This article appeared in a journal published by Elsevier. The attachedcopy is furnished to the author for internal non-commercial researchand education use, including for instruction at the authors institution

and sharing with colleagues.

Other uses, including reproduction and distribution, or selling orlicensing copies, or posting to personal, institutional or third party

websites are prohibited.

In most cases authors are permitted to post their version of thearticle (e.g. in Word or Tex form) to their personal website orinstitutional repository. Authors requiring further information

regarding Elsevier’s archiving and manuscript policies areencouraged to visit:

http://www.elsevier.com/authorsrights

Author's personal copy

Plasmodium falciparum DOZI, an RNA helicase interacts with eIF4E

Mohammed Tarique, Moaz Ahmad, Abulaish Ansari, Renu Tuteja ⁎Malaria Group, International Centre for Genetic Engineering and Biotechnology, P. O. Box 10504, Aruna Asaf Ali Marg, New Delhi 110067, India

a b s t r a c ta r t i c l e i n f o

Article history:Accepted 15 March 2013Available online 2 April 2013

Keywords:ATPaseDEAD boxHelicaseMalaria parasiteTranslationUnwinding

DEAD box RNA helicases play crucial roles in RNA metabolism such as splicing, ribosome biogenesis, RNAtransport, degradation and translation. DDX6/DOZI (development of zygote inhibited) is one of the wellcharacterized member of the DEAD box family and is highly conserved from humans to malaria parasite.DDX6 is involved in a variety of biological processes, which include the sexual development of the protozoanparasite. In the present manuscript we report that P. falciparum DOZI (DDX6 homologue); PfDZ50 containsthe characteristic DNA and RNA binding, nucleic acid-dependent ATPase and RNA unwinding activities. Enzy-matic characterization of truncated derivatives of PfDZ50 such as PfDZ50T1 (domain 1) and PfDZ50T2(domain 2) shows that none of them contains ATPase activity. Furthermore, we confirmed that PfDZ50 inter-acts with PfeIF4E mainly through domain 1. Using in vitro translation assays we show that PfDZ50 inhibitstranslation. With the same assays we further report that externally added PfeIF4E restores ~70% of transla-tion. Using immunofluorescence assays we demonstrate that PfDZ50 is localized mainly in the cytoplasmin the asexual intraerythrocytic developmental stages of P. falciparum. The localization pattern furthersuggests that PfDZ50 appears typically in granular bodies throughout the cytoplasm. Thus these studieswill advance our knowledge regarding the function of PfDZ50/DDX6 in general.

© 2013 Elsevier B.V. All rights reserved.

1. Introduction

The ability to unwind double stranded RNAs or DNA duplexes hasbeen attributed to specific enzymes commonly known as helicases orunwindases. These enzymes work in an energy-dependent mannerthrough the hydrolysis of nucleoside triphosphates (NTPs). Helicasesare ubiquitous and present in almost all organisms ranging frombacteria and viruses, malaria parasite to humans (Tuteja, 2010;Umate et al., 2011). These enzymes are members of the well-characterized DEAD-box family, which was named according to oneof its conserved motifs (Linder et al., 1988; Wassarman and Steitz,1991). The sequence analysis of RNA helicases revealed that theyalso contain almost similar conserved motifs which are present inmost of the DNA helicases (Gorbalenya and Koonin, 1993). It is wellestablished that the DEAD-box helicases are involved in almost allthe metabolic processes involving RNA such as splicing, ribosomebiogenesis, RNA transport, degradation and translation. Helicases

mostly use NTP (usually ATP) binding and hydrolysis to remodelRNA, or RNA–protein complexes and this action results in double-stranded RNA unwinding, and/or displacement of proteins fromRNA (Cordin et al., 2006; Rocak and Linder, 2004). The catalytic coreof helicases is generally composed of two RecA-like domains withnine conserved domains, including the DEAD-box motif and the Qmotif, which has roles in catalysis and substrate binding. TheDEAD-box p54 RNA helicase is a member of a superfamily 2/DDX6helicase subfamily and is highly conserved from trypanosomes,malaria parasite to humans with well-characterized homologues inXenopus laevis (Xp54), Drosophila melanogaster (Me31B), Caenorhabditiselegans (CGH-1), and Saccharomyces cerevisiae (Dhh1) (Mair et al.,2006; Rajyaguru and Parker, 2009; Tuteja, 2010). The DEAD-box p54RNA helicase family is significant due to its involvement in a variety ofbiological processes, which include the sexual development of theprotozoan Plasmodium, the regulation of multiple virulence-associatedgenes in Cryptococcus neoformans, germline apoptosis, and embryoniccytokinesis regulation in C. elegans (Beckham and Parker, 2008; Mair etal., 2006; Rajyaguru and Parker, 2009). It has been reported in previousstudies that DDX6-like proteins can apparently interact with a range ofdifferent protein partners in a variety of cellular situations (Weston andSommerville, 2006). DDX6 is localized in the cytoplasm and thusplays a role at the protein translational level and/or mRNA stability(Abdelhaleem, 2004). It has been reported earlier that the Xenopushomologue Xp54 is an integral component of messenger ribonucleopro-tein (mRNP) particles and it acts as a factor involved in translationalcontrol (Ladomery et al., 1997). The studies reported thus far indicate

Gene 522 (2013) 46–59

Abbreviations: ATP, adenosine tri-phosphate; ATPase, adenosine tri-phosphatase;DAB, 3,3′-diaminobenzidine; DAPI, 40,60-di-amidino-2-phenylindole dihydrochloride;DOZI, development of zygote inhibited; ELISA, enzyme-linked immunosorbent assay;HRP, horseradish peroxidase; IPTG, isopropyl-β-D-thiogalactopyranoside; NTP, nucleo-side triphosphate; PBS, phosphate buffered saline; PCR, polymerase chain reaction;PfeIF4E, Plasmodium falciparum eukaryotic initiation factor 4E; ss-DNA, single strandedDNA; SDS-PAGE, sodium dodecyl sulfate polyacrylamide gel electrophoresis; TLC,thin-layer chromatography.⁎ Corresponding author. Tel.: +91 11 26741358; fax: +91 11 26742316.

E-mail addresses: renu@icgeb.res.in, renututeja@gmail.com (R. Tuteja).

0378-1119/$ – see front matter © 2013 Elsevier B.V. All rights reserved.http://dx.doi.org/10.1016/j.gene.2013.03.063

Contents lists available at SciVerse ScienceDirect

Gene

j ourna l homepage: www.e lsev ie r .com/ locate /gene

Author's personal copy

that this subfamily of helicases serves as a key component in the metab-olism of mRNA (Weston and Sommerville, 2006).

It is well known that Plasmodium is haploid throughout its life cycleand the sexual development of the malaria parasite commences withthe generation of gametocytes in the human host (Tuteja, 2007). In aninteresting study in Plasmodium berghei it has been reported that thetranslationally dormant mRNAs are found in the cytoplasm of femalegametocytes and their stabilization and maintenance depends onthis DEAD-box RNA helicase termed DOZI (development of zygoteinhibited) (Mair et al., 2006) which is a homologue of DDX6. It wasfurther reported that in the absence of DOZI, gametocytes undergolarge scale mRNA destabilization which results in the abortion ofookinete development soon after zygote formation (Mair et al., 2006).In a related study, a messenger ribonucleoprotein (mRNP) fromP. berghei gametocytes defined by DOZI and the Sm-like factor CITH(homolog of worm CAR-I and fly Trailer Hitch) was identified (Mair etal., 2010). Their analysis indicated that mRNP includes 16 commonmajor protein factors including eIF4E and poly(A)-binding protein(Mair et al., 2010).

In the present study we report the detailed biochemical character-ization of the DOZI homologue commonly known as DDX6 fromPlasmodium falciparum 3D7 strain. P. falciparum DOZI homologue,PfDZ50 contains the characteristic nucleic acid-dependent ATPase,RNA and DNA binding and RNA unwinding activities but it has nodetectable DNA unwinding activity. The characterization of truncatedderivatives of PfDZ50 such as PfDZ50T1 (domain 1) and PfDZ50T2(domain 2) shows that none of them contains the enzymatic activityand full-length PfDZ50 is required for the ATPase activity. UsingELISA-based interaction assays we show that PfDZ50 interacts withPfeIF4E. PfDZ50 interacts with PfeIF4E mainly through domain 1 be-cause when the truncated derivatives, PfDZ50T1 (domain 1) andPfDZ50T2 (domain 2) were used, it was observed that the interactionwith PfDZ50T2 is almost negligible as compared to PfDZ50T1, whichshowed almost 50% interaction in comparison to PfDZ50. Using invitro translation assays we further show that PfDZ50 inhibits transla-tion. The supplementation of the lysate with externally added recom-binant PfeIF4E restores ~70% of the translation. It is interesting tonote that PfDZ50 (PfDDX6) is expressed in all the intraerythrocyticdevelopmental stages i.e. ring, trophozoite and schizont stages inthe parasite P. falciparum 3D7 strain. Using immunofluorescenceassays we demonstrate that PfDDX6 is localized mainly in thecytoplasm in the P. falciparum 3D7 strain. To the best of our knowl-edge this is the first study reporting the direct interaction betweenPfDZ50 and PfeIF4E and these studies will advance our knowledgeregarding the function of DDX6 in general.

2. Materials and methods

2.1. Identification and cloning of P. falciparum PfDZ50gene

The P. falciparum helicase gene was amplified using the forwardprimer PfDZ50F1 with BamHI site and the reverse primer PfDZ50R1with HindIII site using P. falciparum 3D7 cDNA as template. The PCRconditions used were 95 °C for 1 min, 54 °C for 1 min and 72 °C for2 min. Thiswas repeated for a total of 35 cycles and at the endone elon-gationwas done at 72 °C for 12 min. The PCRproduct of ~1.3 kbwas gelpurified and cloned into the pGEM-T easy vector from Promega(Madison, WI, USA) and the clones were sequenced by dideoxysequencing reactions (Macrogen, Korea). The nucleotide sequencewas submitted to GenBank and the accession no. is EF070216.1. Forthe amplification of domain 1 (from amino acid 1–254) separately,PfDZ50F1 with BamHI site and PfDZ50R2 with HindIII site were used.For the amplification of domain 2 (from amino acid 255–433) separate-ly, PfDZ50F2with BamHI site and PfDZ50R1with HindIII site were usedand the PCR conditions used were 95 °C for 1 min, 54 °C for 1 min and72 °C for 1 min.

1. PfDZ50F1, 5′-GGGATCCATGAGTTATAAAACCAATT-3′ (BamHI site)2. PfDZ50R1, 5′-CAAGCTTTTAGGTATATAAGGATGGGT-3′ (HindIII site)3. PfDZ50F2, 5′-GGGATCCTTATCAGATGCCCATGAAATAAATC-3′ (BamHI

site)4. PfDZ50R2,5′-CAAGCTTGATTTATTTCATGGGCATCTGATAA-3′

(HindIII site)

The DNA bands were excised using BamHI and HindIII enzymes(New England Biolabs, Beverly, MA, USA) and gel purified forsubcloning into the expression vector pET28a.

2.2. Expression and purification of recombinant protein in Escherichiacoli

The fragment excised from pGEMT easy vector was stitched in thepET28a (Novagen, Madison, WI, USA) at the BamHI and HindIII sites.The expression clones were transformed into E. coli strain BL21 (DE3)pLysS and the expression of recombinant protein was induced by1 mM IPTG. The expressed protein was purified using standardmethods with Ni-NTA (Qiagen, GmbH, Germany) affinity chromatog-raphy. The recombinant his-tagged protein was eluted with 200 mMimidazole in protein buffer (20 mM Tris–HCl (pH 8.0), 250 mM NaCl,10% (v/v) glycerol and protease inhibitor cocktail (Sigma ChemicalCo., St. Louis, MO, USA) and was checked for purity by SDS-PAGE(10% (w/v) polyacrylamide gel) and silver staining using slightmodifications of the standard protocol (Sambrook et al., 1989). Theslight modification included extensive washing of the gel afterfixation and this step reduces the background and increases thesensitivity of the stain.

2.3. Generation of polyclonal antisera

Purified PfDZ50 was used for the preparation of antibodies in miceusing the standard protocols (Sambrook et al., 1989). The polyclonalantibodies were purified as IgG fractions using protein A-Sepharoseas described (Sambrook et al., 1989).

2.4. Preparation of DNA helicase substrate

The helicase activity of PfDZ50 was determined by the standardstrand displacement assay using the partially duplex substrate andthe method described previously (Pradhan and Tuteja, 2007). Tennanograms of the oligodeoxynucleotide was labeled at 5′-end withT4 polynucleotide kinase (5 U) and 1.85 MBq of [γ-32P]ATP (specificactivity 222 TBq/mmol) in standard conditions. The labeled oligode-oxynucleotide was then annealed with 0.5 μg of single-strandedcircular M13mp19 (+) DNA using standard annealing buffer byheating at 95 °C for 1 min, transferring immediately to 65 °C for2 min and then cooling slowly to room temperature. The non-hybridized oligodeoxynucleotide was removed using gel filtrationthrough a Sepharose 4B column (Pharmacia, Sweden). The reactionmixture (10 μl) containing appropriate buffer, the 32P-labeledhelicase substrate (1000 cpm/10 μl) and the purified protein fractionto be assayed was incubated at 37 °C for 60 min. The substrate andproducts were separated by electrophoresis on a nondenaturing 12%PAGE (Tuteja et al., 1993) and the gel was scanned on phosphoimagerand both the substrate and unwound DNA bands were quantified.

2.5. ATPase assay

The hydrolysis of ATP catalyzed by PfDZ50 was assayed by mea-suring the formation of 32P from [γ-32P] ATP. The reaction conditionswere identical to those described for the helicase reaction, except thatthe 32P-labeled helicase substrate was replaced by a mixture of[γ-32P] ATP (specific activity 222 TBq/mmol−1) and cold ATP(1 mM). The reaction was performed for 2 h at 37 °C (or one hour)

47M. Tarique et al. / Gene 522 (2013) 46–59

Author's personal copy

I Ia

Ib

II III

IV

V VI

VI

A

48 M. Tarique et al. / Gene 522 (2013) 46–59

Author's personal copy

in the presence of both enzyme and 50 ng of M13 mp19 ssDNA. Thiswas followed by thin-layer chromatography (TLC) (Pradhan andTuteja, 2007) and the quantitation was done using Image J/GelDocsoftware (AlphaImager) (http://rsbweb.nih.gov/ij/).

2.6. Immunodepletion assay

For this assay aliquots of the purified PfDZ50 were incubated withIgG purified from anti-preimmune and/or anti-PfDZ50 antisera at0 °C for 60 min. The antigen–antibody complexes were removed bythe addition of protein A Sepharose beads. The supernatants wereused for the ATPase activity analysis in the same way as describedabove.

2.7. RNA helicase and RNA-dependent ATPase assays

The RNA helicase assay was done by using the method describedearlier (Pradhan and Tuteja, 2007). The RNA helicase substrate wasprepared by using the RNA oligonucleotides synthesized fromPrimmsrl (Milan, Italy): 13 mer 5′-AUAGCCUCAACCG-3′ and 39 mer5′-GGGAGAAAUCACUCGGUUGAGGCUAUCCGUAAAGCACGC-3′. TheRNA-dependent ATPase assay was performed using a protocol similarto that described above. The reaction was performed for 2 h at 37 °Cboth in the presence and absence of RNA from P. falciparum followedby TLC. The quantitation was done using Image J/GelDoc software(AlphaImager) (http://rsbweb.nih.gov/ij/).

2.8. In vitro RNA binding assay

The RNA binding assay was done by using the method describedpreviously (Cheng et al., 2005; Shankar et al., 2008). BSA (1 μg) anddifferent amounts of PfDZ50 were dot-blotted on pre-charged PVDFmembrane and the membrane was incubated in blocking bufferwhich contained 25 mM NaCl, 10 mM MgCl2, 10 mM HEPES,0.1 mM EDTA, 1 mM DTT and 3% BSA. The 13 mer RNA oligonucleo-tide used for the preparation of RNA helicase substrate was labeledat the 5′-end using the standard procedure and T4 polynucleotidekinase (NEB, England) with 1.85 MBq of [γ-32P] ATP (specific activity222 TBq/mmol) and purified using Sepharose 4B (Pharmacia, Sweden)column chromatography. After blocking the membrane was incubatedfor 2 h in binding buffer containing 10 pmol of 32P-labeled RNA oligonu-cleotide. After binding, the membrane was washed thrice with bindingbuffer and exposed for autoradiography. To check for loading of proteins,increasing amounts of PfDZ50 were dot-blotted on another prechargedPVDF membrane. This membrane was blocked with blocking buffer(1% BSA in Tris buffered saline (TBS)) for 1 h at room temperature andprobed for a further 1 h with alkaline phosphatase conjugated anti-hisantibody (Sigma Chemical Co., St. Louis, MO, USA) in the same buffer.The blot was washed and developed using standard protocol.

2.9. In vitro DNA binding assay

The DNA-binding assay was performed by using the samemethod as described for RNA binding but end-labeled DNA

Fig. 1. A. Comparison of amino acid sequence of P. falciparum (Pf) DOZI (1–433) with DOZI from P. berghei (Pb) (1–433), Cgh-1 from Caenorhabditis elegans (Ce) (1–430), Me31bfrom Drosophila melanogaster (Dm) (1–459), DDX6 from Homo sapiens (Hs) (1–483) and Dhh1 from Saccharomyces cerevisiae (Sc) (1–506). The alignment was done using CLUSTALW2 program (http://www.ebi.ac.uk). The following symbols denote the degree of conservation observed in each column: ‘*’ means that the residues in that column are identical inall sequences in the alignment, ‘:’means that conserved substitutions have been observed and ‘.’means that semi-conserved substitutions are observed. The colors in the alignmentare described as follows: red denotes the amino acid residues AVFPMILW, blue denotes DE, magenta denotes RK and green denotes the residues STYHCNGQ respectively. The con-served motifs are boxed in blue and the name of each motif is written in roman numerals. The red and black boxes indicate domains 1 and 2 respectively. The PlasmoDB number forthe PfDDX6 (PfDZ50) sequence is PF0915w. The PlasmoDB number for P. berghei DOZI is PBANKA_121770 and the accession numbers for Cgh-1 from C. elegans (Ce), Me31b from D.melanogaster (Dm), DDX6 from H. sapiens (Hs), and Dhh1 from S. cerevisiae (Sc) are CCD63592, AAF52881, NP_004388.2 and CAA98734 respectively. B. Comparison of amino acidsequence of P. falciparum (Pf) DOZI (1–433) with eukaryotic initiation factor 4A (eIF4A) from P. falciparum. The PlasmoDB number for eIF4A from P. falciparum is PF3D7_1468700.The alignment was done using CLUSTAL W2 program (http://www.ebi.ac.uk). The other details are as described in the above legend.

49M. Tarique et al. / Gene 522 (2013) 46–59

Author's personal copy

oligodeoxynucleotide of 32 bases with the sequence 5′-TTCGAGCTCGGTACCCGGGGATCCTCTAGAGT-3′ was used. BSA (1 μg) anddifferent amounts of PfDZ50 were dot-blotted on pre-charged PVDFmembrane and the membrane was incubated in blocking bufferdescribed in Section 2.8. After blocking themembranewas incubatedfor 2 h in binding buffer containing 10 pmol of 32P-labeled DNAoligodeoxynucleotide. After binding, the membrane was washedthrice with binding buffer and exposed for autoradiography. Tocheck for loading of proteins, increasing amounts of PfDZ50 weredot-blotted on another precharged PVDFmembrane. This membranewas blocked with blocking buffer (1% BSA in TBS) for 1 h at roomtemperature and probed for a further 1 h with alkaline phosphataseconjugated anti-his antibody (Sigma Chemical Co., St. Louis, MO,USA) in the same buffer. The blot was washed and developed usingthe standard protocol.

2.10. Enzyme-linked immunosorbent assay (ELISA) for protein–proteininteraction study

The 96 well ELISA plates were coated with varying concentrations(1, 2.5, 5, 10 and 20 ng) of purified PfDZ50, PfDZ50T1 (domain 1),PfDZ50T2 (domain 2) or PfeIF4E diluted in bicarbonate/carbonatecoating buffer (100 mM NaHCO3 and Na2CO3, pH 9.6), overnight at4 °C in 100 μl volume. The plate was emptied and washed threetimes with washing buffer phosphate buffered saline (PBS containing0.05% Tween 20) and blocked for 2 h with 5% non-fat milk in PBS. Theblocking buffer was removed and the wells were washed 3 times withthe same washing buffer. The interacting protein (5 ng) was addedand the incubation was continued at 37 °C for a further 2 h. Theplate was emptied and washed three times with the washing buffer.The assembly of the protein complex in the wells was then assessedthrough the use of polyclonal antibodies (1:15,000 dilutions). Theincubation with antibodies was done for 2 h at 37 °C. The plate waswashed three times with washing buffer to remove the unboundantibodies. The plate was further incubated with a 1:3000 dilutionof horse radish peroxidase (HRP) conjugated secondary antibody for2 h at 37 °C. The unbound antibody was removed by washingthe wells with washing buffer. The binding of the secondary antibodyto the protein complex was then detected by applying theo-Phenylenediamine (OPD) substrate (Sigma Chemical Co., St. Louis,MO, USA) in an appropriate buffer. The reaction was stopped andthe plate was read at 490 nm using an ELISA reader.

2.11. In vitro protein–protein interaction/co-immunoprecipitation assay

For the protein–protein interaction assay various full length clonessuch as PfDZ50 and PfeIF4E cloned in pET28a were used. These DNAclones were used to generate each of these radio-labeled proteinsusing the TnT T7 coupled reticulocyte lysate transcription translationsystem from Promega Corporation (Madison, WA, USA). For an inter-action assay, PfDZ50 was mixed with PfeIF4E and the mixture wasincubated at room temperature for 60 min. An aliquot of this mixturewas saved as loading control. Subsequently preimmune serum orpurified PfDZ50 IgG (~1 μg) was added to the mixtures and theincubation was continued for 60 min on ice. Protein A sepharose(Amersham) was added to these mixtures and the incubation wascontinued for 30 min on ice and the complex was removed by centri-fugation. The beads containing the complex were washed twice withPBS, boiled in sample buffer and these samples were analyzed bySDS-PAGE followed by autoradiography.

2.12. Effect of PfeIF4E on the PfDZ50 ATPase activity

To investigate the effect of PfeIF4E on PfDZ50 the RNA ATPase assaywas done in the presence of different concentrations (20 nM, 40 nM,60 nM, 80 nM and 120 nM) of purified PfeIF4E using the method

described in previous sections. The quantitation was done usingImage J/GelDoc software (AlphaImager) (http://rsbweb.nih.gov/ij/).

2.13. Effect of PfDZ50 on translation

The effect of PfDZ50 on the protein translation was checked byusing TnT T7 coupled reticulocyte lysate transcription translationsystem from Promega Corporation (Madison, WA, USA). The controlluciferase DNA (~1 μg) was incubated with TnT T7 coupled reticulo-cyte lysate at 18 °C for 3 h. It was used as a positive control fortranslation. In separate reactions 20 nM and 50 nM of purifiedrecombinant PfDZ50 were also added with luciferase DNA (~1 μg)and incubated at 18 °C for 3 h. The samples were boiled at 100 °Cfor 5 min with protein loading dye and analyzed by SDS-PAGEfollowed by autoradiography. The quantitation was done usingImage J/GelDoc software (AlphaImager) (http://rsbweb.nih.gov/ij/).

2.14. Restoration of translation by PfeIF4E added externally

The restoration of translation was tested by using purified recom-binant PfeIF4E. The control luciferase DNA (~1 μg) was used aspositive control and 50 ng purified PfDZ50 as negative control fortranslation. In different reactions, 20 nM and 50 nM of purifiedrecombinant PfeIF4E, 50 nM of purified PfDZ50, 50 nM of purifiedPfDZ50T1 and 50 nM of purified PfDZ50T2 and (~1 μg) luciferaseDNA were incubated with TnT T7 coupled reticulocyte lysate at18 °C for 3 h. The samples were analyzed by SDS-PAGE followed byautoradiography. The quantitation was done using ImageJ/GelDocsoftware (AlphaImager) (http://rsbweb.nih.gov/ij/).

2.15. Immunofluorescence assay and western blotting

Thin smears of parasitized red blood cells of different develop-mental stages were prepared and fixed in acetone for 5 min followedby chilled methanol for 15 s at room temperature. The fixed slideswere dried and incubated in 10% fetal bovine serum (FBS) in PBS ina humid chamber at 37 °C for 2 h for blocking. The slides werewashed with PBS and incubated with purified IgG of anti-PfDZ50antibodies at 1:200 dilutions in PBS containing 10% FBS for 1 h at37 °C. The slides were then washed four times with PBS for 15 mineach and then incubated for 1 h at 37 °C with secondary antibody(fluorescein isothiocynate-conjugated-anti-mouse IgG (Sigma Chem-ical Co., St. Louis, MO, USA) diluted 1:100 in PBS containing 10%FBS. After washing, the slides were mounted with antifade mixedwith DAPI (Invitrogen). In an immunofluorescence assay 40,60-di-amidino-2-phenylindole dihydrochloride (DAPI) is used for thenuclear staining. The images were collected using a BioRad 2100laser-scanning microscope attached to a Nikon TE 2000 Umicroscope.

For the purpose of western blotting parasites of different stages(ring, trophozoite and schizont and mixed) were released fromsynchronized 3D7 infected erythrocyte cultures by treatment with0.1% (w/v) saponin. The parasite culture was synchronized by usinga previously described method of sorbitol treatment (Lambros andVanderberg, 1979). Cell-free protein extracts from these cultureswere prepared by sonicating the parasite pellets in a buffer containing10 mM Tris–HCl, pH 7.5, 100 mM NaCl, 5 mM EDTA, 1% Triton X-100and protease inhibitor cocktail (Sigma Chemical Co., St. Louis, MO,USA). The extracted proteins were resolved by SDS-PAGE and trans-ferred onto nitrocellulose membrane for the purpose of western blot-ting using the purified IgG of anti-PfDZ50 antibodies. The upper partof the membrane was first incubated overnight at 4 °C with a 1:200or 1:1000 dilution of the purified IgG of anti-PfDZ50 antibodies andthe lower part of the membrane was incubated overnight at 4 °Cwith a 1:500 dilution of the purified IgG of anti-PfH45 antibodies(Pradhan and Tuteja, 2007) and then incubated with the appropriatesecondary antibody (Sigma Chemical Co., St. Louis, MO, USA) coupled

50 M. Tarique et al. / Gene 522 (2013) 46–59

Author's personal copy

to horse radish peroxidase (HRP). The blot was developed using DABtablet (Sigma Chemical Co., St. Louis, MO, USA) according to themanufacturer's instructions.

3. Results and discussion

Similar to other DEAD box family of helicases, DDX6 is alsocomposed of two RecA-like linked domains, domains 1 and 2. Domain1 consists of motif Q, Ia, Ib, II and III and domain 2 contains motif IVto VI (Cordin et al., 2006). Motif III forms intramolecular interactionsand domains 1 and 2 participate in ATP and RNA binding (Cordinet al., 2006). With regard to the location of conserved DEAD boxmotifs, DDX6 helicases are closely related to the eukaryotic transla-tion initiation factor 4A (eIF4A) and the mRNA export factorDbp5 (Pradhan and Tuteja, 2007; Mehta and Tuteja, 2011). All ofthese helicases contain enough sequence information for nucleotidebinding, hydrolysis and nucleic acid unwinding activity. But thelength of N and C-terminal extensions is highly variable (Westonand Sommerville, 2006). These extensions most likely contribute tothe protein–protein interaction function of these helicases.

3.1. Identification and sequence analysis of the helicase

Using the bioinformatics tools we have reported previously thatthe gene with PlasmoDB number PFC0915w (new PlasmoDB numberPF3D7_0320800) is the bonafide homologue of human DDX6 (Tutejaand Pradhan, 2006; Tuteja, 2010). This DEAD-box helicase has beentermed as DOZI (Mair et al., 2006) and it shows a high degree ofhomology with other homologues and also has the characteristicsignature motifs I to VI of DEAD box family (Fig. 1A). The 1302 basepair open reading frame was amplified from the P. falciparum 3D7cDNA. The nucleotide sequence encodes a deduced polypeptidewith a predicted molecular mass of ~50 kDa and is thus designatedas PfDZ50. Sequences similar to PfDZ50 were identified in otherPlasmodium species and protozoan parasites. PfDZ50 contains ~62–96% homology to its other counterparts Cgh-1 from C. elegans,Me31b from D. melanogaster, DDX6 from Homo sapiens, Dhh1 fromS. cerevisiae and DOZI from P. berghei (Fig. 1A). It is noteworthy thatPfDZ50 contains a shorter N-terminal region and therefore it is slight-ly smaller in size as compared to human DDX6 (Fig. 1A). A compari-son of the amino acid sequence of PfDZ50 with eukaryotic initiation

D. Template E. PfDZ50 F. Superimposed

B. P. falciparum

C. Homo sapiens

60-88 91 261 271(Q motif) (ATP binding) (Helicase-C terminal)

(ATP binding) (Helicase-C terminal) 298 308

(Q motif) )

I Ia Ib II III IV V VI

A

431

96-124 127 468

Fig. 2. A. Sequence conservation within the characteristic DDX6 motifs. The height of the amino acids reflects the level of conservation at a given position, tall letters indicate higherconservation. The coloring marks the chemical properties of a given amino acid position: blue—most hydrophobic, green—polar non-charged, magenta—acidic, red—positivelycharged, histidine—pink, glycine—orange and tyrosine is turquoise. Sequence logos were created from the alignment of the six sequences shown in Fig. 1 according to Baileyand Elkan (1994). B and C. The amino acid position and location of each domain. The domain analysis was done by using ‘Scan Prosite’ at http://expasy.org/tools/scanprosite/.The text in bracket is the name of the domain and the numbers are positions of respective domains in the protein. D to F. Structure modeling. PfDZ50 sequence was submittedto the Swissmodel server and the structure was obtained. The molecular graphic images were produced using the UCSF Chimera package from the resource for Biocomputing, Vi-sualization, and Informatics (http://www.cgl.ucsf.edu/chimera) at the University of California, San Francisco (supported by NIH P41 RR-01081). D. Template; E. PfDZ50 and F.superimposed image.

51M. Tarique et al. / Gene 522 (2013) 46–59

Author's personal copy

factor 4A (eIF4A) homologue PfH45 from P. falciparum 3D7 strainshows that except the conserved domains, PfDZ50 contains verylimited homology with PfH45 in other regions of the proteinsequence (Fig. 1B).

A comparison of amino acid sequences of all these motifs in sixDDX6 helicases shown in Fig. 1A was done using the MEME programat http://ws.nbcr.net (Bailey and Elkan, 1994). It is evident from theresults that except at one place in motif I, the level of conservationis very high in all the other motifs (Fig. 2A). The scan prosite analysis(http://swissmodel.expasy.org) of PfDZ50 and HsDDX6 revealed thatboth of these proteins contain the characteristic Q, ATP-binding andhelicase C-terminal motifs (Fig. 2B and C).

3.2. Molecular modeling of PfDZ50 structure

The complete sequence of PfDZ50 was submitted to theSwissmodel homology-modeling server (http://swissmodel.expasy.org/) (Arnold et al., 2006) for the purpose of molecular modeling.PfDZ50 primary sequence shows ~71.85% identity with the yeasthomologue DHH1P. The structural modeling of PfDZ50 was thereforedone using the known crystal structure of the DEAD-box proteinDHH1P (1s2m) as the template (Cheng et al., 2005) (Fig. 2D). Molec-ular graphics images were produced using the UCSF Chimera packagefrom the resource for biocomputing, visualization and informatics atthe University of California, San Francisco (supported by NIH P41

RR-01081) (Pettersen et al., 2004). The ribbon diagram of the tem-plate is shown in Fig. 2D and the predicted structure is shown inFig. 2E. The superposition shows that the structures overlap almostcompletely (Fig. 2F).

3.3. Construction of the truncated derivatives of PfDZ50

In order to check the roles of the domain 1 and domain 2, twotruncated derivatives of PfDZ50 (Fig. 3A) were made. PfDZ50T1(from amino acid 1–254, ~29 kDa) contains domain 1, which includesmotif Q, I, Ia, Ib, II and III (Figs. 1A, 3B). PfDZ50T2 (from amino acid255–433, ~20 kDa) contains domain 2, which includes motif IV,V and VI (Figs. 1A, 3C). The sequences of PfDZ50T1 and PfDZ50T2were submitted to the Swissmodel homology-modeling server(http://swissmodel.expasy.org/) (Arnold et al., 2006) for the purposeof molecular modeling. PfDZ50T1 sequence shows ~67% identity withthe yeast homologue DHH1P. The structural modeling of PfDZ50T1was therefore done using the known crystal structure of theDEAD-box protein DHH1P (1s2m) as the template (Cheng et al.,2005) (Fig. 3). PfDZ50T2 sequence shows ~80% identity with theC-terminal domain of human DDX6 homologue (2waxC) (Tritschleret al., 2009). The structural modeling of PfDZ50T2 was thereforedone using the known crystal structure of the C-terminal domain ofhuman DDX6 homologue (2waxC) as the template (Fig. 3). Moleculargraphics images were produced using the UCSF Chimera package as

GTGKT PTRELA TPGR DEAD SAT RGID HRIGRSGRQ IIF

86 Q 19 I 24 Ia 43 Ib 20 II 27 III 58 IV 55 V 20 VI 43

GTGKT PTRELA TPGR DEAD SATQ86 Q 19 I 24 Ia 43 Ib 20 II 27 III 12

RGID HRIGRSGRIIF

46 IV 55 V 20 VI 43

A

B

C

D

i iii

vi

ii

iv v

Fig. 3. A. Structure of PfDZ50. The conserved sequences of each motif are written inside the boxes. The numbers refer to the amino acids separating the various motifs and the lengthof N- and C-terminal extensions. This figure is not drawn to scale. B. Motif structure of PfDZ50T1 (domain 1). C. Motif structure of PfDZ50T2 (domain 2). D. Structure modeling.PfDZ50T1 and PfDZ50T2 sequences were submitted to Swissmodel server and the structures were obtained. The molecular graphic images were produced using the UCSF Chimerapackage from the resource for Biocomputing, Visualization, and Informatics (http://www.cgl.ucsf.edu/chimera) at the University of California, San Francisco (supported by NIH P41RR-01081). i. Template; ii. PfDZ50T1; iii. superimposed image; iv. template; v. PfDZ50T2 and vi. superimposed image.

52 M. Tarique et al. / Gene 522 (2013) 46–59

Author's personal copy

described above (Pettersen et al., 2004). The ribbon diagram ofthe template for domain 1 is shown in Fig. 3D (i) and the predictedstructure is shown in Fig. 3D (ii). The superposition shows that thestructures overlap partially (Fig. 3D (iii). The ribbon diagram of thetemplate for domain 2 is shown in Fig. 3D (iv) and the predictedstructure is shown in Fig. 3D (v). The superposition shows that thestructures overlap almost completely (Fig. 3D (vi)).

3.4. Expression and purification of PfDZ50 and its truncated derivatives

The biochemical characterization of PfDZ50 and its truncatedderivatives were important in order to assign it as a bonafide helicase.PCR amplification of PfDZ50 gene was done using cDNA from the

P. falciparum 3D7 strain as template because the gene containsthe introns. Therefore the amplified products of PfDZ50, PfDZ50T1(domain 1) and PfDZ50T2 (domain 2) were cloned into appropriatesites in the bacterial expression vector pET28a for the activityanalysis of all of these proteins. The recombinant his-tagged PfDZ50,PfDZ50T1 and PfDZ50T2 were purified through Ni2+-NTA affinitychromatography. The SDS-PAGE analysis followed by silver stainingof the purified proteins showed that all of these proteins such asPfDZ50, PfDZ50T1 and PfDZ50T2 are homogeneous preparations(Fig. 4A, lanes 1, 2 and 3 respectively). These purified preparationswere used for all of the assays described in the following sectionsand PfDZ50 was also used for the production of polyclonal antibodiesin mice.

116

66

45

35

25

18

A B

(nM)

ATP

Pi

EPi

ATP

F

C

Pi

ATP

M 1 2 3

C 1 2 3 4 5 6 7 8 9

PfDZ50T1 - + + + - - -

PfDZ50T2 - + + + - - -

PfDZ50 -

-

-

- -

- -

- - - - - - + + +

Concentration - 20 40 80 20 40 80 20 40 80

1 2 3 4 C

C 1 2 3 B

C 1 2 3 4 C 1 2 3 4 5 6

D

Per

cent

AT

P h

ydro

lysi

s

0

5

10

15

20

25

30

35

40

Per

cent

AT

P h

ydro

lysi

s

0

5

10

15

20

25

30

35

40

Fig. 4. A. Silver-stained gels of purified proteins. Lane M contains the protein molecular weight marker and lanes 1, 2 and 3 contain 0.2 μg of the purified PfDZ50, PfDZ50T1 andPfDZ50T2 proteins respectively. B. ATPase activity of PfDZ50 and its truncated derivatives in the presence of DNA. Lane C, is a reaction without an enzyme, lanes 1–3, 4–6 and7–9 are reactions with increasing concentrations (20, 40 and 80 nM) of purified PfDZ50T1 (domain 1, D1), PfDZ50T2 (domain 2, D2) and PfDZ50 respectively. C. ATPase activityof PfDZ50. Lane C is a reaction without an enzyme, lanes 1–4 are reactions with increasing concentrations (20, 40, 80 and 120 nM) of purified PfDZ50. D. Time dependence(lanes 1–6 are 10, 20, 40, 60, 90 and 120 min, respectively) of ATPase activity of purified PfDZ50. C is the control reaction without an enzyme. In C and D the upper panelshows the quantitative enzyme activity data as percent ATP hydrolysis from the autoradiograms. E. An immunodepletion assay of ATPase activity. Lanes 1 and 2, are reactionswith increasing concentrations of PfDZ50 pretreated with immune IgG and lanes 3 and 4, are reactions with increasing concentrations of PfDZ50 pretreated with pre-immuneIgG. C is the control reaction without an enzyme. F. Lanes 1–3 are DNA helicase assays with increasing concentrations (20, 40 and 80 nM) of purified PfDZ50. C is a reaction withoutan enzyme and B is a heat denatured substrate.

53M. Tarique et al. / Gene 522 (2013) 46–59

Author's personal copy

3.5. Characterization of ATPase and DNA helicase activities of PfDZ50 andits truncated derivatives

The ssDNA-dependent ATPase activity of PfDZ50, PfDZ50T1(domain 1) and PfDZ50T2 (domain 2) was checked using standardassay conditions and varying concentrations (20, 40 and 80 nM) ofpurified enzymes and 1665 Bq [γ-32P] ATP as a substrate in thepresence of 50 ng of M13 ssDNA. The results clearly showed thatPfDZ50T1 (Fig. 4B, lanes 1–3) and PfDZ50T2 did not show any ATPaseactivity (Fig. 4B, lanes 4–6). On the other hand only PfDZ50 exhibitsconcentration dependent ATPase activity (Fig. 4B, lanes 7–9). Theseresults clearly suggest that for the enzymatic activity full-length

protein is required. Therefore further analysis was done only withPfDZ50.

The concentration-dependence of ssDNA-dependent ATPaseactivity was further checked using different concentrations (20,40, 80 and 120 nM) of PfDZ50. The results clearly showed thatPfDZ50 exhibits concentration dependent ATPase activity (Fig. 4C,lanes 1–2) and the activity levels off at a higher concentration(Fig. 4C, lanes 3–4). The ATPase reaction using 80 nM of purifiedPfDZ50 at different time points was carried out in order to study thetime dependence of ATPase activity. The percent release of radioactivephosphate (Pi) from [γ-32P] ATP showed linearity up to 90 min(Fig. 4D, lane 5) and the activity did not increase further on longer

A

C 1 2 3 C 1 2 3

B

C D

C 1 2 3 C 1 2 3

E

Pi

ATP

PfDZ50

Concentration -

C 1 2 3 4 5 6 7 8 9

PfDZ50T1 - + + + - - - - - -

PfDZ50T2 - - - - + + + - - -

- - - - - - - + + +

20 40 80 20 40 80 20 40 80 (nM)

F G

0

5

10

15

20

25

30

35

40

ATP

C 1 2 3 4 5 6

Per

cent

AT

P h

ydro

lysi

s

0

5

10

15

20

25

30

35

40

Pi

C 1 2 3 4

Per

cent

AT

P h

ydro

lysi

s

Fig. 5. RNA binding activity of PfDZ50. A. Western blot probed with anti-his antibody. Different amounts (50, 100 and 200 ng) of purified PfDZ50 (lanes 1–3) were spotted on thecharged PVDFmembrane and the blot was probed with anti-his antibody and developed using the standard procedure. B. RNA binding activity. This experiment was done as writtenin the Materials and methods section. DNA binding activity of PfDZ50. C. Western blot probed with anti-his antibody. Different amounts (50, 100 and 200 ng) of purified PfDZ50(lanes 1–3) were spotted on the charged PVDF membrane and the blot was probed with anti-his antibody and developed using the standard procedure. D. DNA binding activity.This experiment was done as written in the Materials and methods section. E. ATPase activity of PfDZ50 and its derivatives in the presence of RNA. Lane C, a reaction without anenzyme, Lanes 1–3, 4–6 and 7–9 are reactions with increasing concentrations (20, 40 and 80 nM) of purified PfDZ50T1 (domain 1), PfDZ50T2 (domain 2) and purified PfDZ50 re-spectively. F. ATPase activity of PfDZ50 in the presence of RNA. Lane C is a reaction without an enzyme, lanes 1–4 are reactions with increasing concentrations (20, 40, 80 and120 nM) of purified PfDZ50. G. Time dependence (10, 20, 40, 60, 90 and 120 min) of ATPase activity of purified PfDZ50 in the presence of RNA. C is the control reaction withoutan enzyme. In F and G the upper panel shows the quantitative enzyme activity data as percent ATP hydrolysis from the autoradiogram.

54 M. Tarique et al. / Gene 522 (2013) 46–59

Author's personal copy

incubation up to 120 min (Fig. 4D, lane 6). It has been reportedrecently that yeast homologue of DDX6, Dhh1 has weaker RNA-dependent ATPase activity than other well characterized DEAD-boxhelicases (Dutta et al., 2011).

For an immunodepletion assay purified PfDZ50 was allowed toreact separately with IgGs purified from the pre-immune sera andfrom the sera of the mice immunized with PfDZ50 using the methoddescribed in the Materials andmethods section. The immunodepletedsupernatants were checked for ATPase activity. The results showedthat the ATPase activity of PfDZ50 (Fig. 4E, lanes 1–2) was depletedwith the specific antibodies. On the contrary the samples treatedwith pre-immune IgG showed concentration-dependent ATPaseactivity (Fig. 4E, lanes 3–4). These results confirm that the observedATPase activity is specifically due to PfDZ50. The helicase activityof PfDZ50 was determined by the standard strand displacementassay using the partially duplex substrate and the method describedpreviously (Pradhan and Tuteja, 2007). On repeated trials we wereunable to detect any DNA unwinding activity in PfDZ50 (Fig. 4F,lanes 1–3).

3.6. RNA binding activity of PfDZ50

In order to check the efficiency of PfDZ50 to bind to RNA, its RNAbinding efficiency was determined using the RNA binding assay asdescribed previously (Cheng et al., 2005; Shankar et al., 2008). Theresults indicate that PfDZ50 shows concentration-dependent RNAbinding activity (Fig. 5B, lanes 1–3). BSA was used as a control,which showed no binding to the RNA (Fig. 5B, lane C). An identicalblot of PfDZ50 was probed with anti-his antibody, which confirmedconcentration-dependent loading of protein (Fig. 5A, lanes 1–3). Ithas been reported recently that yeast homologue of DDX6, Dhh1also has RNA binding activity (Dutta et al., 2011).

3.7. DNA binding activity of PfDZ50

DNA-binding activity of PfDZ50 was also checked using thesame procedure as described for RNA binding but the labeled RNAwas replaced with labeled DNA oligodeoxynucleotide. The resultsindicate that PfDZ50 shows concentration-dependent DNA bindingactivity also (Fig. 5D, lanes 1–3). BSA was used as a control,which showed no binding to the DNA (Fig. 5D, lane C). An identicalblot of PfDZ50 was probed with anti-his antibody, which confirmedconcentration-dependent loading of protein (Fig. 5C, lanes 1–3).

3.8. RNA-dependent ATPase and RNA helicase activities of PfDZ50

The ATPase activity of purified PfDZ50 was tested in the presence ofRNA (50 ng). The results clearly showed that PfDZ50T1 (Fig. 5E,lanes 1–3) and PfDZ50T2 did not show any ATPase activity (Fig. 5E,lanes 4–6). On the other hand only PfDZ50 exhibits concentrationdependent ATPase activity (Fig. 5E, lanes 7–9). The concentration-dependence of RNA-dependent ATPase activity was further checkedusing different concentrations of PfDZ50. The results clearly showedthat PfDZ50 has concentration and RNA-dependent ATPase activity(Fig. 5F, lanes 1–4). The ATPase reaction using 80 nM of purifiedPfDZ50 at different time points was checked to study the time-dependence of ATPase activity. The percent release of radioactivephosphate (Pi) from [γ-32P] ATP showed linearity up to 90 min(Fig. 5G, lane 5) and the activity did not increase further upon longerincubation (Fig. 5G, lane 6). In an earlier study it was reported thatthe S. cerevisiae homologue of DDX6, Dhh1 has very low ATPase activityin the absence of RNA and a 10 fold stimulation in activity was observedin the presence of RNA (Dutta et al., 2011).

DDX6 is a well established RNA helicase (Weston and Sommerville,2006) therefore we also tested its RNA helicase activity. The RNAhelicase assay was done by using the method described earlier

(Pradhan and Tuteja, 2007). The RNA helicase activity of PfDZ50 waschecked by using the partially duplex RNA–RNA substrate using differ-ent concentrations of the enzyme (50, 100, 150, 200 and 250 nM). Theresults show clearly that PfDZ50 contains concentration dependentRNA helicase activity and unwinds up to ~70% of the substrate at aconcentration of≥150 nMof PfDZ50 (Fig. 6A). The time course analysiswas also performed with the steady state concentration of enzyme for20, 40, 60, 80 and 100 min. The results clearly show that PfDZ50 alsocontains characteristic time dependent RNA helicase activity andactivity gets leveled off at ~70% unwinding (Fig. 6B). It has been

C

A

Boil

C Boil

B

1 2 3 4 5

1 2 3 4 5

Concentration(nM)

Per

cent

unw

indi

ng

20

50 100

150

200

250

40

60

80

100

Per

cent

unw

indi

ng

0

0

20

40

60

80

100

Time(min)

Fig. 6. A. Concentration dependence of RNA helicase activity of PfDZ50. Different con-centrations (50–250 nM) of purified PfDZ50 were used for the assay. The activity isshown as percentage unwinding. B. Time dependence (20–100 min) of RNA helicaseactivity of purified PfDZ50. In A and B the upper panel shows the quantitative enzymeactivity data as percent unwinding in the form of graphs.

55M. Tarique et al. / Gene 522 (2013) 46–59

Author's personal copy

reported recently that the yeast homologue of DDX6, Dhh1 has nodetectable RNA helicase activity (Dutta et al., 2011).

3.9. In vitro protein–protein interaction of PfeIF4E and PfDZ50 usingELISA and co-immunoprecipitation

In previous studies from our laboratory we have reported theidentification and functional characterization of translation initiationcomplex eIF4F components and poly (A)-binding protein fromP. falciparumwhere we had shown that PfeIF4E is involved in transla-tion (Tuteja and Pradhan, 2009; Tuteja, 2009). In the present study

using ELISA based assays we performed the interaction analysis.Purified PfDZ50T1, PfDZ50T2 and PfDZ50 were used for the interac-tion study with recombinant PfeIF4E. The results show that PfeIF4Einteracts with PfDZ50 (Fig. 7A) and furthermore this interactionwas attributable to PfDZ50T1, which contains only domain 1(Fig. 7A). On the other hand PfDZ50T2, which contains only domain2 showed almost no interaction (Fig. 7A). It was reported in HeLacells that the cap-binding translation initiation factor eIF4E is inmolecular contact with the DDX6 homologue rck/p54 in P bodies invivo (Andrei et al., 2005). It has been reported previously that rck/p54 interacts with eIF4E in HeLa cells (Akao et al., 2006). In a previous

Concentration

Abs

orba

nce

at 4

90 n

m

A

PfDZ50

PfeIF4E

B

Pi

CR

elat

ive

Tra

nsla

tion

D E

Rel

ativ

e T

rans

lati

on

Lysate + + + + + PfDZ50T1 - + - - -PfDZ50T2 - - + - -PfDZ50 - - - + + Conc. (nM) 50 50 20 50

Lysate + + + + + +PfDZ50T1 - + - - - -PfDZ50T2 - - + - - -PfDZ50 - + + + + + PfeIF4E - - - + + -Conc. PfeIF4E 50 20 -(nM)

ATP

C 1 2 3 4 5 6 7

Per

cent

AT

P h

ydro

lysi

s

0

5

10

15

20

25

30

35

40

1 2 3 4 5 6 7

1 2 3 4 51 2 3 4 5 6

Fig. 7. A. ELISA based protein–protein interaction study with varying concentrations (1, 2.5, 5, 10 and 20 ng in columns 2–6 respectively) of PfDZ50T1 (domain 1, dark gray bars),PfDZ50T2 (domain 2, light gray bars) and PfDZ50 (shaded bars), and a fixed concentration of PfeIF4E (5 ng). The interaction was done as described in the Materials and methodssection. This experiment was repeated at least three times and the data are shown. B. Co-immunoprecipitation assay (Co-IP) using anti-PfDZ50 IgG, anti-PfeIF4E IgG, andpreimmune IgGs. Lane 1 is an aliquot of the mixture used as loading control (10 μl of PfDZ50 + 10 μl of PfeIF4E) and lane 2 shows in vitro translated PfDZ50 (10 μl), lane 3shows in vitro translated PfeIF4E (10 μl), lanes 4, 5, 6 and 7 are co-immunoprecipitates with anti-PfDZ50 IgG, anti-PfeIF4E IgG, and two preimmune IgGs (preimmune-1 of miceused for PfDZ50 antiserum production and preimmune-2 of mice used for PfeIF4E antiserum production) respectively. C. Effect of PfeIF4E on the RNA-dependent ATPase assay.The assay was done with a fixed concentration of purified PfDZ50 (80 nM) and varying concentrations (20, 40, 60, 80 and 120 nM) of purified PfeIF4E (lanes 3–8). Lane C is thecontrol reaction without an enzyme, lane 1 is PfDZ50 (80 nM) used as a positive control and lane 2 is PfeIF4E alone (50 nM), which did not show any ATPase activity. Theupper panel shows the quantitative enzyme activity data as percent ATP hydrolysis from the autoradiogram. D. Effect of PfDZ50T1, PfDZ50T2 and PfDZ50 on translation. The exper-iment was done as described in the Materials and methods section. Lane 1 is positive control with luciferase DNA (~1 μg). Lanes 2–5 are reactions in the presence of 50 nM ofPfDZ50T1, 50 nM of PfDZ50T2 and 20 and 50 nM of PfDZ50 respectively. The upper panel shows the relative band intensity in the form of relative translation. E. Restoration oftranslation. The translation was restored when PfeIF4E was added externally. Lane 1 is without PfDZ50 as positive control and lanes 2–6 contain 50 nM of PfDZ50. Lanes 2 and3 are the reactions in the presence of 50 nM of PfDZ50T1 and PfDZ50T2. In addition lanes 4 and 5 are reactions in the presence of externally added 50 nM and 20 nM of PfeIF4E.The upper panel shows the relative band intensity in the form of relative translation. The arrows in D and E show the position of translated luciferase protein.

56 M. Tarique et al. / Gene 522 (2013) 46–59

Author's personal copy

study in P. berghei, the protein components of P granules wereidentified and the poly(A) binding protein and the cap bindingprotein PbeIF4E were reported to be associated with PbDOZI (Mairet al., 2010).

In addition to the ELISA based interaction, we also performed theinteraction study through an in vitro co-immunoprecipitation assay(Co-IP) using in vitro translated PfDZ50 and PfeIF4E proteins asdescribed in the Material and methods section. Lane 1 is a mixtureof 10 μl each of in vitro translated PfDZ50 and in vitro translatedPfeIF4E, as loading control while lanes 2 and 3 are 10 μl aliquots of

in vitro translated PfDZ50 and PfeIF4E respectively (Fig. 7B). Co-IPwith anti PfDZ50 IgG clearly shows the PfeIF4E band along withPfDZ50 protein (Fig. 7B, lane 4). Similarly, Co-IP with anti PfeIF4EIgG clearly shows the PfDZ50 band along with PfeIF4E protein(Fig. 7B, lane 5). We have used pre-immune IgGs as a negative controlto void the possibility of nonspecific immune precipitation. It isevident that when the co-IP was done using pre-immune IgGsfrom two different pools none of the proteins were precipitated(Fig. 7B, lanes 6 and 7). These results show that mouse pre-immuneIgGs do not nonspecifically bind with either of the proteins and

A

a

b

c

d

e

130957255

43

34

28

17

C

PfDZ50

M

PfDZ50

1 2 3 M

170

130

100

72

55

45

45

35

25

PfH45

B

1

Fig. 8. A. Localization of PfDZ50 by immunofluorescence staining. The cells were fixed and immunostained. Panel (I) phase, panel (II) image of cell stained with DAPI (for nucleusstaining), panel (III) immunofluorescently stained cell (green, PfDZ50) and panel (IV) super-imposed images. Panels a to e are late ring, early trophozoite, late trophozoite, earlyschizont and late schizont stages of the asexual development of P. falciparum 3D7 strain stained with immune IgG. B. Western blot analysis. Upper panel shows the blot probedwith anti-PfDZ50 antibodies and the lower panel shows the other half of the blot probed with anti-PfH45 antibodies. Lane M is the protein molecular weight marker and lanes1, 2 and 3 are proteins from ring, trophozoite and schizont stages respectively. C. Western blot analysis. Lane M is the protein molecular weight marker and lane 1 is proteinfrom mixed asexual stages of the intraerythrocytic development of the parasite. The arrows in B and C show the PfH45 and PfDZ50 bands.

57M. Tarique et al. / Gene 522 (2013) 46–59

Author's personal copy

firmly establish the specific interaction between PfDZ50 and PfeIF4Eproteins. In a previous study in P. berghei, PbeIF4E was identified inthe immunoprecipitates of PbDOZI (Mair et al., 2010).

3.10. Modulation of PfDZ50 ATPase activity by PfeIF4E

In previous experiments we have shown that PfDZ50 interactswith PfeIF4E. In order to investigate the modulation of ATPase activityof PfDZ50 by PfeIF4E, the ATPase activity of PfDZ50 was tested in theabsence and presence of PfeIF4E. The results clearly show that PfeIF4Einhibits the ATPase activity of PfDZ50 in a concentration-dependentmanner as compared to the activity in the absence of any PfeIF4E(Fig. 7C, lanes 3–7 compared to Fig. 7C, lane 1). It is to be noted thatPfeIF4E alone did not show any ATPase activity (Fig. 7C, lane 2).From the results it is clear that PfeIF4E modulates the ATPase activityof PfDZ50. To the best of our knowledge this is the first studyreporting the inhibition of ATPase activity of a DDX6 homologue byeIF4E.

3.11. PfDZ50 inhibits translation

We studied the effect of PfDZ50 and the truncated derivativesPfDZ50T1 and PfDZ50T2 on translation using in vitro assays. Theresults clearly show that PfDZ50 inhibits translation in a dose-dependent manner (Fig. 7D, lanes 4 and 5 compared to lane 1,which is a control reaction without PfDZ50). On the other hand thetruncated derivatives PfDZ50T1 and PfDZ50T2 had no detectableeffect on the translation (Fig. 7D, lanes 2 and 3 compared to lane 1,which is a control reaction without PfDZ50). These results suggestthat full-length PfDZ50 inhibits translation. It has been shown previ-ously that Xp54 RNA helicase represses mRNA translation in Xenopusoocytes, possibly by sequestering eIF4E from the translationalmachinery (Minshall and Standart, 2004). It was reported thatDhh1p, and RCK/p54, repress translation in vitro, and Dhh1p functionis bypassed in vivo by inhibition of translational initiation (Coller andParker, 2005). Our studies support these previous observations thatmost likely the inhibition of in vitro translation by PfDZ50 is throughits interaction with eIF4E.

3.12. Restoration of translation by PfeIF4E

The results reported in the previous section reveal that addition offull-length PfDZ50 inhibits the in vitro translation reaction. In order tostudy the effect of exogenously added PfeIF4E on translation weperformed this experiment. The results are really supporting thisobservation and it is interesting to note that the translation wasrestored in a dose-dependent manner when recombinant PfeIF4Ewas added externally (Fig. 7E, lanes 4 and 5). Lane 1 is withoutPfDZ50 as positive control and lane 6 is in the presence of PfDZ50alone where the translation is inhibited. On the other hand thetruncated derivatives PfDZ50T1 and PfDZ50T2 were unable to restorethe translation. Our results suggest that the addition of PfeIF4Erestored ~70% of the translation (Fig. 7E, lane 4 compared to lane6). All the above observations suggest that most likely PfDOZI(DDX6 homologue) regulates translation via its effects on eIF4E asreported earlier (Weston and Sommerville, 2006). In a previousstudy it was reported that the C-terminal D2 domain of XenopusXp54 alone is sufficient for translational repression and completeaccumulation in P-bodies, although it is deficient for P-body assembly(Minshall et al., 2009).

3.13. Localization of PfDZ50 by immunofluorescence assay

In order to localize endogenous PfDZ50 protein in variousintraerythrocytic developmental stages of P. falciparum 3D7 strainthe purified antibodies to PfDZ50 and the procedure described inthe Materials and methods section was used. The nucleus of theparasites was stained by DAPI (Fig. 8A, panel II) and the endogenousPfDZ50 (PfDDX6) with fluorescein isothiocynate-conjugated second-ary antibodies which produces green fluorescence (Fig. 8A, panelIII). The results clearly show that PfDZ50 is expressed in all theasexual stages (i.e. the ring, the early and late trophozoite and theearly and late schizont) of the parasite (Fig. 8A, a–e, panels I–IV).These data are in agreement with the expression profile reported in‘PlasmoDB’ (www.plasmodb.org).

Although PfDZ50 shows a diffused localization throughout theparasite co-localization with the nuclear stain DAPI clearly showsthat it is concentrated mainly in the cytoplasmic region and shows a

PABP(AAA)n

eIF4G

eIF4E

eIF4E

Depletion of DDX6 inthe cytoplasm-restorationof translation

Protein translation

PABP(AAA)n

eIF4G

eIF4E

PABP(AAA)n

eIF4G

eIF4E

External supply

DDX6 inhibits translation either by removing eIF4E frommRNA cap or by inhibiting eIF4E led ribosome binding

Fig. 9. A hypothetical model for the inhibition and restoration of translation.

58 M. Tarique et al. / Gene 522 (2013) 46–59

Author's personal copy

granular localization pattern (Fig. 8A, a–e, I–IV). In a previous study amodified P. berghei line expressing a C-terminal green fluorescentprotein (GFP) fusion of endogenous DOZI was generated to studythe cellular localization of DOZI (Mair et al., 2006). It has beenreported that a punctate GFP-fluorescence pattern restricted tothe cytoplasm of female gametocytes was observed in live andfixed cells (Mair et al., 2006). In a previous study the immunostainingof XTC cells (derived from metamorphosing tadpoles) with anti-p54showed a predominantly particulate cytoplasmic localization(Ladomery et al., 1997). It has also been reported in previous studiesthat DDX6 like proteins are mainly concentrated in distinct cytoplasmicfoci commonly known as P granules or P bodies (Weston andSommerville, 2006). It has been reported in previous studies thatDDX6-like proteins can apparently interact with a range of differentprotein partners in a variety of cellular situations and is therefore amultifunctional protein and is thus required in the nucleus as well asthe cytoplasm (Weston and Sommerville, 2006). The western blotresults with the protein extracts from the ring, trophozoite, schizontand mixed stages of intraerythrocytic development of P. falciparum3D7 strain shows that PfDZ50 is expressed in all the asexualintraerythrocytic developmental stages of the parasite (Fig. 8B, upperpanel, lanes 1, 2 and 3 and Fig. 8C, lane 1). For control, the lower partof the same blot was probed with antibodies against PfH45 (Fig. 8B,lower panel, lanes 1, 2 and 3), which as reported previously, is alsoexpressed in all the asexual intraerythrocytic developmental stages ofthe P. falciparum 3D7 strain (Pradhan and Tuteja, 2007).

To the best of our understanding, this is the first report of cloningand characterization of a biochemically and functionally active DDX6homolog PfDZ50 from the malaria parasite P. falciparum 3D7 strain.The results presented in this study further suggest that PfDZ50contains nucleic acid dependent ATPase and RNA helicase activitiesand the full-length protein is required for the enzymatic activity.PfDZ50 is expressed in all the developmental stages of the parasiteand it interacts with PfeIF4E. PfDZ50T1 (domain 1) is partiallyresponsible for the interaction with PfeIF4E. The interaction studiessuggest that PfDZ50 interacts with PfeIF4E and negatively regulatesthe translation (Fig. 9) similar to the human homolog RCK/p54 andthe yeast homologue Dhh1p (Akao et al., 2006; Coller and Parker,2005). PfDZ50 most likely inhibits translation either by removingeIF4E from the mRNA cap or by inhibiting eIF4E led ribosomebinding (Fig. 9). The results presented in this study make significantdevelopment towards understanding the role of DDX6 in RNAmetab-olism of the parasite. Translational repression is an importantphenomenon. In a recent study in P. berghei, it has been reportedthat calcium-dependent protein kinase 1 (CDPK1) is expressed in alllife stages and it translationally activates mRNA species in the devel-oping zygote. Their findings suggest that CDPK1 is a multifunctionalprotein that translationally regulates mRNAs to ensure timely andstage-specific protein expression (Sebastian et al., 2012).

Acknowledgements

The authors sincerely thank Arun Pradhan with help in initial stagesof this work. The work on helicases in RT's laboratory is partially sup-ported by the Department of Biotechnology and Department of Scienceand Technology grants. Infra-structural support from theDepartment ofBiotechnology, Government of India is gratefully acknowledged.

References

Abdelhaleem, M., 2004. Do human RNA helicases have a role in cancer? Biochim.Biophys. Acta 1704, 37–46.

Akao, Y., Matsumoto, K., Ohguchi, K., Nakagawa, Y., Yoshida, H., 2006. Human DEAD-box/RNA unwindase rck/p54 contributes to maintenance of cell growth byaffecting cell cycle in cultured cells. Int. J. Oncol. 29, 41–48.

Andrei, M.A., Ingelfinger, D., Heintzmann, R., Achsel, T., Rivera-Pomar, R., Lührmann, R.,2005. A role for eIF4E and eIF4E-transporter in targeting mRNPs to mammalianprocessing bodies. RNA 11, 717–727.

Arnold, K., Bordoli, L., Kopp, J., Schwede, T., 2006. The SWISS-MODEL workspace: aweb-based environment for protein structure homology modelling. Bioinformatics22, 195–201.

Bailey, T.L., Elkan, C., 1994. Proceedings of the Second International Conference on IntelligentSystems for Molecular Biology. AAAI Press, Menlo Park, California, pp. 28–36.

Beckham, C.J., Parker, R., 2008. P bodies, stress granules, and viral life cycles. Cell HostMicrobe 3, 206–212.

Cheng, Z., Coller, J., Parker, R., Song, H., 2005. Crystal structure and functional analysisof DEAD-box protein Dhh1p. RNA 11, 1258–1270.

Coller, J., Parker, R., 2005. General translational repression by activators of mRNAdecapping. Cell 122, 875–886.

Cordin, O., Banroques, J., Tanner, N.K., Linder, P., 2006. The DEAD-box protein family ofRNA helicases. Gene 367, 17–37.

Dutta, A., Zheng, S., Jain, D., Cameron, C.E., Reese, J.C., 2011. Intermolecular interactionswithin the abundant DEAD-box protein Dhh1 regulate its activity in vivo. J. Biol.Chem. 286, 27454–27470.

Gorbalenya, A.E., Koonin, E.V., 1993. Helicases: amino acid sequence comparisons andstructure–function relationship. Curr. Opin. Struct. Biol. 3, 419–429.

Ladomery, M., Wade, E., Sommerville, J., 1997. Xp54, the Xenopus homologue ofhuman RNA helicase p54, is an integral component of stored mRNP particles inoocytes. Nucleic Acids Res. 25, 965–973.

Lambros, C., Vanderberg, J.P., 1979. Synchronization of Plasmodium falciparum erythrocyticstages in culture. J. Parasitol. 65, 418–420.

Linder, P., et al., 1988. Birth of the D–E–A–D box. Nature 337, 121–122.Mair, G.R., et al., 2006. Regulation of sexual development of Plasmodium by translational

repression. Science 313, 667–669.Mair, G.R., et al., 2010. Universal features of post-transcriptional gene regulation are

critical for Plasmodium zygote development. PLoS Pathog. 6 (2), e1000767.http://dx.doi.org/10.1371/journal.ppat.1000767.

Mehta, J., Tuteja, R., 2011. A novel dual Dbp5/DDX19 homologue from Plasmodiumfalciparum requires Q motif for activity. Mol. Biochem. Parasitol. 176, 58–63.

Minshall, N., Kress, M., Weil, D., Standart, N., 2009. Role of p54 RNA helicase activityand its C-terminal domain in translational repression. P-body localization andassembly. Mol. Biol. Cell 20, 2464–2472.

Minshall, N., Standart, N., 2004. The active form of Xp54 RNA helicase in translationalrepression is an RNA-mediated oligomer. Nucleic Acids Res. 32, 1325–1334.

Pettersen, E.F., et al., 2004. UCSF chimera — a visualization system for exploratoryresearch and analysis. J. Comput. Chem. 25, 1605–1612.

Pradhan, A., Tuteja, R., 2007. Bipolar, dual Plasmodium falciparum helicase 45 expressedin the intraerythrocytic developmental cycle is required for parasite growth. J. Mol.Biol. 373, 268–281.

Rajyaguru, P., Parker, R., 2009. CGH-1 and the control of maternal mRNAs. Trends CellBiol. 19, 24–28.

Rocak, S., Linder, P., 2004. DEAD-box proteins: the driving forces behind RNAmetabolism.Nat. Rev. Mol. Cell Biol. 5, 232–241.

Sambrook, J., Fritsch, E., Maniatis, T.I., 1989. Molecular Cloning: A LaboratoryManual.Cold Spring Harbour Laboratory Press, Cold Spring Harbour, New York.

Sebastian, S., et al., 2012. A Plasmodium calcium-dependent protein kinase controlszygote development and transmission by translationally activating repressedmRNAs. Cell Host Microbe 12, 9–19.

Shankar, J., Pradhan, A., Tuteja, R., 2008. Isolation and characterization of Plasmodiumfalciparum UAP56 homologue: evidence for the coupling of RNA binding andsplicing activity by site directed mutations. Arch. Biochem. Biophys. 478, 143–153.

Tritschler, F., Braun, J.E., Eulalio, A., Truffault, V., Izaurralde, E., Weichenrieder, O., 2009.Structural basis for the mutually exclusive anchoring of P body components EDC3and Tral to the DEAD box protein DDX6/Me31B. Mol. Cell 33, 661–668.

Tuteja, N., Rahman, K., Tuteja, R., Falaschi, A., 1993. Human DNA helicase V,a novel DNA unwinding enzyme from HeLa cells. Nucleic Acids Res. 21, 2323–2329.

Tuteja, R., 2007. Malaria: an overview. FEBS J. 274, 4670–4679.Tuteja, R., 2009. Identification and bioinformatics characterization of translation

initiation complex eIF4F components and poly(A)-binding protein from Plasmodiumfalciparum. Commun. Integr. Biol. 2, 1–16.

Tuteja, R., 2010. Genome wide identification of Plasmodium falciparum helicases: acomparison with human host. Cell Cycle 9, 104–120.

Tuteja, R., Pradhan, A., 2006. Unraveling the ‘DEAD-box’ helicases of Plasmodiumfalciparum. Gene 376, 1–12.

Tuteja, R., Pradhan, A., 2009. Isolation and functional characterization of eIF4Fcomponents and poly(A)-binding protein from Plasmodium falciparum. Parasitol.Int. 58, 481–485.

Umate, P., Tuteja, N., Tuteja, R., 2011. Genome-wide comprehensive analysis of humanhelicases. Commun. Integr. Biol. 4, 118–137.

Wassarman, D.A., Steitz, J.A., 1991. Alive with DEAD proteins. Nature 349, 463–464.Weston, A., Sommerville, J., 2006. Xp54 and related (DDX6-like) RNA helicases: roles in

messenger RNP assembly, translation regulation and RNA degradation. NucleicAcids Res. 34, 3082–3094.

59M. Tarique et al. / Gene 522 (2013) 46–59