Tài liệu Báo cáo khoa học: A tyrosinase with an abnormally high tyrosine hydroxylase/dopa oxidase ratio Role of the seventh histidine and accessibility to the active site docx

Tyrosinases catalyse the hydroxyla-tion of monophenols to o-diphenols cresolase or mono-phenolase activity and the subsequent oxidation of o-diphenols to o-quinones catechol oxidase or d

hydroxylase/dopa oxidase ratio

Role of the seventh histidine and accessibility to the active site Diana Herna´ndez-Romero1, Antonio Sanchez-Amat1 and Francisco Solano2

1 Department of Genetics and Microbiology, 2 Department of Biochemistry and Molecular Biology B, University of Murcia, Spain

Polyphenol oxidases (PPOs) are a broad group of

cop-per enzymes able to catalyze the oxidation of a great

variety of phenols by molecular oxygen [1] Basically,

there are two main types of PPO, laccases and

tyrosin-ases, with significant differences at the polypeptidic

cop-per-binding sites [2] and the spectroscopic properties of

the metal ions [3,4] Both enzymes are widely distributed

in nature The active site of tyrosinases consists of a pair

of coupled copper ions called copper type-3 However,

blue laccases have up to four copper ions at the active

site of three different types, one type-1, one type-2 and

a couple of type-3 Tyrosinases catalyse the hydroxyla-tion of monophenols to o-diphenols (cresolase or mono-phenolase activity) and the subsequent oxidation of o-diphenols to o-quinones (catechol oxidase or dipheno-lase activity) [5,6] (Fig 1) One of the most common monophenolic substrates in a variety of organisms is tyrosine, justifying the activity tyrosine hydroxylase for monophenolase The product of this hydroxylation

is an o-diphenol, dopa, so that the oxidation of this

Keywords

catechol oxidase; copper enzymes;

monophenolase; phenol oxidase; tyrosinase

Correspondence

F Solano, Department of Biochemistry and

Molecular Biology B, School of Medicine,

University of Murcia, Murcia 30100, Spain

Fax: +34 9683 64150

Tel: +34 9683 67194

E-mail address: psolano@um.es

URL: www.um.es/bbmbi

(Received 25 July 2005, revised 6 October

2005, accepted 27 October 2005)

doi:10.1111/j.1742-4658.2005.05038.x

The sequencing of the genome of Ralstonia solanacearum [Salanoubat M, Genin S, Artiguenave F, et al (2002) Nature 415, 497–502] revealed several genes that putatively code for polyphenol oxidases (PPOs) This soil-borne pathogenic bacterium withers a wide range of plants We detected the expression of two PPO genes (accession numbers NP_518458 and NP_519622) with high similarity to tyrosinases, both containing the six conserved histidines required to bind the pair of type-3 copper ions at the active site Generation of null mutants in those genes by homologous recombination mutagenesis and protein purification allowed us to correlate each gene with its enzymatic activity In contrast with all tyrosinases so far studied, the enzyme NP_518458 shows higher monophenolase than o-diphenolase activity and its initial activity does not depend on the pres-ence of l-dopa cofactor On the other hand, protein NP_519622 is an enzyme with a clear preference to oxidize o-diphenols and only residual monophenolase activity, behaving as a catechol oxidase These catalytic characteristics are discussed in relation to two other characteristics apart from the six conserved histidines One is the putative presence of a seventh histidine which interacts with the carboxy group on the substrate and con-trols the preference for carboxylated and decarboxylated substrates The second is the size of the residue isosteric with the aromatic F261 reported

in sweet potato catechol oxidase which acts as a gate to control accessibil-ity to CuA at the active site

Abbreviations

DO, dopa oxidase; dopachrome, 2-carboxy-2,3-dihydroindole-5,6-quinone; PPO, polyphenol oxidase; R3, a wild-type strain of R solanacearum

hydroxylase.

particular catechol to o-dopaquinone is also called dopa

oxidase (DO) activity On the other hand, laccases

oxid-ize mainly p-diphenols and methoxy-substituted

mono-phenols to finally yield, respectively, p-quinones and

dimeric quinonic structures (Fig 1) [7]

Tyrosinases are responsible for vertebrate cutaneous

pigmentation, browning of fruits and vegetables, and

morphogenesis and fruiting body formation in fungi

All of these processes involve melanin formation In

the bacterial kingdom there are some examples of

well-characterized tyrosinases They were first described in

the genus Streptomyces [8,9], but the enzyme has also

been reported in other bacteria such as Sinorhizobium

meliloti [10] and Marinomonas mediterranea [11] The

latter marine bacterium was the first prokaryote

des-cribed that expresses two different PPOs One of them

is a soluble tyrosinase clearly involved in melanin

synthesis [11], and the second is a membrane-bound

laccase [12,13] with residual tyrosinase activity; its physiological role is uncertain In fact, the role and physiological advantages of the coexistence of several PPOs in the same micro-organism remain unknown The synthesis of melanin in micro-organisms has been related to pathogenesis and virulence [14] Melani-zation of the infectious cell seems to offer an advant-age, as microbial melanin could protect the pathogen against the host cell [15], although melanization in the host cells is also proposed to be part of the defense sys-tem against wounding and infection by the pathogen [16] Thus, the timing and location of melanization seems to be essential for the prevalence of one of these two opposing processes In any case, many of the bac-teria that express PPO activities are strains that interact with plants such as Rhizobium meliloti [10], Ralstonia solanacearum [17] and the marine epiphyte Microbulbi-fer degradans [18] R solanacearum has unique and

A

B

C

Fig 1 PPO activities These copper enzymes show a wide range of action on phenolic compounds (A) Tyrosinases show two activities, the hydroxylation of mono-phenols and the oxidation of o-dimono-phenols Different names are used for these activit-ies, as shown in the Figure One of the most common monophenolic substrates is tyrosine, and in that particular case the activities are named tyrosine hydroxylase (TH) and dopa oxidase (DO) (B) In this case, the product of the catalysis, o-dopaquinone

is rapidly converted into dopachrome, the colored product measured in the spectro-photometric assay (C) Laccases are differ-ent PPOs with the capacity to oxidize p-diphenols or methoxy-monophenols.

relevant features for addressing the molecular

deter-minants of bacterial pathogenicity to plants It is a

soil-borne pathogen which naturally infects roots It

exhibits a strong and tissue-specific tropism within the

host, invading and multiplying in the xylem vessels In

addition, this b-proteobacterium has an unusually wide

host range The genome of the strain GMI1000 isolated

from tomato has been sequenced [19] It contains up to

four genes that putatively code for copper PPOs We

have recently proved that at least three of these genes

are expressed and the corresponding protein products

show PPO activity, including two tyrosinase-like

enzymes and one laccase [17]

The monophenolase activity of tyrosinases is usually

coupled to the o-diphenolase activity In fact, it has

been proposed that tyrosinase binds monophenols

at the active site to directly oxidize these substrates to

o-quinones, so that the two activities cannot be

separ-ated [20] In spite of this, o-diphenolase activity can be

determined by just using o-diphenols as the initial

sub-strate of tyrosinases

Tyrosinases show a much higher specific activity for

oxidation of o-diphenols (o-diphenolase activity) than

for hydroxylation of monophenols (monophenolase or

cresolase activity) [5,21] Furthermore, it is quite

common in plants to find PPOs that act exclusively as

o-diphenolases, with none or a very residual

mono-phenolase activity [16] In animals, as well as a true

tyrosinase, there is another protein called Trp1, which

can be considered an o-diphenolase because it shows

low oxidase activity with two o-diphenols, dopa and

5,6-dihydroxyindole-2-carboxylic acid [22,23] This is

the main reason why classical enzymology classifies the

same family of proteins with the pair of type-3

cop-per ions in tyrosinases (monophenol l-dopa-oxygen

oxidoreductase, EC 1.14.18.1) and catechol oxidases

(o-diphenol–oxygen oxidoreductase, EC 1.10.3.1), but

the differentiation between these two types of enzyme

is not clear [4] Looking at the sequences of the two

enzymes, both show absolute conservation of the

histi-dine residues of the CuA and CuB binding regions and

the same Prosite signatures [2,4,6]

The low or zero monophenolase⁄ o-diphenolase ratio

is understandable Chemical oxidation of o-diphenols is

much easier than hydroxylation of monophenols The

noncatalyzed reaction rate for the atmospheric oxygen

oxidation of o-diphenols to o-quinones is several orders

of magnitude faster than that for monophenol

hydroxylation to o-diphenols Pigment cell researchers

should be aware that stock solutions of l-dopa darken

spontaneously because of its oxidization, especially at

neutral or basic pH, but stock l-tyrosine solutions are

stable for long periods

In this paper, we show that one of the two tyrosin-ase-like PPOs produced by R solanacearum displays higher tyrosine hydroxylase (TH) than DO activity To our knowledge, this is the first tyrosinase with this very interesting feature Comparison of the amino acid sequences at the active site with other tyrosinases and catechol oxidases allows us to propose correlations between key residues in the catalytic patterns of these enzymes and whether they act as true tyrosinases (monophenolases plus o-diphenolases) or only o-diphe-nolases

Results

Genes encoding putative tyrosinases

in R solanacearum After genome sequencing of R solanacearum, two genes that putatively code for tyrosinase-like enzymes were detected by a blast search [19] They were named catechol oxidase (gene RSc0337, protein NP_518458) and tyrosinase (gene RSc1501, NP_519622)

When we submitted both sequences to a hierarchical multiple sequence alignment [25], two sets of proteins showing highest sequence similarity were obtained [17] Interestingly, these sets did not overlap The protein NP_518458 was found to be similar to several plant catechol oxidases and a few bacterial proteins (Table 1) Catechol oxidase from sweet potato (Ipo-moea batatas) was not in the top five highest scoring proteins, but it is included in the table because it is the only enzyme of this family that has an available crystal structure [4,26] The similarity to plant catechol oxid-ases supports the initial naming of this protein [19]

On the other hand, the proteins with highest sequence similarity to NP_519622 were several Streptomyces tyrosinases (Table 1) This therefore justifies the nam-ing of this enzyme as tyrosinase Mushroom tyrosinase

is included in Table 1 because it is the most commonly used tyrosinase in model studies It is important to note that the most characteristic signatures in the sequences are present in both proteins, tyrosinases and catechol oxidases; these include the six histidine resi-dues involved in the binding of a pair of copper ions and other conserved residues [6] However, so far it is not possible to predict from this signature the enzy-matic activity that a protein will actually display

Isolation of R solanacearum mutants affected

in tyrosinase-like activities Strains with mutations in the two genes coding for tyrosinase-like activities were constructed by

homo-Table

logous recombination Briefly, the gene RSc1501 was

amplified by PCR from genomic DNA of a spontaneous

RifR R solanacearum wild-type GMI1000 strain which

we called R3 The PCR product with a size of
1.6 kb

was digested with BamHI to obtain a fragment between

the two copper-binding site coding regions and ligated

to pBlueScript pKSII(+) with T4 DNA ligase

(Invitro-gen, San Diego, CA, USA) The ligation mixture was

transformed in Escherichia coli DH5a, and

transform-ants selected for ampicillin resistance The plasmid

obtained (pBRI15) was digested with EcoRI and SacI,

and the internal RSc1501 gene fragment subcloned in

the pFSVK plasmid The resulting plasmid (pCN15)

was transformed in E coli S17-1 (kpir), and

transform-ants selected for kanamycin resistance The plasmid in

this strain was mobilized into spontaneous RifR R3

by conjugation [17] RSc1501 gene disruption in the

transconjugants was confirmed by appropriate PCR

and product analysis (data not shown) One strain,

R3-1501–, was selected for further assays

To obtain mutants affected in the RSc0337 gene, an

internal fragment of 300 bp between the two

copper-binding sites from this gene was amplified using the

appropriate forward and reverse primers Then the

product was cloned in the pFSVK plasmid using

the NcoI and SacI restriction sites The resulting

plas-mid pCN337 was transformed in E coli and mobilized

into R3 as described above for the RSc1501 gene

RSc0337 disruption was also confirmed in the

trans-conjugants by PCR [17], and one strain, R3-337–, was

selected for further studies

PPO activity in R solanacearum and mutants

affected in genes coding for these proteins

R solanacearumshowed monophenolase and

o-dipheno-lase activities, represented by TH and DO, respectively

The conditions for the PPO enzymatic assays differed

with regard to pH and SDS concentration TH activity

was higher at pH 5 and 0.05% SDS, but DO showed a

sharp peak at 0.02% SDS and pH 7 In fact, the rate

of oxidation of l-dopa was much lower at pH 5, but

under these conditions the optimal SDS concentration

was 0.05%, the same as optimal TH conditions [17]

Furthermore, when these activities were determined in

cellular extracts of the mutant strains generated and

compared with the wild-type strain, we found that

each activity was lost in extracts of different mutants

Mutation of the RSc0337 gene resulted in loss of

almost all TH activity, whereas mutation of the

RSc1501 gene resulted in loss of most of the DO

activ-ity, indicating a correspondence between both activities

and the proteins encoded by the respective mutated

genes, which was opposite to that expected from the blasthomologies and designated names (Fig 2) Moreover, the TH activity in both mutants showed

a very different dependence on l-dopa as cofactor to eliminate the characteristic lag period of tyrosinases [8,20,21] Figure 3 shows the rate of TH activity as a function of the concentration of l-dopa cofactor added

to the assay mixture R3-1501– extracts have a high

TH activity, almost independent of the addition of

l-dopa cofactor, and the lag period before reaching the maximal reaction rate without this addition is short (
40–60 s under standard conditions) The TH activity of R3-0337– extracts is quite low and needs to

Fig 2 TH and DO activities in extracts of wild-type R3 R solana-cearum and two mutant strains with mutations in the PPO genes RSc0337 and RSc1501 TH activity was determined at pH 5 and 0.05% SDS, and DO activity at pH 7 and 0.02% SDS.

Fig 3 Dependence of the TH activity of extracts of R3 and the

formed, after subtraction of the blanks in the absence of the

be activated by the addition of l-dopa Its lag period

in the absence of l-dopa is
5 min R3 wild-type

extracts behave much more like R3-1501– than

R3-0337– This pattern agrees with the presence of two

different enzymes with overlapping activities

Purification of two enzymes with different

affinities for monophenols and o-diphenols

Supernatants of bacterial crude extracts obtained from

R3 wild-type and mutant strains were submitted

to enzyme purification These supernatants, routinely

30 mL, were first concentrated 5–6 times using

ultra-filtration membranes (Millipore; cut-off 10 kDa) and

applied to CM-Sephadex A-50 chromatography in

0.05 m sodium phosphate buffer, pH 7, according to

the basic pI predicted from their amino-acid sequence

After elution of unbound proteins, the ionic strength

was increased with a salt gradient of NaCl up to 1.5 m

to elute proteins bound to the anionic gel Fractions of

1.9 mL were collected, the protein content was

monit-ored (A280), and TH and DO activities were assayed

under the respective optimal conditions

The purification profiles of bacterial extracts from

wild-type (R3-wt), mutant strain R3-1501– affected in

the NP_519622 protein and mutant strain R3-0337–

affected in the NP_518458 protein are shown in Fig 4,

and a summary of the purification is shown at Table 2

Apart from a small amount of DO activity found in the

large peak of unbound proteins eluted before

applica-tion of the salt gradient, two PPOs were eluted in the

wild-type strain at high salt concentration,
0.9 and

1.05 m NaCl, respectively The first one had high TH

activity, although it also had detectable DO activity

under the optimal conditions for this activity (0.02%

SDS, pH 7) The second one displayed only DO activity

under these conditions Interestingly, the first peak but

not the second one was found in the extracts of

R3-1501–, and the opposite was observed in extracts of

R3-0337– mutant This behavior clearly suggests that

these peaks are due to different enzymes, and that the

TH activity is due to the NP_518458 protein, whereas

the DO activity is mostly due to the NP_519662 protein

As these proteins were preliminarily named catechol

oxidase and tyrosinase, respectively, this activity profile

strongly indicates that the names should be exchanged

The main stages of the purification process for the

three extracts are summarized in Table 2 The initial

total amounts of protein are not the same because we

started purification from different amounts of material

During the purification process, we obtained 245-fold

and 691-fold purification for the two wild-type PPOs,

and yields of
30% These purification factors were

not so high when we used the mutant extracts as start-ing material The purified peaks of the two PPOs showed purities greater than 90%, as judged by SDS⁄ PAGE, and apparent molecular masses of the active enzymes of
35 and 50 kDa (Fig 5) The respective specific activities ensure minimum turnover numbers of 750 and 1550 min)1 for the TH activity

of the monophenolase and the DO activity of the o-diphenolase, respectively

Affinity for carboxylated and decarboxylated phenolic substrates

To explore the affinity of the active site of the two PPOs for phenolic substrates and possible correlations between the structural requirements for interaction and

Fig 4 Purification profiles in CM-Sephadex chromatography of cel-lular extracts from wild-type and mutant strains After elution of all

respectively, for the profile of UV absorbance (total protein) and

the differences between the two PPOs, the kinetics

parameters of carboxylated⁄ decarboxylated substrates

were calculated We used the couples l-tyrosine⁄

tyram-ine for the monophenolase activity and l-dopa⁄

dop-amine for the diphenolase activity Standard activities

under optimal conditions are shown in Fig 6, and

val-ues for Vmax, Kmand catalytic efficiencies in Table 3

Concerning monophenolase activity, the enzyme

NP_518458 greatly preferred l-tyrosine to tyramine It

showed higher Vmaxand lower Kmfor the carboxylated monophenol, which can be more clearly appreciated if the catalytic efficiency (Vmax⁄ Km) is calculated At pH

7 the affinity for these substrates was slightly lower (data not shown) On the other hand, the enzyme NP_519662 did not show preference for l-tyrosine In fact, this enzyme was a little bit more efficient in tyramine hydroxylation It was almost completely unable to hydroxylate monophenols at pH 5, showing

Table 2 Purification of tyrosinase and catechol oxidase (proteins NP_518458 and NP_519622, respectively) from R solanacearum In all

column A, 49 and 9 are, respectively, the amounts of protein (lg) in the TH and DO activity peaks Yields were calculated with the values in parentheses, which are the three most active fractions from the purification peaks pooled, but maximal purification (n-fold) was calculated from the most active fraction wt, Wild-type.

Column A: wt, R3 extract (contains both enzymes)

1 2

kDa

90

46 35

20

123 mL) All the peaks showed purities of at least 90% Similar

parame-ters are summarized in Table 3.

a marked loss of affinity for the substrate (the Km

increased to
10 mm; data not shown) and low

reac-tion rates

Concerning diphenolase activity, the enzyme

NP_518458 was a poor catalyst, but again it preferred

the carboxylated o-diphenol (l-dopa) over its

decar-boxylated counterpart, dopamine On the other hand,

the NP_519622 protein showed very efficient

dipheno-lase activity, particularly with dopamine Activities

with these o-diphenol substrates were higher than

1000 mUÆmg)1 (Table 3), although the affinity was not

very high To summarize, protein encoded by RSc0337

is an efficient monophenolase, especially with

carboxyl-ated monophenols, but the protein encoded by

RSc1501 is an efficient diphenolase, especially with

decarboxylated o-diphenols

Dopa accumulation in the TH reaction catalysed

by the NP_518458 protein

Figure 7A shows the stoichiometric formation of

2-carboxy-2,3-dihydroindole-5,6-quinone (l-dopachrome)

and l-dopa during the time course of tyrosine

hydroxylation The l-dopa accumulated by the

sponta-neous disproportion of dopaquinone can be titrated at

different periods of time by addition of sodium

perio-date According to the high preference of the enzyme

encoded by the RSc0337 gene for the monophenols

and the general mechanism for the reaction of

tyrosin-ases (Fig 7B), it can be seen that dopa is not

con-sumed by the enzyme through the o-diphenolase cycle,

as it is not a competitor with the monophenolase

cycle

Stability of PPOs

The stabilities of both enzymes, monophenolase

NP_518458 and o-diphenolase NP_519662, were

stud-ied by heating to 60
C and exposure to a relatively high concentration (0.5%) of the chaotropic and dena-turing agent SDS Note that the concentration is at least 10 times higher than the SDS used for optimal assay conditions (Fig 8) It can be observed that the first PPO is very stable to both treatments, but the second one is labile

Discussion

We have found two different genes in R solanacea-rum coding for putative PPO proteins that contain the typical signatures of tyrosinases, including the CuA and CuB binding sites to ligand the copper

Table 3 Kinetic parameters for the two PPOs The enzymes were

obtained from extracts of R solanacearum strains mutated in the

gene encoding the alternative one DaO, Dopamine oxidase; TaH,

tyramine hydroxylase.

Cat efficiency

A

B

addition of excess sodium periodate at several fixed times of reac-tion (B) Catalytic cycles for the monophenolase (up, clockwise) and o-diphenolase (down, anticlockwise) activities MF, Monophenol;

DF, o-diphenol; Q, o-quinone; T, tyrosinase T has three different forms during the cycles: met, resting tyrosinase with Cu(II); oxy, oxygenated form with peroxide bound to Cu(II); deoxy, reduced Cu(I) transient form with high affinity for oxygen The efficiency for both cycles depends basically on the affinity of oxyT for the mono-phenol or o-dimono-phenol The enzymatic product, o-quinone, undergoes

a very fast spontaneous disprorportion to regenerated o-diphenol and the ‘chrome’ (see Fig 1B) Dopa can be chemically oxidized very rapidly to dopachrome by sodium periodate.

type-3 pair [2,6] In principle, it is unclear what

phy-siological advantages there are for bacteria to express

two proteins so similar in terms of enzymatic activity

However, this situation has been found previously in

other bacteria Genome sequencing of Streptomyces

avermitilis also revealed the presence of two

tyrosin-ase-like enzymes, although it was suggested that one

of those genes is not expressed, or shows a very low

level of transcription [27] In addition, we have

repor-ted the existence and expression of a multipotent

lac-case and a tyrosinase in Marinomonas mediterranea

[11,13]

We have now found in R solanacearum that the two

tyrosinase-like genes and the laccase-like gene are

indeed expressed [17] One attractive advantage to

hav-ing more than one PPO is that these proteins may

interact with each other to form a stable and very

effi-cient melanogenic complex It should be taken into

account that melanogenesis is related to virulence of

the infective micro-organism, but it is also related to

defensive roles in the infected cell, so that the place

and time of triggering of melanogenesis must be key to

the success of one of these two opposite processes In

turn, a melanogenic complex has been described in

mammals between tyrosinase and tyrosinase related

protein 1 [28] The latter can behave as an

o-dipheno-lase-like protein but also as a stabilizing protein for

true tyrosinase [29] In R solanacearum, NP_518458

would mainly catalyse the rate-limiting step,

monophe-nol hydroxylation, and NP_519622 would catalyse the

second step, oxidation of o-diphenol to o-quinone, or

alternatively a stabilization of the former enzyme

Studies on possible interactions between the PPOs are

underway in our laboratory On the other hand,

envi-ronmental conditions, for instance acidic or neutral

environmental pH, may also affect the expression of the most appropriate enzyme

Apart from the physiological roles and environmen-tal advantages of having several PPOs in the same organism, we have found that the RSc0337 gene codes for an enzyme with high TH activity and lower DO activity, with optimum assay conditions at pH 5, whereas the RSc1501 gene codes for an enzyme that efficiently oxidizes l-dopa, although it also shows low activity with l-tyrosine, as revealed by the residual

TH activity detected in the R3-0337– mutant Its opti-mal activity is at pH 7 These preferred activities of the two PPOs of R solanacearum are opposite to the names assigned to them when the genome of this bac-terium was sequenced and the function of these con-ceptual proteins was proposed [19] On the basis of blast homology, the NP_518458 protein from the RSc0337 gene was named catechol oxidase, and the NP_519622 protein encoded by the gene RSc1501 was named tyrosinase This was logical according to the mathematical algorithm used for the blast search Score and e values depend on several factors, but mostly the total length of the sequence used for the blast The shorter sequence (412 amino acids), coming from the RSc1501 gene, more closely matches the short sequences (Table 1), which are tyrosinases from Streptomyces species [2], and these homologies led to this enzyme being designated a putative tyrosinase The long sequence (496 amino acids), coming from the RSc0337 gene, more closely matches long bacterial tyrosinases and a series of plant catechol oxidases, which are also long This led to the designation of this protein as a putative catechol oxidase It is clear that matching the whole sequence is not a good way of dis-tinguishing tyrosinases from catechol oxidases

Having clearly established that protein NP_518458

is a tyrosinase (monophenolase) rather than a catechol oxidase (o-diphenolase), we observed that it is a very unusual tyrosinase as it is a more efficient monopheno-lase than o-diphenomonopheno-lase and its TH⁄ DO ratio is clearly higher than 1 In the same way, it does not need

l-dopa cofactor to reach maximal tyrosine hydroxylase activity To our knowledge, this feature is not found in any other reported tyrosinase, from Streptomyces to mammals The turnover number of tyrosinases for DO

is about 100 times higher than for tyrosine hydroxyla-tion [21] In this regard, fungal and bacterial tyrosinases are very similar, showing a higher kcat and activity with o-diphenols than with monophenols [8] More-over, the TH⁄ DO ratio is almost zero in plant catechol oxidases lacking monophenolase activity In general, o-diphenols bind more rapidly to oxy-tyrosinase than monophenols [4,30] However, this tyrosinase from

Fig 8 Stability of proteins NP_518458 (TH activity) and NP_519622

(DO activity) in phosphate buffer, pH 7 Both purified PPOs were

R solanacearum has the opposite kinetic properties In

contrast with all other tyrosinases, the TH⁄ DO data

summarized in Table 4 clearly show that the

monophe-nol is the preferred substrate

Tyrosinases catalyse monophenolase hydroxylation

and⁄ or o-diphenolase oxidation as shown in Fig 7B

Binding of monophenols to resting met-tyrosinase

results in the inactive dead-end complex, but binding

of o-diphenols leads the enzyme to the oxy-tyrosinase

form, the active species for both monophenolase and

o-diphenolase activity [5,30–32] About 85% of resting

mushroom tyrosinase is found in the met form and

15% in the oxy form, so that the o-diphenol formed

by this 15% is enough to recruit the enzyme to the

cat-alytic cycle after a short time, showing the

characteris-tic lag period of tyrosinases before reaching maximal

reaction rate [5,7,30,31] Note that the product of the

reaction, dopachrome, is chemically formed by a redox

disproportion from the true enzymatic product

o-qui-none (Fig 1) R solanacearum tyrosinase seems to be

almost completely in the oxy form, as judged by the

absence of lag period in the absence of l-dopa

cofac-tor This indicates that the dead-end inactive complex

(Fig 7B) is not formed in this particular enzyme

Titration of the amount of l-dopa generated during

its TH activity with sodium periodate shows that this

o-diphenol is stoichiometrically accumulated with

dopachrome (Fig 7A), but this is not so using

mush-room tyrosinase (data not shown) These data confirm

the great preference of oxy-tyrosinase for

monophen-ols, so that the DO activity is not competing with TH

during the course of the reaction, and the chemically

generated l-dopa is not consumed

The structural difference between catechol oxidases

and tyrosinases has not yet been explained Concerning

the crucial regions for catalytic activity and substrate

affinity, the six copper-binding histidines of the two

PPOs do not show any differences (Table 1), but some

distinctions must exist The most reliable way of

exploring this is comparison of crystal structure data

The only data so far available are for sweet potato (Ipomoea batata) catechol oxidase [26] The catalytic copper center is accommodated in a central four-helix bundle located in a hydrophobic pocket, with the six histidines bound to the copper pair This particular enzyme behaves as a catechol oxidase as it does not show monophenolase activity, and the o-diphenol binds to CuB [4,32]

The most likely explanation for the lack of mono-phenolase activity of this PPO is related to the position

of the bulky aromatic residue F261 In sweet potato o-diphenolase, F261 blocks access to CuA [4,26] This aromatic residue acts as a gate, controlling the accessi-bility of phenolic substrates to the hydrophobic pocket where the dinuclear copper center is found In addition, van der Waals interactions between this aromatic resi-due lining the hydrophobic cavity and the aromatic ring of phenolic substrates help to determine the affin-ity of substrates for the enzyme In wild-type and mutated mouse tyrosinase, it was proposed that the absence of this aromatic residue at the equivalent posi-tion may be the reason why it shows monophenolase activity, assuming that residue controls the access of monophenols to CuA [31] Although monophenols and o-diphenols could access CuB, F261 may block the re-orientation of monophenols toward CuA that is nee-ded for its hydroxylation once is bound to CuB [32] It

is very unlikely that minor details can be universally extrapolated to all tyrosinases and catechol oxidases from any source, but there is no doubt that this factor

is important for accessibility to (or involvement of) CuA in the PPO active site in order for it to display monophenolase and o-diphenolase activity or just the latter activity For instance, all catechol oxidases from tomato, potato and beans have the aromatic residue at the equivalent position (Table 1) However, Streptomy-ces tyrosinases usually have the smallest residue, G, there In octopus hemocyanin, L2830 occupies the posi-tion of F261, and this may be responsible for the weak o-diphenolase activity detected in this protein, as an L residue blocks CuA less effectively than F

Our results on the two PPOs found in R solanacea-rum are totally in agreement with this steric hin-drance (Table 1) The product of the RSc1501 gene (NP_519622) has I294, a bulky but not aromatic resi-due, at the equivalent position followed by P295, a rigid residue It shows very low but measurable mono-phenolase activity The product of the RSc0337 gene (NP_518458) has in that place a small residue, A241,

in agreement with the high tyrosine hydroxylase activ-ity shown by this enzyme (Fig 9)

This steric hindrance is one of the bases of the differ-ence between monophenolases and o-diphenolases, but

TaH, tyramine hydroxylase.

Enzyme

Optimum

preferred name

tyrosinase

catechol oxidase