F

Microbes and Infection 10 (2008) 1142e1149 www.elsevier.com/locate/micinf

Original article

Identification of amino acid residues in HIV-1 Vif critical for binding and exclusion of APOBEC3G/F Tomoki Yamashita, Kazuya Kamada, Kazuki Hatcho, Akio Adachi, Masako Nomaguchi* Department of Virology, Institute of Health Biosciences, The University of Tokushima Graduate School, Tokushima 770-8503, Japan Received 20 May 2008; accepted 9 June 2008 Available online 18 June 2008

Abstract To define a region(s) in human immunodeficiency virus type 1 (HIV-1) Vif that involves binding to its target APOBEC3G (A3G), we have generated a series of site-specific proviral vif mutants. Of 30 mutants examined, 15 did not grow at all or grew more poorly than wild-type virus in non-permissive cells. Eight clones with N-terminal mutations located outside of the HCCH motif and BC-box, which are known to be directly crucial for the degradation of A3G, were chosen from these growth-defective mutants and mainly analyzed in detail for functional activity of their mutant Vif proteins. By single-cycle replication and immunoprecipitation/immunoblotting analyses, mutants designated W21A, S32A, W38A, Y40A, and H43A were demonstrated to hardly or poorly bind to and neutralize A3G. Upon transfection, these mutants produced progeny virions containing much more A3G than wild-type clone. Interestingly, while mutants designated E76A and W79A acted normally to inactivate A3G, they were found to exhibit a Vif-defective phenotype against A3F. Another unique mutant designated Y69A incompetent against both of A3G/F was also identified. Our results here have indicated that at least two distinct regions in the N-terminal half of HIV-1 Vif are critical for binding and exclusion of A3G/F. Ó 2008 Elsevier Masson SAS. All rights reserved. Keywords: HIV-1; Vif; APOBEC3G; APOBEC3F

1. Introduction Human immunodeficiency virus type 1 (HIV-1) genome encodes Vif protein that counters the retroviral inhibitory effect by several members of apolioprotein B mRNA-editing enzymecatalytic polypeptide (APOBEC) family of cytidine deaminases (for review, see [1]). APOBEC3G and APOBEC3F (A3G/F) are potent inhibitors of a wide range of retroviruses. In the absence of Vif, A3G/F are incorporated into HIV-1 virions and deaminate cytidine to uracil in the minus-strand viral DNA in the new target cells. Excessive C-to-U editing leads to the generation of hypermutated sequences that contain multiple guanosine to adenosine transitions in their plus strands and are genetically compromised. Moreover, APOBEC3 proteins may also exert

* Corresponding author. Tel.: þ81 88 633 9232; fax: þ81 88 633 7080. E-mail address: [email protected] (M. Nomaguchi). 1286-4579/$ - see front matter Ó 2008 Elsevier Masson SAS. All rights reserved. doi:10.1016/j.micinf.2008.06.003

antiviral activity independently of their cytidine deaminase activity. HIV-1 Vif antagonize A3G/F by inducing their degradation via ubiquitineproteasome pathway. Mechanisms independent of proteasomal degradation may also contribute to the anti-A3G/F effect of Vif. HIV-1 Vif contains a BC-box motif that is critical for association with Cul5-ElonginB-ElonginC and A3G ubiquitination [2,3]. The conserved BC-box is crucial for binding to ElonginC [3]. An HCCH domain, located upstream of the BCbox, is important for selective recruitment of Cul5 [4e6]. In contrast, a functional region(s) of Vif involving in binding to A3G/F and their exclusion from virions is poorly defined [7e9]. Although some amino acids (aa) in the N-terminal region of Vif were demonstrated, by mutational analyses, to be critical for interaction with A3G/F, whether A3G/F are incorporated into mutant virions is not directly analyzed, and remains to be examined. On the other hand, the N-terminal region is rationally thought to be important for the unique species-tropism of HIV-1 [10,11].

T. Yamashita et al. / Microbes and Infection 10 (2008) 1142e1149

In this study, we have newly constructed a series of sitespecific point mutants of the HIV-1 (NL clone) vif gene encoding 192 aa, and examined their biological (multi- and single-cycle replication) and biochemical (immunoprecipitation/Western blotting) characteristics. Our results here have indicated that Nterminal aa (nos. 21, 32, 38, 40 and 43) and central aa (nos. 76 and 79) in Vif are critical for association with A3G/F and for their exclusion from virions, respectively. We also have showed that aa no. 69 in Vif is important for binding and exclusion of A3G/F. 2. Materials and methods 2.1. Plasmid construction Twenty-six proviral vif-mutant clones of HIV-1 were newly constructed from wild-type (WT) pNL4-3 [12] by the QuikChange site-directed mutagenesis kit (Stratagene) as previously described [13]. Oligonucleotides used to introduce mutations into pNL4-3 are shown in Table 1. Construction and characterization of the mutants designated C114A, F115A, R132A and C133A have been previously described [14]. As a negative control clone (DVif), pNL-Nd [14] was used. A Flag-tagged human A3G-expression vector has been described previously [10]. Construction of Flag-tagged human A3F-expression vector was performed as follows. Human A3F was amplified by RTePCR using mRNA from a lymphocyte cell line H9 with forward (50 -GCT CTA GAA TGA ATC ACT TCA GAA ACA CAG-30 ) and reverse (50 -ACG CGT CGA CCT CGA GAA TCT CCT GCA GCT TGC-30 ) primers containing XbaI and SalI sites, respectively. The reactions were 95  C for 1 min, 63  C for 1 min, 72  C for 2.5 min for 10 cycles; 95  C for 1 min, 65  C for 1 min, 72  C for 2.5 min for 20 cycles. The PCR product was then cloned into pcDNA3.1-Flag, an expression vector containing the FLAG-tag sequence (Invitrogen). Amino acid sequence of human A3F from H9 cells differed from the GenBank database (accession number AAH38808) only by a single aa (Y345D). 2.2. Cell culture, transfection, and virion preparation 293T and MAGI cells were maintained in Eagle’s minimal essential medium supplemented with 10% heat-inactivated fetal bovine serum (FBS). H9 cells were cultured in RPMI1640 medium supplemented with 10% FBS. Plasmid DNA transfection into 293T cells was carried out by the calcium-phosphate co-precipitation method [12]. To make virion preparations, culture supernatants were harvested at 48 h post-transfection, filtered through 0.45-mm filters, and viral particles were concentrated by ultracentrifugation through 25% sucrose for 2 h at 80,000  g using SW41 rotor as previously described [15]. 2.3. Single- and multi-cycle replication assays To determine single-cycle replication of mutants, viral samples were prepared from 293T cells co-transfected with each proviral clone and an expression vector of A3G or A3F at 48 h post-transfection. Single-cycle replication was then

1143

determined by the MAGI assay [16]. To determine growth kinetics of the mutants, viruses were prepared from 293T cells transfected with each proviral clone, and inoculated into H9 cells. Viral growth was monitored by the reverse transcriptase (RT) assay of the culture supernatants of infected H9 cells at intervals as previously described [17]. 2.4. Western immunoblot analysis Cell and virion fractions were prepared from 293T cells transfected with proviral clones, and lysed in a lysis buffer (1% Nonidet P-40, 50 mM TriseHCl (pH 7.5), 150 mM NaCl, 1 mM EDTA and 1% protease inhibitor cocktail (Sigma)). The cell and virion lysates were then resolved on SDSePAGE, followed by electrophoretic transfer to polyvinylidene fluoride membranes. The membranes were treated with anti-FLAG (Sigma), anti-Vif (NIH AIDS Research and References Reagent Program) or anti-p24 [15] antibody, and visualized by the ECL plus Western blotting detection system (Amersham Pharmacia Biotech Inc.). 2.5. Immunoprecipitation Transfected 293T cells were lysed in the lysis buffer as above, mixed with anti-FLAG M2 agarose (Sigma), and incubated at 4  C for 3 h. The reaction mixture was then washed three times with TBS buffer (50 mM TriseHCl (pH 7.4), 150 mM NaCl) and eluted by the addition of 3 FLAG peptides (Sigma). After centrifugation, the supernatants were analyzed by immunoblotting. 3. Results 3.1. Generation and characterization of Vif mutants To determine whether each of highly conserved aa residue in HIV-1 Vif is required for viral infectivity, we generated a series of substitution mutants from WT pNL4-3 including 26 new clones (Fig. 1). Target residues were not chosen from C-terminal 19 aa because they are unnecessary for Vif function [18]. Mutant virus stocks were prepared from transfected 293T cells, and inoculated into H9 cells non-permissive for DVif virus. Out of 30 mutants in Fig. 1, 12 clones (W21A, S32A, W38A, Y40A, Y69A, W79A, H108A, C114A, F115A, R132A, C133A and H139A) did not grow at all similarly with the DVif virus. In addition, H43A, E76A and D104A mutants displayed retarded growth kinetics compared with WT virus. The other mutant viruses were found to be similarly infectious for H9 cells with WT virus. In total, aa residues which are critically required for viral replication in nonpermissive H9 cells were located at scattered sites except for the downstream region of the BC box (Fig. 1), indicating that N-terminal and central portions of Vif are important for its function. The HCCH motif and BC box in Vif have been reported to be essential for association with Cul5 and ElonginC, respectively, but not for interaction with A3G [3]. Therefore, to determine a functional domain(s) for binding to A3G, we mainly examined the property of the mutants with

1144

T. Yamashita et al. / Microbes and Infection 10 (2008) 1142e1149

Table 1 Oligonucleotides used to construct HIV-1 vif mutants in this study Clones

Primers

E2A

50 -GATCATCAGGGATTATGGCAAACAGATGGCAGGTG 50 -CACCTGCCATCTGTTTGCCATAATCCCTGATGATC 50 -GGCAGGTGATGATTGTGTTGCAAGTAGACAGGATG 50 -CATCCTGTCTACTTGCAACACAATCATCACCTGCC 50 -GACAGGATGAGGATTAACACAGCGAAAAGATTAGTAAAACACC 50 -GGTGTTTTACTAATCTTTTCGCTGTGTTAATCCTCATCCTGTC 50 -GAGGATTAACACATGGAAAAGAGCCGTAAAACACCATATG 50 -CARARGGTGTTTTACGGCTCTTTTCCATGTGTTAATCCTC 50 -GGAAAAGATTAGTAAAACACGCCATGTATATTTCAAGCAAAGC 50 -GCTTTCCTTGAAATATACATGGCGTGTTTTACTAATCTTTTCC 50 -CACCATATGTATATTGCAAGGAAAGCTAAGGACTG 50 -CAGTCCTTAGCTTTCCTTGCAATATACATATGGTG 50 -CAAGGAAAGCTAAGGACGCGTTTTATAGACATCAC 50 -GTGATGTCTATAAAACGCGTCCTTAGCTTTCCTTG 50 -GCTAAGGACTGGTTTGCTAGACATCACTATGAAAG 50 -CTTTCATAGTGATGTCTAGCAAACCAGTCCTTAGC 50 -GCTAAGGACTGGTTTTATAGACATGCCTATGAAAGTACTAATCC 50 -GGATTAGTACTTTCATAGGCATGTCTATAAAACCAGTCCTTAGC 50 -CTAATCCAAAAATAAGTGCCGAAGTACACATCCC 50 -GGGATGTGTACTTCGGCACTTATTTTTGGATTAG 50 -GTTCAGAAGTACACATCGCACTAGGGGATGCTAAATTAG 50 -CTAATTTAGCATCCCCTAGTGCGATGTGTACTTCTGAAC 50 -CTAAATTAGTAATAACAACAGCCTGGGGTCTGCATACAG 50 -CTGTATGCAGACCCCAGGCTGTTGTTATTACTAATTTAG 50 -GGGTCTGCATACAGGAGCAAGAGACTGGCATTTGG 50 -CCAAATGCCAGTCTCTTGCTCCTGTATGCAGACCC 50 -CAGGAGAAAGAGACGCGCATTTGGGTCAGGGAGTC 50 -GACTCCCTGACCCAAATGCGCGTCTCTTTCTCCTG 50 -CATTTGGGTCAGGGAGTCGCCATAGAATGGAGGAAAAAG 50 -CTTTTTCCTCCATTCTATGGCGACTCCCTGACCCAAATG 50 -GGAGGAAAAAGAGAGCTAGCACACAAGTAGACCC 50 -GGGTCTACTTGTGTGCTAGCTCTCTTTTTCCTCC 50 -GAAAAAGAGATATAGCGCACAAGTAGACCCTGACC 50 -GGTCAGGGTCTACTTGTGCGCTATATCTCTTTTTC 50 -GACCCTGACCTAGCAGCCCAACTAATTCATCTGC 50 -GCAGATGAATTAGTTGGGCTGCTAGGTCAGGGTC 50 -CCTAGCAGACCAACTAATTGCCCTGCACTATTTTGATTGTTTTTC 50 -GAAAAACAATCAAAATAGTGCAGGGCAATTAGTTGGTCTGCTAGG 50 -GACCAACTAATTCATCTGCACGCCTTTGATTGTTTTTCAGAATC 50 -GATTCTGAAAAACAATCAAAGGCGTGCAGATGAATTAGTTGGTC 50 -CTATTTTGATTGTTTTTCAGAAGCCGCTATAAGAAATACC 50 -GGTATTTCTTATAGCGGCTTCTGAAAAACAATCAAAATAG 50 -GAATATCAAGCAGGAGCTAACAAGGTAGGATCTC 50 -GAGATCCTACCTTGTTAGCTCCTGCTTGATATTC 50 -GGTAGGATCTCTACAGTACGCGGCACTAGCAGCATTAATAAAAC 50 -GTTTTATTAATGCTGCTAGTGCCGCGTACTGTAGAGATCCTACC 50 -CAAAACAGATAAAGCCAGCTTTGCCTAGTGTTAGG 50 -CCTAACACTAGGCAAAGCTGGCTTTATCTGTTTTG 50 -GATAAAGCCACCTTTGCCTGCTGTTAGGAAACTGACAGAGG 50 -CCTCTGTCAGTTTCCTAACAGCAGGCAAAGGTGGCTTTATC 50 -GTTAGGAAACTGACAGCGGACAGATGGAACAAGC 50 -GCTTGTTCCATCTGTCCGCTGTCAGTTTCCTAAC

W11L W21A L24A H28A S32A W38A Y40A H43A S53A P58A Y69A E76A W79A S86A Y94A T96A D104A H108A Y111A S118A H139A L148A P162A S165A E171A Nucleotides for mutations are underlined.

mutations in N-terminal half (W21A, S32A, W38A, Y40A, H43A, Y69A, E76A and W79A) in the following analyses. 3.2. N-terminal region (aa 21e43) of HIV-1 Vif is important for suppression of A3G Next, we examined the effect of N-terminal mutations in HIV-1 Vif on the suppression of A3G activity. As clearly

seen in the result of single-cycle replication assay in the presence of A3G (Fig. 2A), while growth-defective mutant clones (W21A, S32A, W38A, Y40A and H43A in Fig. 1) were unable to completely counter A3G, growth-competent positive controls (L24A and H28A) behaved like WT. The mobilityshift observed for the H28A Vif would probably be related to its structure. The leaky mutant H43A still retained the ability to suppress A3G in this assay. These data were consistent

1145

R132A C133A H139A

D104A H108A C114A F115A

Y69A

E76A W79A

S32A W38A Y40A H43A

W21A

T. Yamashita et al. / Microbes and Infection 10 (2008) 1142e1149

6

9

12

15

18

3

21

6

WT Vif S86A Y94A T96A D104A

3

6

9

12

15

18

9

12

18

P162A S165A

21

E171A

L148A

S118A 15

8000 7000 6000 5000 4000 3000 2000 1000 0

WT Vif P58A Y69A E76A W79A

3

Days after infection RT activity (cpm/1.67 µl)

RT activity (cpm/1.67 µl)

Days after infection 10000 9000 8000 7000 6000 5000 4000 3000 2000 1000 0

Y111A

Y94A T96A

WT Vif S32A W38A Y40A H43A S53A

10000 9000 8000 7000 6000 5000 4000 3000 2000 1000 0

21

Days after infection

WT Vif H108A Y111A S118A H139A

3

6

9

12

15

18

Days after infection

6

9

12

15

18

21

Days after infection RT activity (cpm/1.67 µl)

3

8000 7000 6000 5000 4000 3000 2000 1000 0

RT activity (cpm/1.67 µl)

WT Vif E2A W11L W21A L24A H28A

S86A

S53A P58A

W11L

RT activity (cpm/1.67 µl)

RT activity (cpm/1.67 µl)

8000 7000 6000 5000 4000 3000 2000 1000 0

L24A H28A

192

E2A

1

21

10000 9000 8000 7000 6000 5000 4000 3000 2000 1000 0

WT Vif L148A P162A S165A E171A

3

6

9

12

15

18

21

Days after infection

Fig. 1. Growth kinetics of various vif mutants in non-permissive H9 cells. H9 cells (1  106) were infected with an equivalent RT units (5  106) of cell-free viruses prepared from transfected 293T cells, and virus replication was monitored at intervals by RT production in the culture supernatants. The experiment was repeated with similar results. Locations of mutations in HIV-1 Vif (NL clone) and the mutant designations are indicated at the top. Mutants above and below the bar indicate the phenotypes of growth-defective and normal growth, respectively. Mutants in bold lines are growth-incompetent at all in H9 cells like DVif virus. Characterization of the mutants designated C114A, F115A, R132A and C133A has been reported previously [14]. Black, gray and dotted areas in the bar are the HCCH domain, BC-box and Basic domain, respectively.

with a conclusion that the impairment of multicycle replication of the N-terminal mutants (Fig. 1) is due to their inability to suppress A3G activity. We therefore analyzed the potentials of these mutants to control the level of A3G in virus-producing cells and in virions. As shown in Fig. 2B, WT Vif reduced somewhat the level of A3G in cells compared with the mutants, and no A3G was detected in WT virions. In contrast, none of the mutants examined here effectively excluded A3G from virions. Interaction of various Vif proteins with A3G was evaluated by co-immunoprecipitation analysis, as shown in Fig. 2C. A mutant control H108A and WT was associated with A3G but any of the other mutants was not. These data strongly suggest that the N-terminal region in Vif is required for its binding to A3G. 3.3. Glu76 and Trp79 of HIV-1 Vif are important for suppression of A3F but not for that of A3G We further examined the effect of central region mutations (aa 69e79) in Vif on the suppression of A3G activity. As shown in Fig. 3A, although the three mutants studied here were all growth-defective in H9 cells (Y69A, E76A and

W79A in Fig. 1), only the Y69A was unable to suppress A3G activity showing single-cycle replication similar to that of DVif clone. Thus, the expression level of A3G in virus-producing cells was monitored for the mutants (Fig. 3B). WT Vif reduced the expression of A3G relative to that by various mutants as described above. Among the mutants, only W79A was noticed to slightly reduce A3G expression. Interaction of the mutant proteins with A3G was then evaluated by the co-immunoprecipitation assay as described above. As shown in Fig. 3C, while Y69A did not bind to A3G, E76A and W79A mutants did as efficiently as WT. When the amount of A3G in virions was monitored, interestingly, only Y69A was found to be defective for exclusion of A3G among the three mutants (Fig. 3D). These results indicated that the growth-defective nature of E76A and W79A in H9 cells was not due to their defective antiA3G activity. Based on this finding, we examined the ability of E76A and W79A mutants to act against A3F. As shown in Fig. 4A, single-cycle replication of the mutants in the presence of A3F was similarly inefficient with DVif mutant. The expression level in cells of A3F in the presence of mutant Vif proteins was similarly higher than that by WT

T. Yamashita et al. / Microbes and Infection 10 (2008) 1142e1149

1146

75

A3G-Flag

50

p24

H1 08A mo ck

W3 8A Y4 0A H4 3A

Vif

WT

100

W2 1A S3 2 A

B

125

Cell Virion

H1 08A mo ck

W3 8A Y4 0A H4 3A

W2 1A S3 2 A

WT

Vif

Y40A

C

H43A

S32A

W38A

H28A

L24A

Vif

0

W21A

25

WT

Relative infectivity ( )

A

A3G-Flag

Vif

IP Flag

Vif

p24

Cell

Fig. 2. Activity of various vif mutants (W21A, S32A, W38A, Y40A and H43A) against A3G. (A) Effect of A3G on the single-cycle replication of the mutants. Viral samples were prepared from 293T cells co-transfected with each proviral clone and an A3G-expression vector, and their infectivity was determined in MAGI cells. Expression levels of Vif and Gag-p24 as monitored by immunoblotting of lysates from transfected 293T cells are also shown at the bottom. (B) Level of A3G in virus-producing cells and in progeny virions. 293T cells were transfected with the mutant clones indicated, and cell/virion lysates for immunoblotting analysis were prepared 2 days later. H108A was used as a control (see Fig. 1). (C) Ability of WT and mutant Vif proteins to interact with A3G. Lysates of 293T cells cotransfected with each proviral clone and an A3G-expression vector were prepared, and used for immunoprecipitation (IP) of A3G-Flag by anti-Flag-M2 agarose. The interaction of various Vif proteins and A3G was then analyzed by immunoblotting using anti-Flag or anti-Vif anti-serum. Cell lysates were examined by immunoblotting analysis with anti-Vif anti-serum. The experiment in this figure was repeated with similar results.

B

ck mo

9A W7

6A E7

9A

100

WT

Cell Vif

75 A3G-Flag 50 Vif 25 p24 9A

A3G-Flag

ck mo

9A W7

6A

ck mo

9A W7

E7 6A

9A Y6

WT

Vif

Vif

Virion E7

E7 6A

9A

W7

D

IP: Flag

Vif

C

Y6

WT

Vi

f

0

Y6 9A

Relative infectivity ( )

125

WT

A

the leaky mutant E76A (Fig. 1) was very defective for binding activity. As shown in Fig. 4D, we finally analyzed the encapsidation of A3F into mutant virions. As expected, while the WT clone effectively excluded A3F from virions,

Y6

(Fig. 4B). Interaction of mutants with A3F was then evaluated by the co-immunoprecipitation analysis as shown in Fig. 4C. Whereas WT Vif efficiently bound to A3F, Y69A and W79A (Fig. 4C, lane 5) clearly lost the ability. Even

A3G-Flag

p24

Fig. 3. Activity of vif mutants (Y69A, E76A and W79A) against A3G. (A) Effect of A3G on the single-cycle replication of the mutants. Viral samples were prepared from 293T cells co-transfected with each proviral clone and an A3G-expression vector, and their infectivity was determined in MAGI cells. (B) Expression level of A3G in the presence of WT or mutant Vif proteins. Samples were prepared from 293T cells transfected with each proviral clone, and analyzed by immunoblotting using antibodies indicated. (C) Ability of WT and mutant Vif proteins to interact with A3G. Lysates of 293T cells co-transfected with each proviral clone and an A3G-expression vector were prepared, and used for immunoprecipitation (IP) of A3G-Flag by anti-Flag-M2 agarose. The interaction of various Vif proteins and A3G was then analyzed by immunoblotting using anti-Flag or anti-Vif anti-serum. (D) Packaging of A3G into mutant virions. Virions prepared from 293T cells co-transfected with each proviral clone and an A3G-expression vector were examined by immunoblotting analysis using antibodies indicated. The experiment in this figure was repeated with similar results.

T. Yamashita et al. / Microbes and Infection 10 (2008) 1142e1149 120

60

mo ck

W7 9A

6A E7

Y6

80

9A

Cell

B

WT

Relative infectivity (%)

100

Vif

A

1147

A3F-Flag

40 Vif 20 p24

W7 9A

D

mo ck

W7 9A

6A E7

9A Y6

WT

Virion

mo ck

W7 9A

6A E7

9A Y6

WT

Vif

IP: Flag

Vif

C

E7 6A

Y6 9A

WT

Vif

0

A3F -Flag

Vif

p24

A3F-Flag

Fig. 4. Activity of vif mutants (Y69A, E76A and W79A) against A3F. (A) Effect of A3F on the single-cycle replication of the mutants. Viral samples were prepared from 293T cells co-transfected with each proviral clone and an A3F-expression vector, and their infectivity was determined in MAGI cells. (B) Expression level of A3F in the presence of WT or mutant Vif proteins. Samples were prepared from 293T cells transfected with each proviral clone, and analyzed by immunoblotting using antibodies indicated. (C) Ability of WT and mutant Vif proteins to interact with A3F. Lysates of 293T cells co-transfected with each proviral clone and an A3F-expression vector were prepared, and used for immunoprecipitation (IP) of A3F-Flag by anti-Flag-M2 agarose. The interaction of various Vif proteins and A3F was then analyzed by immunoblotting using anti-Flag or anti-Vif anti-serum. (D) Packaging of A3F into mutant virions. Virions prepared from 293T cells cotransfected with each proviral clone and an A3F-expression vector were examined by immunoblotting analysis using antibodies indicated. The experiment in this figure was repeated with similar results.

1

do ma in

HCCH zinc binding domain

sic

In this report, we have performed a systematic molecular genetic study on the A3G/F-binding function of HIV-1 Vif (NL clone). The data obtained have indicated that N-terminal aa (nos. 21, 32, 38, 40 and 43) and central aa (nos. 76 and 79) in Vif are critical for its binding to A3G/F and finally for their

Ba

4. Discussion

exclusion from virions, respectively. We also have demonstrated that aa no. 69 in Vif is important for binding and exclusion of both A3 proteins. Our results here have confirmed and significantly extended the previous reports [7e9]. Taken together, all the available data can be summarized as shown in Fig. 5. As is clear in the figure, several distinct regions in the N-terminal half of HIV-1 Vif are critical for its binding to anti-retroviral innate factors A3G/F. We showed that the N-terminal region (aa 21e38) is necessary for suppression of A3G activity through binding (Fig. 2), in addition to the conclusion previously published [8,9]. We also showed for

BC -b ox

none of the mutants did so. These data strongly suggested that the central region in Vif is required for its binding to A3F.

192 COOH

NH2 11-17

21

38

W11A W21A Q12A S32A V13A W38A D14A R15A M16A R17A

40-44, 48

69 76 79

Y69A Y40A R41A H42A H43A H42/43N Y44A N48A

108

139 144-153

173-184

E76A W79A A3G-binding A3F-binding

Fig. 5. Regions of HIV-1 Vif responsible for binding to A3G/F. Based on the results previously published [7e9] and presented in this study, amino acid residues critical for the binding activity of Vif are summarized. Mutations that abolish or diminish the activity are shown. The eight mutations identified in this study are underlined (bold underlines, newly identified).

1148

T. Yamashita et al. / Microbes and Infection 10 (2008) 1142e1149

the first time that the downstream region (aa 76e79) is important for suppression of A3F activity through binding (Fig. 4). Furthermore, an aa critical for suppression of both proteins was newly identified (Figs. 3 and 4). It has been reported that negatively charged residues Asp128 and Asp130 in A3G are critical for interaction with HIV-1 Vif [19]. Our new result that aa residues 21e43 in HIV-1 Vif play a functional role for interaction with A3G is quite reasonable, considering that the region is rich in positively charged aa residues and that VifeA3G interaction is dependent on electrostatic interaction. While Glu76 and Trp79 residues were not essential for A3G suppression (Fig. 3), they were critical for A3F suppression (Fig. 4). Of note, whereas E76A and W79A almost normally exclude A3G from virions, they degraded A3G more poorly than WT in producer cells and efficiently interacted with it (Fig. 3B and C). Inhibition of A3G incorporation into virions by HIV-1 Vif may not be a direct outcome of the Vif-mediated degradation of A3G in producer cells. It can be also claimed here that the virion-associated A3 protein is a final and fatal factor for modulation of viral infectivity in target cells. We previously reported that the region containing Tyr69 is important for formation of b-strand structure and for stable expression of Vif [13]. Similarly, a mutation of Ile9 required for b-strand structure abolished the ability of Vif to interact with A3G [20,21]. Thus, it is likely that Tyr69 is important for a proper folding of Vif molecule, and accordingly, the mutant Y69A is inactive against both of A3G and A3F. It has been reported that some mutants of HIV-1 Vif still retain selective neutralizing activity against A3F but not A3G in a single-cycle infection [7,22]. Moreover, Asp128 of A3G is critical for recognition by HIV-1 Vif but a mutation of Glu128 in A3F does not affect the recognition [1,23]. Our data also showed that HIV-1 Vif recognizes A3G and A3F by different regions. A3G generates GG-to-AA mutations, whereas A3F causes GA-to-AA mutations [1]. GA-to-AA mutations are frequently observed in viral sequences recovered from HIV-1-infected individuals [24]. Therefore, A3F may be important for restriction of HIV-1 infection in vivo. In fact, our data showed that mutations of Glu76 and Trp79, which are essential for suppression of A3F activity by Vif, abolished viral multi-cycle replication in H9 cells (Fig. 1). In this study, we demonstrated that new regions in HIV-1 Vif are critical for interaction with A3G/F. It is well anticipated that the VifeA3G/F interaction represents a novel and important target of chemical therapy for suppression of HIV1 replication. Understanding more fully the region in Vif required for A3G/F-binding would provide useful information on the design of antiviral drugs. Acknowledgments We thank Ms. Kazuko Yoshida for editorial assistance. We also thank the members of our department for their helpful discussion. This work was supported in part by a Grant-in-Aid for Scientific Research on Priority Areas (19041051) from the Ministry of Education, Culture, Sports, Science and

Technology of Japan (to A.A.), and by a Health Sciences Research Grant (Research on HIV/AIDS (2007e2009)) from the Ministry of Health, Labour and Welfare of Japan (to A.A.). References [1] R.K. Holmes, M.H. Malim, K.N. Bishop, APOBEC-mediated viral restriction: not simply editing? Trends Biochem. Sci. 32 (2007) 118e128. [2] A. Mehle, J. Goncalves, M. Santa-Marta, M. McPike, D. Gabuzda, Phosphorylation of a novel SOCS-box regulates assembly of the HIV-1 VifCul5 complex that promotes APOBEC3G degradation, Genes Dev. 18 (2004) 2861e2866. [3] Y. Yu, Z. Xiao, E.S. Ehrlich, X. Yu, X.F. Yu, Selective assembly of HIV-1 Vif-Cul5-ElonginB-ElonginC E3 ubiquitin ligase complex through a novel SOCS box and upstream cysteines, Genes Dev. 18 (2004) 2867e2872. [4] K. Luo, Z. Xiao, E. Ehrlich, Y. Yu, B. Liu, S. Zheng, X.F. Yu, Primate lentiviral virion infectivity factors are substrate receptors that assemble with cullin 5-E3 ligase through a HCCH motif to suppress APOBEC3G, Proc. Natl. Acad. Sci. USA 102 (2005) 11444e11449. [5] A. Mehle, E.R. Thomas, K.S. Rajendran, D. Gabuzda, A zinc-binding region in Vif binds Cul5 and determines cullin selection, J. Biol. Chem. 281 (2006) 17259e17265. [6] Z. Xiao, E. Ehrlich, Y. Yu, K. Luo, T. Wang, C. Tian, X.F. Yu, Assembly of HIV-1 Vif-Cul5 E3 ubiquitin ligase through a novel zinc-binding domainstabilized hydrophobic interface in Vif, Virology 349 (2006) 290e299. [7] C. Tian, X. Yu, W. Zhang, T. Wang, R. Xu, X.F. Yu, Differential requirement for conserved tryptophans in human immunodeficiency virus type 1 Vif for the selective suppression of APOBEC3G and APOBEC3F, J. Virol. 80 (2006) 3112e3115. [8] R.A. Russel, V.K. Pathak, Identification of two distinct human immunodeficiency virus type 1 Vif determinants critical for interactions with human APOBEC3G and APOBEC3F, J. Virol. 81 (2007) 8201e8210. [9] A. Mehle, H. Wilson, C. Zhang, A.J. Brazier, M. McPike, E. Pery, D. Gabuzda, Identification of an APOBEC3G binding site in human immunodeficiency virus type 1 Vif and inhibitors of Vif-APOBEC3G binding, J. Virol. 81 (2007) 13235e13241. [10] K. Kamada, T. Igarashi, M.A. Martin, B. Khamsri, K. Hatcho, T. Yamashita, M. Fujita, T. Uchiyama, A. Adachi, 2006. Generation of HIV-1 derivatives that productively infect macaque monkey lymphoid cells, Proc. Natl. Acad. Sci. USA 103 (2006) 16959e16964. [11] T. Hatziioannou, M. Princiotta, M. Piatak Jr., F. Yuan, F. Zhang, J.D. Lifson, P.D. Bieniasz, Science 314 (2006) 95. [12] A. Adachi, H.E. Gendelman, S. Koenig, T. Folks, R. Willey, A. Rabson, M.A. Martin, Production of acquired immunodeficiency syndrome-associated retrovirus in human and nonhuman cells transfected with an infectious molecular clone, J. Virol. 59 (1986) 284e291. [13] M. Fujita, H. Akari, A. Sakurai, A. Yoshida, T. Chiba, K. Tanaka, K. Strebel, A. Adachi, Expression of HIV-1 accessory protein Vif is controlled uniquely to be low and optimal by proteasome degradation, Microbes Infect. 6 (2004) 791e798. [14] M. Fujita, A. Sakurai, A. Yoshida, S. Matsumoto, M. Miyaura, A. Adachi, Subtle mutations in the cysteine region of HIV-1 Vif drastically alter the viral replication phenotype, Microbes Infect. 4 (2002) 621e624. [15] H. Akari, T. Fukumori, A. Adachi, Cell-dependent requirement of human immunodeficiency virus type 1 gp41 cytoplasmic tail for Env incorporation into virions, J. Virol. 74 (2000) 4891e4893. [16] J. Kimpton, M. Emerman, Detection of replication-competent and pseudotyped human immunodeficiency virus with a sensitive cell line on the basis of activation of an integrated beta-galactosidase gene, J. Virol. 66 (1992) 2232e2239. [17] R.L. Willey, D.H. Smith, L.A. Lasky, T.S. Theodore, P.L. Earl, B. Moss, D.J. Capon, M.A. Martin, In vitro mutagenesis identifies a region within the envelope gene of the human immunodeficiency virus that is critical for infectivity, J. Virol. 62 (1988) 139e147. [18] A. Sakurai, A. Jere, A. Yoshida, T. Yamada, A. Iwamoto, A. Adachi, M. Fujita, Functional analysis of HIV-1 vif genes derived from Japanese

T. Yamashita et al. / Microbes and Infection 10 (2008) 1142e1149 long-term nonprogressors and progressors for AIDS, Microbes Infect. 6 (2004) 799e805. [19] H. Huthoff, M.H. Malim, Identification of amino acid residues in APOBEC3G required for regulation by HIV-1 Vif and virion encapsidation, J. Virol. 81 (2007) 3807e3815. [20] M.J. Wichroski, K. Ichiyama, T.M. Rana, Analysis of HIV-1 viral infectivity factor-mediated proteasome-dependent depletion of APOBEC3G: correlating function and subcellular localization, J. Biol. Chem. 280 (2005) 8387e8396. [21] C. Pace, J. Keller, D. Nolan, I. James, S. Gaudieri, C. Moore, S. Mallal, Population level analysis of human immunodeficiency virus type 1

1149

hypermutation and its relationship with APOBEC3G and vif genetic variation, J. Virol. 80 (2006) 9259e9269. [22] V. Simon, V. Zennou, D. Murray, Y. Huang, D.D. Ho, P.D. Bieniasz, Natural variation in Vif: differential impact on APOBEC3G/3F and a potential role in HIV-1 diversification, PLoS Pathog. 1 (2005) e6. [23] B. Liu, P.T. Sarkis, K. Luo, Y. Yu, X.F. Yu, Regulation of Apobec3F and human immunodeficiency virus type 1 Vif by Vif-Cul5-ElonB/C E3 ubiquitin ligase, J. Virol. 79 (2005) 9579e9587. [24] M.T. Liddament, W.L. Brown, A.J. Schumacher, R.S. Harris, APOBEC3F properties and hypermutation preferences indicate activity against HIV-1 in vivo, Curr. Biol. 14 (2004) 1385e1391.