Virology 435 (2013) 463–471

Contents lists available at SciVerse ScienceDirect

Virology

0042-68

http://d

n Corr

Price Ce

E-m

journal homepage: www.elsevier.com/locate/yviro

Contribution of SUMO-interacting motifs and SUMOylation to theantiretroviral properties of TRIM5a

Alberto Brandariz-Nunez a, Amanda Roa a, Jose Carlos Valle-Casuso a, Nikolaos Biris b,Dmitri Ivanov b, Felipe Diaz-Griffero a,n

a Department of Microbiology and Immunology, Albert Einstein College of Medicine Bronx, NY 10461, USAb Department of Biochemistry and Cancer Therapy and Research Center, University of Texas Health Science Center at San Antonio, TX 78229, USA

a r t i c l e i n f o

Article history:

Received 9 August 2012

Returned to author for revisions

29 August 2012

Accepted 28 September 2012Available online 16 October 2012

Keywords:

TRIM5arh

HIV-1

SUMOylation

Restriction

SUMO-interacting motif

22/$ - see front matter & 2012 Elsevier Inc. A

x.doi.org/10.1016/j.virol.2012.09.042

espondence to: Albert Einstein College of M

nter 501, New York, NY 10461, USA. Fax: þ1

ail address: Felipe.Diaz-Griffero@einstein.yu.e

a b s t r a c t

Recent findings suggested that the SUMO-interacting motifs (SIMs) present in the human TRIM5a(TRIM5ahu) protein play an important role in the ability of TRIM5ahu to restrict N-MLV. Here we

explored the role of SIMs in the ability of rhesus TRIM5a (TRIM5arh) to restrict HIV-1, and found that

TRIM5arh SIM mutants IL376KK (SIM1mut) and VI405KK (SIM2mut) completely lost their ability to

block HIV-1 infection. Interestingly, these mutants also lost the recently described property of

TRIM5arh to shuttle into the nucleus. Analysis of these variants revealed that they are unable to

interact with the HIV-1 core, which might explain the reason that these variants are not active against

HIV-1. Furthermore, NMR titration experiments to assay the binding between the PRYSPRY domain of

TRIM5arh and the small ubiquitin-like modifier 1(SUMO-1) revealed no interaction. In addition, we

examined the role of SUMOylation in restriction, and find out that inhibition of SUMOylation by the

adenoviral protein Gam1 did not alter the retroviral restriction ability of TRIM5a. Overall, our results do

not support a role for SIMs or SUMOylation in the antiviral properties of TRIM5a.

& 2012 Elsevier Inc. All rights reserved.

Introduction

Several newly discovered proteins endogenously expressed inprimates show the ability to dominantly block retroviral infectionand cross-species transmission by interfering with the early phase ofviral replication (Kirmaier et al., 2010; Sayah et al., 2004; Stremlauet al., 2004). Of particular interest are members of the tripartitemotif (TRIM) family of proteins (Reymond et al., 2001). The splicingvariant alpha of TRIM5 from rhesus macaque (TRIM5arh) is a�53 kDa cytosolic protein that potently restricts HIV-1(Keckesovaet al., 2004; Song et al., 2005a). TRIM5arh blocks HIV-1 and certainother retroviruses soon after viral entry, but prior to reversetranscription (Keckesova et al., 2004; Stremlau et al., 2004). Theretroviral capsid protein (CA) is the viral determinant for suscept-ibility for restriction by TRIM5a (Owens et al., 2003). Studies on thefate of the HIV-1 capsid in the cytosol of infected cells havecorrelated restriction with a decrease in the amount of cytosolicparticulate capsid (Diaz-Griffero et al., 2007a, 2008; Perron et al.,2007; Roa et al., 2012; Stremlau et al., 2006), suggesting thatTRIM5arh acts by inducing premature uncoating in target cells.

ll rights reserved.

edicine, 1301 Morris Park,

718 632 4338.

du (F. Diaz-Griffero).

TRIM5arh is composed of four distinct domains: RING, B-box 2,coiled-coil and PRYSPRY (Reymond et al., 2001). The RING domainof TRIM5arh is an E3 ubiquitin ligase (Diaz-Griffero et al., 2006a;Kar et al., 2008; Kim et al., 2011; Langelier et al., 2008; Lienlafet al., 2011; Maegawa et al., 2010; Yamauchi et al., 2008). The E3-ligase activity of TRIM5arh correlates with the ability of TRIM5arh

to block HIV-1 (Lienlaf et al., 2011; Roa et al., 2012). The B-box2 domain of TRIM5arh and other TRIM proteins, such as TRIM63,self-associates into dimeric complexes that are important forTRIM5a higher-order self-association and capsid binding avidity;these B-box 2 domain functions are essential for full and potentrestriction of HIV-1 (Diaz-Griffero et al., 2007b, 2009; Ganser-Pornillos et al., 2011Javanbakht et al., 2005; Mrosek et al., 2008;Perez-Caballero et al., 2005). The coiled-coil domain enablesTRIM5arh dimerization (Kar et al., 2008; Langelier et al., 2008),which is critical for interaction of the PRYSPRY domain with theHIV-1 capsid (Diaz-Griffero et al., 2006b; Sebastian and Luban,2005; Stremlau et al., 2006). The PRYSPRY domain is the capsidrecognition module that dictates the specificity of restriction(Biris et al., 2012; Nakayama et al., 2005; Sawyer et al., 2005;Song et al., 2005b; Stremlau et al., 2005; Yap et al., 2005).

Recent discoveries have demonstrated that TRIM5arh has theability to shuttle into the nucleus and then exported back into thecytoplasm by a CRM-1 dependent mechanism (Diaz-Grifferoet al., 2011). This work demonstrated that the use of the CRM-1

Fig. 1. Contribution of TRIM5arh SUMO-conjugation and SUMO-interactingmotifs(SIM) to restriction of HIV-1. Dog Cf2Th cells were transduced with the

LPCX vector expressing HA-tagged wild-type and mutant TRIM5arh proteins. The

A. Brandariz-Nunez et al. / Virology 435 (2013) 463–471464

inhibitor, Leptomycin B (LMB) (Fornerod et al., 1997; Kudo et al.,1998; Kuersten et al., 2001; Mohr et al., 2009; Watanabe et al.,1999; Wolff et al., 1997), induced the accumulation of TRIM5arh

in the nuclear compartment; however, the ability of TRIM5arh

to restrict HIV-1 infection was not altered by the use of LMB(Diaz-Griffero et al., 2011). Because the use of LMB leads toaccumulation of TRIM5arh in the nucleus while de novo synthesisof TRIM5arh occurs in the cytoplasm, it is difficult to knowwhether the shuttling of TRIM5arh through the nucleus is impor-tant for HIV-1 restriction. Therefore, the contribution of thenuclear shuttling properties of TRIM5arh to HIV-1 restriction isnot understood.

The addition of small ubiquitin-related modifiers (SUMO) tocellular proteins, a process known as SUMOylation (Johnson,2004), is linked to the targeting of cytosolic proteins to thenuclear compartment (Wang et al., 2012). Recent work suggestedthat SUMO-interacting motifs (SIMs) in the human TRIM5a(TRIM5ahu) protein are involved in the ability of TRIM5ahu torestrict N-tropic murine leukemia virus (N-MLV) (Arriagada et al.,2011). This work showed that mutation of the SIMs of TRIM5ahu

inactivates its restriction capability (Arriagada et al., 2011). Thesestudies, however, did not test whether TRIM5ahu bearing mutatedSIMs lost its ability to shuttle into and out of the nucleus.

To understand the role of TRIM5arh nuclear-shuttling andSUMOylation in HIV-1 restriction by TRIM5arh, we performedthe following studies: (1) contribution of SUMO-conjugatingmotifs (SCM) and SIM to the ability of TRIM5arh to restrict HIV-1 was examined by independent mutagenesis of these motifs, (2)ability of TRIM5arh SCM and SIM variants to shuttle into thenucleus by the use of LMB, (3) ability of TRIM5arh variants to bindin vitro assembled HIV-1 capsid–nucleocapsid complexes andoligomerization, (4) ability of the SUMOylation inhibitor Gam1to affect N-MLV restriction by cells that endogenously expressTRIM5ahu, and (5) ability of the purified PRYSPRY domain ofTRIM5arh to interact with SUMO-1.

TRIM5arh SUMO-conjugation motif (SCM1) and the three SUMO-interacting

motifs (SIM1, SIM2 and SIM3) variants are shown with their respective amino

acid position (A). Stable cell lines were selected with 5 mg/ml puromycin, and the

expression levels of mutant and wild-type TRIM5arh proteins were assayed by

Western blotting using anti-HA and anti-GAPDH antibodies (B). Stable cell lines

were challenged with different amounts of HIV-1-GFP (C). The percentage of GFP-

positive cells was measured 48 h later by FACS. The results of three independent

experiments were similar; the results of a single experiment are shown.

Results

Role of SUMO-interacting and SUMO-conjugation motifs in HIV-1

restriction by TRIM5arh

Previous work has reported that N-MLV restriction byTRIM5ahu requires the presence of intact SUMO-interactingmotifs in the PRYSPRY domain of TRIM5ahu (Arriagada et al.,2011). To test the role of SUMO-interacting motifs (SIMs) andSUMO-conjugation motifs (SCMs) in the ability of TRIM5arh torestrict HIV-1, we independently mutated the K10R (SCM1mut),the IL376KK (SIM1mut), the VI405KK (SIM2mut) and the IV430KK(SIM3mut) (Fig. 1A). The TRIM5arh variants were stably expressedin dog Cf2Th cells (Fig. 1B), and the ability of TRIM5arh variants torestrict HIV-1 was tested (Fig. 1C). In agreement with observa-tions studying TRIM5ahu SCMmut and SIMmut motifs, we foundthat TRIM5arh SIM1mut and SIM2mut did not restrict HIV-1suggesting the involvement of SIMs in restriction (Fig. 1C). Bycontrast, TRIM5arh SCM1mut and SIM3mut blocked HIV-1 infec-tion as potent as wild type. These results suggested that SIMs areimportant for the ability of TRIM5arh to block HIV-1 infection.

Role of SIM1 and SIM2 in the ability of TRIM5arh to shuttle into and

out of the nucleus

We have previously described the ability of TRIM5arh toshuttle into and out of the nucleus (Diaz-Griffero et al., 2011).Because of SUMO’s involvement in the nuclear import of cellularproteins (Wang et al., 2012), we tested the role of SIMs in the

ability of TRIM5arh to shuttle into and out of the nucleus. For thispurpose, we studied the localization of TRIM5arh variants byimmunofluorescence microscopy in the presence of leptomycin B(LMB), which is a specific inhibitor of nuclear export (Fornerodet al., 1997; Kudo et al., 1998; Kuersten et al., 2001; Mohr et al.,2009; Watanabe et al., 1999; Wolff et al., 1997). As shown inFig. 2, TRIM5arh SIM1mut and SIM2mut were not retained in thenuclear compartment in the presence of LMB for 5 or 12 h whencompared to wild type TRIM5arh (Supplementary Table 1). Theseexperiments suggested that TRIM5arh-SIM1mut and TRIM5arh-SIM2mut have lost their ability to traffic into the nucleus. Bycontrast, TRIM5arh SCM1mut and SIM3mut were retained in thenucleus similar to wild type TRIM5arh (Fig. 2 and SupplementaryTable 1). These experiments suggested that SIM1 and SIM2 areinvolved in the translocation of TRIM5arh into the nucleus.

Binding of TRIM5arh variants to in vitro assembled HIV-1 capsid–

nucleocapsid (CA–NC) complexes

The PRYSPRY domain of TRIM5arh mediates the interactionbetween TRIM5arh and the HIV-1 core (Ohkura et al., 2006; Songet al., 2005c; Stremlau et al., 2006). Because TRIM5arh SIM1mutand SIM2mut are located in the PRYSPRY domain, which could

Fig. 2. Intracellular distribution of TRIM5arh SIM mutant proteins in the presence of leptomycin B. Human HeLa cells expressing the indicated TRIM5arh variants were

treated with 20 ng/ml of leptomycin B (LMB) or DMSO for either 5 or 12 h. Treated cells were fixed and immunostained using anti-HA antibodies (green). Cellular nuclei

were stained by using DAPI (blue). Representative figures are shown. The results of three independent experiments were similar (Supplementary Table 1); the result of a

single experiment is shown.

Fig. 3. Binding of TRIM5arh SIM mutant proteins to in vitro assembled HIV-1capsid–nucleocapsid (CA–NC) complexes. Human 293T cells were transfected

with plasmids expressing the indicated wild-type and mutant HA-tagged

TRIM5arh proteins. Forty-eight hours after transfection, cells were lysed. The

lysates were incubated at room temperature for 1 h with in vitro assembled HIV-1

CA–NC complexes. The mixtures were applied to a 70% sucrose cushion and

centrifuged. INPUT represents the lysates analyzed by Western blotting before

being applied to the 70% cushion. The input mixtures were Western blotted using

anti-HA antibodies. Similarly, the pellets from the 70% cushion (BOUND) were

analyzed by Western blotting using anti-HA and anti-p24 antibodies. The results

of three independent experiments were similar; the result of a single experiment

is shown.

A. Brandariz-Nunez et al. / Virology 435 (2013) 463–471 465

affect the ability of TRIM5arh to bind the HIV-1 core (Stremlauet al., 2006); we examined the ability of the different TRIM5arh

variants to bind in vitro assembled HIV-1 CA–NC complexes(Fig. 3), as previously described (Lienlaf et al., 2011; Roa et al.,2012; Stremlau et al., 2006). Interestingly, TRIM5arh SIM1mutand SIM2mut lost their ability to bind the HIV-1 core whencompared to wild type TRIM5arh. These results explain theinability of TRIM5arh SIM1mut and SIM2mut variants to blockHIV-1 infection (Fig. 1C). By contrast, TRIM5arh SCM1mut and

TRIM5arh SIM3mut bound in vitro assembled HIV-1 CA–NCcomplexes similar to wild type TRIM5arh. These results suggestedthat the inability of TRIM5arh SIM1mut and SIM2mut to blockHIV-1 infection is due to their failure to bind in vitro assembledHIV-1 CA–NC complexes.

Oligomerization of TRIM5arh variants

Next we examined the ability of TRIM5arh variants to oligo-merize by cross-linking using glycolbis-succinimidylsuccinate(EGS), as previously described (Diaz-Griffero et al., 2007b; Karet al., 2008; Langelier et al., 2008). As shown in Fig. 4, cross-linking experiments suggested that all TRIM5arh variants studiedhere efficiently form oligomers. This observation is consistentwith previous results demonstrating that the coiled-coil is theonly domain required for oligomerization (Javanbakht et al.,2006).

Role of SUMOylation in the ability of TRIM5ahu to restrict N-MLV

Our results suggested that TRIM5arh SIM1mut and SIM2mutdo not restrict HIV-1 because of their inability to bind the HIV-1core; however, these results do not exclude the possibility thatSUMOylation is a necessary process for the capacity of TRIM5a torestrict retroviruses, as previously suggested (Arriagada et al.,2011). To understand the role of SUMOylation in restriction, wetook advantage of the avian adenoviral protein Gam1, whichinhibits protein SUMOylation in human cells by degradation ofSUMO E1 and E2 enzymes (Boggio et al., 2004, 2007; Colomboet al., 2002). To understand the role of SUMOylation in N-MLVrestriction by cells endogenously expressing TRIM5ahu, we chal-lenged the fibrosarcoma human HT-1080 cell line with differentamounts of N-tropic murine leukemia virus bearing the greenfluorescent protein as a reporter (N-MLV–GFP) in the presence ofGam1 (Fig. 5A and B). Expression of the SUMOylation inhibitor

Fig. 4. Oligomerization of TRIM5arh SIM mutant proteins. Lysates from 293T

cells expressing wild-type and mutant TRIM5arh proteins were cross-linked with

increasing concentrations of ethylene glycol-bis(succinimidyl succinate) (EGS),

and were subjected to Western blotting using anti-HA antibodies, as described in

Materials and methods. ‘‘m’’ and ‘‘d’’ stands for monomer and dimer, respectively.

Similar results were obtained in three independent experiments and a represen-

tative experiment is shown.

A. Brandariz-Nunez et al. / Virology 435 (2013) 463–471466

Gam1 did not released the restriction imposed by TRIM5ahu toN-MLV–-GFP infection (Fig. 5A and B). As control, we performedsimilar infections in the presence of the mutant Gam1-L258A/L265A, which is unable to inhibit the SUMOylation pathway(Boggio et al., 2004; Chiocca et al., 1997; Colombo et al., 2002).To rule out the possibility that inhibition of SUMOylation isaffecting N-MLV infection, we challenged HT-1080 cells withB-tropic MLV bearing a GFP reporter (B-MLV–GFP) in the presenceof Gam1 (Fig. 5A and B). The expression of Gam1 in HT-1080 cellsdid not affect the infection of B-MLV–GFP suggesting that Gam1did not interfere with the process of infection (Fig. 5A). Becauseinhibition of SUMOylation by expression of Gam1 destroysnuclear bodies that contain promyelocytic leukemia (PML) pro-tein (Colombo et al., 2002), we tested whether expression ofGam1 in human HT-1080 cells displayed the same effect (Fig. 5C).As expected, expression of Gam1 destroyed PML-containingnuclear bodies in HT-1080 cells when compared to Gam1-L258A/L265A (Fig. 5C and Supplementary Fig. 1). Overall theseresults showed that inhibition of SUMOylation by expression ofGam1 does not affect the ability of TRIM5ahu to restrict N-MLVsuggesting that TRIM5ahu does not require SUMOylation forrestriction of N-MLV.

Inhibition of SUMOylation by Gam1 did not affect the ability of

TRIM5arh to traffic into the nucleus

Because SUMOylation is important for nuclear import ofcellular proteins (Wang et al., 2012), we tested whether SUMOy-lation is necessary for the ability of TRIM5arh to shuttle into thenucleus (Diaz-Griffero et al., 2011). For this purpose, we examinedthe subcellular localization of transiently transfected HT-1080cells with HA-tagged TRIM5arh (TRIM5arh-HA) and Gam1 in thepresence of LMB (Fig. 6 and Supplementary Table 2). As pre-viously shown (Diaz-Griffero et al., 2011), the treatment withLMB accumulated TRIM5arh-HA proteins in the nuclear compart-ment; however, expression of Gam1 did not diminish or alter theaccumulation of TRIM5arh-HA proteins in the nuclear compart-ment. To show that TRIM5arh does not interfere with thebiological activity of Gam1, we demonstrated that Gam1destroyed PML-nuclear bodies in the presence of TRIM5arh

(Supplementary Fig. 2). These results suggested that SUMOylationis not involved in the ability of TRIM5arh to shuttle into thenucleus.

The PRYSPRY domain of TRIM5arh does not interact with SUMO-1

To test whether the protein SUMO-1 can interact with thePRYSPRY domain of TRIM5arh (residues 292–497),we performedNMR titrations of the 15N-labeled rhesus PRYSPRY domain withthe unlabeled SUMO-1 protein. NMR titrations are sensitive toeven very weak protein–protein interactions and can be used tomap the interaction surfaces of the PRYSPRY domain (Biris et al.,2012). As shown in Fig. 7A, the addition of a five-fold excess ofSUMO-1 to the PRYSPRY domain, did not produce observableshifts on the NMR signals in the 15N –HSQC spectrum. In somecases protein–protein interactions can produce attenuation ratherthan shifts of the peaks in the NMR spectra; however, Fig. 7Bshowed that no signal attenuation was observed in NMR titra-tions of the PRYSPRY domain with SUMO-1. By contrast, signalattenuation is clearly visible in the titration of PRYSPRY with theN-terminal domain of the HIV-1 capsid, a protein that interactswith the PRYSPRY domain very weakly with a Kd �400 mM (Biriset al., 2012). Thus, no interaction between the PRYSPRY domain ofTRIM5arh and SUMO-1 was detected even at SUMO-1 concentra-tion of up to 1.5 mM.

Mapping of SIM residues on the PRYSPRY domain of TRIM5arh

We have recently solved the structure of the PRYSPRY domainof TRIM5arh (Biris et al., 2012). SIM1, SIM2 and SIM3 motifs werepredicted based on the consensus amino acid sequence, but thesehydrophobic amino acid motifs have to be exposed on the proteinsurface in order to mediate protein–protein interactions. Inspec-tion of the TRIM5arh PRYSPRY structure revealed that out of thethree predicted SIM motifs, only SIM3 contains surface-exposedresidues, while SIM1 and SIM2 are buried in the hydrophobic coreof the PRYSPRY structure; therefore, SIM1 and SIM2 cannotmediate protein–protein interactions (Fig. 8).

Discussion

The recent discovery of the ability of TRIM5arh and TRIM5ahu

to shuttle into and out of the nucleus suggested the exitingpossibility that the nuclear shuttling ability of TRIM5a is playinga role in restriction (Diaz-Griffero et al., 2011). Since SUMOylationis important for the ability of some cellular proteins to shuttleinto the nucleus and recent findings suggested a role for SIM inrestriction (Arriagada et al., 2011), here we explored the role ofSCM, SIM and SUMOylation in the ability of TRIM5a to restrictretroviral infection.

In agreement with previous observations (Arriagada et al.,2011), we found that changes TRIM5arh SIM1mut and SIM2mutlost their ability to restrict HIV-1. Interestingly, both TRIM5arh

variants also lost their ability to shuttle into the nucleus. How-ever, these variants were unable to bind in vitro assembled HIV-1CA–NC complexes, which suggest that these TRIM5arh variantsare unable to restrict HIV-1 because of their impaired binding tothe HIV-1 core. The most likely explanation is that TRIM5arh

SIM1mut and SIM2mut contain a misfolded PRYSPRY domain.In agreement, analysis of SIM1 and SIM2 motifs in the structure ofthe PRYSPRY domain revealed that these motifs are buried in thehydrophobic core of the protein, and their mutation to chargedresidues is likely to disrupt the structural integrity of the PRYSPRYdomain (Fig. 8). Taken together, we would like to suggest thatTRIM5arh SIM1mut and SIM2mut are proteins that contain amisfolded PRYSPRY domain, and this is the reason that they areunable to restrict retroviral infection.

Although our results do not support a role for SUMOylation inthe ability of TRIM5arh to restrict HIV-1, this work does not

Fig. 5. Role of SUMOylation in the ability of TRIM5ahu to restrict N-MLV. Human HT-1080 cells were transienly transfected with plasmids expressing Myc-tagged Gam1

or Gam1-L258A/L265A. After 24 h, cells were challenged with increasing amounts of N-MLV-GFP or B-MLV-GFP. Viruses expressing the reproter GFP were normilized in

dog Cf2Th cells. Forthy-eight hours post-transfection, infection was determined by counting the percentage of GFP-positive cells using FACS (A). The expression of Gam1

and Gam1-L258A/L265A were analyzed by Western blotting using anti-Myc and anti-GAPDH antibodies (B). The ability of Gam1 to destroy PML-containing nuclear bodies

was assayed by transiently transfecting HT-1080 cells with plasmids expressing Gam1 or Gam1-L258A/L265A tagged with a Myc epitope. After 24 h, cells were fixed and

immunostained using anti-Myc (green) and anti-PML (red) antibodies (C). Cellular nuclei were stained by using DAPI (blue). Similar results were obtained in three

independent experiments and a representative experiment is shown. (For interpretation of the references to color in this figure legend, the reader is referred to the web

version of this article.)

A. Brandariz-Nunez et al. / Virology 435 (2013) 463–471 467

discard the possibility that TRIM5arh is SUMOylated. We haveattempted to detect TRIM5arh SUMOylation by using a taggedSUMO protein without any success, as tested by other groups(Arriagada et al., 2011). The development of antibodies againstthe endogenous TRIM5arh will allow the exploration of thisinteresting possibility.

To directly test the ability of the PRYSPRY domain of TRIM5arh

to interact with SUMO protein, we tested the ability of recombi-nant PRYSPRY domain from TRIM5arh to interact with SUMO-1 byNMR spectroscopy. However, we were unable to detect anyinteraction suggesting that TRIM5arh does not interact with

SUMO-1. An alternative possibility is that the interaction withSUMO-1 requires the full-length TRIM5arh, which is likely tocontain two PRYSPRY domains in close proximity, as previouslysuggested (Arriagada et al., 2011).

We tested the role of SUMOylation in the restriction ability ofTRIM5ahu by using the SUMOylation inhibitor Gam1 (Boggioet al., 2004, 2007; Colombo et al., 2002). For this purpose, weused the human fibrosarcoma cell line HT-1080 that expressesTRIM5ahu and is able to potently block infection of N-MLV in aTRIM5ahu-dependent manner. Even though Gam1 destroyed theformation of PML nuclear bodies (Fig. 5C), it did not affect the

Fig. 6. Inhibition of SUMOylation by Gam1 does not affect the ability of TRIM5arh to traffic into the nucleus. Human HT-1080 cells were cotransfected with plasmids

expressing Myc-tagged Gam1 and HA-tagged TRIM5arh. After 24 h, cells were treated with 20 ng/ml of leptomycin B (LMB) or DMSO for 5 h, fixed and immunostained

using antibodies anit-myc (green) and anti-HA (red) antibodies. Cellular nuclei were stained by using DAPI (blue). Similar results were obtained in three independent

experiments(Supplementary Table 2), and a representative experiment is shown. (For interpretation of the references to color in this figure legend, the reader is referred to

the web version of this article.)

Fig. 7. NMR titrations of the TRIM5arh PRYSPRY domain with SUMO-1. (A) The superposition of the 15N-HSQC spectra of the 0.3 mM 15N-labeled PRYSPRY sample in the

presence of 1.5 mM SUMO-1 (green) or in the absence of SUMO-1 added (red) is shown. Overlapping signals appear black. (B) NMR titrations of the 15N-labeled PRYSPRY

domain with SUMO-1 (blue) and the N-terminal domain of HIV-1 capsid (red). Addition of 5:1 M excess of SUMO-1 does not show any detectable attenuation of the

PRYSPRY NMR signals. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

A. Brandariz-Nunez et al. / Virology 435 (2013) 463–471468

ability of TRIM5ahu to restrict N-MLV suggesting that SUMOyla-tion is not involved the ability of TRIM5ahu to restrict N-MLV.Altogether these results suggested that the process of SUMOyla-tion is not involved in the ability of TRIM5ahu to restrict N-MLV;however, the use of Gam1 does not affect the levels of SUMO inthe cell, which could be important for restriction.

Interestingly, increasing the levels of SUMO-1 in human 293Tor TE671 cells increases the restriction imposed to N-MLV byTRIM5ahu (Arriagada et al., 2011), arguing for a role of SUMO-1 inrestriction. One possibility is that the SUMOylation of capsiddescribed for Moloney murine leukemia virus is importantfor the ability of TRIM5ahu to bind N-MLV capsid and restrictinfection (Yueh et al., 2006).

Taken altogether these findings do not support the role of SCM,SIMs or SUMOylation in the ability of TRIM5a to restrict retroviral

infection. However, the contribution of the ability of TRIM5a toshuttle into and out of the nucleus still remains unknown. Thefuture finding of TRIM5a variants that do not shuttle into and outof the nucleus and bind the HIV-1 core will be essential touncover a role for nuclear shuttling on retroviral restriction byTRIM5a.

Materials and methods

Plasmid construction

The pLPCX-TRIM5a-HA plasmid, specific primers and a Quik-Change site-directed mutagenesis kit (Stratagene) were used accord-ing to the manufacturer’s specifications to generate recombinant

Fig. 8. Putative SUMO interaction motifs (SIM) mapped onto the TRIM5arh PRYSPRY structure. (A) SIM1, SIM2 and SIM3 mapped onto the TRIM5arh PRYSPRY

structure. The SPRY domain is shown as a cartoon with the protein surface colored in light green. The segments of the protein identified as SUMO interacting motifs SIM-1,

SIM-2 and SIM-3 are colored in cyan, orange and red respectively. Only the SIM-3 segment contains surface-exposed residues, while SIM-1 and SIM-2 are buried in the

hydrophobic core of the protein. (B) Slice through the TRIM5arh PRYSPRY structure showing the volume (green mesh) occupied by residues I376 and L377 of SIM1 (cyan).

The two residues are completely buried within the hydrophobic core and the lysine residues of the SIM1 mutant (magenta) cannot be accomodated within that space. The

resulting steric clash and buried positive charge are incompatible with the stably-folded native structure. (C) The same is true for SIM2. (D) In contrast, SIM3 is exposed on

the protein surface, therefore SIM3mut is not expected to disrupt the overall PRYSPRY structure. (For interpretation of the references to color in this figure legend, the

reader is referred to the web version of this article.)

A. Brandariz-Nunez et al. / Virology 435 (2013) 463–471 469

plasmids that expressed TRIM5a-HA variants K10R, IL376KK,VI405KK and IV430KK. The plasmids expressing Gam1 andGam1-L258A/L265A proteins tagged with Myc epitope, was agenerous gift from Dr. Susanna Chiocca, and has been describedpreviously (Boggio et al., 2004).

Antibodies

Monoclonal antibodies against the HA epitope were fromSigma. Anti-GAPDH rabbit polyclonal antibodies, anti-mouse-Cy2, anti-rabbit-Cy3, anti-mouse Alexa 488, and anti-goat Alexa594 were obtained from Sigma. Anti-p24 HIV-1 capsid wasobtained from Immunodiagnostics. Anti-Myc monoclonal andanti-PML goat polyclonal antibodies were purchased from SantaCruz Biotechnology.

Protein analysis

Cellular proteins were extracted with radioimmunoprecipita-tion assay (RIPA) buffer, as previously described (Lienlaf et al.,2011). Detection of proteins by Western blotting was performedusing anti-HA (Sigma), anti-GAPDH (Sigma), anti-P24 (Sigma),anti-Myc (Santa Cruz Biotechnology) and anti-rabbit and anti-mouse antibodies conjugated to Alexa Fluor 680 (Li-Cor). Bandswere detected by scanning blots with the Li-Cor Odyssey ImagingSystem using the 700 channel.

Binding of TRIM5arh variants to in vitro assembled HIV-1 capsid–

nucleocapsid (CA–NC) complexes

The HIV-1 CA–NC protein was expressed, purified, andassembled as previously described (Lienlaf et al., 2011). HIV-1CA–NC particles were assembled in vitro by diluting the CA–NCprotein to a concentration of 0.3 mM in 50 mM Tris–HCl (pH 8.0),0.5 M NaCl and 2 mg/ml DNA oligo (TG)50. The mixture wasincubated at 4 1C overnight and centrifuged at 8600g for 5 min.These CA–NC complexes were used for the binding assay asfollows: 293T cells were transfected with plasmids expressingwild-type or mutant TRIM5arh proteins. Forty-eight hours aftertransfection, cell lysates were prepared as follows: washed cellswere resuspended in hypotonic lysis buffer (10 mM Tris [pH 7.4],1.5 mM MgCl2, 10 mM KCl, 0.5 mM DTT). The cell suspension wasfrozen and thawed and then incubated on ice for 10 min. Next,the lysate was centrifuged at full speed in a refrigerated Eppen-dorf microcentrifuge (14,000g) for 5 min. To test binding, 5 ml ofCA–NC complexes assembled in vitro were incubated with 200 mlof cell lysate at room temperature for 1 h. A fraction of thismixture was stored (input). The mixture was spun through a 70%sucrose cushion (70% sucrose, 1� PBS, and 0.5 mM DTT) at100,000g in an SW55 rotor (Beckman) for 1 h at 4 1C. Aftercentrifugation, the supernatant was carefully removed and thepellet was resuspended in 1� SDS-PAGE loading buffer (pellet).The level of TRIM5arh proteins was determined by Westernblotting using anti-HA antibodies (Sigma). The level of HIV-1CA–NC protein in the pellet was assessed by Western blottingwith an anti-p24 capsid antibody (Immunodiagnostics).

A. Brandariz-Nunez et al. / Virology 435 (2013) 463–471470

Immunofluorescence microscopy

Transfections of cell monolayers were performed using Lipo-fectamine Plus reagent (Invitrogen), according to the manufac-turer’s instructions. Transfected cells were incubated at 37 1C for24 h, unless otherwise stated. For indirect immunofluorescencemicroscopy, cell monolayers grown on coverslips were trans-fected. Monolayers were washed twice with PBS and fixed for15 min with 4% paraformaldehyde in PBS. Paraformaldehyde-fixed cells were washed twice with PBS, incubated for 4 min inpermeabilizing buffer (0.5% Triton X-100 in PBS), and thenblocked using PBS containing 2% bovine serum albumin for 1 hat room temperature. Subsequently, cells were incubated for 1 hat room temperature with primary antibodies in blocking buffer,as previously described (Diaz-Griffero et al., 2002). After threewashes with PBS, the cells were incubated for 30 min withsecondary antibodies. Cellular nuclei were labeled with 1 mg/mlof DAPI (49, 69-diamidino-2-phenylindole) in PBS. Subsequently,samples were mounted for fluorescence microscopy by using theProLong Antifade Kit (Molecular Probes, Eugene, OR). Images wereobtained with a Zeiss Observer.Z1 microscope using a 63�objective, and deconvolution was performed using the softwareAxioVision V4.8.1.0 (Carl Zeiss Imaging Solutions).

Generation of cells stably expressing TRIM5arh variants

Retroviral vectors encoding wild-type or mutant TRIM5arh

were created using the pLPCX vector (Clontech). Recombinantviruses were produced in 293T cells by cotransfecting the pLPCXplasmids with the pVPack-GP and pVPack-VSV-G packagingplasmids (Stratagene). The pVPack-VSV-G plasmid encodes thevesicular stomatitis virus G envelope glycoprotein, which allowsefficient entry into a wide range of vertebrate cells (Yee et al.,1994). Transduced Cf2Th cells were selected in 5 mg/ml ofpuromycin.

Infection with viruses expressing green fluorescent protein (GFP)

Recombinant HIV-1 expressing GFP was prepared as described(Diaz-Griffero et al., 2008). The virus was pseudotyped with theVSV-G glycoprotein. For infections, 5�104 cells seeded in 24-wellplates were incubated at 37 1C with virus for 24 h. Cells werewashed and returned to culture for 48 h unless otherwise stated,and the percentage of GFP positive cells was determined by flowcytometry (Becton Dickinson). Viral stocks were titrated by serialdilution on Cf2Th cells to determine the concentration of infec-tious viruses.

Cross-linking of TRIM5arh

The HA-tagged TRIM5arh variants were expressed transientlyin 293T cells. Cells were washed in phosphate-buffered saline(PBS) and lysed in NP40 lysis buffer (0.5% Nonidet P40 (NP40),1� protease inhibitor (complete EDTA-free, Roche Diagnostics) inPBS) for 45 min at 4 1C. Lysates were centrifuged at 14,000g for15 min at 4 1C. The cleared lysates were not stored or frozen, butrather were directly cross-linked. Approximately 100–200 ml ofcleared lysates was diluted with PBS/1 mM EDTA to a finalvolume of 400 ml. Lysates were cross-linked with varying con-centrations (up to 2 mM) of ethylene glycol-bis(succinimidylsuccinate) (EGS) for 30 min at room temperature and centrifugedbriefly in a tabletop centrifuge. The reaction mix was quenchedwith 0.1 M Tris–HCl, pH 7.5 and briefly centrifuged. The cleared,cross-linked lysates were analyzed by Western blotting usingantibodies against HA epitope (Sigma).

Recombinant proteins preparation

The pET28b expression vector encoding the SUMO1 with aC-terminal 6His tag was a generous gift from Dr. ChristopherLima. Transformed BL21 Rosetta 2 competent cells (Novagen)were grown at 37 1C until cell density reached A600–0.6. Thetemperature was then lowered to 30 1C and protein productionwas induced with 0.75 mM isopropyl-b-D-thiogalactopyranoside(IPTG). After 4 h cells were harvested by centrifugation (6000g)and lysed by sonication. The purification of the C-terminallyHis-tagged SUMO-1 was carried out by metal affinity followedby size-exclusion chromatography on a Superdex 75 (GE lifesciences). The cloning, expression and purification of PRYSPRYdomain from TRIM5arh have been described previously (Biriset al., 2012). All NMR samples were prepared in 50 mM sodiumphosphate pH 7.0, 50 mM NaCl and 10 mM dithiothreitol buffer.

NMR titrations

NMR titrations of 15N-labeled PRYSPRY domain of TRIM5arh

with unlabeled His-tagged SUMO1 and unlabeled N-terminaldomain of the HIV capsid (residues 1–147) were carried out ona Bruker Avance 700 MHz spectrometer at 298 K by recording aseries of 1H–15N TROSY-HSQC spectra containing molar ratiosfrom 1:0 up to 1:5 for SUMO1 and from 1:0 up to 1:7 for theN-terminal domain of HIV-1 capsid.

Acknowledgments

We thank Andre Rosowsky for critical reading of the manu-script. We thank Dr. Andrei Tomashevski for his help withrecombinant protein purification. We thank Susanna Chiocca forproviding reagents. This work was funded by the NIH No. R01AI087390 to F.D.-G, CPRIT Scholar Award to D.I. and a K99/R00Pathway to Independence Award to F.D.-G. from the NationalInstitutes of Health No. 4R00MH086162-02.

Appendix A. Supporting information

Supplementary data associated with this article can be found inthe online version at http://dx.doi.org/10.1016/j.virol.2012.09.042.

References

Arriagada, G., Muntean, L.N., Goff, S.P., 2011. SUMO-interacting motifs of humanTRIM5alpha are important for antiviral activity. PLoS Pathog. 7, e1002019.

Biris, N., Yang, Y., Taylor, A.B., Tomashevski, A., Guo, M., Hart, P.J., Diaz-Griffero, F.,Ivanov, D.N., 2012. Structure of the rhesus monkey TRIM5alpha PRYSPRYdomain, the HIV capsid recognition module. Proc. Nat. Acad. Sci. USA.

Boggio, R., Colombo, R., Hay, R.T., Draetta, G.F., Chiocca, S., 2004. A mechanism forinhibiting the SUMO pathway. Mol. Cell 16, 549–561.

Boggio, R., Passafaro, A., Chiocca, S., 2007. Targeting SUMO E1 to ubiquitin ligases:a viral strategy to counteract sumoylation. J. Biol. Chem. 282, 15376–15382.

Chiocca, S., Baker, A., Cotten, M., 1997. Identification of a novel antiapoptoticprotein, GAM-1, encoded by the CELO adenovirus. J. Virol. 71, 3168–3177.

Colombo, R., Boggio, R., Seiser, C., Draetta, G.F., Chiocca, S., 2002. The adenovirusprotein Gam1 interferes with sumoylation of histone deacetylase 1. EMBORep. 3, 1062–1068.

Diaz-Griffero, F., Gallo, D.E., Hope, T.J., Sodroski, J., 2011. Trafficking of some oldworld primate TRIM5alpha proteins through the nucleus. Retrovirology 8, 38.

Diaz-Griffero, F., Hoschander, S.A., Brojatsch, J., 2002. Endocytosis is a critical stepin entry of subgroup B avian leukosis viruses. J. Virol. 76, 12866–12876.

Diaz-Griffero, F., Kar, A., Lee, M., Stremlau, M., Poeschla, E., Sodroski, J., 2007a.Comparative requirements for the restriction of retrovirus infection byTRIM5alpha and TRIMCyp. Virology.

Diaz-Griffero, F., Kar, A., Perron, M., Xiang, S.H., Javanbakht, H., Li, X., Sodroski, J.,2007b. Modulation of retroviral restriction and proteasome inhibitor-resistantturnover by changes in the TRIM5alpha B-box 2 domain. J. Virol. 81,10362–10378.

A. Brandariz-Nunez et al. / Virology 435 (2013) 463–471 471

Diaz-Griffero, F., Li, X., Javanbakht, H., Song, B., Welikala, S., Stremlau, M., Sodroski, J.,2006a. Rapid turnover and polyubiquitylation of the retroviral restriction factorTRIM5. Virology 349, 300–315.

Diaz-Griffero, F., Vandegraaff, N., Li, Y., McGee-Estrada, K., Stremlau, M., Welikala, S.,Si, Z., Engelman, A., Sodroski, J., 2006b. Requirements for capsid-binding andan effector function in TRIMCyp-mediated restriction of HIV-1. Virology 351,404–419.

Diaz-Griffero, F., Perron, M., McGee-Estrada, K., Hanna, R., Maillard, P.V., Trono, D.,Sodroski, J., 2008. A human TRIM5alpha B30.2/SPRY domain mutant gains theability to restrict and prematurely uncoat B-tropic murine leukemia virus.Virology 378, 233–242.

Diaz-Griffero, F., Qin, X.R., Hayashi, F., Kigawa, T., Finzi, A., Sarnak, Z., Lienlaf, M.,Yokoyama, S., Sodroski, J., 2009. A B-box 2 surface patch important forTRIM5alpha self-association, capsid binding avidity, and retrovirus restriction.J. Virol. 83, 10737–10751.

Fornerod, M., Ohno, M., Yoshida, M., Mattaj, I.W., 1997. CRM1 is an export receptorfor leucine-rich nuclear export signals. Cell 90, 1051–1060.

Ganser-Pornillos, B.K., Chandrasekaran, V., Pornillos, O., Sodroski, J.G., Sundquist, W.I.,Yeager, M., 2011. Hexagonal assembly of a restricting TRIM5alpha protein. Proc.Nat. Acad. Sci. USA 108, 534–539.

Javanbakht, H., Diaz-Griffero, F., Stremlau, M., Si, Z., Sodroski, J., 2005. Thecontribution of RING and B-box 2 domains to retroviral restriction mediatedby monkey TRIM5alpha. J. Biol. Chem. 280, 26933–26940.

Javanbakht, H., Yuan, W., Yeung, D.F., Song, B., Diaz-Griffero, F., Li, Y., Li, X.,Stremlau, M., Sodroski, J., 2006. Characterization of TRIM5alpha trimerizationand its contribution to human immunodeficiency virus capsid binding.Virology 353, 234–246.

Johnson, E.S., 2004. Protein modification by SUMO. Annu. Rev. Biochem. 73,355–382.

Kar, A.K., Diaz-Griffero, F., Li, Y., Li, X., Sodroski, J., 2008. Biochemical andbiophysical characterization of a chimeric TRIM21-TRIM5alpha protein. J.Virol. 82, 11669–11681.

Keckesova, Z., Ylinen, L.M., Towers, G.J., 2004. The human and African greenmonkey TRIM5alpha genes encode Ref1 and Lv1 retroviral restriction factoractivities. Proc. Nat. Acad. Sci. USA 101, 10780–10785.

Kim, J., Tipper, C., Sodroski, J., 2011. Role of TRIM5{alpha} RING domain E3ubiquitin ligase activity in capsid disassembly, reverse transcription blockade,and restriction of simian immunodeficiency virus. J. Virol. 85, 8116–8132.

Kirmaier, A., Wu, F., Newman, R.M., Hall, L.R., Morgan, J.S., O’Connor, S., Marx, P.A.,Meythaler, M., Goldstein, S., Buckler-White, A., Kaur, A., Hirsch, V.M.,Johnson, W.E., 2010. TRIM5 suppresses cross-species transmission of a primateimmunodeficiency virus and selects for emergence of resistant variants in thenew species. PLoS Biol. 8.

Kudo, N., Wolff, B., Sekimoto, T., Schreiner, E.P., Yoneda, Y., Yanagida, M.,Horinouchi, S., Yoshida, M., 1998. Leptomycin B inhibition of signal-mediated nuclear export by direct binding to CRM1. Exp. Cell Res. 242,540–547.

Kuersten, S., Ohno, M., Mattaj, I.W., 2001. Nucleocytoplasmic transport: ran, betaand beyond. Trends Cell Biol. 11, 497–503.

Langelier, C.R., Sandrin, V., Eckert, D.M., Christensen, D.E., Chandrasekaran, V.,Alam, S.L., Aiken, C., Olsen, J.C., Kar, A.K., Sodroski, J.G., Sundquist, W.I., 2008.Biochemical characterization of a recombinant TRIM5alpha protein thatrestricts human immunodeficiency virus type 1 replication. J. Virol. 82,11682–11694.

Lienlaf, M., Hayashi, F., Di Nunzio, F., Tochio, N., Kigawa, T., Yokoyama, S., Diaz-Griffero, F., 2011. Contribution of E3-ubiquitin ligase activity to HIV-1 restric-tion by TRIM5alpha(rh): structure of the RING domain of TRIM5alpha. J. Virol.85, 8725–8737.

Maegawa, H., Miyamoto, T., Sakuragi, J., Shioda, T., Nakayama, E.E., 2010.Contribution of RING domain to retrovirus restriction by TRIM5alpha dependson combination of host and virus. Virology 399, 212–220.

Mohr, D., Frey, S., Fischer, T., Guttler, T., Gorlich, D., 2009. Characterisation of thepassive permeability barrier of nuclear pore complexes. EMBO J. 28,2541–2553.

Mrosek, M., Meier, S., Ucurum-Fotiadis, Z., von Castelmur, E., Hedbom, E., Lustig, A.,Grzesiek, S., Labeit, D., Labeit, S., Mayans, O., 2008. Structural analysis of B-Box2 from MuRF1: identification of a novel self-association pattern in a RING-likefold. Biochemistry 47, 10722–10730.

Nakayama, E.E., Miyoshi, H., Nagai, Y., Shioda, T., 2005. A specific region of 37amino acid residues in the SPRY (B30.2) domain of African green monkey

TRIM5alpha determines species-specific restriction of simian immunodefi-ciency virus SIVmac infection. J. Virol. 79, 8870–8877.

Ohkura, S., Yap, M.W., Sheldon, T., Stoye, J.P., 2006. All three variable regions of theTRIM5alpha B30.2 domain can contribute to the specificity of retrovirusrestriction. J. Virol. 80, 8554–8565.

Owens, C.M., Yang, P.C., Gottlinger, H., Sodroski, J., 2003. Human and simianimmunodeficiency virus capsid proteins are major viral determinants of early,postentry replication blocks in simian cells. J. Virol. 77, 726–731.

Perez-Caballero, D., Hatziioannou, T., Yang, A., Cowan, S., Bieniasz, P.D., 2005.Human tripartite motif 5alpha domains responsible for retrovirus restrictionactivity and specificity. J. Virol. 79, 8969–8978.

Perron, M.J., Stremlau, M., Lee, M., Javanbakht, H., Song, B., Sodroski, J., 2007. Thehuman TRIM5alpha restriction factor mediates accelerated uncoating of theN-tropic murine leukemia virus capsid. J. Virol. 81, 2138–2148.

Reymond, A., Meroni, G., Fantozzi, A., Merla, G., Cairo, S., Luzi, L., Riganelli, D.,Zanaria, E., Messali, S., Cainarca, S., Guffanti, A., Minucci, S., Pelicci, P.G.,Ballabio, A., 2001. The tripartite motif family identifies cell compartments.EMBO J. 20, 2140–2151.

Roa, A., Hayashi, F., Yang, Y., Lienlaf, M., Zhou, J., Shi, J., Watanabe, S., Kigawa, T.,Yokoyama, S., Aiken, C., Diaz-Griffero, F., 2012. RING domain mutationsuncouple TRIM5alpha restriction of HIV-1 from inhibition of reverse tran-scription and acceleration of uncoating. J. Virol. 86, 1717–1727.

Sawyer, S.L., Wu, L.I., Emerman, M., Malik, H.S., 2005. Positive selection of primateTRIM5alpha identifies a critical species-specific retroviral restriction domain.Proc. Nat. Acad. Sci. USA 102, 2832–2837.

Sayah, D.M., Sokolskaja, E., Berthoux, L., Luban, J., 2004. Cyclophilin A retro-transposition into TRIM5 explains owl monkey resistance to HIV-1. Nature430, 569–573.

Sebastian, S., Luban, J., 2005. TRIM5alpha selectively binds a restriction-sensitiveretroviral capsid. Retrovirology 2, 40.

Song, B., Diaz-Griffero, F., Park, D.H., Rogers, T., Stremlau, M., Sodroski, J., 2005a.TRIM5alpha association with cytoplasmic bodies is not required for antire-troviral activity. Virology.

Song, B., Gold, B., O’Huigin, C., Javanbakht, H., Li, X., Stremlau, M., Winkler, C.,Dean, M., Sodroski, J., 2005b. The B30.2(SPRY) domain of the retroviralrestriction factor TRIM5alpha exhibits lineage-specific length and sequencevariation in primates. J. Virol. 79, 6111–6121.

Song, B., Javanbakht, H., Perron, M., Park, D.H., Stremlau, M., Sodroski, J., 2005c.Retrovirus restriction by TRIM5alpha variants from Old World and New Worldprimates. J. Virol. 79, 3930–3937.

Stremlau, M., Owens, C.M., Perron, M.J., Kiessling, M., Autissier, P., Sodroski, J.,2004. The cytoplasmic body component TRIM5alpha restricts HIV-1 infectionin Old World monkeys. Nature 427, 848–853.

Stremlau, M., Perron, M., Lee, M., Li, Y., Song, B., Javanbakht, H., Diaz-Griffero, F.,Anderson, D.J., Sundquist, W.I., Sodroski, J., 2006. Specific recognition andaccelerated uncoating of retroviral capsids by the TRIM5alpha restrictionfactor. Proc. Nat. Acad. Sci. USA 103, 5514–5519.

Stremlau, M., Perron, M., Welikala, S., Sodroski, J., 2005. Species-specific variationin the B30.2(SPRY) domain of TRIM5alpha determines the potency of humanimmunodeficiency virus restriction. J. Virol. 79, 3139–3145.

Wang, Y.E., Pernet, O., Lee, B., 2012. Regulation of the nucleocytoplasmic traffick-ing of viral and cellular proteins by ubiquitin and small ubiquitin-relatedmodifiers. Biol. Cell 104, 121–138.

Watanabe, M., Fukuda, M., Yoshida, M., Yanagida, M., Nishida, E., 1999. Involve-ment of CRM1, a nuclear export receptor, in mRNA export in mammalian cellsand fission yeast. Genes Cells 4, 291–297.

Wolff, B., Sanglier, J.J., Wang, Y., 1997. Leptomycin B is an inhibitor of nuclearexport: inhibition of nucleo-cytoplasmic translocation of the human immu-nodeficiency virus type 1 (HIV-1) Rev protein and Rev-dependent mRNA.Chem. Biol. 4, 139–147.

Yamauchi, K., Wada, K., Tanji, K., Tanaka, M., Kamitani, T., 2008. Ubiquitination ofE3 ubiquitin ligase TRIM5 alpha and its potential role. FEBS J. 275, 1540–1555.

Yap, M.W., Nisole, S., Stoye, J.P., 2005. A single amino acid change in the SPRYdomain of human Trim5alpha leads to HIV-1 restriction. Curr. Biol. 15, 73–78.

Yee, J.K., Friedmann, T., Burns, J.C., 1994. Generation of high-titer pseudotypedretroviral vectors with very broad host range. Methods Cell Biol. 43 (Pt. A),99–112.

Yueh, A., Leung, J., Bhattacharyya, S., Perrone, L.A., de los Santos, K., Pu, S.Y.,Goff, S.P., 2006. Interaction of moloney murine leukemia virus capsidwith Ubc9 and PIASy mediates SUMO-1 addition required early in infection.J. Virol. 80, 342–352.