expression of an mg -dependent hiv-1 rnase h construct for...

ANTIMICROBIAL AGENTS AND CHEMOTHERAPY, Oct. 2011, p. 4735–4741 Vol. 55, No. 100066-4804/11/$12.00 doi:10.1128/AAC.00658-11Copyright © 2011, American Society for Microbiology. All Rights Reserved.

Expression of an Mg2�-Dependent HIV-1 RNase HConstruct for Drug Screening�†

Richard V. Farias,2‡ Deborah A. Vargas,1 Andres E. Castillo,1 Beatriz Valenzuela,1§ Marie L. Cote,2¶Monica J. Roth,2 and Oscar Leon1*

Programa de Virología, Instituto de Ciencias Biomedicas, Facultad de Medicina Universidad de Chile, Independencia 1027,Santiago, Chile,1 and Department of Biochemistry, University of Medicine and Dentistry of New Jersey-Robert

Wood Johnson Medical School, 675 Hoes Lane, Piscataway, New Jersey 088542

Received 12 May 2011/Returned for modification 21 June 2011/Accepted 11 July 2011

A single polypeptide of the HIV-1 reverse transcriptase that reconstituted Mg2�-dependent RNase H activityhas been made. Using molecular modeling, the construct was designed to encode the p51 subunit joined by alinker to the thumb (T), connection (C), and RNase H (R) domains of p66. This p51-G-TCR construct waspurified from the soluble fraction of an Escherichia coli strain, MIC2067(DE3), lacking endogenous RNase HIand HII. The p51-G-TCR RNase H construct displayed Mg2�-dependent activity using a fluorescent nonspe-cific assay and showed the same cleavage pattern as HIV-1 reverse transcriptase (RT) on substrates that mimicthe tRNA removal required for second-strand transfer reactions. The mutant E706Q (E478Q in RT) waspurified under similar conditions and was not active. The RNase H of the p51-G-TCR RNase H construct andwild type HIV-1 RT had similar Kms for an RNA-DNA hybrid substrate and showed similar inhibition kineticsto two known inhibitors of the HIV-1 RT RNase H.

The HIV reverse transcriptase (RT) has been a majortarget for antiviral drug development. The enzyme containsthree activities: RNA- and DNA-dependent DNA polymer-ase activities and RNase H activity. Currently, all clinicallyapproved RT inhibitors target the polymerase function.RNase H cleaves RNA within a RNA-DNA hybrid and isessential for viral replication (7, 32, 33, 36). RNase H in-hibitors are currently being developed as antiviral agents(35). Several classes of HIV RNase H inhibitors, which actthrough distinct mechanisms, have been identified, includ-ing tropolones (5), N-hydroxyimides (16, 24), dihydroxy ben-zoyl naphthyl hydrazones (DHBNH) (17), diketo acids (4,38, 45), and dimeric lactones (11).

The HIV-1 RT is heterodimer consisting of a p66 subunitand a p51 subunit (2, 26). The RNase H (R) domain is uniqueto the C terminus of p66. The shape of the asymmetric het-erodimer has been related to a right hand with subdomainsfingers (F), palm (P), thumb (T), and connection (C), in addi-tion to RNase H. RNase H cleavages have been further clas-sified as polymerase dependent or polymerase independent,where polymerase-dependent cleavages are defined as thosethat are coupled approximately 18 nucleotides from the posi-tioning of the 3� OH in the polymerase active site (14, 46).

Two structures of RNase H molecules complexed with

RNA-DNA have been obtained. The first is with the Bacillushalodurans RNase H (31); the second is with HIV RT incomplex with a noncleavable 3� polypurine tract (PPT) hybrid(37). Through structural and biochemical studies, limited re-gions of RT have been identified to be in close contact withnucleic acid substrates. Within RT, both DNA-DNA (19) andRNA-DNA (37) substrates display a bend of approximately 40°between the polymerase active site and the p66 thumb. Withrespect to RNase H, two specific regions are of interest. Thefirst is the minor groove binding track (MGBT) within the p66thumb domain (residues 258 to 266 of helix H) (3, 27). Thesecond is the RNase H primer grip domain, which contacts theDNA primer strand (37). Biochemical studies have suggestedthe possibility of alternative binding orientations of the RNA-DNA hybrid with RT (13). Cross-linking studies on an HIV-1RNase H domain also agree with this hypothesis (15). Re-cently, single-molecule fluorescence resonance energy transferwas used to examine the interactions between RT and nucleicacid substrates in real time. These studies show that the RNA-DNA hybrid can bind in two orientations (28) and that theorientation of the RT depends on the composition of thesubstrate (RNA-DNA versus DNA-DNA) (1).

Previous attempts to express the HIV-1 RNase H as a cat-alytically active domain have been possible only in the pres-ence of Mn2�, and many of these attempts required an alter-native substrate binding domain (16, 21, 39, 40, 42, 47). Therole of metal binding in the RNase H domain of HIV has beena controversial area. Reports of one and two metal bindingsites have been described both biochemically and structurally(10, 12, 29, 34). The presence of Mn2� has been shown toinduce conformational changes within the RNase H as well asactivate known RNase H catalytic site mutations (E478Q) (8,15, 29, 34). In addition, many of the candidate drugs identifiedhave the potential to bind to metals (5, 24, 38). It has been

* Corresponding author. Mailing address: Programa de Virología,Instituto de Ciencias Biomedicas, Facultad de Medicina Universidadde Chile, Independencia 1027, Santiago, Chile. Phone: 56-29786919.Fax: 56-29786317. E-mail: [email protected].

† Supplemental material for this article may be found at http://aac.asm.org/.

‡ Present address: New Jersey Center for Biomaterials, RutgersUniversity, 145 Bevier Road. Piscataway, NJ.

§ Present address: Laboratorio de Inmunología, Facultad deQuimica y Biología, Universidad de Santiago, Santiago, Chile.

¶ Present address: William Paterson University, Wayne, NJ 07470.� Published ahead of print on 18 July 2011.

4735

on October 16, 2018 by guest

http://aac.asm.org/

Dow

nloaded from

http://aac.asm.org/

postulated that the Mn2�-dependent activity might not reflectthe catalytic activity of the native protein (9).

Mg2�-dependent RNase H activity could be reconstitutedusing specific combinations of RT subdomains, with optimalactivity obtained with p51 plus a TCR construct initiating atQ222 (40). In the study described in this report, this combina-tion of p51 plus TCR has been generated within a singlepolypeptide (p51-G-TCR). Expression and purification of thep51-G-TCR RNase H construct indicated that Mg2�-depen-dent RNase H activity was maintained. This HIV-1 RNase Hconstruct has the potential to greatly assist in mechanistic,structural, as well as drug discovery studies aimed at targetingthe RNase H function of HIV-1 RT.

MATERIALS AND METHODS

Materials. pRT4, encoding HIV-IIIB HXB2 p66, was a gift of Bradley Pres-ton, University of Washington, Seattle, WA. HIV-1 RT was purchased fromCalbiochem, and Escherichia coli RNase H was purchased from New EnglandBioLabs.

The E. coli strain MIC2067(DE3) (F� rnhA339::cat rnhB716::kam) was a giftof Shigenori Kanaya of the Osaka University Graduate School of Engineering,Department of Material and Life Sciences.

Generation of p51-linker-TCR. The region encoding p51 was PCR amplifiedusing pRT4 as a template, 5� EX primer (5�-GGAATTCCATATGCCCATTAGCCCTATTGAGACTGTACC-3�) incorporating an NdeI restriction site (un-derlined), 3� EX primer (5�-CGCGGATCCTTAGAAGGTTTCTGCTCCTACTATGGG-3�) incorporating a BamHI restriction site (underlined), and aninverted stop (TAA) codon. The 1,344-bp product was digested with NdeI andBamHI, extracted with a QIAquick gel extraction kit (Qiagen), and ligated into6HisT-pET11 vector (T indicates the thrombin cleavage site), resulting in theplasmid pHTP51.

Generation of linker region E. The linker region E was made by overlappingPCR. Codons were optimized for expression in E. coli. A 10:1 mixture of Taq andPfx (Invitrogen) was used for the PCR amplification. DNA primers for constructE included unit A (5�-GGGGTACCAGTTAGAGAAAGAACCCATAGTAGGAGCAGAAACCTTCGGCAA-3�), G1 (5�-CCAGCTGCACTTCCCAACGCATACGCGCATATTTGCCGAAGGTTTCTGCTCC-3�), G2 (5�-CGTTGGGAAGTGCAGCTGGGCATTCCGCATCCGGCGGTGGGCCTGAAAAAACCGG-3�), and unit B (5�-CGCGGATCCGCCAGCTGTCTTTTTCTGGCAGCACAATCGGCTGCACGGTCCATTTATCCGGTTTTTTCAGGCCCAC-3�).

Individual PCRs included unit A-G1 and G2-unit B, yielding AE1 and E2B,respectively. The products AE1 and E2B were denatured, annealed at 57°C, andextended for 8 cycles prior to the addition of primers L5EX (5�-GGGGTACCAGTTAGAGAAAGAACCC-3�) and L3EX (5�-CGCGGATCCGCCAGCTGTC-3�). The resulting 173-bp product for linker E was isolated from the gel(QIAquick kit), digested sequentially with KpnI and BamHI, and ligated intopHTP51 similarly digested with KpnI and BamHI, yielding pHTP51E.

Incorporation of TCR into pHTP51E. The plasmid pTCR was generated frompTCCR (43) by removal of the 450-bp KpnI fragment generated within theduplicated connection domain. The �925-bp PvuII/BamHI fragment encodingthe p66 TCR subdomains was isolated for insertion into pHTP51E. Partial PvuIIdigestion of pHTP51E was performed, and the full-length linear product was gelisolated and digested with BamHI. The �7,000-bp fragment was gel isolated andligated with the �925-bp PvuII/BamHI TCR fragment described above. Theresulting plasmid, p51ETCR, was used for expression studies. Another construct,p51-G-TCR, containing a linker 2 amino acids longer was made by site-directedmutagenesis as described by Kirsch and Joly (22) using p51-E-TCR as thetemplate and the oligonucleotides 5�-GGTACCAGTTAGAGAAAGAACCCATAGTAGGAGCAG-3� and 5�-CCGGTTTTTT CAGGCCCACCGCCGGATGCGG-3�. The resulting 120-bp product was gel isolated and used as amegaprimer in the next round of PCR using p51-E-TCR as the template. A 10:1mixture of Taq and Pfu (Stratagene) was used for amplification. After amplifi-cation, the reaction mixture was digested with 10 U of DpnI for 1 h at 37°C. DNAwas precipitated with ethanol and ammonium acetate, washed with 70% ethanol,dried at room temperature, and dissolved in distilled water to transform E. coliDH5� cells.

Linker sequencing. The 3� linker sequencing primer was 5�-GCACAATCGGCTGCACGG-3�. The 5� linker sequencing primer was 5�-CCTTCGGCAAATATGCGCG-3�. DNA sequencing was performed at the University of Medicine

and Dentistry of New Jersey-Robert Wood Johnson Medical School DNA Syn-thesis and Sequencing Core Facility.

Construction of E706Q P51-G-TCR mutant. E706 in construct G correspondsto E478 in HIV-1 RT. E706Q P51-G-TCR was made as described by Ko and Ma(25). Construct G was amplified using pET p51-G-TCR as template with twopairs of DNA primers. Pair 1 was the T7 promoter (5�-TAATACGACTCACTATAGGG) and a reverse mutagenic primer (5�-AACTCTTCTCTGAGTCTTCTGATTTGTTGTGTC). Pair 2 was forward mutagenic primer E706Q fwd (5�-AACTCTTCTCAGTTACAAGCAATTTATCTAGC) and the T7 terminator(5�-GCTAGTTATTGCTCAGCGG). Mutagenic primers contain the sequencefor an E-to-Q change and a restriction site for EarI.

The amplification reaction mixture contained 1.25 U of GoTaq Flexi DNApolymerase (Promega), 200 �M deoxynucleoside triphosphates, 1 �M primers,and 100 ng of p51-G-TCR in 50 �l of 1� reaction GoTaq Flexi buffer (Promega).The amplified DNA fragments were purified using a Wizard PCR Preps kit(Promega), digested with EarI, and purified with the same kit. The fragmentswere then ligated with T4 ligase, and the 2,513-bp ligation product was purifiedby electrophoresis on 1% agarose and then ligated to pGEM-T Easy and used totransform E. coli JM109 cells. Colonies containing plasmids with the E706Qmutation were identified by DNA sequencing. Purified plasmid was digested withNdeI and BamHI, and the released 2,378-bp product was purified by electro-phoresis on a 1% agarose gel and ligated to the 6HisT-pET11 vector, yieldingE706Q P51-G-TCR.

Molecular modeling. The molecular modeling was performed using the Oprogram (20). The surface potential was determined using the program Graph-ical Representation and Analysis of Surface Properties (GRASP) (30).

Complementation assay. Plasmids were transformed into competentMIC2067(DE3) cells and plated onto LB agar with chloramphenicol, kanamycin,and carbenicillin (30 �g/ml, 50 �g/ml, and 100 �g/ml, respectively).MIC2067(DE3) cells were grown at either the permissive (30°C) or the nonper-missive (42°C) temperature. BL21(DE3) cells were cultured overnight in LBbroth at 37°C, and BL21 cells containing the pHTP51ETCR plasmid were cul-tured overnight in LB-carbenicillin (100 �g/ml) broth at 37°C.

Protein expression and purification. Procedures for expression and purifica-tion of recombinant proteins are described in the supplemental material.

32P-labeled RNA-DNA hybrid production. Preparation and labeling of thehybrid 17-mer DNA-RNA or DNA-DNA substrate mimic were performed aspreviously described (41).

Dynamic light scattering. Dynamic light scattering measurements were per-formed at 20°C using 100 �l of sample in Uvettes disposable cuvettes (Eppen-dorf) with a Zetasizer Nano-S apparatus (Malvern). The samples were dilutedwith 20 mM HEPES, pH 7.4, 1.5 mM dithiothreitol, 400 mM KCl, 0.1% NP-40,20% glycerol. The viscosity value was 1.7500 cP, corresponding to a 20% glycerolsolution at 20°C. These analyses were performed at the Institute Pasteur ofMontevideo, Montevideo, Uruguay.

RNase H fluorescence assay. The RNase H fluorescence assay was carried outessentially as described previously (35). The reaction was carried out at 37°C ina 96-well plate in a final volume of 100 �l. The reaction mixture contained 0.25�M DNA-RNA substrate, 50 mM Tris, pH 8.0, 60 KCl mM, and 5 mM MgCl2.RNase H activity was detected by the increase of fluorescence at 525 nm due tosubstrate hydrolysis using an excitation wavelength of 490 nm. Fluorescence wasmeasured using a Fluoroskan II apparatus, and data analysis was carried outusing the program Spectrosoft (MTX Lab Systems).

RNase H inhibition assays. To determine the susceptibility of RNase H con-struct G to known HIV-1 RNase H inhibitors, we tested the tropolone derivative�-thujaplicinol over a concentration range from 6.9 � 10�10 M to 6.9 � 10�4 Mand an N-hydroxyimide, ARK-2456, over a concentration range of from 7.05 �

10�10 M to 7.05 10�4 M. Both compounds were dissolved in dimethyl sulfoxide(DMSO) and added to the RNase H reaction mixtures. The DMSO concentra-tion in the assay was 5%. The fluorescence assay was used under the conditionsdescribed above. Fifty percent inhibitory concentrations (IC50s) were determinedby nonlinear regression using GraphPad Prism software, version 5. �-Thujapli-cinol was obtained from Molecular Diversity Preservation International (Basel,Switzerland), and ARK-2456 was obtained from the Florida Center for Hetero-cyclic Compounds.

tRNA removal assay. The activity of the collected fractions was determined ina reaction mixture (10 �l) containing 50 mM Tris-HCl, 50 mM KCl, 8 mM MgCl2or MnCl2, 1 pmol of substrate, and 1 to 2 pmol of the enzyme. After 30 min at37°C, 10 �l of 2� formamide stop buffer was added and the products of thereaction were separated by electrophoresis on a 20% acrylamide–7 M urea gel.

4736 FARIAS ET AL. ANTIMICROB. AGENTS CHEMOTHER.


http://aac.asm.org/

Dow

nloaded from

http://aac.asm.org/

RESULTS

An important tool for studying the HIV RT RNase H as wellas for drug screening is an RNase H construct that maintainsthe specificity and biochemical properties of the wild-type(WT) enzyme. Reconstitution studies using reverse transcrip-tase subdomains had indicated that Mg2�-dependent RNase Hactivity could be robustly restored in the presence of two com-ponents: p51 plus a p66 C-terminal construct encoding TCR(Q222) (40). Molecular modeling was utilized to design con-structs which incorporated this combination of RT subdomainswithin a single polypeptide.

Design criteria. Figure 1 outlines the constructs that weredeveloped and that joined the p51 polypeptide with the p66TCR domains through a linker region. Deletion and biochem-ical studies of p51 had indicated that the C terminus influencesRNase H cleavage and RT properties (6). The C terminus ofp51 terminates on the back face of the protein, on the oppositeside of the polymerase and RNase H active sites. This greatlyfacilitated the modeling of the linker, as it avoided generatinga linker that would interfere with the pathway of substratebinding. The linker region between the p66 thumb and p51connection spans a direct linear distance of 62 Å. The linkerwas designed to maintain a solvent-accessible hydrophilic faceto improve solubility. Linker G consists of 33 amino acidsderived from residues 85 to 102 of HIV-1 p66, a flexible region.Codons for the linker were optimized for expression in E. coli,and the construct contains an N-terminal His6 tag, followed bythe thrombin cleavage site. Linker E is 2 amino acids shorterthan linker G.

Molecular modeling and surface potential of p51-linker-TCR constructs. Figure 2B compares the molecular model ofthe p51-E-TCR constructs with that of the WT HIV-1 RTheterodimer (Fig. 2A). In the p51-E-TCR model, the linker isflexible and the side chains of the amino acids have no inter-action with any of the residues of the p51-TCR backbone.

Of key importance was whether the RNase H constructsmaintained surface charges that mimicked the WT protein. To

analyze this, the surface potential of the p51-E-TCR (Fig. 2D)construct was compared with that of the WT HIV-1 RT het-erodimer (Fig. 2C). The surface potential of the p51-E-TCRconstruct mimics rather well the homologous region of theHIV heterodimer. Of interest is the deep, highly positive re-gion (blue) prior to the RNase H, indicating that the primergrip and minor groove binding track of the p66 thumb arelikely to be maintained. Changes in the individual constructcompared with the WT RT are due to the energy minimiza-tions performed on each molecular model. In addition, dock-ing experiments with a 23-mer–22-mer RNA-DNA hybrid

FIG. 1. Schematic diagram of single-polypeptide HIV-1 RNase H constructs. (A) The structure of the HIV-1 reverse transcriptase p66/p51heterodimer is shown on the top, where subdomains are labeled F (fingers), P (palm) T (thumb), C (connection), and R (RNase H). Theboundaries of the subdomains are marked above the sequence and are based on HXB2. Domains of p66 and p51 are color coded, where p51 FPdomains are gray, p66 FP domains are white, the p66 thumb domain is blue, the p51 thumb domain is turquoise, the p66 connection is yellow, thep51 connection is orange, and RNase H is green. The general scheme for the p51-linker-TCR construct is outlined at the bottom using the colorscheme defined for p66/p51. The construct encodes an N-terminal His6 tag, followed by a thrombin cleavage site. Key restriction sites used in theassembly of the DNA components are indicated in their relative coding positions. (B) Sequences of linkers E and G. The sequence of HXB2 fromposition Y427 is included. The linker sequences are positioned between p51 F440 and p66 P247 and are indicated in bold.

FIG. 2. Molecular modeling and surface potential of p51-linker-TCR RNase H constructs. (A) Ribbon diagram of WT HIV-1 RT(1RTD); (B) ribbon diagram of p51-E-TCR. The coloring of the do-mains follows that defined in the legend to Fig. 1; the linker region isindicated in purple. (C and D) Grasp analysis of the surface chargepotentials, where blue is a positive charge and red is a negative charge.(C) WT HIV-1 RT; (D) p51-E-TCR.

VOL. 55, 2011 Mg2�-DEPENDENT HIV-1 RNase H CONSTRUCT FOR DRUG SCREENING 4737


http://aac.asm.org/

Dow

nloaded from

http://aac.asm.org/

(1HYS) indicated that the substrate would maintain access tothe MGBT of the p66 thumb as well as the RNase H primergrip (data not shown).

Complementation studies. E. coli strain MIC2067(DE3) (F�

rnhA339::cat rnhB716::kam) lacks RNase H1 and HII and dis-plays temperature-sensitive (ts) growth at 42°C that can berelieved by the introduction of plasmids carrying an RNase Hgene. The ability of the p51-E-TCR construct (Fig. 1) to com-plement MIC2067 growth at 30°C and 42°C was tested (Fig. 3).At 30°C (Fig. 3B), growth was detected in all strains, as pre-dicted. At 42°C (Fig. 3C), the growth of BL21(DE3) cells wasnormal; however, MIC2067(DE3) was ts for growth. With in-troduction of HIV p51, lacking the RNase H domain, theMIC2067(DE3) remained ts. Considerable growth was ob-served when the p51-E-TCR construct was introduced into theMIC2067(DE3) cells. This indicates that the construct pro-vides sufficient RNase H activity to complement E. coli growthat 42°C. These results are extremely supportive of the sugges-tion that molecularly designed HIV-1 RNase H single poly-peptide constructs provide a soluble, functional RNase H.

Protein expression. Plasmid encoding p51-E-TCR was as-sembled through a generalized scheme. Briefly, the linker re-gions were assembled using overlapping PCR and inserted intothe p51 construct through the KpnI-BamHI sites (Fig. 1A).The TCR region was then inserted into linker-positive isolatesthrough the PvuII-BamHI sites, generating p51-E-TCR. p51-G-TCR was obtained by adding 2 amino acids in the linker bysite-directed mutagenesis (Fig. 1B). Plasmids were introducedinto E. coli MIC6067(DE3) and induced for expression byisopropyl-�-D-thiogalactopyranoside. Initially, purification ofthe constructs indicated that p51-G-TCR was more stable thanp51-E-TCR (see Fig. S1 in the supplemental material); there-fore, construct p51-G-TCR was used in the following studies.

Whole-cell extracts of individual colonies from the controlpET15B plasmid, p51-G-TCR, and p51-G-TCR E706Q wereanalyzed on SDS-polyacrylamide gels by Coomassie blue stain-ing and Western blotting using a His6 tag detection reagent.

Comparison of the control pET15B extract with the p51-G-TCR and p51-G-TCR E706Q extracts showed a protein prod-uct which corresponded with the predicted full-length 93-kDaprotein. This observation was confirmed by Western blotting(see Fig. S2 in the supplemental material).

Protein purification. MIC2067(DE3) cells harboring a plas-mid containing the p51-G-TCR construct were autoinduced(43) to increase production of soluble proteins. Furthermore,to make sure that the RNase H activity is due to the expressionof the p51-G-TCR protein, we replaced the amino acid residueE706 in p51-G-TCR that corresponds to E478 in HIV-1 RTwith glutamine. E478Q HIV-1 RT lacks RNase H activity inMg2� (8). p51-G-TCR and p51-G-TCR E478Q were purifiedfrom the soluble fraction and chromatographed on nickel ni-trilotriacetic acid (Ni2�-NTA) resin. The purity of the proteinswas greater than 90%, as judged by SDS-PAGE (see Fig. S3 inthe supplemental material). Analysis of the p51-G-TCR pro-tein solutions at 0.25 and 2.0 mg/ml by dynamic light scatteringindicated that more than 98% of the protein is a monomer(results not shown).

Characterization of magnesium-dependent RNase H activ-ity of p51-G-TCR. The RNase H activity of the RNase Hconstructs was determined in the presence of Mg2� by thefluorescence assay as described in Materials and Methods. Asshown in Fig. 4, the RNase H activity of p51-G-TCR wassimilar to that of HIV-1 RT, whereas E706Q was inactive, asexpected, ruling out potential contamination with E. coliRNase H. To compare the kinetic properties of the p51-G-TCR protein and HIV-1 RT, we determined the Km using the

FIG. 3. Complementation assay in E. coli. Growth of E. coli strain BL21(DE3) or MIC2067 (DE3) in the presence of plasmid pHTP51 orP51-E-TCR. (A) Schematic outlines indicating the strains and plasmids within the six quadrants; (B) growth at the permissive temperature (30°C);(C) growth at the MIC2067 nonpermissive temperature (42°C).

FIG. 4. RNase H activity of purified recombinant proteins. TheRNase H activity of HIV-1 RT, p51-G-TCR, and E706Q p51-G-TCRin 5 mM Mg2� was measured with the RNase H fluorescence assay asdescribed in Materials and Methods. The assays were performed intriplicate and repeated three times. Error bars indicates means SDs.



http://aac.asm.org/

Dow

nloaded from

http://aac.asm.org/

fluorescence assay. Under our conditions, the Kms for the RNaseH activity of p51-G-TCR and HIV-1 RT were 0.036 0.020 and0.025 0.012 �M, respectively, suggesting that the structure ofthe substrate binding site was maintained in p51-G-TCR.

tRNA removal assay. Fractions eluting from the Ni2�-NTAcolumn were assayed for Mg2�- and Mn2�-dependent RNaseH activity using a specific cleavage reaction. Figure 5B com-pares the activities of E. coli RNase H, HIV-1 RT, and p51-G-TCR on the RNA-DNA substrate shown in Fig. 5A. Thespecific cleavage patterns of the substrate that were detectedfor HIV-RT and p51-G-TCR in Mg2� were distinct from thecleavage pattern of E. coli RNase H. The cleavage patterns for theHIV-1 RT and p51-G-TCR construct correspond to the releaseof the 11-mer RNA, corresponding with the authentic cleavageidentified in vivo. Of considerable interest is the fact that therelative activity of the isolated HIV-1 RNase H (p51-G-TCR) wascomparable to that of the full-length HIV RT. Construct G main-tained Mn2�-dependent activity and resulted in processive cleav-age of the substrate into smaller products (Fig. 5C).

Effects of �-thujaplicinol and ARK-2452 on p51-G-TCRRNase H activity. �-Thujaplicinol and ARK-2452 are knowninhibitors of HIV-1 RT RNase H (16, 18). In order to assess

the correct folding of the inhibitor binding sites in p51-G-TCR,we determined the IC50 using wild-type HIV-1 RT as a control.The results are shown in Fig. 6 and summarized in Table 1.�-Thujaplicinol (Fig. 6A and B) and ARK-2452 (6C and D)inhibited both HIV-1 RT and p51-G-TCR with IC50s in thesame order of magnitude. Taken together, these results indi-cate that the essential catalytic properties of the RNase Hdomain were maintained in p51-G-TCR.

DISCUSSION

This study defines the first HIV-1-based construct to displayMg2�-dependent RNase H activity in the absence of a poly-merase domain. Multiple hypotheses as to why the minimalisolated HIV-1 RNase H constructs do not maintain RNase Hactivity in Mg2� have previously been proposed. These includelimitations in the substrate binding, the presence of disorderedloops, extensive mobility and flexibility of the domain, andmetal-induced conformations. Within isolated HIV-1 RNaseH domains, cross-linking studies have indicated that the Mn2�

plays a structural role as well as a catalytic role within RNaseH. Nuclear magnetic resonance analysis of the isolated domain

FIG. 5. RNase H activity in Mg2� or Mn2�. (A) RNase H substrate. The substrate mimics the replicative intermediate formed during tRNAremoval. The RNA sequence is indicated in bold. The asterisk marks the position of the 5� radiolabel. The arrow marks the initial RNase Hcleavage site, as defined in vivo and in vitro (39). (B and C) Time course of RNase H cleavages using the substrate defined in panel A. (B) RNaseH activity in 10 mM Mg2�. Results obtained with 2 pmol of either the E. coli RNase H, HIV-1 RT, or p51-G-TCR construct are shown. Times(t) are indicated. The position of the initial cleavage product is indicated. (C) RNase H activity of p51-G-TCR in the presence of 8 mM Mn2�.



http://aac.asm.org/

Dow

nloaded from

http://aac.asm.org/

indicated that divalent metals stabilize the structure of theisolated RNase H domain. Incorporation of the p51 thumb andthe substrate binding domain of the two connections (TCCR)also was not able to sustain Mg2�-dependent RNase H activity.Activity remained Mn2� dependent (47).

The p51-G-TCR Mg2�-dependent RNase H construct dif-fers from the prior TCCR construct by incorporating the p66thumb and MGBT as well as the p51 finger-palm domain.Complementation studies (40) indicated that restoration ofMg2� activity required the presence of subdomains of p51 andthe T subdomain of p66. Accordingly, our results with thep51-G-TCR construct confirm these observations, indicatingthat the RNase H active site is stabilized by these RT sub-domains, providing the conformation needed for catalysis withMg2�. The inclusion of the p66 thumb and MGBT assists with thepositioning and binding of substrate. Motifs within the finger-palm have been identified to be involved in RNA-DNA binding,in particular, p51 Lys 22, which is reported to interact only withthe RNA-DNA PPT duplex and not DNA-DNA. Mutants withmutations within the MGBT have shown differential utilization ofDNA and RNA templates and have been shown to function asprotein sensors of the substrate minor groove.

Nonnucleoside RT inhibitors (NNRTIs) bind to the base of thethumb and have been shown to enhance RNase H cleavage andincrease the angle between the polymerase active site-NNRTIbinding site and the RNase H active site by 10°. In the RT-PPTcrystal structure, it was noted that minor alterations in the sub-strate trajectory of 2 Å could reposition an RNA-DNA hybrid outof the RNase H catalytic core. Thus, the overall positioning of thesubstrate within the RNase H active site appears to be critical forthe maintenance of Mg2�-dependent RNase H activity.

The mechanism by which the substrate binding regulates thereverse transcriptase activity remains unclear; apparently, theactivities of RT, DNA synthesis, and RNA hydrolysis are de-termined by its binding orientation on substrates. It has beenreported that RT adopted opposite binding orientations onduplexes containing DNA or RNA primers (1, 13). The RNaseH construct that has Mg2�-dependent activity in the absence ofpolymerization will be useful to determine the substrate ori-entation during tRNA removal.

The RNase H activity within the viral RT is being developed asa target for drug development. Initial screens have utilized thefull-length RT. Due to the complexity of the enzymes, inhibitorscan affect RNase H activity directly through the RNase H activesite or indirectly through positioning of the RNA-DNA template.The use of truncated RT constructs lacking the polymerase activesite and NNRTI binding sites could assist in localizing the effectsof lead compounds in secondary screens. The crystal structures ofthe complexes of HIV-1 RT with �-thujaplicinol (18), a naphthy-ridinone derivative (44), and pyrimidol carboxylic acids (23) havebeen solved. The p51-G-TCR construct could be used for similarstructural analysis of RNase H inhibitors in the absence of apolymerase active site.

FIG. 6. Dose-response inhibition of the RNase H activity of wild-type RT and p51-G-TCR by �-thujaplicinol and ARK-2456. RNase H activitywas determined by the fluorescence assay at different concentrations of the inhibitors. (A and B) �-Thujaplicinol; (C and D) ARK-2456; (A andC) HIV RT; (B and D) p51-G-TCR. Each point indicates the mean value of four assays.

TABLE 1. IC50s for RNase H activity ofp51-G-TCR and HIV-1 RT

InhibitorIC50 (M) SDa

p51-G-TCR HIV-1 RT

�-Thujaplicinol 1.76 � 10�7 0.086 � 10�7 2.57 � 10�7 0.027 � 10�7

ARK-2456 1.86 � 10�7 0.023 � 10�7 4.70 � 10�7 0.026 � 10�7

a Data are the means standard deviations for four independent experiments.



http://aac.asm.org/

Dow

nloaded from

http://aac.asm.org/

In conclusion, the p51-G-TCR construct should be valuablefor drug screening and a useful model for basic studies onRNA-DNA recognition and polymerase-independent RNaseH cleavage.

ACKNOWLEDGMENTS

We thank Gaurav Patel for generating pTCR.This work was supported by grant Fondecyt 1080137, awarded to

O.L., and National Institutes of Health grant R01 GM070837, awardedto M.J.R. A Conicyt fellowship was awarded to D.A.V.

REFERENCES

1. Abbondanzieri, E. A., et al. 2008. Dynamic binding orientations direct activ-ity of HIV reverse transcriptase. Nature 453:184–189.

2. Arnold, E., et al. 1992. Structure of HIV-1 reverse transcriptase/DNA com-plex at 7Å resolution showing active site locations. Nature 357:85–89.

3. Bebenek, K., et al. 1995. Reduced frameshift fidelity and processivity ofHIV-1 reverse transcriptase mutants containing alanine substitutions in helixH of the thumb subdomain. J. Biol. Chem. 270:19516–19523.

4. Beilhartz, G. L., and M. Gotte. 2010. HIV-1 RNase H: structure, catalyticmechanism and inhibitors. Viruses 2:900–926.

5. Budihas, S. R., et al. 2005. Selective inhibition of HIV-1 reverse transcrip-tase-associated RNase H activity by hydroxylated tropolones. Nucleic AcidsRes. 33:1249–1256.

6. Cameron, C. E., M. Ghosh, S. F. Le Grice, and S. J. Benkovic. 1997. Muta-tions in HIV reverse transcriptase which alter RNase H activity and decreasestrand transfer efficiency are suppressed by HIV nucleocapsid protein. Proc.Natl. Acad. Sci. U. S. A. 94:6700–6705.

7. Champoux, J. J., E. Gilboa, and D. Baltimore. 1984. Mechanism of RNAprimer removal by the RNase H activity of avian myeloblastosis virus reversetranscriptase. J. Virol. 49:686–691.

8. Cirino, N. M., et al. 1995. Divalent cation modulation of RNase functions ofhuman immunodeficiency virus reverse transcriptase. Biochemistry 34:9936–9943.

9. Cowan, J. A. 1997. Metal-mediated hydrolysis of biological phosphate esters.A critical analysis of the essential metal on stoichiometry for magnesium-dependent nuclease activation. J. Biol. Inorg. Chem. 2:168–176.

10. Cowan, J. A., et al. 2000. Metal-ion stoichiometry of the HIV-1 RT RNaseH domain: evidence for two mutually exclusive sites leads to new mechanisticinsights on metal-mediated hydrolysis in nucleic acid biochemistry. J. Biol.Inorg. Chem. 5:67–74.

11. Dat, N., et al. 2007. A dimeric lactone from Ardisia japonica with inhibitoryactivity for HIV-1 and HIV-2 RNase H. J. Nat. Prod. 70:839–841.

12. Davies, J. F., III, Z. Hostomska, Z. Hostomsky, S. R. Jordan, and D. A.Matthews. 1991. Crystal structure of the RNase H domain of HIV-1 reversetranscriptase. Science 252:88–95.

13. DeStefano, J. J. 1995. The orientation of binding of human immunodefi-ciency virus reverse transcriptase on nucleic acid hybrids. Nucleic Acids Res.13:1901–1908.

14. Furfine, E. S., and J. E. Reardon. 1991. Human immunodeficiency virusreverse transcriptase RNase H: specificity of tRNA(Lys3)-primer excision.Biochemistry 30:7041–7046.

15. Guaitiao, J., R. Zuniga, M. J. Roth, and O. Leon. 2004. Lysine directed cross-linking of viral DNA-RNA:DNA hybrid substrate to the isolated RNase Hdomain of HIV-1 reverse transcriptase. Biochemistry 43:1302–1308.

16. Hang, J. Q., et al. 2004. Activity of the isolated HIV RNase H domain andspecific inhibition by N-hydroxyimides. Biochem. Biophys. Res. Commun.317:321–329.

17. Himmel, D., et al. 2006. HIV-1 reverse transcriptase structure with RNase Hinhibitor dihydroxy benzoyl naphthyl hydrazone bound at a novel site. ACSChem. Biol. 1:702–712.

18. Himmel, D. M., et al. 2009. Structure of HIV-1 reverse transcriptase with theinhibitor beta-thujaplicinol bound at the RNase H active site. Structure17:1625–1635.

19. Jacobo-Molina, A., et al. 1993. Crystal structure of human immunodeficiencyvirus type 1 reverse transcriptase complexed with double-stranded DNA at 3.0Å resolution shows bent DNA. Proc. Natl. Acad. Sci. U. S. A. 90:6320–6324.

20. Jones, T. A., J. Y. Zou, S. W. Cowan, and M. Kjeldgaard. 1991. Improvedmethods for building protein models in electron density maps and the loca-tion of errors in these models. Acta Crystallogr. A 47:110–119.

21. Keck, J. L., and S. Marqusee. 1995. Substitution of a highly basic helix/loopsequence into the RNase H domain of human immunodeficiency virus re-verse transcriptase restores its Mn2�-dependent activity. Proc. Natl. Acad.Sci. U. S. A. 92:2740–2744.

22. Kirsch, R. D., and E. Joly. 1998. An improved PCR-mutagenesis strategy for

two-site mutagenesis or sequence swapping between related genes. NucleicAcids Res. 26:1848–1850.

23. Kirschberg, T. A., et al. 2009. RNase H active site inhibitors of humanimmunodeficiency virus type 1 reverse transcriptase: design, biochemicalactivity, and structural information. J. Med. Chem. 52:5781–5784.

24. Klumpp, K., et al. 2003. Two-metal ion mechanism of RNA cleavage by HIVRNase H and mechanism-based design of selective HIV RNase H inhibitors.Nucleic Acids Res. 31:6852–6859.

25. Ko, J. K., and J. Ma. 2005. A rapid and efficient PCR-based mutagenesismethod applicable to cell physiology study. Am. J. Physiol. Cell Physiol.288:C1273–C1278.

26. Kohlstaedt, L. A., J. Wang, J. M. Friedman, P. A. Rice, and T. A. Steitz. 1992.Crystal structure at 3.5 Å resolution of HIV-1 reverse transcriptase com-plexed with an inhibitor. Science 256:1783–1790.

27. Latham, G., et al. 2000. Vertical-scanning mutagenesis of a critical trypto-phan in the “minor groove binding track” of HIV-1 reverse transcriptase.Major groove DNA adducts identify specific protein interactions in theminor groove. J. Biol. Chem. 275:15025–15033.

28. Liu, S., E. A. Abbondanzieri, J. W. Rausch, S. F. Le Grice, and X. Zhuang.2008. Slide into action: dynamic shuttling of HIV reverse transcriptase onnucleic acid substrates. Science 322:1092–1097.

29. Mueller, G. A., K. Pari, E. F. DeRose, T. W. Kirby, and R. E. London. 2004.Backbone dynamics of the RNase H domain of HIV-1 reverse transcriptase.Biochemistry 43:9332–9342.

30. Nicholls, A., K. A. Sharp, and B. Honig. 1991. Protein folding and associa-tion: insights from the interfacial and thermodynamic properties of hydro-carbons. Proteins 11:281–296.

31. Nowotny, M., S. A. Gaidamakov, R. J. Crouch, and W. Yang. 2005. Crystalstructures of RNase H bound to an RNA/DNA hybrid: substrate specificityand metal-dependent catalysis. Cell 121:1005–1016.

32. Omer, C. A., and A. J. Faras. 1982. Mechanism of release of the avianretrovirus tRNATrp primer molecule from viral DNA by RNase H duringreverse transcription. Cell 30:797–805.

33. Omer, C. A., R. Resnick, and A. J. Faras. 1984. Evidence for involvement ofan RNA primer in initiation of strong-stop plus DNA synthesis duringreverse transcription in vitro. J. Virol. 50:465–470.

34. Pari, K., G. A. Mueller, E. F. DeRose, T. W. Kirby, and R. E. London. 2003.Solution structure of the RNase H domain of the HIV-1 reverse transcrip-tase in the presence of magnesium. Biochemistry 42:639–650.

35. Parniak, M. A., K. L. Min, S. R. Budihas, S. F. Le Grice, and J. A. Beutler.2003. A fluorescence-based high-throughput screening assay for inhibitors ofhuman immunodeficiency virus-1 reverse transcriptase-associated RNase Hactivity. Anal. Biochem. 322:33–39.

36. Resnick, R., C. A. Omer, and A. J. Faras. 1984. Involvement of retrovirusreverse transcriptase-associated RNase H in the initiation of strong-stop (�)DNA synthesis and the generation of the long terminal repeat. J. Virol.51:813–821.

37. Sarafianos, S. G., et al. 2001. Crystal structure of HIV-1 reverse transcriptasein complex with a polypurine tract RNA:DNA. EMBO J. 20:1449–1461.

38. Shaw-Reid, C. A., et al. 2003. Inhibition of HIV-1 RNase H by a novel diketoacid, 4-[5-(benzoylamino)thien-2-yl]-2,4-dioxobutanoic acid. J. Biol. Chem.278:2777–2780.

39. Smith, J. S., and M. J. Roth. 1993. Purification and characterization of anactive human immunodeficiency virus type 1 RNase H domain. J. Virol.67:4037–4049.

40. Smith, J. S., K. Gritsman, and M. J. Roth. 1994. Contributions of DNApolymerase subdomains to the RNase H activity of HIV-1 reverse transcrip-tase. J. Virol. 68:5721–5729.

41. Smith, C. M., O. Leon, J. S. Smith, and M. J. Roth. 1998. Sequence require-ments for removal of tRNA by an isolated human immunodeficiency virustype 1 RNase H domain. J. Virol. 72:6805–6812.

42. Stahl, S. J., J. D. Kaufman, S. Vikic-Topic, R. J. Crouch, and P. T. Wingfield.1994. Construction of an enzymatically active RNase H domain of humanimmunodeficiency virus type 1 reverse transcriptase. Protein Eng. 7:1103–1108.

43. Studier, F. W. 2005. Protein production by auto-induction in high densityshaking cultures. Protein Expr. Purif. 41:207–234.

44. Su, H. P., et al. 2010. Structural basis for the inhibition of RNase H activityof HIV-1 reverse transcriptase by RNase H active site-directed inhibitors.J. Virol. 84:7625–7633.

45. Tramontano, A., and R. Di Santo. 2010. HIV-1 RT-associated RNase Hfunction inhibitors: recent advances in drug development. Curr. Med. Chem.17:2837–2853.

46. Wohrl, B. M., C. Tantillo, E. Arnold, and S. F. Le Grice. 1995. An expandedmodel of replicating human immunodeficiency virus reverse transcriptase.Biochemistry 34:5343–5356.

47. Zuniga, R., S. Sengupta, C. Snyder, O. Leon, and M. J. Roth. 2004. Expres-sion of the C-terminus of HIV-1 reverse transcriptase p66 and p51 subunitsas a single polypeptide with RNase H activity. Protein Eng. Des. Sel. 17:581–587.



http://aac.asm.org/

Dow

nloaded from

http://aac.asm.org/

expression of an mg -dependent hiv-1 rnase h construct for...

Documents