of haemoglobin as a haem donor was more efficient improving the refolding yield typically by a factor of 2-3 fold

Purification of enzyme from folding mixtures

The three folding reactions for each mutant enzyme were concentrated together using the large spiral wound Amicon concentrator. This was then concentrated down to around 30ml using the Amicon stirred cell in the cold room. Each concentrated protein solution was dialysed overnight at 4⁰C against 4 litres of NaAc, pH 4.3, containing ImM CaCl2. After centrifuging at 15000 rpm for 20 minutes to remove aggregates the solutions were dialysed again, overnight at 4°C against 1 OmM Na succinate. PH 6.0, I mM CaCl2- This allowed the low pH from the first dialysis to be adjusted for anion exchange chromatography. Finally, the enzymes were purified on the MonoQ column of the FPLC ion exchange chromatography machine using the following buffers :

Buffer A: 1 OmM Na succinate, pH 6.0, I mM CaCl2

Buffer B: 1 OmM Na succinate, pH 6.0, ImM CaC^₅ IM NaCl.

The purified enzymes were gel filtered into 1 OmM Na succinate, pH 6.0, containing no CaCl2 using a PD- 10 column to remove excess salt. The purified preparation was then beaded in liquid nitrogen and stored at -80°C.

General peroxidase assays with ABTS

A number of peroxidase assays were performed on the enzymes at various stages of purification. For all of these assays the following recipe was used:

100 μl 1 OmM Hydrogen peroxide (Final concentration 1.0 mM)

100 μl 10 mM ABTS (Final concentration 1.0 mM)

790 μl 5 I mM Na2HPO4, 24 mM citrate, pH 5.0 buffer

10 μl enzyme, diluted if necessary

Veratryl alcohol (VA) assays for lignin peroxidase activity

VA assays were performed using the triple mutant CIP using various concentrations of hydrogen peroxide, VA and enzyme in order to find the optimum levels of each of these components. The pH of the 3OmM phosphate, 5OmM citrate assay buffer was varied between 3, optimum for LiP and 5optimum for CIP. Results and Discussion

Site directed mutagenesis

In the first round mutagenesis a Trp residue was introduced at position 1.79 in CIP₅ which is equivalent to position 171 in LiP. The site-directed mutagenesis also introduced a BspEl site, which was used to check that the ends of the linear product were intact. pFLAGl already contained a BspEϊ site but this was blocked in one direction by Dam methylation, a successful mutagenesis was indicated by cutting at one site. Twelve clones were digested with the restriction enzyme BspEl to screen for appropriate mutations. Nine out of twelve clones were positive and two were selected for DNA sequencing. In the second round of mutagenesis two changes, R258E and R272D, were introduced into D 179W CIP obtained from the first round of mutagenesis. Second round mutagenesis clones were digested with Spel to screen for positive mutations nine out of twelve clones were positive and two were chosen for DNA sequencing. For both the single and triple mutants the expected DNA sequence was obtained.

Protein expression from the engineered synthetic gene

SDS PAGE analysis of the solubilised inclusion body material after expression and isolation from E. coli extracts showed over-expression of a 43Kda protein recombinant CiP for both the single and triple mutants.

UV/Vis spectroscopy of CiP variants

The triple mutant has a very similar UV/Vis characteristics to that of the WT and is a typical high spin haem system. The D 179W single mutants appears at least partially low spin and there is evidence of unligated haem. Taken together with the low folding yields for this mutant and in contrast to the triple mutant it suggests that the single mutant is not well tolerated by the structure.

Measurements of lignin peroxidase activity

In order to determine whether the engineering experiments had created any of the functionality of lignin peroxidase, enzyme assays with veratryl alcohol were conducted and the results are shown in Table 1. All use the same enzyme concentration and the progress curves for the reactions of the triple mutant at various peroxide concentrations are shown in Fig 4. A summary of results is presented in Fig 5 and a summary of kinetic parameter in Table 3.

Table 1 of Peroxide/VA assay results

Fig 4 shows that as the concentration of hydrogen peroxide in the assay is increased, the rate of the reaction decreases indicating that the triple mutant CIP is highly susceptible to excess hydrogen peroxide, probably due to the formation of compound III or due to inactivation at the level of

Compound I. An activation phase is evident in these assays and is observed when either VA or enzyme is added to the assay first and is not readily explainable. Pre-incubation of the enzyme with peroxide decreases activity markedly presumably due to Compound III formation and or / inactivation.

Figure 5 (rates in Table 1) shows that as VA concentration increases, so does the rate of the reaction. A control experiment containing no enzyme was performed and no activity was observed. This shows that the reaction is not occurring spontaneously and does in fact need the engineered enzyme. A second control was performed using the single mutant CIP and a minimal amount of VA activity was achieved, thus indicating that the surrounding acidic residues in the triple mutant are important for optimal VA activity as suggested earlier on the basis of data for LiP. A final control was performed- using wild type CIP prepared This was unable to catalyse any significant VA oxidation during the time course studies, indicating that the VA activity does result from the engineered changes.

Figure 6 shows the rate of turnover observed with wild type LiP compared to D 179W:R258E:R272D CIP variant at varying concentrations of VA. As VA concentration increases so too does the turnover of both enzymes in a Michaelis fashion. However, wild type LiP quickly becomes saturated and reaches its maximum rate whereas the turnover of D 179W:R258E:R272D CIP continues to rise, indicating that the latter to have a much higher apparent K_m for VA.

Although the activity of D 179W:R258E:R272D CIP continues to rise after that of LiP has reached its maximum, the k_cat of D 179W:R258E:R272D CIP appears to be lower than that of LiP .

Table 3 Apparent K_m and K_cat values for D179W:R258E:R272D CIP compared to that of wild-type LiP

Table 3 shows the apparent k_m and k_cat values for D 179W:R258E:R272D CIP as calculated from a Michaelis-Menten fit to the data in Fig 6. A comparable data set is shown for LiP . The k_cat of D 179W:R258E:R272D CIP is approximately 10 times lower than that of LiP and the k_m is around 60 times higher, making it around 500 times less effective in the oxidation of VA compared to WT LiP.

Conclusion

Two mutant CIP genes were constructed: D 179 W, which contained the Trp residue identified in lignin peroxidase as the site of redox catalysis for VA oxidation activity and D 179W:R258E:R272D, which contained the tryptophan plus two of the surrounding, acidic residues found in LiP. The equivalent of D 168 in LiP is already present in CiP giving a total of three negative charges in the W179 environment, closely mimicking the arrangement found in LiP.

Lignin peroxidase activity was successfully created in both the mutant CIP proteins. However, D 179W:R258E:R272D had more lignin peroxidase activity than D 179W (about 10% of wild type LiP activity), indicating that the acidic residues are important for optimum activity. Although D 179W:R258E:R272D CIP possessed better lignin peroxidase activity than D 179W CIP, it was still 500 times less efficient than wild type LiP, indicating that D 179W:R258E:R272D CIP may need additional changes for optimum lignin peroxidase activity.

This work shows that the electron transfer pathway between the haem group and the critical redox active Trp is generic to CiP and LiP. This is an unexpected finding and shows that the activity can be similarly transferred to other peroxidases.

REFERENCES

Blodig W, Doyle WA, Smith AT, Winterhalter K, Choinowski T & Piontek K (1998) Autocatalytic Formation of a Hydroxy Group at Cβ of Trp l 71 in Lignin Peroxidase. Biochemistry 37, 8832-8838. Blodig W₅ Smith AT₅ Winterhalter K & Piontek K ( 1999) Evidence from Spin-Trapping for a Transient Radical on Tryptophan Residue 171 of Lignin Peroxidase. Arch Biochem Biophys, 370, 88-92.

Blodig W₅ Smith AT₅ Doyle WA & Piontek K (2001 ) Crystal structures of pristine and oxidatively processed lignin peroxidase and the W171F mutant that eliminates the Redox Active Tryptophan 171. Implications for the reaction mechanism. J. MoI. Biol. 305, 851 -861.

Choinowski T₅ Blodig W, Winterhalter KH & Piontek K (1999) The crystal structure of lignin peroxidase at 1.70 A resolution reveals a hydroxyl group on the C^β of tryptophan 171 : A novel radical site formed during the redox cycle. J. MoI. Biol 286, 809-827.

Doyle WA & Smith AT (1996) Expression of lignin peroxidase H8 in Escherichia coli: folding and activation of the recombinant enzyme with Ca2+ and haem. J. Biochem.315, 15- 19.

Doyle, WA, Blodig, W, Veitch, NC, Piontek, K & Smith, AT (1998) Two substrate interaction sites in lignin peroxidase revealed by site-directed mutagenesis. Biocheistry. 37, 15097-15105.

Dunford BH (1999) Heme Peroxidases. John Wiley & Sons, Inc. New York.

Goodwin DC, Aust SD & Grover TA (1995) Evidence for veratryl alcohol as a redox mediator in lignin peroxidase-catalysed oxidation. Biochemistry 34, 5060-5065.

Harvey PJ₃ Schoemaker HE & Palmer JM (1986) Veratryl alcohol as a mediator and the role of radical cation in lignin biodegradation by Phanerochaete chrysosporium. .FEBS Lett 195, 242-246.

Khindaria A, Yamazaki I & Aust SD (1996) Stabilization of the veratryl alcohol cation radical by lignin peroxidase. Biochem 35, 6418-6424. Pasti-Grigsby MB₅ Paszczynski A, Goszczynski S, Crawford DL & Crawford RL (1992) Influence of aromatic substitution patterns on azo dye degradability by Streptomyces spp. and Phanerochaete chrysosporium. Appl. Environ. Microbiol 58, 3605-3613.

Perez-Boada M, Doyle WA₅ Ruiz-Duenas FJ₅ Martinez MJ₅ Martinez AT & Smith AT (2002) Expression of Pleurotus eryngii versatile peroxidase in Escherichia coli and optimisation of in vitro folding. Enzyme Microb. Technol. 30, 518-524.

Piontek K₅ Smith AT & Blodig W (2001) Lignin peroxidase structure and function. Bio Soc Trans 29, 1 1 1 - 1 16.

Sambrook J₅ Fritsch EF & Maniatis T (1989) Molecular Cloning: A Laboratory Manual. 2^nd ed, CSHL Press New York

Smith, AT₅ Santama, N, Dacey, S, Edwards, M₅ Bray, RC, Thorneley, RNF

& Burke, JF (1990) Expression of a Synthetic Gene for Horseradish Peroxidase C in Escherichia coli and Folding and Activation of the

Recombinant Enzyme with Ca^⁺ and Heme. Journal of Biol. Chem. 265, 13335- 13343.

Smith, AT & Veitch, NC (1998) Substrate binding and catalysis in heme peroxidases. Current Opinion in Chemical Biology 2, 269-278.

Tien M &Kirk TK (1998) Lignin peroxidase of Phanerochaete chrysosporium. Methods Enzymol. 161 , 238-249

Welinder KG (1992) Superfamily of plant, fungal and bacterial peroxidases. Curr Opin Struct Biol, 2, 388-393. EXPERIMENTS II MATERIALS AND METHODS

Materials

All chemicals used were reagent grade and were purchased from Sigma Aldrich Co. Ltd. or BDH Laboratory Supplies unless otherwise stated. Veratryl alcohol from Sigma was kindly vacuum-distilled by Dr. Klaus Piontek (ETH) to free it of a trace contaminant. Hydrogen peroxide solutions were prepared daily and the concentration determined using E₂₄o_nm ⁼ 46.5 InJVT¹Cm^'1. Molecular biology enzymes were supplied by either New England Biolabs Inc. or Boehringer Mannheim.

Recombinant LiP Expression

The generation of an engineered form of Phanerochaete chrysosporium LiP H8 cDNA (EMBL Y00262), containing both the mature and pro-sequence regions, and its subsequent cloning into the commercially available expression vector pFLAGl (International Biotechnologies Inc.) has been described previously. The resulting plasmid termed pFLAGl- LipP has already been shown to express wild-type recombinant LiP H8 protein in E. coli, that can be refolded to fully active enzyme, LiPH8* [2].

Oligonucleotides Primers for Mutagenesis For each mutant gene to be engineered, two primers were designed, one (the mutagenic primer) overlapped the area to be mutated and contained the intended base change(s), and the second (the reference primer) annealed to the other DNA strand so that the two primers sat exactly back-to-back on opposite strands of the template DNA, pFLAGl-LipP [2]. Oligonucleotides for mutagenesis are listed in Appendix 1.

Plasmid Amplification Mutagenesis

We have described a novel PCR-based method [5] which was previously used to generate the LiP E146G, W171F and W171S site-directed mutations in the LiP H8 cDNA, while cloned into the pFLAGl-LipP vector (Blodig W et al., Biochemistry, 1998 Jun 23; 37(25): 8832-8; PMID: 9636023). For each mutant, a PCR reaction was carried out containing: 10 ng pFLAGl-LipP as template, the manufacturers buffer for cloned Pfu (Stratagene) containing 2 mM Mg²⁺, 100 μM of each dNTP, 1.0 μM of the appropriate mutagenic and reference primers and 2.5 U Pfu polymerase (Stratagene). After a 'hot start' of 95 °C for 10 min. and 25 cycles of 95 ⁰C for 1 min., 55 ⁰C for 1 min. and 72 °C for 20min. (Perkin Elmer GeneAmp 9600) a linear DNA fragment was produced, whose size in each case was equal to that of linearized pFLAGl-LipP. To circularise this DNA, the PCR product was first cleaned by gel- extraction (Geneclean II, BIOlOl Inc.), phosphorylated using T4 polynucleotide kinase and finally ligated using T4 DNA ligase [DNA manipulations were as described in Sambrook, J., Fritsch, E.F. and Maniatis, T. (1989) Molecular Cloning: A Laboratory Manual, 2nd edn., CSHL Press, New York]. The ligation mixture was then transformed into E. coli DH5α cells. Alternatively the gene clean step was omitted and the wild-type template was destroyed by addition of the methylase specific restriction enzyme Dpnl.

This new method of site-directed mutagenesis was found to have advantages over other methods in that it does not require the presence of a restriction enzyme site near to the intended site of mutation. Primers were designed so that a new restriction enzyme site was generated or one already present was regenerated only when a ligation event produced an intact gene, i.e. the restriction enzyme's recognition site overlapped the 5' end of each primer. This was necessary because of a spurious exonuclease activity, possibly arising from Pfu, which resulted in deletions at the ligation site in a high proportion of clones. In each case a unique restriction enzyme site was generated only if a mutant PCR product ligated to give an intact gene. In this case, wild-type plasmids and mutant plasmids bearing deletions were both excluded on the basis of their failure to digest with the appropriate enzyme. The correct integrity of all mutant genes (and the wild-type) were confirmed by complete sequencing of the coding sequencing of the coding sequence on one strand. No extraneous mutations were introduced by the PCR procedure {Pfu is a high-fidelity polymerase). Mutational efficiencies were found to vary from 40 to 80% of the total clones resulting from a single ligation experiment.

Expression, Folding and Purification of Recombinant LiP

Expression, folding and purification of recombinant wild-type and mutant LiP was performed as in Doyle and Smith [2], except that the E. coli strain W3110 was substituted for DH5α. Typical yields of 2-3 mg of fully activated enzyme / litre E. coli were obtained. The expected N-terminal sequence (minus the initiating Met) was confirmed and the molecular mass of LiP was determined to be 37,400+100 D (Single peak obtained by MALDI-TOF mass spectrometry). Enzyme samples were stored in 10 mM Na succinate, pH 6.0, at -80 ⁰C. Absorption spectra were recorded on a Shimadzu UV 1601 spectrophotometer at 25⁰C ± 0.5⁰C, with a spectral bandwidth of 2 nm and a scan speed of 370 nm min^"1, in the same buffer. The concentrations of LiP and the mutant enzymes were calculated from the absorption at 409 nm, using an extinction coefficient of 168 mM^cm^"1, determined previously for the native H8 enzyme. R.Z. (reinheitszahl) (A_4OgZA_28O) values were always greater than 3.5 and generally above 4.0.

¹H-NMR ofLiPHδ* ¹H-NMR experiments were recorded using a Varian 500 MHz instrument. Sample preparation was as previously described using solution conditions of 20 mM KH₂PO₄, D₂O, pH 7.0 to obtain the resting state or 20 mM KH₂PO₄, 15 mM KCN, D₂O, pH 7.0 to obtain the cyanide-ligated state of the enzyme, respectively. The concentration of the recombinant LiPH8* sample used for NMR was 0.1 mM. Standard acquisition and processing parameters or one- and two-dimensional ¹H-NMR of peroxidase samples were employed. All experiments were carried out at 30 ⁰C and referenced to 1,4-dioxan as an internal standard with a resonance at 3.74 ppm relative to 2,2-dimethyl-2-silapentane-5-sulphonate.

Steady-State Enzyme Assays andK_m Determinations All steady-state assays were carried out in 5 mM phosphate, 5 mM citrate, with the ionic strength maintained at 50 mM by the addition of sodium sulfate. The pH values of the standard buffers ranged from 2.5 to 7.0, at 0.5 or 1.0 pH intervals. Assays were conducted as follows, at 25 °C ± 0.5 ⁰C, using 5-30 pmols of enzyme to initiate the reaction. Initial rates were taken over the first 5-20 s of the assay. VA oxidation assay: 2 mM VA, 400 μM H₂O₂, production of veratryl aldehyde was followed at 310 nm (ε₃₁₀ = 9.3 mM^cm^'1); ABTS oxidation assay; 500 μM ABTS, 100 μM H₂O₂, accumulation of ABTS cation radical was followed at 405 nm (ε₄₀₅ = 36.6 mVT'cm^'1); 4-[(3,5-Difluoro-4-hydroxyphenyl)azo]benzene- sulfonate (DFAD) bleaching assay: 50 μM DFAD, 100 μM H₂O₂, bleaching was followed at 375 nm, an isosbestic wavelength for the protonated and deprotonated forms of the dye (S₃₇₅= 14 InMf¹Cm^"1). Activities are expressed throughout as catalytic centre activities (s^'1). The pH dependence and apparent Michaelis-Menten Kinetics were to a single ^K_3. equation and standard saturation function using Sigma Plot V8.0 (SPSS). Pre-Steady State Experiments

Transient kinetics were monitored on an SXl 9MV stopped-flow spectrophotometer (Applied Photophysics)., fitted with a diode array detector, at 25 °C ± 0.2 °C. Buffer conditions were 5 mM phosphate, 5 mM citrate, and pH 4.0. Ionic strength was kept constant at 50 rnM. Compound I formation was followed at 400 nm and its decay at 412 nm. The resulting time- dependent spectra were analysed using the manufacturer's software (Pro-Kineticist, Applied Photophysics).

Stopped flow fluorescence studies

The change in fluorescence when Trp residues in LiP reacted with the N- Bromosuccinamide (NBS) was followed in an SX19MV stopped flow spectrophotometer using an excitation wavelength of 294nm and a 320 nm emission cut off filter. The enzyme concentration was 4.0 μM and the reactions were run at pH 3.0 in 5mM Phosphate / citrate buffer. An NBS concentration of 80 μM after mixing was used. Data were imported into sigma Plot vs 8.0 (SPSS) for display and analysis using bi exponential functions.

RESULTS AND DISCUSSION

The role of the acidic microenvironment of Trpl71: - overcoming the thermodynamic barrier for VA oxidation

The acidic environment of Trpl71 stands out immediately from Fig 7. There are four acidic residues within a 5A radius of Trpl71, including Glul68, Aspl65, Asp264 and Glu250. Two interact directly with the Trpl71. Glu250 is hydrogen bonded to the NH of the indole and GIu 168 to the Cβ-OH of Trpl71. These charges and Trpl71 are conserved in all ligninases and partially in one new manganese peroxidase sequence that has limited (high K_m) VA oxidation activity [see 9]. Although these residues will carry only a partial negative charge at physiological pH (4.5) they could provide the acidic microenvironment that has been postulated to be necessary to stabilise VA^*®. Moreover, the negative electrostatic potential created by these acidic residues should help decrease the redox potential for the oxidation of Trpl71 and VA. This will render their oxidation by the haem more thermodynamically favourable (see below for discussion of redox potentials). If a positive charge is created during the oxidation of VA at this site then one of the acidic groups with the lowest pKa, will deprotonate, thereby stabilising a VA^* : LiP intermediate by charge compensation. To put it another way, the local co-operative ionisation equilibrium will shift to a Io wer pK thereby acting as a charge buffer. We would envisage partial charge dissipation of the postulated [VA-Trpl71]"^Θ on Glu250 and Glul68 with a smaller contribution from D165N. The results of mutagenesis experiments in which the negatively charged residues depicted in Fig 7 are individuallymutated are shown in Table 4 together with the substrate profile for a family of methoxy-benzenes of gradually increasing electropositivity in Table 5.

Table 4 Steady state kinetic parameters obtained with veratryl alcohol as substrate for the charge mutants depicted in Fig 2

Table 4 shows that the negatively charged environment of Trpl71 also modulates the reactivity of the enzyme towards veratryl alcohol. All substitutions increased the apparent K_m for VA, except D264N. The most pronounced change was obtained with El 68Q₃ which interestingly hydrogen bonds to the Cβ OH of Trpl71. Data has also been obtained for a variety of methoxy benzenes spanning the redox interval 1.4- 1.7V. Interestgingly, E250Q appears to loose oxidative capability, it can no longer oxidize 1,2,3 tmethoxy benzene (half potential 1.67V). On the other hand the D165N variant can oxidize 1,3 dimethoxy benzene (E° > 1.74 V) which cannot be oxidised by the wild type, suggesting that the relative oxidation potential of this mutant may be greater than that of wild type. The precise mechanistic significance of this is not understood. However we have confirmed that the effects seen with E168Q for VA oxidation are greater than those obtained with any other mutant including those at the haem edge which control the oxidation of negatively charged dyes, except of course mutations at Trpl71, all of which (Tyr, Ser, Phe and His) eliminate the oxidation of VA.

Table 5 Oxidation of a hierarchical series of methoxy benzenes by Wl 71 environment charge mutants.

We have confirmed that a Trp residue at 171 is essential for the oxidation of VA and all methoxy benzenes with potentials greater than 1.4V (Table

5). The charge located at El 68 is especially important for maintaining the kinetic effectiveness of the enzyme, presumably because it aids in the stabilization of an unfavourable transition state possibly involved in formation of a [VA-Trp l 71 ]^*Θ complex (see above). Furthermore, there are indications that the substrate range of the enzyme can be extended to oxidise the more electropositive substrate 1 ,3 DMB.

Modification of the accessibility properties of the 'reactive haem-edge ': Reverse engineering of lignin peroxidase to resemble the open 'classical¹ haem edge architecture of Coprinus cinereus peroxidase.

We had previously shown that the oxidation of simple model azo dyes at the restricted haem-edge of LiP is in part controlled by the negative charge at Glul 46 [5] . The decreased negative electrostatic potential and enlarged access channel of the Glul46Gly mutant increased the pH optimum for the oxidation of negatively charged azo dyes by 2.5-3.0 pH units (c/. [I ]) and turnover numbers at pH 6.0 increase by some 300-fold compared to the wild type. We have extended this work to incorporate additional changes designed to create a more open haem-edge substrate oxidation site based on that seen in the structural homolog Coprinus cinereus peroxidase (CIP), whose structure was solved by molecular replacement from LiP. Unlike LiP, CIP has a high activity for traditional peroxidase substrates such as phenols, anilines and dyes but cannot oxidise highly electropositive methoxy benzenes of the type used in Table 5 [5] as it lacks the key Trp residue and its haem-iron high oxidation state intermediates lack the necessary oxidative power. These changes were made in LiP to provide an enzyme with both attributes and therefore have more widespread applications in bio bleaching reactions.

The residues provided by the E-F turn, D-E loop and F-G turn in lignin peroxidase provide contacts to the haem group and substrate access channel in all members of the plant peroxidase family. In lignin peroxidase the D-E loop and F-G turn are stiffer and more occluded than that of CIP due to the absence of GIy at two key positions and the presence of Pro at the 83 position. The LiP access channel is also negatively charged due to a GIu at the 146 position and an Asp at the 183 position, not found in any other peroxidase. The GIu at the 146 position was addressed in previous work and shown to control the oxidation of small negatively charged azo dyes [5] . The mutants selected below were chosen to (a) mimic the open more flexible D-E loop of CIP [H82P :P83A] (b) remove additional steric hindrance to the access channel [I85G] and

(c) remove the negative charge and steric hindrance from the 183 position [D183G/N] . Each of these changes progressively mimics the structure found in CIP, whose substrate profile is quite different to that of LiP [3] . The nature of the interaction between D 183 and the haem propionate is very unusual in peroxidases. The equivalent linkage is usually provided by a polar or basic residue and in CEP by a GIy. The interaction of these two negatively charged groups in LiP could result in a significant increase in the apparent pK observed for Aspl 83 and may in part account for the low pH optimum observed for many substrates and may contribute to the enhanced oxidation potential of the haem center at low pH.

The hydrogen bonding at Nl 83 to the haem propionate is intact and the mutant is essentially iso-structural with WT confirming that the increase in ABTS activity is likely to be due to the loss of the negative charge at the 183 position. The largest active site change is a small movement of the catalytic His 47 (0.2A)₅ this may in part explain the increased rate of compound I formation in this mutant (10.8XlO⁵M⁴S^"1) compared to the wild-type (5.OxIO⁵M^-1S^'1). Structural perturbation of the CIP mimic D183G was more severe with a localised rearrangement of the back bone in the F-G region leading to destabilisation of the C-terminal region.

CONCLUSIONS

We have constructed and characterised a large number of mutations (see Appendix 1) in recombinant lignin peroxidase in order to probe the unique properties of lignin peroxidase an enzyme capable of generating and controlling extraordinarily oxidising potentials.

REFERENCES 1. Blodig, W., Smith, A.T., Doyle, W.A & Piontek, K (2001). J. MoL Biol. 305, 851-

861.

2 Doyle, WA. & Smith, AT. (1996) Biochem. J. 315, 15-19.

2. Smith, AT. & Veitch NC. (1998) Curr. Opin. Chem. Biol. 2, 269. 3. Veitch, N.C & Smith, A.T (2001) Adv. in Inorg. Chemistry 51, 107-161.

4. Veitch, N.C & Smith, A.T (2001) Horseradish peroxidase Adv. in Inorg. Chemistry 51, 107-161.

5. Doyle, WA₅ Blodig, W₅ Veitch, NC, Piontek, K. & Smith, AT. (1998) Biochemistiy 43, 15097.

6. Blodig, W, Doyle, WA, Smith, AT, Winterhalter, K, Choinowski, T. & Piontek, K. (1998) Biochemistry 37, 8832.

7. Blodig, W., Smith, A.T., Winterhalter, K & Piontek, K. (1999) Arch. Biochem. Biophy. 370, 86-92. 8. Hiner A.N.P., Hermandez-Ruiz, J., Rodrigues Lopez, J.N., Garcia-Canovas, F.,

Brissett, N.C, Smith A.T., Arnao M.B., & Acosta, M. (2002) J. Biol. Chem. 277, 26879-

26885.

9. Perez-Boada, M., Doyle, W.A., Ruiz-Duenas, FJ., Martinez, M.J. Martinez, A.T &

Smith A.T. (2002) Enz. Microbial. Tech. 30, 518-524.

Appendix 1

Description of the Coprinus cinereus peroxidase (CIP) synthetic gene

The first 8 amino acid residues from the mature CIP sequence were deleted, as the crystal structure shows that this region is largely unstructured. The first nine nucleotides of the synthetic horseradish peroxidase gene ATGTTAACT, encoding a synthetic N-terminus of MLT, were added. The rest of the sequence was codon optimised for E. coli I yeast expression. Particular care was taken to remove two GC-rich islands at the 5' end of the gene. The removal of these considerably improved E. coli expression of the protein. The full nucleotide sequence of the wild-type CIP synthetic gene and translated protein sequence is given in SEQ ID NOs: 42 and 43. All amino acid residue position numbers following use the current numbering taken from the mature CIP sequence as published for the crystal structure.

Annotation of mutations.

D179W single mutant: From the wild-type synthetic gene below, the Asp codon at nucleotide position 517 was replaced with TGG encoding a Trp at the 179 position (note amino acid numbering, above), and the Ser and Pro codons at nucleotide positions 499 and 502 were replaced with AGT and CCG respectively, both the latter being silent changes which introduced a BspEl restriction enzyme site.

D179W:R258E:R272D triple mutant. This mutant incorporated the nucleotide changes described above for D 179 W, plus the following changes were also made. The Arg codon at 754 was replaced with GAG (GIu) and the Arg codon at position 796 was replaced by GAT (Asp). Further silent changes were made to the codons at nucleotide positions 769, to ACT, and 772, to AGT. These changes created an Spel restriction enzyme site.

In a method of determining the effect of a substance upon the catalytic activity of the triple mutant peroxidase (SEQ ID NO: 1), a sample of the substance is first mixed with the peroxidase, and the mixture is then contaced with a veratryl alcohol.

The catalytic activity of the peroxidase in the oxidation of the veratryl alcohol is then measured, and the results obtained are compared with control results obtained with the same veratryl alcohol and the peroxidase in the absence of the substance. The results of the comparison thus evidence the effect of the substance upon the catalytic activity of the peroxidase.

It will be appreciated that it is not intended to limit the scope of the invention to the above examples only, other embodiments being readily apparent to one of ordinary skill in the art without departing from the scope of the appended claims.

Claims

1. A peroxidase capable of oxidising veratryl alcohol, wherein said peroxidase is a peroxidase not normally capable of oxidising veratryl alcohol, and wherein a residue equivalent to tryptophan 171 in LiP, but which is not tryptophan, is substituted by tryptophan, and, if necessary, providing one or more acidic amino acid residues such that there are at least two acidic amino acid residues in sufficient proximity to the indole ring of the tryptophan to be able to enhance the stability of any charge on the indole ring and/ or substrate or intermediate formed therewith.

2. A peroxidase according to claim 1, which is derived from Coprinus cinereus.

3. A peroxidase according to claim 1 or 2, wherein there are at least 3, and preferably 4, acid residues in proximity to the tryptophan residue.

4. Coprinus cinereus peroxidase, or a modified version thereof, substituted at position 179 with tryptophan and at one or more of positions 173, 258 and 272 with an acidic amino acid residue.

5. Coprinus cinereus peroxidase according to claim 4, wherein position 258 is substituted with a glutamate, and positions 272 and 173 substituted each with an aspartate.

6. Coprinus cinereus peroxidase according to claim 4 or 5, wherein position 275 is substituted with a phenylalanine.

7. Coprinus cinereus peroxidase according to claim 4, 5 or 6, wherein positions 174 and/ or 175 have been mutated to glutamate and leucine, respectively, and/ or at least one further amino acid residue has been inserted between residues 171 and 172.

8. A method of oxidising veratryl alcohol, comprising the steps of:

(i) contacting said veratryl alcohol with a peroxidase according to any of claims 1-7; and (ii) allowing said peroxidase to catalyse oxidation of said veratryl alcohol.

9. A method of determining the effect of at least one substance upon the catalytic activity of a peroxidase according to any of claims 1-7, comprising the steps of:

(i) contacting a veratryl alcohol with said peroxidase and said at least one substance;

(ii) measuring the catalytic activity of said peroxidase in the oxidation of said veratryl alcohol; and

(iii) comparing the results of step (ii) with control results in order to determining the effect of said least one substance upon said catalytic activity of said peroxidase.

10. A method according to claim 9, said control results being obtained by measuring the catalytic activity of said peroxidase in the oxidation of said veratryl alcohol in the absence of said at least one substance.