Protein Science
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Erratum (v12,p2383)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Kajander, T.
Right arrow Articles by Goldman, A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Kajander, T.
Right arrow Articles by Goldman, A.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?
Protein Science (2003), 12:1855-1864.
Copyright © 2003 The Protein Society

The structure of Pseudomonas P51 Cl-muconate lactonizing enzyme: Co-evolution of structure and dynamics with the dehalogenation function

Tommi Kajander1,3, Lari Lehtiö1,3, Michael Schlömann2 and Adrian Goldman1

1 Research Program in Structural Biology and Biophysics, Institute of Biotechnology, University of Helsinki, FIN-00014 Helsinki, Finland
2 Interdisziplinäres Ökologisches Zentrum, TU Bergakademie Freiberg, D-09599 Freiberg, Germany

Reprint requests to: Adrian Goldman, Research Program in Structural Biology and Biophysics, Institute of Biotechnology, University of Helsinki, P.O. Box 65, FIN-00014 Helsinki, Finland; e-mail: adrian.goldman{at}helsinki.fi; fax: 358-9-191-59940.

(RECEIVED April 9, 2003; FINAL REVISION June 5, 2003; ACCEPTED June 6, 2003)

Article and publication are at http://www.proteinscience.org/cgi/doi/10.1110/ps.0388503.

3 These authors contributed equally to this work. Back


   Abstract
TOP
 Abstract
Introduction
 Results and Discussion
 Materials and methods
 References
 
Bacterial muconate lactonizing enzymes (MLEs) catalyze the conversion of cis,cis-muconate as a part of the ß-ketoadipate pathway, and some MLEs are also able to dehalogenate chlorinated muconates (Cl-MLEs). The basis for the Cl-MLEs dehalogenating activity is still unclear. To further elucidate the differences between MLEs and Cl-MLEs, we have solved the structure of Pseudomonas P51 Cl-MLE at 1.95 Å resolution. Comparison of Pseudomonas MLE and Cl-MLE structures reveals the presence of a large cavity in the Cl-MLEs. The cavity may be related to conformational changes on substrate binding in Cl-MLEs, at Gly52. Site-directed mutagenesis on Pseudomonas MLE core positions to the equivalent Cl-MLE residues showed that the variant Thr52Gly was rather inactive, whereas the Thr52Gly-Phe103Ser variant had regained part of the activity. These residues form a hydrogen bond in the Cl-MLEs. The Cl-MLE structure, as a result of the Thr-to-Gly change, is more flexible than MLE: As a mobile loop closes over the active site, a conformational change at Gly52 is observed in Cl-MLEs. The loose packing and structural motions in Cl-MLE may be required for the rotation of the lactone ring in the active site necessary for the dehalogenating activity of Cl-MLEs. Furthermore, we also suggest that differences in the active site mobile loop sequence between MLEs and Cl-MLEs result in lower active site polarity in Cl-MLEs, possibly affecting catalysis. These changes could result in slower product release from Cl-MLEs and make it a better enzyme for dehalogenation of substrate.

Keywords: Cl-muconate lactonizing enzyme; conformational changes; mechanism; structure/function studies; crystallography; mutagenesis; dehalogenation


   Introduction
TOP
 Abstract
 Introduction
Results and Discussion
 Materials and methods
 References
 
Muconate lactonizing enzymes (MLEs) are involved in the breakdown of lignin-derived aromatics, catechol and protocatechuate, to citric acid cycle intermediates as a part of the ß-ketoadipate pathway in soil microbes. Some bacterial species are also capable of dehalogenating chloroaromatic compounds by the action of chloromuconate lactonizing enzymes (Cl-MLEs; Schmidt and Knackmuss 1980). The bacterial MLEs belong to the enolase superfamily, several structures from which are known (see Goldman et al. 1987; Neidhart et al. 1990; Babbitt et al. 1995; Thompson et al. 2000; Levy et al. 2002). Two MLE structures from Pseudomonas putida (PpMLE) and Ralstonia eutropha (ReCl-MLE) have been solved previously (Goldman et al. 1987; Hoier et al. 1994; Helin et al. 1995; Kleywegt et al. 1996). The ReCl-MLE structure was solved at only 3 Å, thus providing limited information on the structural details of Cl-MLEs.

The catechols are important intermediates in the aerobic catabolism of aromatics, including lignin. Further, several chlorinated species are degraded this way. For instance, degradation of PCB through chlorobenzoates leads to chlorocatechols and 3-Cl-muconate production. It has also been documented that protoanemonin formation, presumably by MLE from 3-Cl-muconate (Fig. 1Go; Blasco et al. 1995), can result in significant slowdown of PCB mineralization, because protoanemonin is bacteriotoxic (Blasco et al. 1995; Abraham et al. 2002). In addition, some of the bacterial strains possessing the Cl-MLE enzyme and the plasmid for chlorocatechol degradation, such as JMP34 of Ralstonia, can survive on and mineralize the herbicide 2,4-dichlorophenoxyacetate (Don and Pemberton 1981). This pathway is therefore important in bioremediation of polluted soil by bacteria.



View larger version (16K):
[in this window]
[in a new window]
  		Figure 1. The MLE and Cl-MLE reactions with 2-Cl-cis,cis-muconate and 3-Cl-cis,cis-muconate. Formation of the enolate intermediate is shown only in A. The MLEs prefer the 1,4-cycloisomerisation of 2-Cl-muconate, leading to 2-Cl-muconolactone (A), whereas Cl-MLEs produce mostly 5-Cl-muconolactone that is converted by them to trans-dienelactone and HCl (B). Furthermore, Cl-MLEs avoid formation of bacteriotoxic protoanemonin from 3-Cl-muconate by dehalogenation with deprotonation from 4-muconolactone (C).

 
Based on the existing structures of the PpMLE and ReCl-MLE and sequence comparisons, site-directed mutagenesis experiments have also been conducted to explore how substrate binds and is dehalogenated by Cl-MLEs (Vollmer et al. 1998). In these experiments, it was found that Cl-MLEs bind the 2-Cl-muconate substrate mainly in one of two possible defined orientations, resulting in 3,6-cycloisomerisation and production of 5-Cl-muconolactone, from which Cl- can be abstracted (Fig. 1Go; Vollmer et al. 1998). MLEs, on the other hand, bind and cycloisomerase the substrate both ways, are very inefficient in cycloisomerization of Cl-muconates, and are unable to dehalogenate them (Schmidt and Knackmuss 1980; Vollmer et al. 1998, 1999). However, none of the mutations, which were all in the active site, produced an enzyme capable of dehalogenation, although some did preferentially cycloisomerize in the 1,4- or 3,6-direction (Vollmer et al. 1998). Similarly, the Cl-MLE from Rhodococcus opacus is unable to dehalogenate 2-Cl-muconate although it solely binds the substrate in the orientation that produces 5-Cl-muconolactone, as do the other Cl-MLEs (Solyanikova et al. 1995). The structures of two MLE active site variants with preference for 1,4- and 3,6-cycloisomerisation, respectively, have been determined, and it has been shown how the substituted substrates might be converted in the observed way because of the mutations in the active site (Schell et al. 1999). Importantly, it is clear that the lactone ring of the product needs to rotate before dehalogenation by proton abstraction can occur in order for the correct stereoisomer of dienelactone to be produced from 2-Cl-muconate (Fig. 1Go; Schell et al. 1999). It has not, however, been possible to link the dehalogenating activity of Cl-MLE to any specific residues in the active site.

We have solved the structure of Pseudomonas P51 Cl-MLE at 1.95 Å resolution and compared it to the previously solved PpMLE and ReCl-MLE structures. The P51 Cl-MLE structure allows us to examine conformational changes on closure of the N-terminal mobile loop over the active site, which presumably occurs on substrate binding during catalysis. Based on the structures and sequence alignment of MLEs and Cl-MLEs, differences in core packing of MLEs versus Cl-MLEs were observed. A cavity, a single peptide layer away from the active site, at the boundary between the catalytic {alpha}/ß-barrel and the N-terminal domain, is twice as large in Cl-MLEs as in MLEs. This may allow dehalogenation to occur, through the generation of additional flexibility that allows for the required ring rotation prior to dehalogenation. This was further studied through site-directed mutagenesis on PpMLE to replace MLE residues by Cl-MLE residues, followed by kinetic analysis of these variants. The MLE variants Thr52Gly and Thr52Gly-Phe103Ser show that creation of the cavity results in loss of lactonization activity, whereas introducing another cavity-creating mutation leads to gain in activity. We suggest that this is so because the second mutation generates the Gly–Ser hydrogen bond present in the Cl-MLEs, which has an important role in defining the structure and function of Cl-MLEs.


   Results and Discussion
TOP
 Abstract
 Introduction
 Results and Discussion
Materials and methods
 References
 
The P51 Cl-MLE structure
Overall, the octameric P51 Cl-MLE structure is highly similar to the PpMLE and ReCl-MLE structures (Fig. 2Go; Goldman et al. 1987; Hoier et al. 1994; Helin et al. 1995; Kleywegt et al. 1996). It has an N-terminal {alpha}+ß-domain (residues 1–134), an unusual {alpha}/ß-barrel catalytic domain (residues 136–330) with secondary structure (ß{alpha})7ß, and a C-terminal subdomain (residues 331–370)—replacing the last {alpha}-helix of the MLE {alpha}/ß-barrel and folding back on the N-terminal domain (Goldman et al. 1987). The P51 Cl-MLE aligns with PpMLE and ReCl-MLE with root mean square deviations of 1.05 Å (358 atoms) and 0.661 Å (360 atoms) for the C{alpha}-atoms of PpMLE and ReCl-MLE, respectively (excluding the mobile loop; see below). There was one molecule per asymmetric unit in the I422 crystal. One additional Mn2+ ion is bound to the back of the {alpha}/ß-barrel (Fig. 2Go), at the turn 237–241, directly coordinated to Thr235 and Asn238 carbonyl groups (2.2 Å to both), and indirectly coordinated to the Asp246 side chain carboxylate group (4.0 Å) and Val240 carbonyl O (4.4 Å) through water molecules. This site was occupied probably because of the high MnCl2 concentration in the crystallization solution and is not known to be biologically important.



View larger version (40K):
[in this window]
[in a new window]
  		Figure 2. Main chain B-factor distributions. B factors are given for PpMLE (A; scale, 10–40; blue to red), P51 Cl-MLE (B; scale, 20–50; blue to red), and both structures (C), in red for P51 Cl-MLE and in blue for PpMLE. Residue numbers are according to PpMLE. The bound Mn2+ ions are drawn as gray spheres.

 
In general, the quality of the structure and the fit to the electron density is very good (Fig. 3Go). The N-terminal mobile loop that covers the active site is traceable in an open conformation, but has weak density. It is in contact with helix 82–94 (with Arg92, Ala89, and Ala93) of an adjacent monomer of the biological octamer. This may also be a relevant conformation in solution conditions, but it is clear that the loop is highly mobile as B-factors for main chain at residues 19–26 are ~78 to 100 Å2 (100 Å2 was set as the upper limit in refinement). Also, only the side chains of Pro21, Leu22, Phe26, and His30 are visible in the loop (here, and in what follows, we use the MLE numbering system). Very weak density is also observed at Trp59, unlike in the PpMLE structure, in which this loop region is well ordered. This region has also much higher B-factors than in the PpMLE structure: The main chain in P51 Cl-MLE at region 53–60 has B-factors of 40 to 76 Å2 versus 15 to 36 Å2 in the equivalent MLE region (Fig. 2Go), whereas the average main chain B-factors for P51 Cl-MLE were 39.5 Å2 and 21 Å2 for PpMLE. The chain is otherwise well ordered throughout its whole length, except for the last two C-terminal residues.



View larger version (59K):
[in this window]
[in a new window]
  		Figure 3. A representative section of P51 Cl-MLE fit to electron density as observed in the {sigma}-weighted composite omit map. The fit is shown for the P51 Cl-MLE active residues around the Mn2+ metal cofactor.

 
The P51 Cl-MLE active site is similar to that of PpMLE. No bound cis,cis-muconate was found in the active site, despite its inclusion in the crystallization solution. Also, despite the 100 mM Cl--ion concentration, we did not observe bound Cl-, unlike in the ReCl-MLE active site. The presence of the Mn2+ ion in the active site was obvious, and the novel second site occupied was verified by calculation of an anomalous difference Fourier map (data not shown).

Comparison of structures of muconate and Cl-muconate lactonizing enzymes
The active sites
The active sites in PpMLE, ReCl-MLE, and P51 Cl-MLE are very similar in terms of the key catalytic acid and base residues: Lys169, Lys167, Lys273, and Glu327 are all conserved. Of these, Lys169 and Glu327 are the suggested catalytic base and acid, whereas the roles of Lys273 and Lys167 seem to be electrostatic stabilization of the substrate carboxylate (Helin et al. 1995; Schell et al. 1999). In Cl-MLEs, an additional catalytic protonation step is required for dehalogenation, which is most likely performed by Lys169 (Schell et al. 1999; Kaulmann et al. 2001). The metal ligands Asp198, Asp249, and Glu224, as well as the nearby Glu250, occupy approximately the same positions as in PpMLE and ReCl-MLE. Indeed, it has been shown that the residue differences in the active site seem to account mainly for differences in substrate specificity and orientation of binding (Vollmer et al. 1998; Schell et al. 1999). Individual amino acids that line the active site and are different in PpMLE and ReCl-MLE have been swapped in both proteins (Ile54Val, Tyr59Trp, Phe329Ile, Ala272Ser, Lys276Asn, and Leu333Val). However, none were found to affect dehalogenation significantly (Hoier et al. 1994; Vollmer et al. 1998; U. Kaulmann and M. Schlömann, pers. comm.). Consequently, residues outside the active site must be involved in dehalogenation by Cl-MLEs.

The mobile active site loops in MLEs and Cl-MLEs
The only other major sequence differences between Cl-MLEs and MLEs near the active sites can be found in the N-terminal mobile loop region (PpMLE 20–31; Fig. 4Go). In the PpMLE structure this loop is disordered in all four different crystal forms found so far (Goldman et al. 1987; Hasson et al. 1998; T. Kajander and A. Goldman, unpubl.); in P51 Cl-MLE the loop is in an open conformation (see above); and in ReCl-MLE it is closed over the active site.



View larger version (81K):
[in this window]
[in a new window]
  		Figure 4. A multiple sequence alignment of the characterized bacterial MLEs and Cl-MLEs. The numbering for the Cl-MLE structures is given above the alignment; for MLE, below the alignment. The top four sequences are the dehalogenating Cl-MLEs. Sequences for which structures are available: clmle p51 = P51 Cl-MLE, clmle alceu = Ralstonia eutropha Cl-MLE (former Alcaligenes eutrophus), and mle psepu = Pseudomonas putida PRS2000 MLE (PpMLE). The conserved differences between Cl-MLEs and MLEs in the active site or near the cavities are indicated (shaded), and active site positions are labeled with a plus sign (+). The N-terminal mobile loop region (20–31) is underlined for the sequences for which structures are known. The internal sequence identity in the MLE and Cl-MLE families is in both cases >60%. Between families, it is 40%–45%. The Rhodococcus species are 40% identical to the others and most similar to each other (sequence IDs: mle ropac and clmle ropac).

 
To understand potential differences between MLEs and Cl-MLEs, we modeled the mobile loop into the PpMLE structure based on the loop conformation in ReCl-MLE (Fig. 5Go). Because the loop is the same size and is in a similar position relative to the active site in the two proteins, it must assume at least approximately the same conformation in both Cl-MLEs and MLEs when it closes over the active site during catalysis. Modeling clearly shows that the interface of the loop and the active site is more polar in PpMLE than in Cl-MLEs (Fig. 5Go): His22 and Leu24 of MLE will be buried into the closed active site, whereas the equivalent amino acids in Cl-MLEs are Ile22 and Met24 (Figs. 4, 5GoGo). The more hydrophobic Cl-MLE loop may favor more contact with the protein rather than the solvent environment, whereas in MLE the middle part of the loop is more hydrophilic, in particular at His22 and Lys23. This may explain in part why the loop in Cl-MLEs tends to be more ordered than in MLEs (Figs. 4, 5GoGo).



View larger version (66K):
[in this window]
[in a new window]
  		Figure 5. Modeling of the MLE mobile loop 20–31 and comparison with Cl-MLE loop. ReCl-MLE crystal structure (A) and PpMLE crystal structure (B) with modeled loop 20–31.

 
In ReCl-MLE, the closure of the loop over the active site may be due to the presence of a Cl- in the active site bound to the Mn2+ ligand, which partially neutralizes the charge in the active site (Hoier et al. 1994; Kleywegt et al. 1996). Calculations on the electrostatic potential distribution indicate that active site without the substrate has strong positive potential in both ReCl-MLE and in the PpMLE closed model without the substrate (Fig. 6Go). Including the Cl- present in the ReCl-MLE crystal structure neutralizes the active site cavity effectively in ReCl-MLE but only partially in the MLE model (Fig. 6Go). This strongly indicates that the loop is normally closed only when substrate is bound, as otherwise the interaction between the loop and the {alpha}/ß-barrel will not be favorable.



View larger version (75K):
[in this window]
[in a new window]
  		Figure 6. Electrostatic potential distribution and cavity surfaces on ReCl-MLE (A,B) and MLE (C,D). The buried active site cavity is indicated with an arrow. (A) ReCl-MLE without the buried Cl anion in active site. (B) The inclusion the Cl anion into the active site results in neutralization of the strongly positive potential. (C) MLE with the loop modeled in closed conformation, and without the Cl anion, also has strong positive potential in the buried active site. (D) The MLE model with the Cl anion included as in ReCl-MLE. Here the positive potential is only partially neutralized because of the effect of His22 on the loop. Electrostatic potential is displayed from -20 kT/e to 20 kT/e.

 
The loop polarity could affect the rate of elimination of the Cl- leaving group on dehalogenation. In the case of 3-Cl-muconate and its lactonization product 4-Cl muconolactone, the polar MLE His22 could provide electrostatic pull for the Cl substituent to help it leave (Fig. 1Go). This would result in dehalogenation by the formation of protoanemonin and CO2 and Cl- (Fig. 1Go). The differences in the loop between Cl-MLEs and MLEs might also affect the rate of catalysis. The possibly more favorable interaction of the loop with the protein might slow down the rate of product release. Finally, as the product lactone is less polar than the substrate, it may bind more tightly to the Cl-MLE active site with the hydrophobic loop "cap" (Figs. 1, 4, 5GoGoGo) than to the more polar MLE active site, contributing to the possible slower rate of release.

Evolution of the hydrophobic core structure in Cl-MLEs
Most interestingly, in all MLEs the core of the protein close to the active site is rather loosely packed and is consistently looser in the Cl-MLEs than in the non-dehalogenating MLEs. ReCl-MLE and P51 Cl-MLE have a large cavity separated by a single ß-strand from the active site, whereas in PpMLE there are two smaller ones. Calculated cavity volumes are 105 Å3 (P51 Cl-MLE), 210 Å3 (ReCl-MLE), and 50 + 33 Å3 (PpMLE; Figs. 6, 7GoGo). The increase in cavity volume from MLE to Cl-MLE is due in part to the following changes: Thr52 to Gly, Phe103 to Ser, Ser312 to Ala, and Glu304 to Asp (Figs. 4, 7GoGo). Conversely, the MLE Leu34 is replaced by Tyr in Cl-MLEs. The replacement of the half-buried MLE Phe103 with Ser in Cl-MLEs results in a deep cleft between the 53–63 loop and the rest of the protein, leaving this second active site loop more mobile in Cl-MLEs, as can be observed in the B-factors of the P51 Cl-MLE high-resolution structure (Fig. 3Go). In addition, an important specific stabilizing interaction is lost in Cl-MLEs in comparison with MLE: The Thr52Gly change eliminates the Thr52-Glu50 hydrogen bond stabilizing the buried charge of Glu50 (Kajander et al. 2000; D.M. Cohen and P.C. Kahn, pers. comm.). This is important in linking the 53–63 loop to the {alpha}-helix 306–319, as Glu50 functions here as a helix cap. This is partially compensated for in Cl-MLEs by formation of hydrogen bonds from Tyr34 to Leu303 on the {alpha}/ß-barrel, and from a buried water molecule to Glu50. These changes in residues would be expected to increase the cavity volume in Cl-MLE by 120 Å3 (Creighton 1993). Clearly, the calculated cavity size does not match with the observed cavity size in P51 Cl-MLE. The increase in cavity volume from MLE to the open conformation of P51 Cl-MLE is only ~22 Å3. However, if we compare MLE with the ReCl-MLE structure the increase in cavity volume is remarkably 127 Å3. This follows from conformational changes.



View larger version (29K):
[in this window]
[in a new window]
  		Figure 7. The hydrophobic core packing in MLEs and Cl-MLEs near the active site. MLE (A), P51 Cl-MLE (B; open conformation), and ReCl-MLE (C; closed conformation). h indicates helix 306–319, the arrow indicates the active site capping 53–63 loop, and red lines represent hydrogen bonds (for Glu50 this also indicates the helix cap formed with the helix 306–319), and the second green mesh (labeled with *) in the ReCl-MLE structure represents the volume of the closed active site.

 
The cavities in all three MLEs are large; solvated interdomain cavities in proteins are on average ~44 Å3 in size (Hubbard and Argos 1994). As cavities in the protein interior are destabilizing (Eriksson et al. 1992), Cl-MLEs may therefore be less stable than MLEs because the cavities in Cl-MLEs are larger. Furthermore, the B-factor plots (Fig. 2Go) also indicates that in the 52–60 region and around Met302, the B-factors are higher for Cl-MLE, that is, around the cavities. The main chain B-factors of the 52–60 region are as high as 78 Å2 in P51 Cl-MLE, and the Trp59 side chain is disordered, whereas the equivalent region in MLE has B-factors of 15 to 38 Å2 (Fig. 2Go). The end of the ß-strand on the {alpha}/ß-barrel at Met302 is also more mobile than in MLE (Fig. 2Go).

Conformational changes upon loop closure and substrate binding
As a result of loop closure in ReCl-MLE, Gly52 shifts by ~1 Å relative to P51 Cl-MLE and MLE (Figs. 7, 8GoGo). In Cl-MLEs, Gly52 allows the ß-strand (residues 44–54) to twist upon loop closure over the active site without loss of the hydrogen bond to Ser103; indeed, it is then hydrogen-bonded both to the Gly52 peptide carbonyl oxygen and to the amide hydrogen (Fig. 7Go), whereas in MLE, Thr52 is hydrogen-bonded to Glu50. Thus, the conformational change is possible only in Cl-MLEs, and the large cavity so generated can be present only in Cl-MLEs. The change in conformation shows that the Cl-MLE hydrophobic core near the domain interface is more flexible than in MLEs. The Cl-MLE structure could be stabilized by the closure of the N-terminal mobile loop over the active site, and if the closed conformation in Cl-MLE was more stable than in MLE, product release would be slower. This would enable productive dehalogenation, without the release of substrate, whereas the increased flexibility near the active site would allow the required rotation of the lactone ring prior to dehalogenation.



View larger version (47K):
[in this window]
[in a new window]
  		Figure 8. Loop closure and conformation differences between P51 Cl-MLE and ReCl-MLE. C{alpha}-trace for P51 Cl-MLE structure is shown in blue and for ReCl-MLE in red. C{alpha}’s of glycines corresponding to the Thr52 in MLE are shown as dots.

 
MLE core variants and implications on structure and dynamics during catalysis
To test the effects of the differences in packing order in the MLEs versus Cl-MLEs, we constructed Cl-MLE–like, cavity-creating MLE mutants in PpMLE and measured their activities. The mutants made were Thr52Gly, Glu304Asp, Ser312Ala, the double mutants Thr52Gly-Phe103Ser and Glu304Asp-Ser312Ala, and the triple mutant Thr52Gly-Phe103Ser-Glu304Asp. The two most interesting variants made were Thr52Gly and Thr52Gly-Phe103Ser. The others had no significant further effects on cis,cis-muconate turnover, and none had increased activity toward Cl-muconates (data not shown), which is not surprising as active site changes are also required for substrate specificity (Vollmer et al. 1998). The activity measurement results showed Thr52Gly to be rather inactive (Table 1Go), whereas the double mutant Thr52Gly-Phe103Ser regained significant MLE activity (Table 1Go). Both variants behaved as wild type in purification and on native PAGE (data not shown).


View this table:
[in this window]
[in a new window]
  		Table 1. Kinetics of cis,cis-muconate conversion for PpMLE mutants
 
Both mutations affect the kcat for cis,cis-muconate turnover markedly, whereas Km (~150 µM) is similar for both (Table 1Go). Both residues are outside the active site but nevertheless affect the catalytic rate of MLE substantially without much affecting affinity. Presumably, in the double mutant, a hydrogen bond is generated between Gly52 and Ser103, analogous to Cl-MLEs. It is also interesting that creating this Gly–Ser interaction increases kcat for cis,cis-muconate (by fourfold). Although the activity in the Thr52Gly-Phe103Ser double mutant is <10% of the wild-type MLE, it is very similar to the kcat for P51 Cl-MLE with its preferred substrate 2,4-diCl-muconate (Table 1Go; Vollmer et al. 1999). We suggest that this Gly–Ser interaction and the increase in cavity volume are required for dehalogenation in Cl-MLEs by providing flexibility, for the rotation of the lactone ring for dehalogenation, while maintaining structural integrity. The observed decrease in the rate of lactonization by these mutants (Table 1Go) probably follows from increased structural flexibility. The effects are thus consistent with the idea that the Thr–Phe (MLE) or the Gly–Ser (Cl-MLE) interactions are not involved in determining substrate affinity per se, but rather with allowing bound substrate to undergo lactonisation.

Conclusions
The comparison of Cl-MLE and MLE structures reveals conformational changes upon loop closure over the active site that most likely occur on substrate binding. The structural changes observed in Cl-MLEs compared with MLEs may be energetically unfavorable and may also reduce the lactonization activity. In addition to the required changes at the active site to account for substrate specificity, the lactone ring of the muconolactone product must be able to rotate before dehalogenation of 2-Cl-muconate in Cl-MLEs can occur (Schell et al. 1999). This requires extra space or flexibility in the active site structure. This appears to be achieved by the increased flexibility in the core of the {alpha}/ß-barrel, which may make further breathing motions possible. This is consistent with the notion that cavities can have functional roles (Hubbard and Argos 1996). Finally, we suggest that the reduction in catalytic efficiency of the lactonization step (kcat) in Cl-MLEs may be required to match the lactonization rate with ring rotation and dehalogenation rate (Table 1Go). This may have been achieved by the increased structural flexibility around the large Cl-MLE cavity, also presumably required for ring rotation. Most importantly, product release may be slowed down by stabilizing the closed conformation of the active site loops, as the N-terminal mobile loop is more hydrophobic and the other 53–63 loop has a more favored closed conformation in Cl-MLEs than in MLEs. We anticipate the differences in the mobile loop polarity to be the other major factor responsible for dehalogenation in Cl-MLEs.


   Materials and methods
TOP
 Abstract
 Introduction
 Results and Discussion
 Materials and methods
References
 
Mutagenesis, protein purification, and kinetics
Mutagenesis was done using the Stratagene Quickchange mutagenesis kit. The oligonucleotides for PCR for introducing the Thr52Gly point mutation were 5'-CGG-TGA-GGC-CGG-CAC-CAT-CGG-TGG-C (forward) and 5'-GCC-ACC-GAT-GGT-GCC-GGC-CTC-ACC-G (reverse); and for the Phe103Ser mutation, 5'-GGC-CAA-GGG-CAA-CAC-TAG-CGC-CAA-GTC-GG (forward) and 5'-CCG-ACT-TGG-CGC-TAG-TGT-TGC-CCT-TGG-CC (reverse). Recombinant P51 Cl-MLE, as well as the PpMLE recombinant variants, were purified from Escherichia coli, as published (Schell et al. 1999; Vollmer et al. 1999). Kinetics of substrate conversion were measured essentially as previously reported (Vollmer et al. 1998). Measurements were done at room temperature with a Hitachi U-2010 spectrophotometer at 260 nm with 10-, 5-, and 1-mm cuvettes using {varepsilon} = 16800 M-1 cm-1 for cis,cis-muconate conversion in 20 mM Tris-HCl (pH 7.5), 1 mM MnCl2, and 10 to 1300 µM cis,cis-muconate. The decrease in absorbance was followed for 40 sec, and the slope was recorded. Three measurements were made for each of the 11 substrate concentrations (10, 20, 50, 100, 200, 300, 400, 700, 800, 1000, and 1300 µM), and the experimental curve was fitted with SigmaPlot 4.00 (SPSS Inc.).

Crystallization and data collection
The Pseudomonas P51 Cl-MLE was crystallized by vapor diffusion in sitting drops from 100 mM HEPES (pH 7.5), 100 mM MnCl2, and 12% (w/v) PEG-400. The drop contained 3 µL of 10 mg/mL protein (in 20 mM Tris at pH 7.5, 2 mM MnCl2), 1 µL of 1 mM cis,cis-muconate, and 3 µL of the well solution. Crystals appeared after 1 day and grew to their full size within a week. The final size was ~0.8 x 0.6 x 0.4 mm. Before flash freezing, the crystal was quickly dipped into well solution complemented with 30% (v/v) glycerol and 1 mM cis,cis-muconate. Data were collected on the 711 beam line (Max-Lab II), and indexed and integrated by using DENZO and Scalepack (Table 2Go; Otwinowski and Minor 1997). The high resolution data to 1.95 Å were collected separately, as the low-resolution observations otherwise were oversaturated due to the longer exposure required for the high-resolution data range (2.3 to 1.95 Å).


View this table:
[in this window]
[in a new window]
  		Table 2. Data collection and refinement statistics of the P51 Cl-MLE structure
 
Structure determination and refinement
The structure was solved by molecular replacement by using the ReCl-MLE structure (Kleywegt et al. 1996) as a search model, with nonconserved residues truncated to alanine. The model was built again by using the P51 Cl-MLE sequence, and refined with CNS (Brünger et al. 1998). The minimization, however, did not work, and R factors stayed ~30% (Rcryst/Rfree = 29/32%) despite substantial rebuilding. To circumvent the problems in refinement, the calculated phases from the structural model were used with ARP/wARP 5.1 (Perrakis et al. 1999) in the warpNtrace mode, namely, starting from scratch with the calculated phases and maps from the P51 model. This resulted in R factors of Rcryst/Rfree = 21/25%. Further refinement together with anisotropic B-factor corrections and bulk solvent modelling were done with CNS (Brünger et al. 1998). The final model had R factors of Rcryst/Rfree = 20.1/23% after grouped B-factor refinement, with good stereochemistry (Table 2Go).

Modeling and calculations
The mobile loop of PpMLE was built based on the conformation of the equivalent positions in the ReCl-MLE structure. First the MLE loop sequence was mutated into the ReCl-MLE structure and the geometry was refined on graphics by using O (Jones et al. 1991), then MLE and ReCl-MLE were aligned, and the loop coordinates were pasted into the MLE coordinate file and manually checked on the graphics. Some of the residue rotamers and side chain torsion angles were then adjusted by hand. The region exchanged to the Cl-MLE conformation contained residues 20–31. The model structure was further subjected to 200 steps of conjugate gradient minimization with CNS (Brünger et al. 1998). Cavity volumes were calculated with SURFNET (Laskowski 1995) by using a probe radius of 1.4 Å. The electrostatic calculations and visualization were done with GRASP (Nicholls et al. 1991), with simple charging: full charges for the side chains of Arg, Lys, Glu, and Asp and for Mn2+ and Cl-, with the histidines charged (with 0.5 charge on N{varepsilon} and N{delta} of the imidazole ring) and backbone neutral. The inner (protein) dielectric was raised to eight. Other figures were prepared with MOLSCRIPT (Kraulis 1991), Bobscript (Esnouf 1997), and Raster3D (Merritt and Bacon 1997). Sequence alignment was done with the GCG Wisconsin package (Accelrys Inc.).


   Acknowledgments
 
We thank the Max Lab protein crystallography beam line 711 personnel (Lund, Sweden) and Dr. Pirkko Heikinheimo for help with data collection. We also thank Dr. Ursula Kaulmann for stimulating discussions on the Cl-MLE catalysis and help with MLE mutagenesis and sequencing. This work was supported by the Academy of Finland, grants 69520 and 63252. Coordinates and structure factors have been deposited in the Protein Data Bank with the accession code 1NU5.

The publication costs of this article were defrayed in part by payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 USC section 1734 solely to indicate this fact.


   References
TOP
 Abstract
 Introduction
 Results and Discussion
 Materials and methods
 References
 
Abraham, W.R., Nogales, B., Golyshin, P.N., Pieper, D.H., and Timmis, K.N. 2002. Polychlorinated biphenyl-degrading microbial communities in soils and sediments. Curr. Opin. Mirobiol. 5: 246–253.[CrossRef][Medline]

Babbitt, P.C., Mrachko, G.T., Hasson, M.S., Huisman, G.W., Kolter, R., Ringe, D., Petsko, G.A., Kenyon, G.L., and Gerlt, J.A. 1995. A functionally diverse enzyme superfamily that abstracts the alpha protons of carboxylic acids. Science 267: 1159–1161.[Abstract/Free Full Text]

Blasco, R., Wittich, R.M., Mallavarapu, M., Timmis, K.N., and Pieper, D.H. 1995. From xenobiotic to antibiotic: Formation of protoanemonin from 4-chlorocatechol by enzymes of the 3-oxoadipate pathway. J. Biol. Chem. 270: 29229–29235.[Abstract/Free Full Text]

Brünger, A.T., Adams, P.D., Clore, G.M., DeLano, W.L., Gros, P., Grosse-Kunstleve, R.W., Jiang, J.S., Kuszewski, J., Nilges, M., Pannu, N.S., et al. 1998. Crystallography and NMR system: A new software suite for macromolecular structure determination. Acta Crystallogr. D 54: 905–921.[CrossRef][Medline]

Creighton, T.E. 1993. Proteins: Structure and molecular properties. W.H. Freeman and Company, New York.

Don, R.H. and Pemberton, J.M. 1981. Properties of six pesticide degradation plasmids isolated from Alcaligenes paradoxus and Alcaligenes eutrophus. J. Bacteriol. 145: 681–686.[Abstract/Free Full Text]

Eriksson, A.E., Baase, W.A., Zhang, X.J., Heinz, D.W., Blaber, M., Baldwin, E.P., and Matthews, B.W. 1992. Response of a protein structure to cavity-creating mutations and its relation to the hydrophobic effect. Science 255: 178–183.[Abstract/Free Full Text]

Esnouf, R.M. 1997. An extensively modified version of MolScript that includes greatly enhanced coloring capabilities. J. Mol. Graph. Model 15: 132–134.[CrossRef][Medline]

Goldman, A., Ollis, D.L., and Steitz, T.A. 1987. Crystal structure of muconate lactonizing enzyme at 3 Å resolution. J. Mol. Biol. 194: 143–153.[CrossRef][Medline]

Hasson, M.S., Schlichting, I., Moulai, J., Taylor, K., Barrett, W., Kenyon, G.L., Babbitt, P.C., Gerlt, J.A., Petsko, G.A., and Ringe, D. 1998. Evolution of an enzyme active site: The structure of a new crystal form of muconate lactonizing enzyme compared with mandelate racemase and enolase. Proc. Natl. Acad. Sci. 95: 10396–10401.[Abstract/Free Full Text]

Helin, S., Kahn, P.C., Guha, B.L., Mallows, D.G., and Goldman, A. 1995. The refined X-ray structure of muconate lactonizing enzyme from Pseudomonas putida PRS2000 at 1.85 Å resolution. J. Mol. Biol. 254: 918–941.[CrossRef][Medline]

Hoier, H., Schlömann, M., Hammer, A., Glusker, J.P., Carrell, H.L., Goldman, A., Stezowski, J.J., and Heinemann, U. 1994. Crystal structure of chloromuconate cycloisomerase from Alcaligenes eutrophus JMP134 (pJP4) at 3 Å resolution. Acta Crystallogr. D50: 75–84.

Hubbard, S.J. and Argos, P. 1994. Cavities and packing at protein interfaces. Protein Sci. 3: 2194–2206.[Abstract]

———. 1996. A functional role for protein cavities in domain: Domain motions. J. Mol. Biol. 261: 289–300.[CrossRef][Medline]

Jones, T.A., Zou, J.Y., Cowan, S.W., and Kjeldgaard, M. 1991. Improved methods for building proteins models in electron density maps and the location of errors in these models. Acta Crystallogr. A47: 110–117.

Kajander, T., Kahn, P.C., Passila, S.H., Cohen, D.C., Lehtio, L., Adolfsen, W., Warwicker, J., Schell, U., and Goldman, A. 2000. Buried charged surface in proteins. Structure 8: 1203–1214.[Medline]

Kaulmann, U., Kaschabek, S.R., and Schlömann, M. 2001. Mechanism of chloride elimination from 3-chloro- and 2,4-dichloro-cis,cis-muconate: New insight obtained from analysis of muconate cycloisomerase variant CatB-K169A. J. Bacteriol. 183: 4551–4561.[Abstract/Free Full Text]

Kleywegt, G.J., Hoier, H., and Jones, T.A. 1996. A re-evaluation of the crystal structure of chloromuconate cycloisomerase. Acta Crystallogr. D 52: 858–863.

Kraulis, P. 1991. MOLSCRIPT: A program to produce both detailed and schematic plots of protein structures. J. Appl. Cryst. 24: 946–950.

Laskowski, R.A. 1995. SURFNET: A program for visualizing molecular surfaces, cavities and intermolecular interactions. J. Mol. Graph. 13: 323–330.[CrossRef][Medline]

Levy, C.W., Buckley, P.A., Sedelnikova, S., Kato, Y., Asano, Y., Rice, D.W., and Baker, P.J. 2002. Insights into enzyme evolution revealed by the structure of methylaspartate ammonia lyase. Structure 10: 105–113.[Medline]

Merritt, E.A. and Bacon, D.J. 1997. Raster3D: Photorealistic molecular graphics. Methods Enzymol. 277: 505–524.[Medline]

Neidhart, D.J., Kenyon, G.L., Gerlt, J.A., and Petsko, G.A. 1990. Mandelate racemase and muconate lactonizing enzyme are mechanistically distinct and structurally homologous. Nature 347: 692–694.[CrossRef][Medline]

Nicholls, A., Sharp, K.A., and Honig, B. 1991. Protein folding and association: Insights from the interfacial and thermodynamic properties of hydrocarbons. Proteins 11: 281–96.[CrossRef][Medline]

Otwinowski, Z. and Minor, W. 1997. Processing of X-ray diffraction data collected in oscillation mode. Methods Enzymol. 276: 307–326.

Perrakis, A., Morris, R., and Lamzin, V.S. 1999. Automated protein model building combined with iterative structure refinement. Nat. Struct. Biol. 6: 458–463.[CrossRef][Medline]

Schell, U., Helin, S., Kajander, T., Schlömann, M., and Goldman A. 1999. Structural basis for the activity of two muconate cycloisomerase variants toward substituted muconates. Proteins 34: 125–136.[CrossRef][Medline]

Schmidt, E. and Knackmuss, H.-J. 1980. Chemical structure and biodegradability of halogenated aromatic compounds: Conversion of chlorinated muconic acids into maleoylacetic acid. Biochem. J. 192: 339–347.[Medline]

Solyanikova, I.P., Maltseva, O.V., Vollmer, M.D., Golovleva, L.A., and Schlömann, M. 1995. Characterization of muconate and chloromuconate cycloisomerase from Rhodococcus erythropolis 1CP: Indications for functionally convergent evolution among bacterial cycloisomerases. J. Bacteriol. 177: 2821–2826.[Abstract/Free Full Text]

Thompson, T.B., Garrett, J.B., Taylor, E.A., Meganathan, R., Gerlt, J.A., and Rayment, I. 2000. Evolution of enzymatic activity in the enolase superfamily: Structure of o-succinylbenzoate synthase from Escherichia coli in complex with Mg2+ and o-succinylbenzoate. Biochemistry 39: 10662–10676.[CrossRef][Medline]

Vollmer, M.D., Hoier, H., Hecht, H.-J., Schell, U., Gröning, J., Goldman, A., and Schlömann, M. 1998. Substrate specificity and product formation of muconate cycloisomerases: An analysis of wild type enzymes and engineered variants. Appl. Env. Microbiol. 64: 3290–3299.[Abstract/Free Full Text]

Vollmer, M.D., Schell, U., Seibert, V., Lakner, S., and Schlömann, M. 1999. Substrate specificities of the chloromuconate cycloisomerases from Pseudomonas sp. B13, Ralstonia eutropha JMP134 and Pseudomonas sp. P51. Appl. Microbiol. Biotechnol. 51: 598–605.[CrossRef][Medline]


Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us   Add to Digg Digg   Add to Reddit Reddit   Add to Technorati Technorati    What's this?



This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Erratum (v12,p2383)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Kajander, T.
Right arrow Articles by Goldman, A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Kajander, T.
Right arrow Articles by Goldman, A.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS