Pro-515 of the dynamin-like GTPase MxB contributes to HIV-1 inhibition by regulating MxB oligomerization and binding to HIV-1 capsid

Interferon-regulated myxovirus resistance protein B (MxB) is an interferon-induced GTPase belonging to the dynamin superfamily. It inhibits infection with a wide range of different viruses, including HIV-1, by impairing viral DNA entry into the nucleus. Unlike the related antiviral GTPase MxA, MxB possesses an N-terminal region that contains a nuclear localization signal and is crucial for inhibiting HIV-1. Because MxB previously has been shown to reside in both the nuclear envelope and the cytoplasm, here we used bioinformatics and biochemical approaches to identify a nuclear export signal (NES) responsible for MxB's cytoplasmic location. Using the online computational tool LocNES (Locating Nuclear Export Signals or NESs), we identified five putative NES candidates in MxB and investigated whether their deletion caused nuclear localization of MxB. Our results revealed that none of the five deletion variants relocates to the nucleus, suggesting that these five predicted NES sequences do not confer NES activity. Interestingly, deletion of one sequence, encompassing amino acids 505–527, abrogated the anti-HIV-1 activity of MxB. Further mutation experiments disclosed that amino acids 515–519, and Pro-515 in particular, regulate MxB oligomerization and its binding to HIV-1 capsid, thereby playing an important role in MxB-mediated restriction of HIV-1 infection. In summary, our results indicate that none of the five predicted NES sequences in MxB appears to be required for its nuclear export. Our findings also reveal several residues in MxB, including Pro-515, critical for its oligomerization and anti-HIV-1 function.

In this study, given that MxB is found in both the nuclear envelope and the cytoplasm (36,38,(44)(45)(46)(47), we used online software to predict the putative nuclear export signal (NES) in MxB that might have led to the cytoplasmic location of MxB and further tested their nuclear export functions by deleting each of the five candidates. All of these five deleted MxB mutants were still found in the cytoplasm, suggesting that none of these five sequences bears NES activity. To our surprise, we found that the MxBdel(505-527) mutant lost both its ability to impair HIV-1 infection and the subcellular localization pattern of WT MxB. Further study showed that amino acids 515-519 (especially Pro-515) were critical for oligomerization of MxB and capsid-binding ability and thus essential for MxB antiviral activity.

Deletion of amino acids 505-527 diminishes antiviral ability of MxB
Although it was reported that MxB had an NLS at its N terminus, MxB was found in the cytoplasm (36,38,(44)(45)(46)(47). It is unclear whether MxB has an NES to assist its translocation from nucleus to cytoplasm. Using online software to predict the putative NES in MxB (LocNES) (51), several high-score regions with overlapping sequences were suggested (Table 1). Thus, we created five DNA constructs, each deleting one of these potential regions, named the 51-68, 98 -123, 199 -223, 374 -389, 505-527 deletions. A schematic representation of these five deletions is shown in Fig. 1A. These five mutants, WT MxB, or N-terminal (residues 1-25) deletion were transfected into HeLa cells. As shown in Fig. 1B, MxB was localized at the nuclear envelope and in the form of cytoplasmic puncta/granules, whereas the N-terminal (residues 1-25) deletion, lacking the NLS, was dispersed in the cytoplasm. All of the five NES deletion candidates were seen in the cytoplasm, suggesting that these five sequences do not bear NES function.
In the meantime, we measured the antiviral activity of these MxB deletions. Cells were transfected with these DNA con-structs and then infected with vesicular stomatitis virus glycoprotein (VSV-G) pseudotyped HIV-1 NL4 -3-⌬E-YFP . Infected cells were scored by flow cytometry. The results showed that the MxBdel(505-527) mutant exhibited dramatically decreased antiviral activity, similar to the MxBdel(1-25) mutant (Fig. 1C, top). The expression levels of these deletions were comparable with WT MxB (Fig. 1C, bottom). Previous reports showed that the stalk domain of MxB contains leucine zipper repeats, contributes to MxB dimerization, and is critical for restricting HIV-1 (36,38,41,42,45). To test whether the loss of antiviral activity of MxBdel(505-527) was due to the lack of dimer formation, the WT or MxB mutants tagged with a FLAG epitope were tested for association with Myc-MxB by immunoprecipitation. Only the dimer interface mutant 574/ 651D lost the ability to oligomerize, and the five deletions did not disrupt MxB homodimer formation (Fig. 1D).

Pro-515 is important for the antiviral activity of MxB
To understand the role of the individual amino acids within region 515-519 in MxB inhibiting HIV-1, we mutated each amino acid to alanine (Fig. 3A). Results of immunofluorescence experiments showed that mutants MxB515A, MxB517A, MxB518A, and MxB519A exhibited similar patterns of distribution at the nuclear envelope with cytoplasmic puncta (Fig.  3B). These MxB mutant-expressing HEK293 cells were infected with VSV-G pseudotyped HIV-1 NL4 -3-⌬E-YFP . Results of flow cytometry showed that the antiviral activity of MxB decreased when Pro-515 was mutated, whereas its expression levels were similar to that of MxB (Fig. 3C). Co-immunoprecipitation assays showed that MxB mutants MxB515A, MxB517A, MxB518A, and MxB519A interacted with Myc-

Pro-515 of MxB contributes to its antiviral activity
MxB as efficiently as the WT (Fig. 3D), which suggests that these mutants did not disrupt MxB homodimer formation.

Mutations (515-519)A and MxB515A inhibit the formation of oligomer
The three-dimensional density map (Protein Data Bank ID 5UOT) of the MxB helical assembly has been determined by Alvarez et al. (43) at 4.6 Å resolution, through cryo-EM and real-space helical reconstruction. MxB was shown to assemble into highly ordered long helical tubes like other dynamin family members. Oligomerization of MxB has been shown to be essential for binding to HIV-1 core and inhibiting HIV-1 infection (36,38,(41)(42)(43). The cryo-EM structure of the MxB assembly reveals that amino acids 515-519 are located on the S␣2 helix of
To determine the oligomerization ability of MxB(515-519)A and MxB515A, a chemical cross-linking assay and gel filtration Pro-515 of MxB contributes to its antiviral activity chromatography were performed. As determined by crosslinking with disuccinimidyl suberate (DSS), the interface 2 mutant MxB574/651D lost its oligomerization ability completely (Fig. 5A). Similar to interface 3 mutant MxB449/495D, MxB(515-519)A was defective in oligomer formation, and oligomerization of MxB515A decreased as well. Furthermore, the results of gel filtration chromatography showed that the highermolecular weight oligomerization of MxB449/495D, MxB574/ 651D, MxB(515-519)A, and MxB515A decreased with concomitant increase in lower-molecular-mass complexes (Fig.  5B), which correlates with their incapability to inhibit HIV-1 infection. These data suggest that loss of oligomerization has led to the deficiency of MxB(515-519)A and MxB515A in inhibiting HIV-1.

The 515-519A and 515A mutants lose the ability to bind HIV-1 capsid
It was previously reported that MxB dimerization and binding to HIV-1 capsid are necessary for inhibiting HIV-1 infection (36,38,(41)(42)(43)(44). To measure the binding of (515-519)A to HIV-1 capsid, we performed step gradient analyses to detect the interaction between p24 and MxB (WT, MxB449/495D, and MxB(515-519)A). WT MxB showed strong interaction with p24. Interface 3 mutant MxB449/495D reduced MxB binding to HIV-1 capsid, and the capsid binding ability of (515-519)A was also dramatically reduced (Fig. 6, A-C). These data suggest that the lack of binding to HIV-1 capsid underlies the loss of HIV-1 inhibition by (515-519)A. Association of 515A mutant with HIV-1 capsid was also disrupted, as opposed to moderate or no adverse effect by the other mutants, MxB517A, MxB518A, and MxB519A (Fig. 6, D and E).

Discussion
The N-terminal 25 amino acids are essential for the anti-HIV-1 activity of MxB by regulating MxB subcellular localization and binding to HIV-1 capsid (15-17, 35, 36, 38, 45, 46). The amino acids 11 RRR 13 are involved in association with HIV-1 capsid; mutation of 11 RRR 13 impairs the anti-HIV-1 function  (35,36). Amino acids 20 KY 21 contribute to the NLS function, yet mutation of Lys 20 does not affect MxB inhibition of HIV-1 infection, casting doubt on the role of NLS in MxB anti-HIV-1 activity (35). Because MxB is localized to both the nucleus and the cytoplasm (36,38,(44)(45)(46)(47), it is speculated that MxB has a NES to assist its translocation. In this study, using online software to predict the putative NES of MxB, we identified five regions and deleted each of these five sequences to test their potential NES function. Unfortunately, these five sequences do not bear NES function.
MxB is able to assemble into highly ordered helical tubes. Using molecular dynamics flexible fitting, the cryo-EM struc-

Pro-515 of MxB contributes to its antiviral activity
ture of MxB assembly reveals that six MxB dimers are interconnected one by one to form a rung, which further forms a onestart helical tube (43). The ability to dimerize has been shown to be critical for its antiviral activity, as the dimer interface mutations ablate the ability to inhibit HIV-1 (36,38,(41)(42)(43)(44). Mutations (F495D, R449D, and E484K) in one of the oligomerization interfaces, interface 3, lose both the ability to form oligomers and the ability to restrict HIV-1 (43). Here we observed that the (515-519)A mutation disrupts the anti-HIV-1 activity of MxB. Based on the cryo-EM structure, amino acids 515-519 are located adjacent to oligomer interface 3. The (515-519)A and 515A mutations may thus affect the hydrophobic interactions at interface 3 and affect rung stacking and oligomer formation. Indeed, results of chemical cross-linking and gel filtration chromatography showed that (515-519)A and 515A mutations diminish MxB oligomerization. We have also tested the capsid binding ability of these mutations and observed that (515-519)A and 515A did not bind to HIV-1 capsid. This indicates possible association between MxB oligomerization and its binding to HIV-1 capsid. Oligomerization of MxB has been reported to play a pivotal role in inhibiting HIV-1 and herpesviruses, likely through a coordinated binding to viral capsid and thus blockade of viral DNA from entering the nucleus (26 -28, 36, 38, 41-44). Through investigating a series of MxB mutants, we have identified new amino acids in MxB, including amino acids 515-519, which participate in MxB oligomerization and contribute to the antiviral function of MxB. Our data further strengthen the importance of MxB oligomerization in viral restriction.

Plasmids and cells
The FLAG-tagged MxB in the pQCXIP expression vector was described previously (15). The Myc-tagged MxB, MxB deletions, and mutations were generated with PCR or the KOD-Plus mutagenesis kit (TOYOBO). Plasmid DNA was transfected into cells using polyethyleneimine (Sigma-Aldrich) in accordance with the manufacturer's instructions. HEK293T cells, HeLa cells, and HEK293 cells were cultured in Dulbecco's modified Eagle's medium containing 10% fetal bovine serum (Gibco), 1% penicillin and streptomycin (100ϫ; Solarbio). HEK293 cells stably expressing MxB were constructed by pQCXIP plasmids expressing WT or mutant MxB proteins transfection and puromycin (0.6 g/ml) selection.

Pro-515 of MxB contributes to its antiviral activity
FLAG antibody (Cell Signaling Biotechnology). Flow cytometry was used to quantitate the infection of HIV-1.

Western blotting
Cells were lysed in radioimmune precipitation assay buffer comprised of 25 mM Tris/HCl (pH 7.4), 150 mM NaCl, 1% Nonidet P-40, 0.25% sodium deoxycholate, 1 mM EDTA, and a proteinase inhibitor mixture (Sigma-Aldrich). Lysates were cleared by centrifugation. Proteins were separated by SDS-PAGE and then transferred onto the nitrocellulose membranes (Millipore). After blocking in 5% (w/v) nonfat skim milk for 1 h, membranes were incubated with anti-FLAG antibody (Sigma-Aldrich), anti-Myc antibody (Sigma-Aldrich), anti-actin antibody (Sigma-Aldrich), or anti-p24 antibody (Sino Biological) at 4°C for overnight. The corresponding IRDye TM secondary antibodies (Odyssey) were then applied to the membranes. After extensive washing, the membranes were scanned by the Odyssey IR imaging system (LI-COR). Protein signals were quantified using the ImageJ automated digitizing program (National Institutes of Health).

Immunoprecipitation assay
Immunoprecipitation was conducted as described previously (48). Briefly, cells were lysed in radioimmune precipitation assay buffer 48 h post-transfection. After centrifugation, cell lysates were incubated with anti-FLAG antibodyconjugated agarose (Sigma-Aldrich) at 4°C for overnight. After extensive washing, agarose was boiled in SDS-PAGE sample buffer. Then Western blotting was performed as described above.

Immunofluorescence staining and confocal microscopy
For immunostaining, transfected HeLa cells were fixed in 4% paraformaldehyde for 10 min and then permeabilized by 0.1% Triton X-100 for 10 min. After incubation in blocking buffer, staining was performed with anti-FLAG antibody (Sigma-Aldrich) for 1 h, followed by anti-mouse Alexa Fluor 488conjugated antibody (Thermo Scientific) incubation. DAPI was used for nuclei staining. Images were acquired with a Leica TCS SP5, DMI6000 confocal microscope (Leica Microsystems).

Cross-linking assay
As described previously (42), FLAG-tagged MxB or mutants transfected HEK293T cells were harvested and lysed with 0.5% Nonidet P-40/PBS. Cell lysates were cleared by centrifugation at 16,000 ϫ g for 10 min. Supernatants were incubated with DSS (Thermo Scientific) at different final concentrations for 1 h at room temperature. The mixtures were then incubated with SDS-PAGE sample buffer for 30 min at 37°C. Samples were resolved by 6% SDS-PAGE and detected by Western blotting.

Gel filtration chromatography
For gel filtration chromatography, 1 ϫ 10 7 HEK293T cells expressing MxB or mutants were collected and lysed in buffer containing 20 mM Tris/HCl (pH 7.4), 150 mM NaCl, and 14 mM CHAPS. Cell lysates were separated using a Superdex TM 200 column (GE Healthcare) at an elution rate of 0.5 ml/min. Samples from each fraction were analyzed by Western blotting.

Data availability
All data are contained within the article.