A novel DNA repair enzyme containing RNA recognition, G-patch and specific splicing factor 45-like motifs in the protozoan parasite Toxoplasma gondii Najoua Dendouga 1 , Isabelle Callebaut 2 and Stanislas Tomavo 1 1 Equipe de Parasitologie Mole ´ culaire, Laboratoire de Chimie Biologique, CNRS UMR 8576, Universite ´ des Sciences et Technologies de Lille, France; 2 Equipe Syste ` mes Mole ´ culaires et Biologie Structurale, Laboratoire de Mine ´ ralogie-Cristallographie, CNRS UMR 7590, Universite ´ Paris, France We report the cloning and functional charaterization of the full-length cDNA and gene encoding a Toxoplasma gondii DNA repair enzyme designated TgDRE. The gene is com- posed of three exons separated by two introns of 780 and 630 bp, and encodes a protein with a predicted molecular mass of 49.6 kDa. The native TgDRE protein, with a molecular mass of 60 kDa, is only detected in the virulent tachyzoite stage of T. gondii. However, the transcript is present in both asexual parasite stages, virulent tachyzoite and avirulent encysted bradyzoite. When an Escherichia coli mutant lacking ruvC endonuclease and recG helicase was transformed with TgDRE cDNA, a significant increase in resistance to DNA-damaging agents, such as UV light and mitomycin C, was observed. Moreover, database searches revealed that TgDRE orthologues were present in the genome sequences of the related apicomplexa parasites Plasmodium falciparum and Plasmodium yoelii,aswellasin those of Arabidopsis thaliana, Drosophila melanogaster, Caenorhabditis elegans and Homo sapiens. This novel family of proteins is characterized by the presence of human splicing factor SF45-like, RNA recognition (RRM) and glycine-rich (G-patch) motifs. The presence of these motifs suggests that T. gondii TgDRE might also be involved in other biological functions such as RNA metabolism in addition to DNA-repair. Keywords: Toxoplasma gondii; DNA repair enzyme; RRM; G-patch; splicing factor motif. Among protozoan parasites of the Apicomplexa phylum, are numerous pathogens such as Toxoplasma gondii (an important opportunistic pathogen associated with AIDS and congenital birth defects), Plasmodium falciparum (causative agent of malaria), Eimeria (agent of coccidiosis) and Cryptosporidium (opportunistic intestinal agent). In mammalian nonfeline hosts, T. gondii is found in two haploid asexual forms, the replicating tachyzoite and the slowly dividing quiescent encysted bradyzoites. T. gondii infection is asymptomatic in most adults, but the parasite persists during the lifetime of infected hosts as slowly dividing encysted bradyzoites. When the protective immu- nity fails, as in AIDS or in transplanted patients [1], the encysted bradyzoites can transform into actively replicating and virulent tachyzoites. The tachyzoites differentiate into encysted bradyzoites in response to the immune system attack during disease progression. They will remain in the brain and other organs during the lifetime of infected hosts. The reactivation of encysted bradyzoites into actively replicating and cytolytic tachyzoites can trigger a disease with severe clinical syndromes leading in many cases to the death of the patients. While the actively replicating tachy- zoites can be successfully treated by different chemothera- peutic agents, none of these drugs are capable of inhibiting the encysted bradyzoites which is the source of toxoplas- mosis reactivation. Therefore, a better understanding of the biology of this obligate intracellular parasite and the molecular mechanisms underlying T. gondii differentiation is useful in controlling the infection. DNA repair proteins have been well characterized in human, yeast, plant and bacteria and play important roles in preserving the genetic information that ensure the normal cellular function and development [2–4]. Although all organisms contain various DNA repair mechanisms, the repair of a specific DNA damage is often conserved from bacteria to human, and in many cases the proteins are highly similar [5]. Little is known about DNA repair mechanisms and the proteins involved in apicomplexan parasites such as T. gondii and Plasmodia. However, preliminary characterization of some DNA repair proteins have been described in Plasmodium falciparum [6,7]. In addition, two expressed sequence tags (EST) homologous to DNA repair proteins have also been identified in T. gondii [8]. Independently, we identified a T. gondii cDNA fragment Correspondence to S. Tomavo, Equipe de Parasitologie Mole ´ culaire, Laboratoire de Chimie Biologique, CNRS UMR 8576, B  aatiment C9, Universite ´ des Sciences et Technologies de Lille, 59655 Villeneuve d’Ascq, France. Fax: + 33 3 20 43 65 55, Tel.: + 33 3 20 43 69 41, E-mail: Stan.Tomavo@univ-lille1.fr Abbreviations:TgDRE,Toxoplasma gondii DNA repair enzyme; RRM, RNA recognition motif; HCA, hydrophobic cluster analysis; MMC, DNA-damaging agent mitomycin C; G-patch, glycine rich motif; SF45, splicing factor 45; DRT111, DNA repair/tolerance 111 enzyme; EST, expressed sequence tag; IPTG, isopropyl thio-b- D - galactoside. Note: the nucleotide sequences reported in this paper has been submittedtoGenBankwithaccessionnumbersAF402310 (genomic DNA) and AF402311 (cDNA). (Received 19 March 2002, revised 2 May 2002, accepted 13 May 2002) Eur. J. Biochem. 269, 3393–3401 (2002) Ó FEBS 2002 doi:10.1046/j.1432-1033.2002.02993.x encoding a polypeptide homologous to the DNA repair enzyme (DRT111) of Arabidopsis thaliana [9,10]. Here, we report on the isolation of the gene locus, full-length cDNA and the functional analysis of the corresponding protein of T. gondii that displays strong similarities with the P. falci- parum, P. yoelii and A. thaliana proteins. We demonstrate that T. gondii DNA repair protein is capable of comple- menting an Escherichia coli mutant lacking ruvC endonuc- lease and recG helicase. In addition, analyses of primary structure indicate that a T. gondii DNA repair protein, designated TgDRE (Tg DNA repair enzyme) for consis- tency with the Arabidopsis thaliana enzyme, belongs to a large family of proteins containing RNA recognition motifs (RRM), glycine-rich motifs (G-patch) and a specific motif named SF45 because of its similarity to the human splicing factor 45 protein, which was described as a component of the spliceosome [11]. The presence of the two first motifs suggests that TgDRE may also be involved in RNA metabolism. EXPERIMENTAL PROCEDURES Parasite and host cell cultures The T. gondii 76K strain was used throughout this study. Tachyzoites were grown in human foreskin fibroblasts (HFF) using Dulbecco’s modified Eagle’s medium (Bio- whittaker) supplemented with 10% fetal bovine serum (Dutscher), 2 m M glutamine (Sigma) and 0.05 mgÆmL )1 gentamicin (Sigma). Southern blotting, Library screening and sequencing For Southern blots, genomic DNA was extracted from T. gondii tachyzoites according to standard procedures [12]. The specific probes for Southern blots were generated using the PCR DIG labelling mix (Roche Diagnostics). Detection was performed using the chemiluminescence method (Roche Diagnostics). A T. gondii genomic library (kindly provided by D. Roos, University of Pennsylvania, Philadelphia, USA) constructed in lambda DASHII vector (Stratagene) using DNA fragment obtained after BamHI digestion was screened by plaque hybridization using a 380-bp digoxigenin-labeled probe [9]. Two fragments, 7000- and 1700-bp, were subcloned into pBluescript-SK+ and sequenced using an ALFexpress automated sequencer (Amersham Pharmacia Biotech). Rapid amplification of cDNA ends (RACE) and sequencing To obtain full-length cDNAs, a RACE-PCR technique was performed using a Marathon cDNA Amplification Kit (Clontech) with the adaptor primer 2 and specific oligonu- cleotides N34AS 5¢-CTTCACCTGGAGGAGATTTCC AAA-3¢ for 5¢ RACE and N34S 5¢-GGGAGGGTC TCGGCGTCAACAAAC-3¢ for 3¢ RACE. Nested PCR was performed using internal oligonucleotides S10 5¢-GT CGAGATGTTGGTTGTCGGAGACC-3¢ for 5¢ RACE and PR2AS 5¢-GACCGTTACCACTGATTGCGGCTG 3¢ for 3¢ RACE. The PCR products were cloned into the TA cloning vector (Invitrogen) which were used to transform E. coli strain DH5a cells. The purified plasmids were sequenced as described above. Measurement of mRNA levels Total RNA was isolated from encysted bradyzoites, tach- yzoites, uninfected mice brain cells and HFF cells, as previously described [13,14]. cDNA was synthesized using reverse transcriptase and serial dilutions of the cDNA were used to amplify ORFs of T. gondii a-tubulin [15] or that of TgDRE. Primers were as follows: a-tubulin 5¢-ATGAGAG AGGTTATCAGCATC-3¢ and 5¢-TTAGTACTCGTCAC CATAGCC-3¢; for TgDRE, 34S13 sens: 5¢-ATGCTGGA CTCTCTCTACGGGGAT-3¢ and 34AS15 antisens: 5¢-TT AGTCGAGGGGTTTGTCTGC-3¢. PCR products were electrophoresed on agarose gels, stained with ethidium bromide, scanned and quantified by densitometry using the program NIH IMAGE (http://rsb.info.nih.gov/nih-image/). Expression of glutathion S -transferase (GST)-TgDRE fusion protein and immunological analysis The cDNA fragment of TgDRE was cloned into the expression vector pGEX-6P-3 (Pharmacia). After induc- tion of transformed E. coli (BL21 strain) using isopro- pylthio-b- D -galactoside, recombinant proteins were purified on glutathione–Sepharose 4B column using preScission TM protease, according to manufacturer’s instructions (Pharmacia). Polyclonal antiserum was gen- erated in Balb/c mice using the purified recombinant protein. IgG specific to TgDRE was purified from this antiserum by incubating the polyclonal antibodies to recombinant TgDRE blotted onto nitrocellulose. After, one hour of incubation at room temperature and washing with NaCl/P i , the bound-IgG was eluted with 3.5 M potassium thiocyanate (KSCN), dialyzed and stored at )20 °C with 20% BSA. Western blots were performed using the purified anti-TgDRE IgG diluted at 1 : 100 or with a monoclonal antibody specific to T. gondii actin diluted at 1 : 100 and rabbit secondary antibodies con- jugated to peroxidase diluted at 1 : 10 000 followed by cheminilescence detection. Functional complementation in E. coli mutants lacking ruvC endonuclease recG helicase E. coli RuvC/RecG-deficient N3398 strain [16,17] and AB1157 wild-type strain (generously provided by J.B Hays, Oregon State University, USA) were transformed with the pGEX-6P-3 containing the TgDRE ORF or pGEX-6P-3 alone. After isopropyl thio-b- D -galactoside (IPTG)-induc- tion, the bacteria were plated onto Luria–Bertani plates containing 0.1 lgÆmL )1 of DNA-damaging agent mitomy- cin C (MMC) and incubated at 30 °Cfor3days.The transformed E. coli RuvC/RecG-deficient N3398 strain and AB1157 wild-type strain were subjected to UV light after IPTG-induction. All experiments were performed in darkness and the plates were then incubated overnight at 30 °C. Database searches and sequence analysis Searches within nonredundant databases were performed using PSI - BLAST [18] at NCBI. The Smart and Pfam databases were searched using RPS - BLAST . Hydrophobic clusteranalysis(HCA)wasalsousedinordertodetect 3394 N. Dendouga et al. (Eur. J. Biochem. 269) Ó FEBS 2002 distant but significant sequence similarities between the members of the SF45 family [19]. RESULTS Cloning of the gene encoding T. gondii DNA repair protein A 380-bp cDNA fragment encoding a putative DNA repair protein (designated TgDRE) isolated from a T. gondii subtractive library [9] was used to probe Southern blots performed using genomic DNA digested with several restriction enzymes. Figure 1A shows the blot washed under stringent conditions. Most enzymes tested gave a single band except those with a predicted internal restriction site. The hybridization data is consistent with a single copy TgDRE gene in the nuclear genome. In order to determine TgDRE genomic organization, a T. gondii genomic library was screened using the 380-bp cDNA probe. Two BamHI-digested fragments of 7000- and 1700- bp were isolated, subcloned and sequenced. The nucleo- tide sequence analysis of the genomic DNA revealed the presence of two introns and three exons, which was confirmed by the isolation and sequencing of the full- length TgDRE cDNA. The comparison of cDNA and genomic DNA nucleotide sequences allowed the deter- mination of the exact position of the two introns which were 780- and 630-bp long (Fig. 1B). These introns contained the typical GT/AG consensus splicing signal. The excision of the two introns from the primary transcript results in a mature TgDRE mRNA of 1401 nucleotides, which probably represents most, if not all, of the entire full-length cDNA. This was confirmed by the reconstruction of full-length cDNA achieved by 5¢ and 3¢ RACE, which gave a nucleotide sequence of 2002 bp with an ORF of 1401 bp (nucleotides 325–1725). This 2002-bp cDNA contains a ORF of 1401 bp that encodes a protein of 466 amino acids with a calculated molecular mass of 49.6 kDa. In addition, RT-PCR performed with primers encompassing the start and stop codons demon- strates a predominant mRNA species of  1.4 kb (Fig. 2A,B). Fig. 1. Genomic organization of T. gondii TgDRE locus. (A) Southern blot analysis using genomic DNA from tachyzoites of T. gondii 76K strain digested with several restriction enzymes indicated at the top. The blot was hybridized with a 380-bp DNA probe indicated by the letter P in (B). (B) Genomic organization of TgDRE locus and its comparison with cDNA map revealed two introns designated i 1 and i 2 . Boxes correspond to exons deduced from cDNAs sequences. The positions of the start and stop codons are indicated by an arrow and asterisk, respectively. Fig. 2. Comparative analysis of TgDRE mRNA level in the two devel- opmental stages of T. gondii. (A) Results of RT-PCR with the primers corresponding to the full-length ORF of T. gondii TgDRE by using total RNA from tachyzoites, T, and in vivo encysted bradyzoites (B), RNA from uninfected human fibroblasts, H, or from naive mice brain cells (C). H and C represent the negative controls. The quantity of cDNA used corresponds to the equivalent of 10 4 parasites for both tachyzoite and bradyzoite stages. RT-PCR were also performed by using the tubulin primers used as positive controls. (B) Semi-quanti- tative RT-PCR was used for the amplification of the ORFs of TgDRE and the housekeeping gene a-tubulin using serial dilutions of cDNAs prepared from tachyzoites and bradyzoites isolated from brain of infected mice. Ó FEBS 2002 A novel T. gondii DNA repair enzyme (Eur. J. Biochem. 269) 3395 Developmental expression of TgDRE gene measured by RT-PCR In order to determine the transcript level of TgDRE in both virulent tachyzoite and avirulent encysted bradyzoite, RT-PCR was performed. To ensure that equal quantities of each mRNA were being compared, the amplification of a-tubulin ORF was used as control (Fig. 2A,B). The RT-PCR revealed a band of 1.4 kb corresponding to TgDRE in both tachyzoite and bradyzoite stages of T. gondii. The cloning and sequencing of this RT-PCR product confirmed its identity. In addition, no RT-PCR products could be amplified from the negative controls corresponding to cDNA prepared from human and uninfected mice brain cells (Fig. 2A, lanes H and C). It appears that TgDRE transcript is slightly more predom- inant in encysted bradyzoites than in tachyzoites. We used semiquantitative RT-PCR to show that TgDRE transcript in encysted bradyzoites represents twofold to threefold higher level than that in tachyzoites as evaluated by densitometry (Fig. 2B), suggesting that TgDRE gene is either overexpressed at the transcriptional level or its mRNA is more stable in encysted bradyzoites than in tachyzoites of T. gondii. Characterization of TgDRE protein expressed in the two developmental stages of T. gondii To investigate whether these mRNA abundances increase the expression of TgDRE protein, a GST fusion protein consisting of amino-acid residues 1–466 of the ORF fused to the C-terminus of the 30-kDa GST protein was constructed. SDS/PAGE analysis of the IPTG-induced E. coli revealed a predominant 90-kDa full-length protein in both total SDS-lysate and sonicated soluble extract (Fig. 3A, lanes 3 and 4). The size of this recombinant fused protein appears higher in both total extract lysate than expected (a difference of 10 kDa) from the apparent molecular mass of 49.6 kDa predicted by TgDRE primary structure plus the 30-kDa GST protein. The recombinant TgDRE protein without the fused GST was also obtained by affinity chromatography on a GST column followed by the preScission protease digestion (Fig. 3A, lane 5). Again, the pure recombinant TgDRE migrated as a 55-kDa protein instead of the apparent molecular mass of  50 kDa. One additional lower molecular mass protein species is present in the purified preparation. We believe this to be due to proteolytic degradation of the 90-kDa protein as a result of contaminating cellular proteases. The purified recombin- ant protein was used to raise polyclonal antisera in mice. The Western blot (Fig. 3A, lane 6) shows the strong immunoreactivity of the affinity-purified IgG from the polyclonal antiserum against the electroeluted pure recombinant TgDRE of 55 kDa, demonstrating that a specific anti-TgDRE serum was generated. However, Western blots of this purified antiserum specific to TgDRE revealed a faint specific native protein of 60 kDa in the tachyzoites (data not shown). This 60-kDa protein can be readily detected when 20- to 200-fold amount of proteins from tachyzoites were loaded in the SDS/PAGE (Fig. 3B, lanes 1–3). It is interesting to note that a native protein of 60-kDa was detected in the parasite instead of 55-kDa as expected for the recombin- ant protein (Fig. 3A, lanes 5 and 6). Because it is difficult to obtain an equivalent amount of protein in encysted bradyzoites, we were unable to identify the TgDRE protein in this developmental parasite stage. We conclu- ded that the higher level of transcript detected in RT-PCR in encysted bradyzoites does not correspond to an increased level of TgDRE protein, and that the protein level is very low in the virulent tachyzoite. This conclusion was supported by the control experiments where we measured the quality/quantity of total protein extracts by immunoblots using monoclonal antibodies directed against actin, the tachyzoite-specific protein SAG1 and the bradyzoite-specific antigen (data not shown). Collec- tively, our data demonstrate that a 60-kDa protein can be recognized by the anti-TgDRE IgG in the virulent tachyzoite of T. gondii. Fig. 3. Developmental expression of TgDRE protein in T. gondii. (A) SDS/PAGE analysis of the GST fusion recombinant TgDRE protein. Lane 1, molecular mass markers; lane 2, total SDS lysate from uninduced E. coli containing the pGEX-6P-3 vector alone; lane 3, total SDS lysate of the same E. coli containing pGEX-6P-3 vector with TgDRE ORF after induction by IPTG for 3 h; lane 4, soluble fraction obtained after centrifu- gation of induced E. coli disrupted by sonication; lane 5, affinity purified recombinant TgDRE after cleavage with the preScission protease and elution from the glutathione column; lane 6, immunoblot of anti-TgDRE serum against the purified specific recombinant TgDRE. (B) Western blot analysis of protein extracts corresponding to 5 · 10 7 (1), 1 · 10 8 (2) and 6 · 10 8 (3) equivalent tachyzoites of T. gondii 76K strain. 3396 N. Dendouga et al. (Eur. J. Biochem. 269) Ó FEBS 2002 TgDRE cDNA is capable of functionally complementating E. coli mutant lacking ruvC endonuclease and recG helicase Similarity searches revealed that TgDRE displays signifi- cant homology to DNA repair/toleration protein (DRT111) of A. thaliana [10]. As the DRT111 protein was previously shown to partially complement E. coli mutants lacking RuvC and RecG helicase activities, we decided to investi- gate whether TgDRE was capable of correcting these activities in E. coli mutant phenotype. The transformation of this E. coli mutant lacking RuvC and RecG enzyme activities with the pGEX-6P-3 alone or pGEX containing TgDRE ORF was performed in the presence of mitomycin C. The in vivo effect of TgDRE activity on E. coli RuvC – RecG – revealed a significant increased resistance of trans- formed mutants to the DNA-crosslinking agent mitomy- cin C (Fig. 4). The resistance was increased by as much as fourfold relative to mutants transformed with the pGEX alone and treated with mitomycin C under the same experimental conditions (Fig. 4). However, it appears that TgDRE did not fully correct the mutant phenotype because the surviving bacteria in the wild type E. coli is higher in the presence of this chemical DNA-damaging agent (Fig. 4). The partial correction of the E. coli mutant by TgDRE cDNA is estimated at 40% less than that of the wild-type bacteria in the presence of mitomycin C. In addition, we also investigated whether TgDRE is capable of correcting the DNA-damaging sensitive phenotype in the same E. coli RuvC – RecG – mutant after UV light treatment because it has been previously shown that this bacteria mutant is also partially sensitive to this treatment. The transfection with TgDRE cDNA also increased the resistance to UV light treatment in this mutant as compared to that transformed with the pGEX alone (data not shown). Taken together, these data show convincingly that TgDRE cDNA encode a functional DNA repair protein which can partially correct the E. coli RuvC – RecG – -sensitive phenotype. TgDRE belongs to a large family of proteins sharing G-patch, RNA recognition motifs and a specific SF45 motif. An initial similarity search revealed a clear similarity ( 30% sequence identity) with DNA repair/toleration protein (DRT111) of A. thaliana, in addition to that observed with two proteins from the related apicomplexa parasites, Plasmodium falciparum and Plasmodium yoelii (Fig. 5). No similarity with other known DNA repair enzymes could be found in the databases, suggesting that these proteins may represent a novel family of DNA repair enzymes. Moreover, searches against domain databases (Pfam, Smart) revealed the presence of two well-defined motifs at the C-terminus of the TgDRE sequence: a G-patch motif, ranging from amino acid 296–340 (Fig. 6A) followed by a RRM module in the amino acid range 362–441 (Fig. 6B). G-Patch is a predicted glycine-rich nucleic binding domain found in the splicing factor 45 and other DNA-binding proteins [11,20], whereas RRM (RNA recognition motif) was already described in numerous proteins [21,22]. Further analyses of the N-terminal moiety of the TgDRE sequence using PSI - BLAST combined with HCA, revealed a conserved and as yet undescribed motif upstream these two domains (Fig. 6C). This motif is present in a limited set of proteins in addition to the DNA repair/toleration DRT111 protein from A. thaliana [8] and the two above mentioned proteins of Plasmodium, the human splicing factor 45 [11] and several hypothetical proteins from sequenced genomes of D. melanogaster and C. elegans. All these proteins share the same architecture with TgDRE as they also possess C-terminal G-patch and RRM domains (Fig. 7). However, the conserved motif appears specific for this family of orthologous proteins here named SF45 motif (the SF45- family-specific motif), is located at different distances from the N-terminus and from the G-patch/RRM couple (Fig. 7). The N-terminal parts of the SF45 family, located before the G-patch and RRM domains, are rich in low complexity sequences, typical of nonglobular regions and could not be well aligned (see also Fig. 5). Several other small conserved motifs were also detected (A, B and C in Fig. 7), but none were found in all of the sequences of the SF45 family. Based on the similarity to the human splicing factor 45 and their cellular localization, we speculate that the SF45 family, whose members are schematically repre- sented in Fig. 7, may also have functions in the nuclear spliceosome and thus in RNA metabolism. DISCUSSION In the present study, we report the isolation and expression of a cDNA encoding a novel DNA repair enzyme in T. gondii designated TgDRE. TgDRE-like sequences are also present in the genomes of two related apicomplexan, P. falciparum and P. yoelii, and belongs to the SF45 family, a large family of proteins found in several genomes, from human (SF45 protein) to plants (DRT111). The highest similarities of apicomplexan TgDRE sequences are ob- served with those of A. thaliana, suggesting that they could be of algal origin. This assumption is supported by the recent discovery of a nonphotosynthetic remnant plastid ÔapicoplastÕ in apicomplexan parasites [23–25]. Moreover, cDNAs encoding glucose-6-phosphate isomerase (Glc6PI), enolases and other enzymes isolated from T. gondii and P. falciparum were shown to have significant identity to enzymes from non or photosynthetic organisms [13,25]. We also showed that TgDRE mRNA is encoded by a single copy gene and several attempts to disrupt the gene by knockout strategy were unsuccessful (D. Dendouga & S. Tomavo unpublished data), suggesting that TgDRE Fig. 4. Resistance of bacteria expressing TgDRE to the DNA dam- aging agent mitomycin C. The wild-type E. coli strain AB 1157 RuvC + RecG + was transformed with the pGEX expression vector alone (WT). The E. coli mutant strain N 3398 RuvC – RecG – was transformed with the pGEX expression vector alone (mutant) or with the pGEX vector containing the TgDRE ORF (complemented). All bacteria were transformed with 10 ng of plasmids and then spread onto agar-plates containing 0.1 lgÆmL )1 of mitomycin C. Ó FEBS 2002 A novel T. gondii DNA repair enzyme (Eur. J. Biochem. 269) 3397 might be essential for parasite growth as expected for a protein involved in DNA repair and conservation of genome integrity. It is interesting that a similar gene knock-out approach to disrupt a T. brucei splicing enzyme containing RRM motif also failed [26]. TgDRE and orthologous sequences do not show obvious similarities with other proteins specialized in DNA repair. Interestingly, they also contain motifs typical of proteins involved in RNA metabolism. TgDRE mRNA and protein expression have been evaluated using RT-PCR and Western blotting, respectively. Although the TgDRE transcript was over- expressed in encysted bradyzoites, the protein could only be detected in the virulent tachyzoite. In this case, the level of the protein seems to be unusually low compare to that reported for other Toxoplasma proteins studied [13,14]. In addition, the native TgDRE protein detected in the parasite migrates as 60 kDa protein instead of 50 kDa. At this time, we have no explanation for the discrepancy between the apparent molecular mass and the native size of TgDRE in the parasite. According to the functional similarities that can be predicted from the sequence similarities observed between TgDRE and A. thaliana DRT111 enzyme, we have been able to partially complement RuvC – RecG – E. coli mutants. Similar results have been described for the A. thaliana DRT111 enzyme [8]. The fact that TgDRE corrects only partially the E. coli mutant phenotype can be interpreted as the result of phylogenetical distance between bacteria and the protozoan parasite T. gondii. Alternatively, other bio- logical functions can be ascribed to TgDRE because sequence analysis revealed the presence of some striking motifs such as G-patch and RRM motifs, which are presumed to be involved in RNA metabolism or splicing occurring in the spliceosome. The G-patch domain has been described as a short conserved region of about 40 amino acids, and is present in a number of putative RNA-binding proteins including a tumor suppressor (human LUCA15) [27], many RNA-processing proteins (45-kDa splicing factor) [11], the type-D retroviral polyproteins [20], and several hypothetical proteins from S. cerevisae, A. thaliana and C. elegans [20]. The RRMs are also found in a variety of RNA binding proteins, including heterogeneous nuclear ribonucleoproteins (hnRNPs), proteins implicated in regu- lation of alternative splicing, and components of small nuclear ribonucleoproteins (snRNPs) [28]. In the protozoan parasite Entamoeba histolytica, database searches for enhancer-binding proteins did not detect any motifs commonly associated with DNA-binding proteins, but Fig. 5. Alignment of the TgDRE sequence of T. gondii and Plasmodium with that of the A. thaliana DRT11. Sequences can be aligned in four distinct regions, corresponding to a small conserved domain present at the N-terminus (overlined, A), a SF45 motif, specific of members of the splicing factor 45 family (overlined and boxed), a G-patch motif (shaded box) and a RRM domain (boxed). The automatic alignment first obtained was refined in the N-terminal moiety as the SF45 motif, which is located at variable distances from the conserved motifs A and G-patch, was not aligned correctly. The SF45 motif was identified after refined analysis using PSI-BLAST and HCA (see Fig. 6). The symbol (*) indicates identical residues; (:) indicates well conserved residues. T. gondii DNA repair (TgDRE, this study), P. falciparum (90938047 chr14–1180), P. yoelii (chrPy1 °c275), A. thaliana (DRT111, M98455). 3398 N. Dendouga et al. (Eur. J. Biochem. 269) Ó FEBS 2002 surprisingly revealed sequences containing regions of simi- larities with RRM [29]. The specific contribution that the different motifs of TgDRE perform in the maintenance of the genome and/or in the RNA metabolism in T. gondii remains to be determined. Further work is needed to identify how T. gondii DNA repair proteins and their constitutive motifs (SF45, G-patch and RRM) function at the molecular level. The identification of the SF45 family opens many new avenues of future investigations into the unexplored molecular mechanisms involved in DNA repair and RNA metabolism in the protozoan parasites such as T. gondii and the Plasmodia. Further analyses should contribute to our understanding of the biochemistry of DNA repair proteins in medically relevant parasites, and thus may be helpful in the identification of other molecules which are critically important for parasite survival. Fig. 6. Alignment of the different domains of the SF45 family. (A) The G-patch motifs of the SF45 family (sequences above the line), aligned with those of the human RNA-binding protein 5 [GenBank identifier (gi) 13124794] and of the human DNA-binding SON protein (gi 15320061). (B) The RRM motifs of the SF45 family, aligned with those of the human Ro ribonucleoprotein-binding protein 1 (gi 6684440). (C) The SF45 motifs, specific of members of the SF45 family. The members of this family correspond to the here-described TgDRE, two Plasmodium sequences (chr14– 1180 (P. falciparum)andchrPy1°c275 (P. yoelii), A. thaliana DRT111 and F26G16.10 proteins (gi 1169200 and 15220757), D. melanogaster CG17540 protein (gi 7289585), C. elegans F58B3.7 protein (gi 7504595) and H. sapiens SF45 protein (gi 5454082). Sequence identities are indicated white on a black background, whereas sequence similarities are shaded (white and black letters for hydrophobic and nonhydrophobic positions, respectively). Ó FEBS 2002 A novel T. gondii DNA repair enzyme (Eur. J. Biochem. 269) 3399 ACKNOWLEDGEMENTS We would like to thank Drs Michael Kibe and Kim Binderup for critical reading of the manuscript. This work was supported by grants to S. Tomavo by the Centre National de la Recherche Scientifique (CNRS) through the ÔAction The ´ matique Incitative sur Programme et Equipe (ATIPE)Õ. N. D. was supported by fellowships from the Ministe ` re de l’Enseignement Supe ´ rieur et de la Recherche (MESR). REFERENCES 1. Luft, B.J. & remington, J.S. (1988) Toxoplasmic encephalitis in AIDS. J. Infect. Dis. 157, 1–6. 2. Eisen, J.A. & Hanawalt, P.C. (1999) A phylogenomic study of DNA repair genes, proteins, and processes. Mut. Res. 435,171– 213. 3. Marti, T.M., Kunz, C. & Fleck, O. (2002) DNA mismatch repair and mutation avoidance pathways. J. Cell Physiol. 191, 28–41. 4. Li, A., Schuermann, D., Gallego, F., Kovalchuk, I. & Tinland, B. (2002) Repair of damaged DNA by Arabidopsis cell extract. Plant Cell. 14, 263–273. 5. Cromie, G.A., Connelly, J.C. & Leach, D.R. (2001) Recombina- tion at double-strand breaks and DNA ends: conserved mechan- isms form phage to humans. Mol. Cell. 8, 1163–1174. 6. Haltiwanger, B.M., Matsumoto, Y., Nicolas, E., Dianov, G.L., Bohr, V.A. & Taraschi, T.F. (2000) DNA base excision repair in human malaria parasites is predominantly by a long-patch path- way. Biochemistry 39, 763–772. 7. Haltiwanger, B.M., Karpinich, N.O. & Taraschi, T.F. (2000) Characterization of class II apurinic/apyrimidinic endonuclease activities in the human malaria parasite, Plasmodium falciparum. Biochem. J. 345, 85–89. 8. Manger, I.D., Hehl, A.B., Parmley, S., Sibley, L.D., Marra, M., Hillier, L., Waterson, R. & Boothroyd, J.C. (1998) Expressed sequence tag analysis of the bradyzoite stage of Toxoplasma gondii: identification of developmentally regulated genes. Infect. Immun. 66, 1632–1637. 9. Yahiaoui, B., Dzierszinski, F., Bernigaud, A., Slomianny, C., Camus, D. & Tomavo, S. (1999) Isolation and characterization of a subtractive library enriched for developmentally regulated transcripts expressed during encystation of Toxoplasma gondii. Mol. Biochem. Parasitol. 99, 223–235. 10. Pang, Q., Hays, J.B. & Rajagopal, I. (1993) Two cDNAs from the plant Arabidopsis thaliana that partially restore recombination proficiency and DNA-damage resistance to E. coli mutants lacking recombination-intermediate-resolution activities. Nucleic Acids Res. 21, 1647–1653. 11. Neubauer, G., King, A., Rappsilber, J., Calvio, C., Watson, M., Ajuh,P.,Sleeman,J.,Lamond,A.&Mann,M.(1998) Mass spectrometry and EST-database searching allows characterization of the multi-protein spliceosome complex. Nat. Genet. 20, 46–50. 12. Sambrook, J., Fritsch, E.F. & Maniatis, T. (1989) Molecular cloning: a Laboratory Manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York. 13. Dzierszinski, F., Popescu, O., Toursel, C., Slomianny, C., Yahiaoui,B.&Tomavo,S.(1999)Theprotozoanparasite Toxoplasma gondii expresses two functional plant-like glycolytic enzymes. Implications for evolutionary origin of apicomplexans. J. Biol. Chem. 274, 24888–24895. 14.Toursel,C.,Dzierszinski,F.,Bernigaud,A.,Mortuaire,M.& Tomavo, S. (2000) Molecular cloning, organellar targeting and developmental expression of mitochondrial chaperone HSP60 in Toxoplasma gondii. Mol. Biochem. Parasitol. 111, 319–332. 15. Nagel, S.D. & Boothroyd, J.C. (1988) The alpha- and beta- tubulins of Toxoplasma gondii are encoded by single copy genes containing multiple introns. Mol. Biochem. Parasitol. 29, 261–273. 16. Lloyd, R.G. (1991) Conjugational recombination in resolvase- deficient ruvC mutants of Escherichia coli K-12 depends on recG. J. Bacteriol. 173, 5414–5418. 17.Hays,J.B.,Martin,S.J.&Bhatia,K.(1985)Repairof nonreplicating UV-irradiated DNA: cooperative dark repair by Escherichia coli uvr and phr functions. J. Bacteriol. 161, 602–608. 18. Altschul, S.F., Madden, T.L., Schaffer, A.A., Zhang, J., Zhang, Z., Miller, W. & Lipman, D.J. (1997) Gapped BLAST and PSI- BLAST, a new generation of protein database search programs. Nucleic Acids Res. 25, 3389–3402. Fig. 7. Modular architecture of proteins of the SF45 family. Protein sequences are drawn to scale. Motifs A, B and C (stars; the corres- ponding alignments are shown in the lower part of the figure) correspond to conserved domains, but none were found in all of the sequences of the SF45 family. The SF45, G-patch and RRM motifs are shown boxed. The total lengths of the proteins are indicated at the end of each sequence. 3400 N. Dendouga et al. (Eur. J. Biochem. 269) Ó FEBS 2002 19. Callebaut, I., Labesse, G., Durand, P., Poupon, A., Canard, L., Chomilier, J., Henrissat, B. & Mornon, J.P. (1997) Deciphering protein sequence information through hydrophobic cluster ana- lysis (HCA), current status and perspectives. Cell Mol. Life Sci. 53, 621–645. 20. Aravind,L.&Koonin,E.V.(1999)G-patch,anewconserved domain in eukaryotic RNA-processing proteins and type D retroviral polyproteins. Trends Biochem. Sci. 24, 342–344. 21. Burd, C. G. & Dreyfuss, G. (1994) Conserved structures and diversity of functions of RNA-binding proteins. Science 265, 615–621. 22. Kenan, D.J., Query, C.C. & Keene, J.D. (1991) RNA recognition, towards identifying determinants of specificity. Trends Biochem. Sci. 16, 214–220. 23. Wilson, R.J., Denny, P.W., Preiser, P.R., Rangachari, K., Roberts, K., Roy, A., Whyte, A., Strath, M., Moore, D.J., Moore, P.W. & Williamson, D.H. (1996) Complete gene map of the plastid-like DNA of the malaria parasite Plasmodium falciparum. J. Mol. Biol. 261, 155–172. 24. McFadden, G.I., Reith, M.E., Munholland, J. & Lang-Unnasch, N. (1996) Plastid in human parasites. Nature 381, 482. 25. Ko ¨ hler, S., Delwiche, C.F., Denny, P.W., Tilney, L.G., Webster, P., Wilson, R.J., Palmer, J.D. & Roos, D.S. (1997) A plastid of probable green algal origin in Apicomplexan parasites. Science 275, 1485–1489. 26. Manger, I.D. & Boothroyd, J.C. (2001) Targeted disruption of an essential RNA-binding protein perturbs cell division in Trypano- soma brucei. Mol. Biochem. Parasitol. 116, 239–245. 27. Gure, A.O., Altorki, N.K., Stockert, E., Scanlan, M.J., Old, L. J. & Chen, Y.T. (1998) Human lung cancer antigens recognized by autologous antibodies, definition of a novel cDNA derived from the tumor suppressor gene locus on chromosome 3p21.3. Cancer Res. 58, 1034–1041. 28. Query, C.C., Bentley, R.C. & Keene, J.D. (1989) A common RNA recognition motif identified within a defined U1 RNA binding domain of the 70K U1 snRNP protein. Cell 57, 89–101. 29. Schaenman, J.M., Gilchrist, C.A., Mann, B.J. & Petri Jr, W.A (2001) Identification of two Entamoeba histolytica sequence- specific URE4 enhancer-binding proteins with homology to the RNA-binding motif RRM. J. Biol. Chem. 276, 1602–1609. Ó FEBS 2002 A novel T. gondii DNA repair enzyme (Eur. J. Biochem. 269) 3401 . A novel DNA repair enzyme containing RNA recognition, G-patch and specific splicing factor 45-like motifs in the protozoan parasite Toxoplasma gondii Najoua. found in the databases, suggesting that these proteins may represent a novel family of DNA repair enzymes. Moreover, searches against domain databases (Pfam,