pubs.acs.org/crystal Published on Web 12/31/2010 r 2010 American Chemical Society

DOI: 10.1021/cg1013363

2011, Vol. 11516–519

Published as part of the Crystal Growth & Design virtual special issueon the 13th International Conference on the Crystallization of BiologicalMacromolecules (ICCBM13)."

Development of aCrystallization Protocol for theDbeA1Variant ofNovel

Haloalkane Dehalogenase from Bradyrhizobium elkani USDA94

T. Prudnikova,† R. Chaloupkova,§ Y. Sato,# Y. Nagata,# O. Degtjarik,† M. Kuty,†,‡

P. Rezacova, ) J. Damborsky,§ and I. Kuta Smatanova*,†,‡

†Institute of Physical Biology, University of South Bohemia in Ceske Budejovice, Zamek 136, 373 33Nove Hrady, Czech Republic, ‡Institute of Systems Biology and Ecology, Academy of Science of CzechRepublic, Zamek 136, 373 33 Nove Hrady, Czech Republic, §Loschmidt Laboratories, Department ofExperimental Biology and Research Centre for Toxic Compounds in the Environment, Faculty ofScience, Masaryk University, Kamenice 5/A4, 625 00 Brno, Czech Republic, #Department ofEnvironmental Life Sciences, Graduate School of Life Sciences, Tohoku University, 2-1-1 Katahira,Sendai 980-8577, Japan, and )Institute of Molecular Genetics, Academy of Sciences of Czech Republic,Videnska 1083, 142 20 Prague 4, Czech Republic

Received October 11, 2010; Revised Manuscript Received November 23, 2010

ABSTRACT: TheDbeA1variant of a novel haloalkane dehalogenaseDbeA (EC3.8.1.5) fromBradyrhizobium elkaniUSDA94was constructed to study the structure-function relationships between DbjA and DbeA enzymes. A DbeA1 variant carries aunique fragment of nine amino acids transplanted from the sequentially closely related enzyme DbjA from BradyrhizobiumjaponicumUSDA110 toDbeA. Here we report development of the crystallization protocol for DbeA1 and soaking experimentswith the ligands 1-fluoropentane and 1,3-dichloropropane.

Introduction

The haloalkane dehalogenases (EC 3.8.1.5) are microbialenzymes that are potentially useful for biodegradationof numerous important environmental pollutants. Theseenzymes catalyze the hydrolytic dehalogenationof halogenatedaliphatic hydrocarbons to the corresponding alcohols.1,2 Sev-eral crystallographic structures such as DhlA from Xantho-bacter autotrophicus GJ103 (PDB code: 1ede); DhaA fromRhodococcus rhodochrous NCIMB 130644,5 (1bn6); LinBfrom Sphingobium japonicum UT266 (1cv2); DmbA fromMycobacterium tuberculosis H37Rv7 (2qvb); and DbjAfrom Bradyrhizobium japonicumUSDA1108 (3a2m) showa two domain organization: the main domain and the capdomain with the active site localized in between these twodomains, where the conserved catalytic triad D, H, E/D, andtwo halide-stabilizing amino acids N/W and W are located.

A novel haloalkane dehalogenase DbeA, isolated fromBradyrhizobium elkani USDA94,9 is closely related to theDbjA enzyme from Bradyrhizobium japonicum USDA110,10

sharing71%sequence identity. TheDbjAprotein11 carries theunique nine amino acid insertion in theN-terminal part of thecap domain. We speculate that this insertion may be impor-tant for higher enzymatic activity, different temperature andpH-dependence, and a different specificity toward iodinatedand brominated aliphatic hydrocarbons, compared to theDbeA enzyme.12 A DbeA variant, herein named DbeA1,

carries fragment 143VAEEQDHAE151, originating from aunique sequence of the DbjA, inserted between D142 andA143 of DbeA. The DbeA1 variant was constructed to studythe importance of the insertion for enzyme function. Here wereport the preparation of crystals optimal for the soakingexperiments of DbeA1 with two halogenated compounds(1-fluoropentane and 1,3-dichloropropane) aimed at studyingthe active site architecture of the DbeA1 enzyme, and under-standing the structure-function relationships of DbeA andDbjA enzymes.

Experimental Section

Crystallization. The freshly isolated and purified sample ofDbeA1 protein was prepared by the procedure previously describedby Prudnikova et al. 200913 and stored at 193 K. This enzyme atconcentration of 2-9mg 3mL-1 in 100mMTris-HCl buffer pH 7.5was used for crystallization experiments. The Emerald BioStruc-tures CombiClover 96-well plates (Emerald BioSystems, BainbridgeIsland, USA) were utilized for the initial screening of crystallizationconditions of the DbeA1 protein using a sitting drop vapor-diffusionmethod.14 The hanging drop procedure applied for opti-mization of crystallization conditions was carried out in 24 Linbrowell plates (HamptonResearch,USA) and the EasyXtal tool fromQiagen Ltd., UK. The reservoir contained from 400 to 1000 μLof crystallization reagents and droplets, which consisted of 2 or4 μL of a total volume with various precipitant to protein ratiosof 1:1, 1:3, and 3:1.

As precipitants, the commercial JBScreen Classic Kits No. 1-10covering 240 different conditions (Jena Bioscience GmbH, Jena,Germany), theMDLcrystal screen kit (MolecularDimensions Ltd.,Suffolk, UK), the Hampton Research Crystal Screen kit (HamptonResearch, Aliso Viejo, USA), and the Wizard III kit (Emerald

*Corresponding author. Tel: þ420608106109, fax: þ420386361219,e-mail: ivanaks@seznam.cz; ivas@ufb.jcu.cz.

Article Crystal Growth & Design, Vol. 11, No. 2, 2011 517

BioSystems Inc., Washington, USA) were tested to screen for novelcrystallization conditions of DbeA1.

Macroseeding experiments15 were carried out by lowering pro-tein concentration to 1-2 mg 3mL-1 and decreasing the PEGconcentration down from the initial 25 (w/v) % to 20 (w/v) %.

The Douglas Instruments USA Vapor Batch 96-well plates wereused to perform the microbatch under oil crystallization.16 Two tothree millimeter layers of paraffin and/or silicon oils were used tocover the 2-4 μL protein-precipitant drops.

Diffraction Data Collection and Soaking Experiments. Two- tofour-day-old crystals of DbeA1 protein were mounted into thenylon loops (HamptonResearch,AlisoViejo,USA) and transferredto the haloalkane solutions (1% total concentration) and dissolvedin the crystallization mother liquor. After 4-24 h of soaking,crystals were flash-cooled in a liquid nitrogen stream withoutadditional cryoprotection. The loop size, used to mount crystals,was optimized for the crystal dimension to minimize the amount ofmother liquor around the crystal. The diffraction data for theDbeA1 protein soaked with 1,3-dichloropropane were collectedat the beamline X13 of the DORIS storage ring with a 0.81 Amonochromatic fixed wavelength and a MARCCD (165 mm)detector (DESY/EMBL Hamburg, Germany). A complete dataset of 360 diffraction images with a 0.5� oscillation angle (starting0� angle and exposure time 15 s) and a crystal-to-detector distance of200mmwas collected forDbeA1 crystals with 1,3-dichloropropane.The X12 beamline with a MARMosaic-CCD (225 mm) detector(DESY/EMBL Hamburg, Germany) was used for the diffractionmeasurement of the crystals soaked with 1-fluoropentane. In total,450 images with a 0.5� oscillation angle (starting angle in 90� and10 s exposure) and a crystal-to-detector distance of 200 mm werecollected. Both sets of diffraction data were processed using theHKL-3000 package.17

Results and Discussion

The novel haloalkane dehalogenase DbeA from Bradyrhi-zobium elkani USDA94 was first crystallized by Prudnikovaet al. 2009.13 The same crystallization conditions were testedfor the DbeA1 variant. Initially, only heavy amorphous pre-cipitations or quasimicrocrystals (Figure 1A) were observed.The next step of a crystallogenesis was based onmoving closerto the metastable zone on the phase transition diagram todecrease oneormore crystallization components’ concentration.

The reduction of theDbeA1 concentration and the varying ofPEG 3350 and MgCl2 concentrations yielded a one-dimen-sional (1D) needle cluster within 1-6 days (Figure 1B). Thesecrystalswere grown fromprecipitants consisting of 14.5-17%(w/v) PEG 3350, 115-125 mM MgCl2 in 100 mM Tris-HCl7.5 buffer and using the 4-5.5 mg 3mL-1 protein concentra-tion. In order to change the ionic strength of themother liquorvarious concentrations of PEG 3350 (5-20 w/v%) in combi-nation with 10 to 200 mMMgCl2 in 100 mMTris-HCl bufferpH7.5were applied to produce singleDbeA1protein crystals.Only small needle-shaped crystals were observed in precipi-tant cocktails containing the following PEG/salt combina-tions: 13%/10mM; 10%/15mM; 15%/10mM; 10%/18 mM;and 10%/25 mM of PEG 3350/MgCl2 (Figure 1C). Thecrystal size and shape were further optimized by changingthe crystallization temperature to 277K and by a substitutionof magnesium chloride with calcium acetate (Figure 1D).

The commercial screening kits (see Experimental Section)wereused to screen for novel conditions yieldingbetter qualitycrystals at both temperatures 277 and 292 K. Colorless, well-shapedmicrocrystals of theDbeA1with dimensions of 0.02�0.04 � 0.08 mm were grown within 5-8 days from thereservoir solution composed of 100 mM Tris-HCl bufferpH 7.5, 25-28% (w/v) PEG 4000 or 6000, and 130-140 mMcalcium acetate (Figure 1E). However, these crystals were notsuitable for the crystallographic soaking experiments, andthey needed size and shape improvement. For this purpose,the macroseeding experiment15 in the hanging drop vapor-diffusion procedure and the microbatch under oil16 weretested. Larger 3D single crystals with a dimension of about0.12 � 0.18 � 0.45 mm were obtained using a precipitantcocktail composed of 20% (w/v) PEG 4000, 130 mM calciumacetate, and 100 mMTris-HCl buffer pH 7.5 by the hangingdrop macroseeding and protein to precipitant ratio of 3:1(Figure 1F).

One- to three-day-old single crystal was used for soakingexperiments followed by the diffraction data collection atthe synchrotron radiation source. The DbeA1 crystals with

Figure 1. The DbeA1 variant protein crystals obtained during consequent steps of crystallization: (A) quasimicrocrystals, (B) 1D needleclusters, (C) 1D single needles, (D) 2D small needle-shaped crystals, (E) 3D microcrystals, and (F) 3D single crystals.

518 Crystal Growth & Design, Vol. 11, No. 2, 2011 Prudnikova et al.

substrates soaked in the active site exhibited C2 space groupwithcell dimensionsa=132.776A,b=77.550 A, c=78.073A,R=γ=90�,β=92.30�, anda=131.836 A,b=74.984 A, c=76.984 A,R=γ=90�,β=91.97� for 1-fluoropentane and1,3-dichloropropane, respectively.

Evaluation of the crystal-packing parameters indicatedthe presence of two molecules of the DbeA1 protein withabout 56% solvent content and a Matthew’s coefficient (Vm)of 2.8.

Diffraction data sets were collected for 1-fluoropentane at aresolution of 1.65 A and for 1,3-dichloropropane to 2.15 Awith a soaking timeof 24 and15h, respectively (Figure 2). Thecomplete data collection statistics are summarized in Table 1.

The diffraction datawill be used to determine the haloalkanedehalogenase structures by molecular replacement using thecoordinates of refined structureof theDbeAapoenzyme.9Thestructure determination will provide important informationabout the active site arrangement and structure-functionrelationships between the DbeA and DbjA proteins.

Conclusions

The protein crystallization can be realized by changingvarious physical-chemical parameters such as temperature,protein concentration, or a precipitant composition, pH, etc.Varying the concentration of one or more components of theprecipitant system can easily influence the sample crystal-lizability. The modification of the PEG-salt composition,varying ratio of compositions, and testing of alternativeseeding methods can improve the crystal quality. In the caseof the DbeA1 variant, using acetate ion9 and PEG withdifferentmolecularmasses, followed by application ofmacro-seeding procedures, dramatically changed the crystal

morphology and resulted in the production of good quality3D crystals effective for further crystallographic experiments.

Acknowledgment. Authors thank Tomas Mozga for sam-ple preparation andDevonMaloy for critical proofreading ofthe manuscript. This research was supported by theMinistry ofEducation of the Czech Republic (LC06010,MSM6007665808,MSM0021622412,MSM0021622413, and CZ.1.05/2.1.00/01.0001) and by the Academy of Sciences of the CzechRepublic(AV0Z60870520, AV0Z50520514, and IAA401630901). Wethank the EMBL for access to the X13 and X12 beamlines atthe DORIS storage ring of DESY in Hamburg.

References

(1) Janssen, D. B.; Dinkla, I. J. T.; Poelarends, G. J.; Terpstra, P.Environ. Microbiol. 2005, 7, 1868–1882.

(2) Kulakova,A.N.; Larkin,M. J.;Kulakov, L.A.Microbiology 1997,143, 109–115.

(3) Verschueren, K. H.; Franken, S. M.; Rozeboom, H. J.; Kalk,K. H.; Dijkstra, B. W. J Mol. Biol. 1993, 232, 856–872.

(4) Newman, J.; Peat, T. S.; Richard, R.; Kan, L.; Swanson, P. E.;Affholter, J. A.; Holmes, I. H.; Schindler, J. F.; Unkefer, C. J.;Terwilliger, T. C. Biochemistry 1999, 38, 16105–16114.

(5) Stsiapanava, A.; Koudelakova, T.; Lapkouski, M.; Pavlova, M.;Damborsky, J.; Kuta-Smatanova, I. Acta Crystallogr. 2008, F64,137–140.

(6) Marek, J.; Vevodova, J.; Kuta-Smatanova, I.; Nagata, Y.;Svensson, L. A.; Newman, J.; Takagi, M.; Damborsky, J. Bio-chemistry 2000, 39, 14082–14086.

(7) Mazumdar, P. A.; Hulecki, J. C.; Cherney, M. M.; Garen, C. R.;James, M. N.. Biochim. Biophys. Acta 2008, 1784, 351–362.

(8) Prokop, Z.; Sato, Y.; Brezovsky, J.; Mozga, T.; Chaloupkova, R.;Koudelakova, T.; Jerabek, P.; Stepankova, V.; Natsume, R.;Leeuwen, J. G. E.; Janssen, D. B.; Florian, J.; Nagata, Y.; Senda,T.; Damborsky, J. Angew. Chem. Int. Ed 2010, 49, 6111–6115.

Figure 2. The X-ray diffraction images of DbeA1 crystals, soaked with (A) 1-fluoropentane (diffracted to the resolution of 1.65 A) and with(B) 1,3-dichloropropane (diffracted to the resolution of 2.15 A), respectively.

Table 1. X-ray Data-Collection Statistics for the DbeA1 Crystals Soaked with 1-Fluoropentane and 1,3-Dichloropropanea

data set 1-fluoropentane 1,3-dichlorpropane

Beamline DESY X12 DESY X13detector MAR-MOSAIC 225 mm MARCCD 165 mmresolution range (A) 34.00-1.65 (1.70-1.65) 16.00 - 2.15 (2.23-2.15)unit-cell parameters (A, �) a = 132.77, b = 77.55, c = 78.07 a = 131.83, b = 74.98, c = 76.98

R = γ = 90, β = 92.30 R = γ = 90, β = 91.97space group C2 C2unique reflections 90,991 (8,523) 39,127 (3,303)redundancy 4.4 (3.7) 3.7 (3.1)completeness (%) 96.41 (84.32) 96.38 (98.9)Rmerge

b(%) 6.7 (36.6) 5.9 (22.9)<I/σ(I)> 21.7 (2.8) 18.55 (3.52)

aNumbers in parentheses refer to the highest resolution shell. b Rmerge=P

hklP

|Ii (hkl)- ÆI (hkl)æ|/P

hklP

i Ii (hkl), where Ii (hkl) is an individualintensity of the ith reflection hkl and ÆI (hkl)æ is the average intensity of the reflection hkl with summation over all data.

Article Crystal Growth & Design, Vol. 11, No. 2, 2011 519

(9) Prudnikova, T.; Rezacova, P.; Mozga, T.; Prokop, Z.; Sato, Y.;Kuty, M.; Nagata, Y.; Damborsky, J; Kuta-Smatanova, I.;Chaloupkova, R., in preparation.

(10) Sato, Y.; Monincova,M.; Chaloupkova, R.; Prokop, Z.; Ohtsubo,Y.; Minamisawa, K.; Tsuda,M.; Damborsky, J.; Nagata, Y. Appl.Environ. Microbiol. 2005, 71, 4372–4379.

(11) Sato, Y.; Natsume, R.; Tsuda, M.; Damborsky, J.; Nagata, Y.;Senda, T. Acta Crystallogr. 2007, F63, 294–296.

(12) Koudelakova, T.; Chovancova, E.; Brezovsky, J.; Monincova,M.;Fortova, A.; Jarkovsky, J.; Damborsky, J., submitted.

(13) Prudnikova, T.; Mozga, T.; Rezacova, P.; Chaloupkova, R.; Sato,Y.; Nagata, Y.; Brynda, Y.; Kuty, M.; Damborsky, J.; Kuta-Smatanova, I. Acta Crystallogr. 2009, F65, 353–356.

(14) Ducruix, A.; Giege, R. Crystallization of Nucleic Acids and Pro-teins, 2nd ed.; Oxford University Press: New York, 1999.

(15) Bergfors, T. M. Protein Crystallization; International UniversityLine: La Jolla, CA, 1999.

(16) Chayen, N. E. J. Cryst. Growth 1999, 196, 434–441.(17) Minor,W.; Cymborowski,M.; Otwinowski, Z.; Chruszcz,M.Acta

Crystallogr. 2006, D62, 859–866.