Rhodococcus sp. Strain CR-53 LipR, the First Member of a New ...

Rhodococcus sp. Strain CR-53 LipR, the First Member of a New Bacterial Lipase Family (Family X) Displaying an Unusual Y-Type Oxyanion Hole, Similar to the Candida antarctica Lipase Clan Arnau Bassegoda, F. I. Javier Pastor, and Pilar Diaz Department of Microbiology, Faculty of Biology, University of Barcelona, Barcelona, Spain

Bacterial lipases constitute the most important group of biocatalysts for synthetic organic chemistry. Accordingly, there is substantial interest in developing new valuable lipases. Considering the lack of information concerning the lipases of the genus Rhodococcus and taking into account the interest raised by the enzymes produced by actinomycetes, a search for putative lipaseencoding genes from Rhodococcus sp. strain CR-53 was performed. We isolated, cloned, purified, and characterized LipR, the first lipase described from the genus Rhodococcus. LipR is a mesophilic enzyme showing preference for medium-chain-length acyl groups without showing interfacial activation. It displays good long-term stability and high tolerance for the presence of ions and chemical agents in the reaction mixture. Amino acid sequence analysis of LipR revealed that it displays four unique amino acid sequence motifs that clearly separate it from any other previously described family of bacterial lipases. Using bioinformatics tools, LipR could be related only to several uncharacterized putative lipases from different bacterial origins, all of which display the four blocks of consensus amino acid sequence motifs that contribute to define a new family of bacterial lipases, namely, family X. Therefore, LipR is the first characterized member of the new bacterial lipase family X. Further confirmation of this new family of lipases was performed after cloning Burkholderia cenocepacia putative lipase, bearing the same conserved motifs and clustering in family X. Interestingly, all lipases grouping in the new bacterial lipase family X display a Y-type oxyanion hole, a motif conserved in the Candida antarctica lipase clan but never found among bacterial lipases. This observation contributes to confirm that LipR and its homologs belong to a new family of bacterial lipases.

L

ipases are glycerol ester hydrolases acting on acyl glycerols to liberate free fatty acids and glycerol. They catalyze reactions involving insoluble lipid substrates at the lipid-water interface and preserve their catalytic activity in organic solvents (23), acting as powerful tools for catalyzing not only hydrolysis but also various reverse reactions such as esterifications or transesterifications in anhydrous organic solvents (16, 23). Moreover, microbial lipases catalyze reactions with high specificity, regioselectivity, and enantioselectivity, constituting the most important group of biocatalysts for synthetic organic chemistry and other biotechnological applications (4, 18, 34, 35). Accordingly, there is substantial interest in developing new lipases for use in food, biomedical, or chemical industries (18). Despite the large number of microbial lipases identified, cloned, and characterized in the last decades (3, 11, 12, 29–31, 36, 37, 39), there are still some cultivable microbial species which are promising sources of new lipases that have not yet been explored. In this respect, many rhodococci display the ability to degrade different alkanes or show tolerance to hydrocarbons, being capable of producing several compounds with medical, industrial, and nutritional applications, such as beta-carotenes and fatty acidcontaining extracellular polysaccharides. They can also transform and degrade a wide range of chemicals, thus showing an enormous potential as a source of enzymes (25). In the last few years, different enzymes from rhodococci with biotechnological applications have been cloned and characterized (25), but no information concerning the lipases of the genus Rhodococcus is yet available. Only the amino acid sequences of two putative enzymes showing certain similarity to lipases have been elucidated (33), although no biochemical data have yet been provided. Most known bacterial lipases have been grouped by Arpigny

1724

aem.asm.org

and Jaeger (2) into eight families on the basis of conserved amino acid sequence motifs and biological properties. The increasing number of newly described bacterial lipases led in 2002 to the revision of true lipases included in family I, which underwent a reorganization (20). However, no new bacterial lipase families were described at that time. More recently, new families of bacterial lipases have been proposed by several authors (26, 27, 28) on the basis of phylogenetic criteria but not on the basis of the presence of conserved blocks of sequence motifs and/or biological function, as was the case for the original bacterial lipase classification (2). Rhodococcus sp. strain CR-53 was previously isolated in our laboratory from a subtropical soil sample (40). Physiological tests plus analysis of the 16S rRNA gene of the strain revealed a high level of similarity (99%) to other strains described to be Rhodococcus erythropolis, although a perfect match was not obtained (11). A remarkable trait of the strain was its high lipolytic activity (11, 40), thus being a good candidate for lipase prospection. Considering the lack of information concerning the lipases of the genus Rhodococcus and taking into account the interest raised by the enzymes produced by this actinomycete, we describe here the isolation, sequence analysis, and enzymatic characterization of a novel lipase, LipR, from Rhodococcus sp. CR-53. The discovery of

Received 27 July 2011 Accepted 22 December 2011 Published ahead of print 6 January 2012 Address correspondence to Pilar Diaz, [email protected]. Copyright © 2012, American Society for Microbiology. All Rights Reserved. doi:10.1128/AEM.06332-11

0099-2240/12/$12.00

Applied and Environmental Microbiology

p. 1724 –1732

New Family X Rhodococcus Lipase

LipR and the knowledge of its particular properties provided evidence to propose the existence of a new family of bacterial lipases. Following the criteria and the nomenclature established by Arpigny and Jaeger (2), LipR would be the first member assigned to a newly proposed bacterial lipase family designated family X. MATERIALS AND METHODS Strains, plasmids, and growth conditions. Wild-type strain Rhodococcus sp. CR-53 was grown in Luria-Bertani medium for 3 days at 20°C, pH 6.8, under aerobic conditions (11). Escherichia coli strain Origami (Novagen) was routinely cultured overnight at 37°C in Luria-Bertani broth or on Luria-Bertani agar plates and was used as the host strain for cloning and expression of lipase-encoding genes. Plasmids pGaston (19) and pET101/ D-TOPO (Invitrogen) were used as expression vectors. Strain Burkholderia cenocepacia J2315 (kindly provided by H. Gomes Leitao) was cultured overnight at 37°C in Luria-Bertani broth or on Luria-Bertani agar plates. DNA manipulation and cloning. DNA manipulations were carried out according to Sambrook and Russell (41). Plasmid DNA was purified using commercial kits (Illustra PlasmidPrep; GE Healthcare, United Kingdom). Restriction nucleases (Roche, Suisse) and thermostable polymerases Taq and Pfu (Biotools, Spain) were used according to the manufacturers’ instructions. PCR amplifications were performed in a GeneAMP PCR system 2400 (Perkin Elmer) using different cycling periods, seldom including a hot-start procedure (41). For lipR isolation, degenerated primers FWintLipRdeg (5=-GGN TAY TCB GGN GGN GCS ATC GCS AC-3=) and BKintLipRdeg (5=-GAT RAA CAT MGG DAT SGT CGG-3=) were used to amplify a central consensus region, using Rhodococcus sp. CR-53 genome as a template. The whole sequence of lipR was then determined by arbitrary primed PCR (AP-PCR) (7) starting from the known sequence of the amplified central DNA fragment of the gene. Specific primers FWLipRNdeI (5=-AAA ACA TAT GGC ATC AAA GAT TCT CTT TCA GGG-3=; NdeI site in bold) and BKLipRHindIII (5=-GAT CAA GCT TTC AGC CCG CCG CGT G-3=; HindIII site in bold) were designed for PCR amplification of the complete lipR open reading frame (ORF). Amplified DNA was purified through chromatography (Illustra PlasmidPrep; GE Healthcare) and ligated to NdeI/HindIII-digested plasmid pGaston. The resulting recombinant plasmid, pGastonLipR, was transformed into E. coli Origami, producing recombinant clone Origami/ pGastonLipR. For B. cenocepacia lipBX isolation, the specific primers FWLipBX (5=-CACC ATG TCC TCC AGA CGT TTC ATG ATG G-3=) and BKLipBX (5=-CTA GTG CAG CGT CTC GGG CGC GA-3=) were used, and the ligation and cloning procedure of the pET101/D-TOPO manual was applied (Invitrogen). To obtain the nucleotide sequences of DNA, PCR-amplified fragments were analyzed using an ABI Prism BigDye Terminator (version 3.1) cycle sequencing kit (Applied Biosystems) and the analytical system CEQ 8000 (Beckman-Coulter), available at the Serveis Científico Tècnics of the University of Barcelona. DNA samples were routinely analyzed by agarose gel electrophoresis (41) and stained with GelRed, 0.27% (vol/vol). Nucleic acid concentration and purity were measured using an ND-100 NanoDrop spectrophotometer (3). Bioinformatics tools. BLAST searches were routinely performed for DNA or protein sequence analysis (1). Alignments were performed using the MAFFT (Multiple Alignment Fast Fourier Transform) server (http:// mafft.cbrc.jp/alignment/server) (22). The phylogenetic tree of LipR was constructed using the software MEGA4 (44). The BioEdit sequence alignment editor (version 7.0.1) (17) was used for restriction pattern determination. The web tool ORF Finder (http://www.ncbi.nlm.nih.gov/projects/ gorf/) was used to identify the open reading frames, and the software ContigExpress was used to assemble DNA sequences (Vector NTI, version 8; Invitrogen, Carlsbad, CA). Identification of putative signal peptide and transmembrane regions was performed through SignalP (version 3.0) software (9) (http://www.cbs.dtu.dk/services/SignalP/). Secondary structure prediction was achieved using the PSIPRED protein structure prediction server (6) (http://bioinf.cs.ucl.ac.uk/psipred/). Structure threedimensional (3D) homology models were generated with the software

March 2012 Volume 78 Number 6

YASARA (version 9.6.28; www.yasara.org) using default settings. PyMOL software (PyMOL molecular graphics system, version 1.2r3pre; Schrödinger, LLC) was used to visualize the 3D protein models. Enzyme classification and family assignment were performed with the CLANS (Cluster Analysis of Sequences) program (http://toolkit.tuebingen.mpg.de/clans) (15). P values for high-scoring segment pairs (HSPs) better than 0.05 were considered to be significant. The ExPASy proteomics server (http://us.expasy.org/tools/protparam.html) was used to analyze the protein physicochemical parameters (ProtParam tool). Activity assays and protein analysis. Qualitative lipase activity assays were performed either on rhodamine-triolein-supplemented agar plates (24) or on paper, using methylumbelliferone (MUF)-derivative substrates, as previously described (8, 32). Routinely, activity determination was performed by a colorimetric microassay (3, 38) based on measuring the release of para-nitrophenol (pNP) from pNP-derivative fatty acid substrates. One unit of activity was defined as the amount of enzyme that released 1 ␮mol of pNP per minute under the assay conditions used. Substrate specificity assays were performed on several pNP-derivative substrates at pH 7. Optimum temperature was determined over the range from 4 to 70°C at pH 7, using 1 mM pNP-caproate as a substrate (38). Optimum pH was established by analysis of the activity over a range from pH 3 to 10 with Britton-Robinson buffer (5) at 40°C. To subtract the effect of pH on color release after incubation, 100 ␮l 2 M Tris HCl (pH 7) was added to the reaction mixture to equal the different pH of each reaction. For inhibition studies, assays were performed on pNP-caproate in the presence of several metal ions, assayed at 1 and 10 mM concentrations. In order to determine thermal and pH stability, enzyme samples were preincubated over a temperature range of 4 to 80°C and over a pH range of 3.0 to 10.0, and the residual activity was analyzed under the standard reaction conditions (3, 38). Enzyme kinetic parameters were calculated using the optimum substrate, pH, and temperature from the activity-versussubstrate concentration curves by regression analysis, performed using the BioFitWeb tool (http://biofitweb.cox-thurmond.net/FittingRoom/ FittingTools.html). LipR purification. A three-step column chromatography protocol was used to purify Rhodococcus sp. CR-53 LipR from Origami/pGastonLipR crude cell extracts, prepared as described previously (36). Purifica¨ KTA fast protein liquid chromatograph tion was performed with an A (FPLC; Amersham Biosciences). Prior to purification, 10-foldconcentrated crude cell extracts were precipitated overnight with 30% (wt/vol) (NH4)2SO4 at 4°C and suspended in 20 mM Tris-HCl (pH 7.0) to get a 25-fold-concentrated sample. A molecular exclusion column, HiLoad 16/60 Superdex prep grade, was used as a first step, using 20 mM Tris HCl (pH 7) elution buffer supplemented with 0.02% sodium azide and a flow rate of 1 ml/min with a maximum pressure of 0.5 MPa. Collected fractions were screened for activity using the fluorimetric paper assay (8, 32), and those showing lipase activity were confirmed by enzymatic assay (39) plus SDS-PAGE and zymogram analysis (8, 32). Concentrated active fractions were pooled and injected into a Tricorn Mono Q 5/50 GL ion exchange chromatography column. An NaCl gradient was applied for elution with 20 mM Tris HCl (pH 7) buffer supplemented with 0.02% sodium azide, using a flow rate of 2 ml/min and a maximum pressure of 4 MPa. Eluted fractions were collected and lipase activity was checked as before. A third purification step involving molecular exclusion chromatography was performed using a Tricorn Superdex 200 10/300 GL column and 20 mM Tris HCl (pH 7) elution buffer supplemented with 0.02% sodium azide, at a flow rate of 1 ml/min and a maximum pressure of 1.5 MPa. Collected fractions were checked for activity, pooled, and used as essentially purified LipR.

RESULTS

LipR gene isolation and cloning. Exhaustive database searches of putative Rhodococcus lipases discovered the presence of an ORF (gene identifier, 3721105; GenBank accession no. YP_345621) in a sequenced linear plasmid from strain R. erythropolis PR4 (43),

aem.asm.org 1725

Bassegoda et al.

TABLE 1 Summary of LipR purification procedure and yield Fraction

Activity (U/ml)

Fraction vol (ml)

Total activity (U)

Protein concn (mg/ml)

Sp act (U/mg)

Purification (fold)

Yield (%)

Crude cell extract Ammonium sulfate precipitation From size exclusion chromatography From ion exchange chromatography Purified LipR

0.017 0.079 0.080 0.064 0.014

50 4 2 1 2

0.86 0.32 0.16 0.064 0.029

3.5 15.76 1.95 0.62 0.05

0.004 0.005 0.041 0.104 0.29

1 1.02 8.37 21.25 60.63

100 36.87 18.63 7.52 3.46

which was annotated as a putative lipase. A BLAST search revealed similarities with putative, uncharacterized lipases from a Streptomyces sp. (50 to 56%), Burkholderia cenocepacia (50%), Nocardia farcinica (38%), and Candida albicans (30%). A multiple alignment of these sequences and the sequence of R. erythropolis PR4 ORF YP_345621 was performed to search for consensus sequences or conserved motifs. Two centrally located regions, -GY-S-G-G-A-I-A-T and P-T-I-P-M-F-I-, were selected to design the degenerated primers FWintLipRdeg and BKintLipRdeg, which allowed amplification of a 403-bp DNA fragment from the Rhodococcus sp. CR-53 genome. The amplified DNA fragment displayed 78% sequence identity with the putative lipase encoded by ORF YP_345621. From the known sequence of the amplified DNA fragment, AP-PCR was performed to unveil the sequence of the complete ORF encoded by Rhodococcus sp. CR-53. Nucleotide sequence analysis revealed that the new 1,350-bp ORF codes for a putative protein of 449 amino acids displaying 76% identity with the putative lipase encoded by R. erythropolis PR4 ORF YP_345621. The new amplified gene, named lipR, was further ligated to vector pGaston and cloned into E. coli Origami for overexpression and correct folding of the encoded LipR protein. To avoid expression problems caused by codon bias in recombinant clone Origami/pGaston-LipR, the alternative GTG start codon of the native sequence was replaced by a conventional ATG start codon. The presence of lipase activity in the newly produced recombinant clone was confirmed through a colorimetric activity microassay using crude cell extracts (38). It must be stated that while the cloning of lipR was being performed, the complete genome sequence of R. erythropolis PR4 was published (http://www.ncbi.nlm.nih.gov/genomeprj/20395), providing new data. The newly identified and cloned gene lipR from Rhodococcus CR-53 showed 96% nucleotide sequence identity to the annotated ORF YP_002769179 from strain R. erythropolis PR4, an ORF located in the chromosome of the strain. Moreover, this ORF encodes a protein annotated as a putative lipase showing 98% identity to LipR. Therefore, strain R. erythropolis PR4 bears at least two genes coding for putative lipases, one of them located in the linear plasmid pREL1 (ORF YP_345621) (43) and a second one (ORF YP_002769179) placed in the chromosome, with the latter being equivalent to Rhodococcus sp. CR-53 LipR. LipR purification and characterization. For characterization purposes, LipR was purified by fast protein liquid chromatography from crude cell extracts of the recombinant Origami/pGastonLipR clone. The purification process rendered a 60.6-fold purified LipR, with a yield of 3.46% and a specific activity of 0.3 U · mg⫺1 protein (Table 1). Purified LipR migrated as a single band with an estimated molecular size of 43 kDa on SDSpolyacrylamide and zymogram gels. The observed differences be-

1726

aem.asm.org

tween the molecular mass deduced (46.4 kDa) and that observed on gels (43 kDa) can be justified by removal of the signal peptide in strain Origami. The isoelectric point (pI) of the purified lipase was determined by isoelectric focusing gels coupled with zymogram analysis. A lipolytic band with a pI of between 5.2 and 5.5, in agreement with the deduced pI for LipR after removal of the signal peptide, was observed (data not shown). LipR substrate specificity assays were performed using several pNP derivatives, with the highest activity (100%) exhibited on pNP-caprate, a medium-chain-length substrate. LipR also displayed high activity against pNP-butyrate, pNP-valerate, and pNP-caprylate (Table 2). Much lower activity was observed for pNP derivatives longer than 10 carbon atoms, although activity could be observed on insoluble long-chain-length triolein and on MUF derivatives like MUF-stearate or MUFoleate (8) (data not shown). The effect of temperature and pH on the activity of LipR was also determined using pNP-caprate as a substrate. The enzyme displayed an optimum pH and temperature of 7 and 40°C, respectively (Fig. 1 A and B). Analysis of the kinetic parameters of LipR on pNP-caprate showed a typical Michaelis-Menten behavior with no interfacial activation. The calculated apparent Vmax and Km values found for the enzyme were 1.1 ⫾ 0.039 U · mg⫺1 and 0.14 ⫾ 0.029 mM, respectively. Concerning stability, LipR displayed an interestingly high activity after 1 h of incubation over a wide range of temperatures, from 4 to 60°C, exhibiting at least 60% of the highest activity at all these temperatures (data not shown). When thermal stability was tested for longer periods, the enzyme maintained almost full activity after 24 h of incubation at 20°C and retained 93% and 43% residual activities after incubation at 4°C for 15 and 30 days, respectively. Residual activities of 82% and 42% were detected after 15 and 30 days of incubation at 20°C, respectively. The enzyme was also very stable over a pH range of from 6 to 8 (data not shown). The effect of different ions and chemicals on the activity of the

TABLE 2 Substrate profile of LipRa pNP derivative

Sp act (mU/mg protein)

pNP-butyrate (C4:0) pNP-valerate (C5:0) pNP-caprylate (C8:0) pNP-caprate (C10:0) pNP-laurate (C12:0) pNP-palmitate (C16:0)

341 (45)b 587 (78) 706 (94) 753 (100) 198 (26) 106 (14)

a Activity values are the means of at least three independent assays using E. coli Origami/pGaston-LipR cell extracts. b Data in parentheses are percentages of the total.



FIG 1 Selected aspects of LipR characterization assayed on pNP-caprate. (A) Optimum temperature; (B) optimum pH; (C) effect of several ions on LipR activity. The activity values are the means of at least three independent determinations.

purified enzyme was analyzed using pNP-caprate as a substrate. LipR activity was drastically reduced in the presence of Fe2⫹, Cu2⫹, and Zn2⫹, being almost completely inhibited when the concentration of these ions was 10 mM (Fig. 1C). On the other hand, when a 10 mM concentration of Mg2⫹, Mn2⫹, or Ca2⫹ was used in the reaction mixture, there was an activity increase of about 30% (Fig. 1C). The enzyme was strongly inhibited by low concentrations of SDS (0.1%), and a 50% residual activity was observed in the presence of 1% Triton X-100, while no activity was detected at 5% concentrations of this detergent (data not shown). LipR sequence and structure analysis. Analysis of the amino acid sequence of LipR revealed a predicted molecular mass of 46 kDa, a deduced pI of 4.33, and a high content of nonpolar amino acids (59.3%). The presence of a signal peptide with cleavage between positions 28 and 29 (VAA-QP) indicates the extracellular location of the enzyme. Secondary structure analysis of LipR revealed the typical ␣/␤ fold of lipases and the presence of the characteristic nucleophile elbow, constituted by the pentapeptide GlyTyr-Ser-Gly-Gly with the catalytic serine residue (Ser212) embedded in it. A LipR 3D homology model constructed using Candida antarctica lipase A (CALA; Protein Data Bank accession no. 2veo) as a template allowed confirmation of the catalytic triad Ser212, Asp372, and His404 (Fig. 2A). Furthermore, the presence of a disulfide bond between cysteines Cys388 and Cys432 was also detected (Fig. 2A). LipR family assignment. Classification of the newly isolated


LipR into one of the bacterial lipase families previously described (2, 20) was performed. The web tool CLANS (Cluster Analysis of Sequences) was used for this purpose by comparing all sequences that define each bacterial lipase family with the sequence of LipR. Surprisingly, LipR displayed no significant similarities to any of the bacterial lipase sequences that define the bacterial lipase families described to date. That made it difficult to assign LipR to any existing lipase family. Further BLAST searches revealed several sequences showing relevant similarity to LipR. Among these, of course, were the putative lipases from R. erythropolis PR4 mentioned above (98% and 76% sequence identities) and other noncharacterized, putative lipases from R. erythropolis SK121 (75%), Streptomyces sp. (50 to 60%), Catenulispora acidiphila (56%), Burkholderia cenocepacia (50%), Nocardia farcinica (38%), Mycobacterium sp. (31%), and Candida albicans (30%). A phylogenetic tree (Fig. 3) was generated using LipR and the sequences displaying over 30% sequence identity to LipR listed above, including also the lipases of the C. albicans lipase-like family (abH38.2). As can be observed in Fig. 3, lipases of the C. albicans lipase-like family (uncolored) grouped in a cluster different from that of LipR, which included only putative bacterial lipases (blue), most of them from Gram-positive bacteria, but also containing a putative lipase from Burkholderia cenocepacia. Other phylogenetic clusters grouped putative bacterial lipases derived from genomes or metagenomes of Mycobacterium species (green) or other Rhodococcus species (orange). Therefore,

aem.asm.org 1727

Bassegoda et al.

FIG 2 3D homology model of LipR with the catalytic residues, the conserved motifs, the disulfide bond, and the oxyanion hole highlighted. (A) The amino acids of the conserved blocks among family X bacterial lipases are mostly located on LipR in the region of the active site and are depicted here as blue spheres. Among them, the amino acids forming the consensus motif of the oxyanion hole are depicted as orange spheres. The amino acids forming the catalytic triad are red sticks, and the disulfide bond formed by the two cysteines (Cys388 and Cys432) is shown as yellow sticks. (B) Detail showing the position of the conserved motif Tyr-Asp-Ser-Leu (as spheres) constituting the Y-type oxyanion hole of LipR, which has never been found before among bacterial lipases but which is shared with most lipases of the C. antarctica clan. The two proposed amino acids involved in the oxyanion hole (Tyr110 and Asp111) are depicted as orange sticks, and the catalytic serine (Ser212) and histidine (His404) are depicted as red sticks. The 3D model was obtained through YASARA.

LipR consistently grouped into a phylogenetic lineage constituted by a cluster of putative, uncharacterized bacterial lipases located far from any previously described lipases. These results suggest that the LipR cluster could constitute a new bacterial lipase family. A second multiple-sequence alignment including only the putative bacterial lipases grouped in the LipR cluster and showing over 38% identity to LipR was performed. On the basis of this new alignment, four conserved sequence blocks were found among the aligned sequences (Fig. 4). It could be observed that the different consensus blocks and the most conserved amino acids in all sequences were located on the LipR 3D model structure near the catalytic site (Fig. 2A), indicating that this kind of active site is a highly conserved region. However, no matches of the conserved motifs existed with respect to the previously described lipase families (26, 27, 28), suggesting that neither LipR nor the putative lipase sequences clustered with it and used for the alignment be-

1728

aem.asm.org

longed to any of the known bacterial lipase families. Moreover, the sequence Tyr-Asp-Ser-Leu (Fig. 2B and 4; residues in bold are in the Y-type oxyanion hole), present in one of the conserved blocks surrounding the active-site region and located in a loop near the catalytic serine (Fig. 2B and 4), caught our attention because of its similarity to the particular CALA Y-type oxyanion hole, which has never been described in bacterial lipases (13). This consensus motif contains a conserved tyrosine shared by all members of the C. antarctica lipase A-like superfamily. Lipases containing the Y-type oxyanion hole have been classified as Y-class lipases (46). According to the previous results, LipR would constitute the first characterized member of a new family of bacterial lipases, designated family X. The lack of alignment of LipR with lipases from previously described families, its clustering with only putative bacterial lipases, the presence of four conserved blocks of sequence in all these uncharacterized lipases, plus the finding of a rare oxyanion hole much closer to the oxyanion holes of the C. antarctica lipase clan than to those of bacterial lipases give support to this proposal. Isolation of a new family X lipase from Burkholderia cenocepacia. Most bacteria coding for the putative lipases that constitute family X belong to the Actinomycete group, with identities to LipR ranging from 30 to 60% (see above). Interestingly, several putative lipases from family X belong to the phylogenetically more distant species Burkholderia cenocepacia, with an average identity to LipR of 50%. In order to confirm that other members of the newly discovered family X also display the properties of lipases, the Gram-negative bacterium Burkholderia cenocepacia was chosen as a good candidate for lipase isolation, cloning, and activity confirmation. Specific primers were used for isolation of the putative lipase encoded by Burkholderia cenocepacia. The amplified DNA fragment was sequenced for confirmation, ligated into plasmid pET101, and further cloned in E. coli Origami. Cell extracts from the recombinant clone were assayed for lipase activity using both MUF- and pNP-derivative substrates, together with SDS-PAGE coupled to zymogram analysis. All assays showed that the newly cloned B. cenocepacia gene encoded a real lipase, which was designated LipBX. Like LipR, LipBX displayed preference for pNPcaprate and maintained high activity rates on shorter-chainlength substrates (data not shown). Therefore, both Rhodococcus sp. CR-53 LipR and B. cenocepacia LipBX are lipases belonging to the new bacterial lipase family X, sharing a particular Y-class oxyanion hole with the C. antarctica lipase clan and displaying similar specificities. DISCUSSION

To date, no information about lipases produced by members of the genus Rhodococcus has been available. The availability in our laboratory of strain Rhodococcus sp. CR-53, showing high lipolytic activity (11), made this isolate a good candidate with which to prospect for novel lipases from rhodococci. An exhaustive database search revealed the presence of an ORF (YP_345621) in a sequenced linear plasmid from strain R. erythropolis PR4 (43) that was annotated as a putative lipase. Computational analysis of the protein sequence encoded by ORF YP_345621 showed the presence of an ␣/␤ fold, a catalytic triad, and high contents of nonpolar amino acids, which are traits typical of lipases (14, 21). Different BLAST searches revealed that the product of ORF YP_345621 exhibited relevant sequence identity



FIG 3 Phylogenetic tree including the amino acid sequences of putative, uncharacterized lipases showing over 30% identity to LipR. The blue box includes lipases of the LipR cluster. Green box, cluster constituted by putative lipases from Mycobacterium species; orange box, cluster including putative lipases from different Rhodococcus species. The tree was obtained using the software MEGA4 (44).

with putative lipases from very different bacterial strains, such as Streptomyces sp. (50 to 56% identity), Burkholderia cenocepacia (50% identity), and Nocardia farcinica (38% identity), and some lipases from Candida albicans (30% identity). All of them displayed identity either with putative lipases or with an experimentally described lipase (C. antarctica CALA), indicating that the encoded protein was probably a lipase. In order to identify a lipase-coding gene in the strain Rhodococcus CR-53, two degenerated primers were designed on the basis of consensus regions shared by the product of ORF YP_345621 and the other similar putative lipases described above. This allowed amplification of a small fragment displaying high sequence identity to the putative lipase encoded by ORF YP_34562. Using AP-PCR, the whole ORF of a supposed Rhodococcus sp. CR-53 lipase was identified, showing 74% nucleotide sequence identity to ORF YP_345621 and the most common traits of lipase sequences. Therefore, these results indicated that a new lipase-coding gene from Rhodococcus sp. CR-53 had been identified and its complete sequence had been isolated. The new gene was called lipR, and the encoded protein, probably a secreted lipase, was designated LipR. LipR was cloned in E. coli Origami and purified for further characterization. Activity assays performed on trioleinsupplemented plates or on MUF-oleate and MUF-stearate produced high fluorescence emission (data not shown), while assays on pNP-derivative substrates revealed that LipR showed preference for acyl groups of medium chain length (C10:0 to C8:0), with lower activity on longer- and shorter-chain-length substrates. A higher affinity for shorter-chain-length pNP-derivative substrates


has already been described for other lipases, which bear activity on long-chain substrates but display a higher affinity for shorterchain-length pNP derivatives (3, 11, 12, 21). Analysis of the kinetic parameters of the enzyme using pNP-caprate (C10:0) revealed a Michaelis-Menten plot with no interfacial activation, a typical behavior of esterases and a few true lipases (14). However, comparison of the biochemical properties of LipR with those of other lipases is difficult to perform since no information about other Rhodococcus lipases exists. The optimum temperature and pH shown by LipR (Fig. 1) would be in agreement with the natural environmental conditions of the strain, which was isolated from a subtropical soil sample in Puerto Iguazú, Argentina (40), a region with an average pH of 6 and a range of temperatures of from 15°C to 24°C (45; Argentine Meteorological Service [www.smn.gov.ar]). Moreover, LipR displayed good long-term stability at up to 30°C, decreasing at temperatures over 40°C but remaining fully active after 24 h of incubation at 20°C. Analysis of the effect of different agents on the activity of LipR revealed high tolerance for the presence of ions and chemical agents in the reaction mixture, supporting the idea that a soil isolate should be able to tolerate changing environmental conditions. In fact, LipR activity was drastically reduced only by Fe2⫹, Cu2⫹, and Zn2⫹ (Fig. 1), frequently described to be inhibitors of lipase activity (3, 16). On the contrary, LipR activity was usually increased by the presence of divalent cations, a fact previously reported for other lipases (3, 12, 16, 30, 36). Classification of LipR into one of the bacterial lipase families previously described (2) revealed that LipR has no significant sim-

aem.asm.org 1729

Bassegoda et al.

FIG 4 Blocks of sequences conserved in the LipR cluster of putative bacterial lipases assigned to bacterial lipase family X. The aligned sequences were the putative lipases forming the LipR cluster. YP_002769179, putative lipase from R. erythropolis PR4; ZP_04383612, putative triacylglycerol lipase from R. erythropolis SK121; YP_345621, putative lipase from R. erythropolis PR4; ZP_04386725, triacylglycerol lipase from R. erythropolis SK121; ZP_07281779, putative triacylglycerol lipase from Streptomyces sp. strain AA4; ZP_06908675, putative triacylglycerol lipase from S. pristinaespiralis ATCC 25486; ZP_04693904, putative lipase from S. roseosporus NRRL 15998; YP_003111889, putative triacylglycerol lipase from Catenulispora acidiphila DSM 44928; ZP_07716111, putative triacylglycerol lipase from Aeromicrobium marinum DSM 15272; YP_002235365, putative exported lipase from Burkholderia cenocepacia J2315; ZP_04942934, putative triacylglycerol lipase from B. cenocepacia PC184; YP_625098, putative triacylglycerol lipase from B. cenocepacia AU 1054; YP_001778248, putative triacylglycerol lipase from B. cenocepacia MC0-3; YP_370846, putative triacylglycerol lipase from Burkholderia sp. strain 383. ZP_07149178, putative triacylglycerol lipase from Corynebacterium resistens DSM 45100; YP_119470, putative lipase from Nocardia farcinica IFM 10152. The amino acids constituting the conserved pentapeptide are underlined. Alignment was performed with the web tool MAFFT. Identical amino acids are on a black background, whereas a gray background indicates identical or equivalent amino acids.

ilarities with the bacterial lipase sequences that define any of the bacterial lipase families described to date. In fact, LipR displays unique amino acid sequence motifs that clearly separate it from the other families of bacterial lipases previously described (2). However, the different BLAST searches revealed that LipR has relevant sequence identity with other nonclassified, uncharacterized putative lipases from different origins, including some lipases from the C. antarctica clan (C. antarctica lipase A-like superfamily abH38) (46). In order to perform a new classification trial, a multiple-sequence alignment and a derived phylogenetic tree were prepared using the sequences from putative lipases showing over 30% identity to LipR (Fig. 3). The derived phylogenetic tree showed a well-defined cluster constituted by LipR and several sequences of putative lipases from different bacterial species (R. erythropolis, a Streptomyces sp., B. cenocepacia, Corynebacterium resistens, Nocardia farcinica, Burkholderia cenocepacia, and Aeromicrobium marinum), all of them displaying over 38% sequence identity to LipR. Interestingly, lipases of the C. albicans lipase-like family (abH38.2) grouped in a different cluster (Fig. 3) far from

1730

aem.asm.org

that of LipR but within the same root. The fact that other putative Rhodococcus lipases constituted a differentiated cluster (Fig. 3) also separated from that of the Mycobacterium putative lipases is also remarkable. The observed clustering diversity suggests that new, yet unknown, bacterial lipase families could still exist. Alignment of the LipR cluster showed that it is defined by different conserved amino acid sequence motifs. All these consensus motifs were identified on the 3D homology model of LipR and were found to be located close to the active site of the enzyme. Among the consensus motifs surrounding the active-site region, the sequence Tyr-Asp-SerLeu, placed in a loop near the catalytic serine, caught our attention for its similarity with the CALA oxyanion hole. This type of oxyanion hole is called Y type (13) and is shared by all members of the C. antarctica lipase A-like superfamily (abH38), which includes the C. albicans lipase-like family (abH38.2) (46). Different structural studies revealed that the CALA oxyanion hole is constituted by an amino acid pair located in a loop close to the catalytic serine (10, 42, 46). An aspartic acid (10, 42) and a tyrosine (13, 46), both in the same loop, are the two candidate residues proposed to constitute the oxyanion



hole. In the case of LipR, both residues can be found in a loop close to the catalytic serine and are included in a highly conserved amino acid sequence motif, strongly suggesting that they are essential for substrate binding. Accordingly, LipR displays the same oxyanion hole as that of CALA, classified as a Y-type oxyanion hole. Moreover, LipR has sequence identity with some secretory lipases from C. albicans, also classified as members of the C. albicans lipase-like family (abH38.2) (2, 46). Interestingly, none of the lipases used to define the different bacterial lipase families (2) has a Y-type oxyanion hole. This fact again provides evidence of the differences existing between LipR and all members of the bacterial lipase families described to date and justifies the difficulty with classification of LipR into any of the previously described families. Taking into consideration all these evidences, we propose that LipR be the first member of a new family of bacterial lipases that should be included in the previous classification (2) as new family X. In order to increase the reliability of this new family, the putative lipase from Burkholderia cenocepacia J2315, included in the LipR cluster, was isolated and cloned, confirming its lipolytic activity with a profile similar to that of LipR. ACKNOWLEDGMENTS We thank J. H. Gomes Leitao for kindly providing strain Burkholderia cenocepacia J2315. The Serveis Cientifico-Tècnics of the University of Barcelona are also acknowledged for technical assistance in sequencing. This work was financed by the Scientific and Technological Research Council (CICYT, MICINN, Spain), grants CTQ2007-60749/PPQ and CTQ2010-21183-C02-02/PPQ, by the IV Pla de Recerca de Catalunya (Generalitat de Catalunya), grant 2009SGR-819, and by the Generalitat de Catalunya to the Xarxa de Referència en Biotecnologia (XRB). A. Bassegoda acknowledges a fellowship from the Spanish Ministry of Science and Education (AP2006-02941).

REFERENCES 1. Altschul SF, et al. 1997. Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. Nucleic Acids Res. 25:3389 – 3402. 2. Arpigny JL, Jaeger K-E. 1999. Bacterial lipolytic enzymes: classification and properties. Biochem. J. 343:177–183. 3. Bofill C, et al. 2010. Differential behaviour of Pseudomonas sp. 42A2 LipC, a lipase showing greater versatility than its counterpart LipA. Biochimie 92:307–316. 4. Bornscheuer UT, Bessler C, Srinivas R, Krishna SH. 2002. Optimizing lipases and related enzymes for efficient application. Trends Biotechnol. 20:433– 437. 5. Britton HTS. 1952. Hydrogen ions. Chapman & Hall, London, United Kingdom. 6. Bryson K, et al. 2005. Protein structure prediction servers at University College London. Nucleic Acids Res. 33:W36 –W38. 7. Das S, Noe JC, Paik S, Kitten T. 2005. An improved arbitrary primed PCR method for rapid characterization of transposon insertion sites. J. Microbiol. Methods 63:89 –94. 8. Diaz P, Prim N, Pastor FIJ. 1999. Direct fluorescence-based lipase activity assay. Biotechniques 27:696 – 698, 700. 9. Emanuelsson O, Brunak S, von Heijne G, Nielsen H. 2007. Locating proteins in the cell using TargetP, SignalP, and related tools. Nat. Protoc. 2:953–971. 10. Ericsson DJ, et al. 2008. X-ray structure of Candida antarctica lipase A shows a novel lid structure and a likely mode of interfacial activation. J. Mol. Biol. 376:109 –119. 11. Falcocchio S, Ruiz C, Pastor FIJ, Saso L, Diaz P. 2005. Identification of a carboxylesterase-producing Rhodococcus soil isolate. Can. J. Microbiol. 51:753–758. 12. Falcocchio S, Ruiz C, Pastor FIJ, Saso L, Diaz P. 2006. Propionibacterium acnes GehA lipase, an enzyme involved in acne development, can be successfully inhibited by defined natural substances. J. Mol. Catalysis B Enzymatic 40:132–137.


13. Fischer M, Thai QK, Grieb M, Pleiss J. 2006. DWARF—a data warehouse system for analyzing protein families. BMC Bioinformatics 7:495. 14. Fojan P, Jonson PH, Petersen MTN, Petersen SB. 2000. What distinguishes an esterase from a lipase: a novel structural approach. Biochimie 82:1033–1041. 15. Frickey T, Lupas A. 2004. CLANS: a Java application for visualizing protein families based on pairwise similarity. Bioinformatics 20:3702– 3704. 16. Gupta R, Gupta N, Rathi P. 2004. Bacterial lipases: an overview of production, purification and biochemical properties. Appl. Microbiol. Biotechnol. 64:763–781. 17. Hall T. 1999. BioEdit: a user-friendly biological sequence alignment editor and analysis program for Windows 95/98/NT. Nucleic Acids Symp. Ser. 41:95–98. 18. Hasan F, Shah AA, Hameed A. 2006. Industrial applications of microbial lipases. Enzyme Microb. Technol. 39:235–251. 19. Henke E, Bornscheuer UT, Schmid RD, Pleiss J. 2003. A molecular mechanism of enantiorecognition of tertiary alcohols by carboxylesterases. Chem. Eur. J. Chem. Biol. 4:485– 493. 20. Jaeger K-E, Eggert T. 2002. Lipases for biotechnology. Curr. Opin. Biotechnol. 13:390 –397. 21. Jaeger KE, Dijkstra BW, Reetz MT. 1999. Bacterial biocatalysts: molecular biology, three-dimensional structures, and biotechnological applications of lipases. Annu. Rev. Microbiol. 53:315–351. 22. Katoh K, Toh H. 2008. Recent developments in the MAFFT multiple sequence alignment program. Brief. Bioinformatics 9:286 –298. 23. Klibanov AM. 1990. Asymmetric transformations catalyzed by enzymes in organic-solvents. Accounts Chem. Res. 23:114 –120. 24. Kouker G, Jaeger KE. 1987. Specific and sensitive plate assay for bacterial lipases. Appl. Environ. Microbiol. 53:211–213. 25. Larkin MJ, Kulakov LA, Allen CC. 2005. Biodegradation and Rhodococcus—masters of catabolic versatility. Curr. Opin. Biotechnol. 16:282–290. 26. Lee M-H, Lee C-H, Oh T-K, Song JK, Yoon J-H. 2006. Isolation and characterization of a novel lipase from a metagenomic library of tidal flat sediments: evidence for a new family of bacterial lipases. Appl. Environ. Microbiol. 72:7406 –7409. 27. Lee MH, et al. 2010. A new esterase EstD2 isolated from plant rhizosphere soil metagenome. Appl. Microbiol. Biotechnol. 88:1125–1134. 28. Park SY, et al. 2007. A new esterase showing similarity to putative dienelactone hydrolase from a strict marine bacterium, Vibrio sp. GMD509. Appl. Microbiol. Biotechnol. 77:107–115. 29. Prim N, Blanco A, Martinez J, Pastor FIJ, Diaz P. 2000. estA, a gene coding for a cell-bound esterase from Paenibacillus sp. BP-23, is a new member of the bacterial subclass of type B carboxylesterases. Res. Microbiol. 151:303–312. 30. Prim N, Bofill C, Pastor FIJ, Diaz P. 2006. Esterase EstA6 from Pseudomonas sp. CR-611 is a novel member in the utmost conserved cluster of family VI bacterial lipolytic enzymes. Biochimie 88:859 – 867. 31. Prim N, Pastor FIJ, Diaz P. 2001. Cloning and characterization of a bacterial cell-bound type B carboxylesterase from Bacillus sp. BP-7. Curr. Microbiol. 42:237–240. 32. Prim N, Sanchez M, Ruiz C, Pastor FIJ, Diaz P. 2003. Use of methylumbeliferyl-derivative substrates for lipase activity characterization. J. Mol. Catalysis B Enzymatic 22:339 –346. 33. Rathbone DA, Holt PJ, Lowe CR, Bruce NC. 1997. Molecular analysis of the Rhodococcus sp. strain H1 her gene and characterization of its product, a heroin esterase, expressed in Escherichia coli. Appl. Environ. Microbiol. 63:2062–2066. 34. Reetz MT. 2002. Directed evolution as a means to create enantioselective enzymes. Abstr. Papers Am. Chem. Soc. 224:ORGN 302. 35. Rosenau F, Jaeger KE. 2000. Bacterial lipases from Pseudomonas: regulation of gene expression and mechanisms of secretion. Biochimie 82: 1023–1032. 36. Ruiz C, Blanco A, Pastor FIJ, Diaz P. 2002. Analysis of Bacillus megaterium lipolytic system and cloning of LipA, a novel subfamily I.4 bacterial lipase. FEMS Microbiol. Lett. 217:263–267. 37. Ruiz C, Falcocchio S, Pastor FIJ, Saso L, Diaz P. 2007. Helicobacter pylori EstV: identification, cloning, and characterization of the first lipase isolated from an epsilon-proteobacterium. Appl. Environ. Microbiol. 73: 2423–2431. 38. Ruiz C, et al. 2004. Activation and inhibition of Candida rugosa and Bacillus-related lipases by saturated fatty acids, evaluated by a new colorimetric microassay. Biochim. Biophys. Acta 1672:184 –191.

aem.asm.org 1731

Bassegoda et al.

39. Ruiz C, Pastor FIJ, Diaz P. 2003. Isolation and characterization of Bacillus sp. BP-6 LipA, a ubiquitous lipase among mesophilic Bacillus species. Lett. Appl. Microbiol. 37:354 –359. 40. Ruiz C, Pastor FIJ, Diaz P. 2005. Isolation of lipid- and polysaccharidedegrading micro-organisms from subtropical forest soil, and analysis of lipolytic strain Bacillus sp. CR-179. Lett. Appl. Microbiol. 40:218 –227. 41. Sambrook J, Russell D. 2001. Molecular cloning: a laboratory manual, 3rd ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. 42. Sandström AG, Engström K, Nyhlén J, Kasrayan A, Bäckvall J-E. 2009. Directed evolution of Candida antarctica lipase A using an episomally replicating yeast plasmid. Protein Eng. Des. Sel. 22:413– 420.

1732

aem.asm.org

43. Sekine M, et al. 2006. Sequence analysis of three plasmids harboured in Rhodococcus erythropolis strain PR4. Environ. Microbiol. 8:334 –346. 44. Tamura K, Dudley J, Nei M, Kumar S. 2007. MEGA4: Molecular Evolutionary Genetics Analysis (MEGA) software version 4.0. Mol. Biol. Evol. 24:1596 –1599. 45. Vadell EM, Cavender JC. 2007. Dictyostelids living in the soils of the Atlantic Forest, Iguazu region, Misiones, Argentina: description of new species. Mycologia 99:112–124. 46. Widmann M, Juhl PB, Pleiss J. 2010. Structural classification by the Lipase Engineering Database: a case study of Candida antarctica lipase A. BMC Genomics 11:123.
