PDF (2 MB)

Structure

Short Article Identification of Phe187 as a Crucial Dimerization Determinant Facilitates Crystallization of a Monomeric Retroviral Integrase Core Domain Meytal Galilee1 and Akram Alian1,* 1Faculty of Biology, Technion–Israel Institute of Technology, Haifa 320003, Israel *Correspondence: [email protected] http://dx.doi.org/10.1016/j.str.2014.08.001

SUMMARY

Retroviral DNA integration into the host genome is mediated by nucleoprotein assemblies containing tetramers of viral integrase (IN). Whereas the fully active form of IN comprises a dimer of dimers, the molecular basis of IN multimerization has not been fully characterized. IN has consistently been crystallized in an analogous dimeric form in all crystallographic structures and experimental evidence as to the level of similarity between IN monomeric and dimeric conformations is missing because of the lack of IN monomeric structures. Here we identify Phe187 as a critical dimerization determinant of IN from feline immunodeficiency virus (FIV), a nonprimate lentivirus that causes AIDS in the natural host, and report, in addition to a canonical dimeric structure of the FIV IN core-domain, a monomeric structure revealing the preservation of the backbone structure between the two multimeric forms and suggest a role for Phe187 in ‘‘hinging’’ the flexible IN dimer.

INTRODUCTION Integrase (IN) of the human immunodeficiency virus-1 (HIV-1) mediates the integration of viral DNA into the host genome during the viral life cycle and thus has been targeted in anti-HIV research and AIDS (Krishnan and Engelman, 2012). IN cleaves a dinucleotide from each 30 -end of the reverse-transcribed viral double-stranded DNA, in a process termed 30 -processing, to prepare the DNA for integration into the host genome in a consequent second step termed strand transfer, the latter step being the target of the current IN strand transfer inhibitors (INSTIs; Krishnan and Engelman, 2012). Like HIV-1 and other retroviral INs, the feline immunodeficiency virus (FIV) IN functions as a multimer, in which discrete domains from each protomer functionally complement each other, and it contains the three common functional domains: the N-terminal domain (NTD) containing the ‘‘HH-CC’’ zinc-binding motif, the catalytic core (CCD) including the canonical catalytic triad D66D118E154, and the C-terminal domain (CTD) with an SH3-like fold (Figure 1A; Jaskolski et al., 2009; Li et al.,

2011; Shibagaki et al., 1997). The high sequence conservation between the IN CCD domains of FIV and HIV-1 (44% identity) suggested the FIV IN CCD structure and INSTIs binding site would be nearly the same as for HIV IN; a prediction that has been validated by the effective binding of INSTIs to FIV and the inhibition of FIV replication by INSTIs in vitro at concentrations comparable to those reported for HIV-1 (Savarino et al., 2007). The seminal structures of IN-DNA complexes from prototype foamy virus (PFV) provided the groundbreaking structural details for IN oligomeric organization within the functional IN-DNA integration complexes, confirming the critical tetrameric state of assembly (Hare et al., 2010; Maertens et al., 2010). Tetramers comprise a dimer of dimers, where the individual dimers preserve a common CCD-CCD dimerization interface, and each dimer is required to bind to one of the viral DNA ends. Whereas the fully active form of IN is evidently a tetramer, various other oligomeric states, including monomers and dimers, do exist in solution and the correct precursors in the tetramer assembly pathway are yet to be determined (Alian et al., 2009; Bojja et al., 2013; Hare et al., 2010; Lesbats et al., 2008; Li et al., 2006; Maertens et al., 2010; Pandey et al., 2011). Diverse models for dimeric and multimeric assemblies, particularly distinctive between IN unbound (apo) and DNA-bound forms, have been proposed (Alian et al., 2009; Ballandras et al., 2011; Barsov et al., 1996; Johnson et al., 2013; Krishnan et al., 2010; Lesbats et al., 2008; Yin and Craigie, 2010). A ‘‘reaching dimer’’ configuration of the apo-IN has been reported lacking the canonical CCD-CCD interface and is stabilized mainly by CTD-CTD interactions (Bojja et al., 2013), a variation that presumably affects global protein structure by affecting the relative positions of the individual domains (Maillot et al., 2013). The principal role of CCD in IN dimerization is well recognized, most remarkably by its consistent crystallization in an analogous dimeric form in all crystallographic structures (Jaskolski et al., 2009; Li et al., 2011; except for one slightly variant dimeric configuration, Ballandras et al., 2011). Experimental evidence as to the level of similarity between IN monomeric and dimeric conformations is missing because of the lack of IN structures in its monomeric form. The canonical CCD-CCD dimer interface has been shown to predominantly involve hydrophobic contacts among a1, a5, and a6 (Jaskolski et al., 2009; Serrao et al., 2012). Notably, all available crystal structures of retroviral IN CCD, in either single or two-domain constructs, contain a linker region (residues 191–210 in HIV-1 IN) and no construct comprising the CCD alone (52–190 in HIV-1 IN) has been crystallized. This

1512 Structure 22, 1512–1519, October 7, 2014 ª2014 Elsevier Ltd All rights reserved

Structure Monomeric Structure of Retroviral Integrase

B A

C

D

E

Figure 1. Domain Organization and Size Exclusion Chromatography Analysis of FIV and HIV-1 IN Variants (A) Schematic representation of the various FIV IN variants used. FL: full length (1–281); CCD: catalytic core domain of two variants: short (S; 61–188) and extended (E; 61–212); C-NTD: N-terminal-core two-domain (1–188), C-CTD: C-terminal-core two-domain (61–281). Theoretical molecular weights (kDa) of FIV and HIV-1 variants are shown (right). (B) SEC analysis of short (fiv-S) and extended (fiv-E) variants of F187K mutant CCD from FIV and HIV-1 (hiv-S: 52–186; hiv-E: 52–210), and WT FIV CCD (fiv-E-WT). (C) SEC analysis of full-length FIV IN WT (fiv-FL) and F187K mutant (fiv-FL-F187K) compared to F185K mutant of full-length HIV-1 IN (hiv-FL). (D) SEC analysis of WT FIV C-CTD, and F187K mutants of C-CTD (C-CTD-F187K) and C-NTD (C-NTD-F187K). Molecular mass markers (sizes in kilodaltons) are shown in gray. Insets: apparent molecular masses of corresponding peaks as determined by partition coefficient [Kav = (elution volume void volume) / (total volume void volume)]. (E) In vitro 30 -processing activities of WT full-length FIV IN and F187K mutant as compared to control containing DNA without IN (Ctrl). Error bars represent SD of three independent replicates.

is probably because, in addition to practicality and convenience in construct preparations (Jaskolski et al., 2009), the region linking the CCD to CTD (C-terminal linker [CTL]) was found, unlike the NTD and CTD, to function in cis to the CCD (Engelman et al., 1993; Shibagaki et al., 1997). Functional importance of residue composition, and finger flexibility, of the CTL region (186– 196 from HIV-1 IN) has been implicated in viral infectivity and in IN functional tetramerization, nuclear localization, enzymatic activities, and conformational changes upon DNA binding (Berthoux et al., 2007; Cellier et al., 2013; Hare et al., 2009a; Zhao

et al., 2008). Still, if the CTL plays a crucial role in the formation of IN dimers, the precursors of functional tetramers, it remains to be clarified. To foster our understanding of IN multimerization, particularly to highlight common and distinct features in the dimerization process, we investigated dimerization of IN using an FIV model system. The FIV model has greatly advanced our understanding of several basic questions in retrovirology (e.g., nuclear import, two-receptor entrance, replication in non-dividing cells, species-specificity, and broad-based inhibitor studies) and can

Structure 22, 1512–1519, October 7, 2014 ª2014 Elsevier Ltd All rights reserved 1513


Table 1. Data Collection and Refinement Statistics Monomeric CCD

Dimeric CCD

Space group

P 21

P6522

Mol/ASU Cell dimensions a, b, c (A˚)

1

1

28.25, 66.64, 32.62

52.92, 52.92, 193.57

a, b, g ( )

90.00, 103.43, 90.00

90.00, 90.00, 120.00

Resolution (A˚)a

33.32–1.08 (1.118–1.08) 45.83–1.84 (1.91–1.844)

Rsym or Rmerge

0.03683 (0.4863)

0.09952 (0.3539)

I/sI

15.94 (2.28)

29.66 (6.74)

Data Collection

Completeness (%) 95.5 (88.4)

90.6 (88.2)

Redundancy

3.4 (3.3)

15.2 (13.8)

Resolution (A˚)

1.1

1.84

No. unique reflections

48,083 (4402)

13,341 (1263)

Rwork/Rfree (5%)

0.140/0.176

0.185/0.225

Protein

1,195

1,188

Water

186

111

Protein

14.0

25.4

Water

33.9

38.5

Bond lengths (A˚)

0.02

0.018

Bond angles ( )

1.94

1.82

Favored

100

100

Outliers

0

0

PDB code

4MQ3

4PA1

Refinement

No. atoms

B-factors

Rmsds

Ramachandran (%)

a

Number in parentheses is for highest resolution shell.

further be exploited for anti-AIDS studies in a settings where the lentivirus is used to infect its natural, inexpensive and uncontroversial host (Bienzle, 2014). We found that a single F187K mutation, analogous to the commonly used F185K solubility mutation in HIV-1 IN (Jaskolski et al., 2009) monomerizes an otherwise dimeric FIV IN, an exciting finding that we further exploited to obtain a crystal structure of IN CCD in the monomeric form, providing experimental evidence for the preservation of the IN CCD backbone between monomeric and canonical dimeric forms. RESULTS AND DISCUSSION Phe187 Is Essential for FIV IN Dimerization Unlike HIV-1 IN for which the CCD (52–210, F185K) exists in the dimeric state, the FIV IN CCD (61–188, F187K) exists as monomers (Figure 1B). Because the FIV IN CCD construct used here lacks the C-terminal linker region (CTL, residues 189–212 in FIV (Figure 1A) and 191–210 in HIV-1 INs), which is present in all reported IN CCD structures, we investigated whether CTL absence caused the observed monomeric state. Thus, we

extended the FIV IN CCD to comprise residues 61-212 and, in parallel, truncated HIV-1 IN CCD to contain only 52–186. The results clearly reveal that the multimerization state of either IN CCD is not affected by the presence or absence of the CTL fragment (Figure 1B). This suggests that the CTL, which is known to function in cis to CCD (Engelman et al., 1993; Shibagaki et al., 1997) and to critically mediate functional IN tetamerization and activity (Cellier et al., 2013; Hare et al., 2009a), does not seem to play a crucial role in dimerization (at least in absence of DNA substrates). Similarly, Lys188 (analogs to HIV-1 K186 that has been implicated in HIV-1 IN multimerization; Berthoux et al., 2007) does not participate in the dimerization of IN because the addition of CTL (containing wild-type [WT] K188) to F187K mutant constructs did not induce dimerization (Figure 1). Lys188 indeed has been shown to stabilize the tetrameric form of IN by interactions with the NTD of the other dimer (Berthoux et al., 2007; Hare et al., 2009a). To further clarify the location of the dimerization determinants of FIV IN, which are apparently distinct from those of HIV-1 IN, we characterized the oligomeric states of full-length FIV IN and truncation variants containing the CCD fused to either CTD (C-CTD) or NTD (C-NTD; Figure 1A). Unlike HIV-1 IN for which the full-length protein exists predominantly in the tetrameric state (Alian et al., 2009; Figure 1C), WT full-length FIV IN exists predominantly in the dimeric form and its F187K variant in the monomeric state (Figure 1C). Similarly, WT C-CTD (residues 61–281) exists as dimers whereas the introduction of F187K into this C-CTD construct rendered it monomeric (Figure 1D). Notably, a Y244A/W245A double mutation within the SH3-like domain, known to disrupt the reaching-dimer configuration in HIV-1 IN (Bojja et al., 2013), did not disrupt the dimeric form of the FIV CCD-CTD. Likewise, the addition of the NTD (residues 1–60) to the F187K variant of CCD (C-NTD, 1–188) was insufficient to induce its dimerization (Figure 1D). Consistently, in all variants analyzed, WT FIV IN exists as dimers whereas a single F187K mutation, analogous to the common F185K solubility mutation in HIV-1 IN, disrupts the dimeric interaction, leading to the formation of monomers. Thus, WT IN of FIV is soluble and stable in its unbound state without a Lys/His solubility mutation at position 187 (185 in HIV-1/2 INs) and its F187K is highly soluble and stable in the monomeric form and at high concentrations. The F187K mutant of full-length IN has been found competent for 30 -processing activity performing the cleavage reaction at comparable levels to the WT (Figure 1E), implying that F187K mutation does not affect the integrity of the CCD fold. Although this may indicate that monomers of FIV IN can catalyze 30 -processing, unlike all other characterized retroviral INs that reportedly require protein multimerization (Engelman et al., 1993; Faure et al., 2005; Fletcher et al., 1997), we cannot preclude multimer assembly upon DNA binding, particularly since FIV IN has also been shown to function in multimers (Shibagaki et al., 1997). Thus, the biological significance of the F187K mutation can best be revealed in activities that critically require the dimerization of IN such as in vitro strand transfer or in vivo pre-integration complex assembly and viral replication. Intriguingly, unlike HIV IN, recombinant FIV IN has been reported inefficient for in vitro strand transfer activity (Vink et al., 1994), a result that we have also confirmed (data not shown). It will be interesting to investigate this feature of FIV IN that may highlight additional requirements



Figure 2. FIV IN CCD Structures (A) Superposition of canonical dimeric forms of IN CCD from FIV (orange/yellow) and HIV-1 (magenta/violet, 1BIS, rmsd: 0.97 A˚). (B) CCD P21 unit cell representation with a top-view along b-axis (left) and side view (b-axis is labeled). a Helices are numbered and D66-Ca (catalytic-site) is represented by a sphere (marked with red asterisk). (C) Structural superposition of IN CCD from FIV (red), HIV-1 (magenta, 1BL3, rmsd: 1.13 A˚), PFV (blue, 3L2V, rmsd: 0.96 A˚), and RSV (yellow, 1C1A, rmsd: 1.13 A˚).

for the integration step of FIV perhaps elucidating alternative pathways to those used by HIV. The effects of F187K on preintegration complex assembly and cellular localization and on viral replication are currently being characterized. Crystal Structure of FIV IN CCD in the Monomeric Form Reveals Backbone Structural Preservation between Monomeric and Dimeric Forms Given that IN CCD is readily crystallizable (Jaskolski et al., 2009), and isolation of FIV IN in two distinct mulitmeric forms, dimer (WT) and monomer (F187K), provided a unique opportunity to further analyze the molecular details of the integrase core domain in monomeric form and compare this to the dimeric form. Indeed soluble and stable monomeric (61–188, F187K) and dimeric (61–212, WT) FIV IN CCD readily crystallized, with the monomeric form giving the highest reported resolution of any lentiviral IN to date (1.1 A˚). Both constructs crystallized with one molecule per asymmetric unit in the P21 and P6522 space groups, respectively (Table 1) and, as expected, the crystal structure from the dimeric solution of CCD formed the canonical dimer with a symmetry-related molecule in crystal (Figure 2A). However, symmetry-related molecules from the monomeric form crystals failed to produce dimers (Figure 2B), providing a crystal structure of IN CCD in the monomeric configuration. The overall backbone structures of the two FIV IN CCD forms are nearly identical to each other (root-mean-square deviation [rmsd] 0.8 A˚) and to that of other dimeric crystal structures from lentiviruses (e.g HIV-1) and retroviruses (e.g., Rous sarcoma retrovirus [RSV] and PFV; Figure 2C). This provides experimental evidence for preservation of the IN CCD backbone between monomeric and canonical dimeric forms; a result especially important in validating the use of single IN chains extracted from dimeric crystal structures for molecular dynamics simulation studies and in silico identification and development of IN anti-dimerization inhibitors (AlMawsawi et al., 2006; Sippel and Sotriffer, 2010; Tintori et al., 2012).

The Catalytic Loop Is Fully Structured in a ‘‘Closed’’ Conformation In both the monomeric and dimeric FIV IN CCD structures, the catalytic-loop (residues 142–155) is fully structured and folded toward the catalytic site in a ‘‘closed’’ conformation comparable to that observed in PFV IN-DNA intasome or in unbound RSV IN (Figure 2C). The loop retained this exact conformation whether monomeric CCD was crystallized at pH 5.5 or 8.5 (see Experimental Procedures), with only Asp66 in the catalytic triad occupying a different conformation at the different pHs (Figure 3A). Of note, dimeric CCD that was crystallized under different conditions (see Experimental Procedures) yielded an unstructured loop similar to that of previously determined IN CCD dimeric structures. Modeling reveals that whereas the loop in the closed conformation would not clash with viral DNA (Figure 3B), a conserved P144 residue (HIV-1 P142, PFV P211) would clash with host acceptor-DNA and inhibitors mandating loop movement to enable substrate binding (Figure 3C). In FIV, the shallower cavity formed by the backbone of smaller G145 (Y212 in PFV and Y143 in HIV-1) apparently provides sufficient space for binding of INSTI because raltegravir binds and inhibits FIV IN at concentrations comparable to those used with HIV-1 IN (Savarino et al., 2007). Possible Role of Phe187 in Hinging the Flexible IN Dimer The backbone of Lys187 (and following Lys188) in the monomeric FIV IN CCD adopts a different conformation than the equivalent residues in all other retroviral IN dimeric structures (including the conformation of native Phe187 from the dimeric FIV IN CCD structure and the reported Phe185 of HIV-2; Hare et al., 2009b; Figure 4A). This could be due to the absence of an opposite protomer as a binding partner or due to the lack of the CTL extension to force a canonical conformation. Still, the extension of CCD with CTL fragment (61–212, F187K) was not sufficient to induce dimerization (Figure 1B). Except for residue 187 (and following 188), the conformation of the residues at the dimeric interface is mostly invariant between the monomeric



Figure 3. FIV IN CCD Loop Structure and Modeling (A) Stereo view (wall-eyed) of the catalytic-site loop from monomeric CCD with FO-FC map (3.0 s) calculated with residues 142–155 omitted. D66 conformations at pH 8.5 (blue omit-map) and pH 5.5 (green omit-map and blue stickball) are shown. (B) Modeling of FIV IN-DNA complex. Original crystal-conformation of the catalytic loop (red) is modeled with viral DNA based on 3L2R. Catalytic triad is labeled and represented with green lines (FIV) or blue stickballs (PFV). P144 from FIV is labeled and shown in green stickballs. The catalytic loop of PFV IN is presented with transparent blue cartoon. Green sphere represents Mg2+ ion. (C) Modeling of FIV IN-DNA-INSTI complex. Amino acid sequence of FIV IN CCD was homology modeled to fit PFV intasome-raltegravir structure (3L2V). Variant residues in PFV IN (blue lines) and FIV IN (green lines) are Q186/N119; T210/I143; S209/G142; Y212/G145; H213/N146; S217/Q150. Blue spheres represent Mn2+ ions from 3L2V.

and dimeric forms, thus accentuating the critical contribution of Phe187 to the dimerization of FIV IN. Several residues at the CCD-CCD canonical dimer interface, implicated as critical for HIV-1 IN dimerization (Serrao et al., 2012), are dissimilar in FIV IN but also in the other retroviral INs (Figure 4B, blue highlights). The HIV-1 IN CCD (52–210) dimer interface buries 25% more area than FIV IN CCD (61–212; 1,500 A˚2 versus 1,200 A˚2) and involves twice as many hydrophobic contacts (99 versus 43), explaining the susceptibility of FIV IN, but not HIV-1, to monomerization by a single F187K mutation. It is well recognized that the tetrameric configuration of apo-IN is dissimilar to that of IN assembled upon DNA binding, and that IN must exist in lower-order oligomeric states in order to interact with the substrate DNA and form functional intasomes (Alian et al., 2009; Barsov et al., 1996; Bojja et al., 2013; Lesbats et al., 2008). Whereas the form of these smaller-multimeric IN entities, likely dimers, and their interaction interfaces and assembly pathways are yet to be defined, we suggest that the residue at the C-terminal tip of CCD a-5 (F187 in FIV) may pin the dimer interface, which hinges the two protomers together during flexible IN rearrangements upon DNA binding. Such a pivot at resi-

due N255 (equivalent to F187 in FIV) is noticeable in the CCD dimeric interface of PFV IN that, upon DNA binding, opens up by 11 A˚ at the N terminus of CCD a-helix-5 and exposes 50% of the buried area (reducing the 1,600 A˚2 in apo-IN [3DLR] to 870 A˚2 in IN-DNA [3L2V]; Figure 4C). Integrity and flexibility of the dimeric interface around this hinge has been implicated in IN multimerization and catalytic activities (Berthoux et al., 2007; Cellier et al., 2013; Hare et al., 2009a). The finding that dimerization of FIV IN can be abrogated by a single amino acid substitution (F187) further validates the proposed strategy for inhibiting proteinprotein interactions by targeting ‘‘functional epitopes’’ or ‘‘hot spots’’ (few residues) at the interacting interfaces, once confined, rather than targeting the entire interacting surfaces (Sippel and Sotriffer, 2010; Toogood, 2002). In conclusion, we report structures of FIV IN; the catalytic core-domain crystallized in the usually obtained dimeric form along with a monomeric form. The monomeric structure obtained from stable monomeric IN in solution finally validates the backbone preservation of IN core-domain between the monomeric and dimeric structures reported here, and reveals that dimerization of FIV IN relies on a single amino acid (Phe187); a result that not only can challenge our current understanding of IN dimerization, but can also facilitate in vivo functional studies with IN-F187K mutant FIV virus and assaying for viral replication and infectivity, IN multimerization and interaction with host proteins (e.g., LEDGF), preintegration complex formation, nuclear localization, and integration. The FIV model is especially suitable for such studies because its biology and disease in cats are common with those of HIV in human and can be used to advance anti-AIDS studies. We propose exploiting the FIV IN as a model system for future in vitro and in vivo efforts



Figure 4. FIV IN CCD Sequence and Structural Comparisons and the Proposed Dimer Hinging Model (A) F187K (red stick) of monomeric FIV CCD superposed to the WT dimeric structure (two protomers in yellow and in orange showing F187, in orange stick, with FO-FC map [3.0 s] calculated while omitting residue F187 [blue map] or F187K [magenta map]); HIV-2 WT F185 (magenta line, 3F9K) and HIV-1 F185K and K186 (cyan lines, 1BIS). Lys188 of FIV CCD is also shown in line-representation for monomeric (red) and dimeric (orange) forms. Helices a1 and a5 are labeled. (B) Sequence alignment of IN CCD from various retroviruses. Conserved residues are white on a red background, and conservative substitutions are red on a white background. Catalytic triad D-D-E residues are labeled with red asterisks. Dissimilar residues at the dimer interface are highlighted with blue. Secondary structure elements from FIV and HIV-1 (1BIS) are shown (blue) above and below their sequences, respectively (coil, helix; arrow, b strand; T, turn). (C) Dimers of PFV IN in apo (3DLR, pale-blue) and DNA bound forms (3L2V, marine-blue) are shown from side view (left). Blue dashed arrow indicates CCD displacement of 11 A˚ upon DNA binding (at A188 of a5 distal end). Inset represents an enlargement of the ‘‘hinge’’ interface comparing open and closed dimers. Interacting residues are shown with sticks; distances in angstroms are indicated with red dashed lines. The PFV IN a helix5 N255 (analogous to F187 of FIV IN) interaction with the backbone oxygen of a helix1 I176 and S175 from the opposite protomer remains constant, 9 A˚ distance, for both apo (closed) and DNAbound (open) forms.

particularly aimed at understanding the molecular basis of IN multimerization and the assembly pathways of functional integration complexes. EXPERIMENTAL PROCEDURES Protein Preparation IN variants of FIV (Petaluma; a gift from Prof. Carsten Mu¨nk, Germany) were subcloned into pET28b (Novagen) with a cleavable His-tag. Wild-type full length and C-CTD two-domain constructs contain a C274S mutation. HIV-1 IN CCD constructs were derived from the previously described HIV-1 IN (SF1; C56S, W131D, F185K, and C280S; Alian et al., 2009). Mutations were introduced using QuikChange Kit (Agilent Technologies Genomics). IN constructs were expressed and purified as previously described (Alian et al., 2009). Protein solutions contain 20 mM HEPES pH 7.4, 1 mM dithiothreitol, 1 M NaCl for full-length or 0.5 M NaCl for the truncation variants, and 1 mM CHAPS for CTD-containing constructs. Apparent molecular masses were

analyzed on a Superose-12 column (GE Healthcare Life Sciences) as compared to molecular size standards (Bio-Rad). Protein Crystallization, Data Collection, and Structure Determination FIV IN CCD (WT 61–212 or F187K 61–188, at 8–9 mg/ml with 1 mM MgCl2) crystals were grown (20 C) using the hanging-drop vapor-diffusion method. Diffracting crystals of monomeric CCD (61–188, F187K) were readily obtained in various sparse-matrix screen conditions: ProComplex-A4, Classics-H11 and H6, PEGs-B5, B11 and B11, JCSG-G1, G4, H10, and H8. Crystals of the monomeric form (from JCSG-G4: 0.2 M trimethylamine N-oxide, 0.1 M Tris-HCl pH 8.5, 20% [w/v] PEG-MME-2000) were used for data collection and structure determination in P21 space group with one molecule per asymmetric unit (Table 1). The alternative conformation of Asp66 was revealed in crystals grown at pH 5.5 in JCSG-H8 (0.2 M NaCl, 0.1 M Bis-Tris pH 5.5, 25% PEG 3350), which also diffracted to 1.1 A˚. Diffraction quality crystals of the dimeric CCD (CCD-E, 61–212) were readily obtained in two conditions: anions-B7 (2.5 M sodium acetate, 0.1M Tris-HCl pH 8.5) with P31 space



group and two molecules per asymmetric unit and unstructured catalytic loop (not shown), and classics-H4 (30% [w/v] PEG-4000, 0.2 M MgCl2, 0.1 M TrisHCl pH 8.5) resulting in P6522 space group with one molecule per asymmetric unit (Table 1). Cryogenic solutions contained a supplement of 20% (v/v) glycerol. Diffraction data for monomeric CCD (61–188, F187K) were collected at beamline 14.1 of BESSY-II synchrotron (Helmholtz-Zentrum Berlin), and processed with XDSAPP (Krug et al., 2012). Diffraction data for dimeric CCD (61–212) were collected in-house on a Rigaku FR-X rotating anode generator and processed using HKL3000 (Minor et al., 2006). The structures were solved by molecular replacement (using 3F9Ka as a search model; Hare et al., 2009b) using BALBES (Long et al., 2008) and resulting models were built using COOT (Emsley et al., 2010) and refined using REFMAC5 (Murshudov et al., 2011; with anisotropic temperature factors for the monomeric form 4MQ3). In Vitro Fluorescence IN 30 -Processing Assays Fluorescence based 30 -processing IN activity assay was performed as described (Merkel et al., 2009). The donor DNA used was prepared by annealing two fragments: 50 -TACAAAATTCCATAGCAGT-6FAM and 50 -ACTGC TATGGAATTTTGTA, and the acceptor DNA was annealed from 50 -BiotinTAT CCG CGA TAA GCT TTA ATG CGG TAG and 50 -Biotin-CTACCGCATTA AAGCTTATCGCGGATA. Sequence and Structural Analysis Structure-guided sequence alignment was made with PROMALS3D (Pei et al., 2008) and annotated using ESPript (Gouet et al., 2003). PyMOL Molecular Graphics System (Schro¨dinger, LLC) was used for preparing structural figures and for structural superposition (using IN structures with Protein Data Bank codes 1BL3c [Maignan et al., 1998] or 1BIS [Goldgur et al., 1998] for HIV-1; 3L2V [Hare et al., 2010] for PFV; and 1C1Aa [Yang et al., 2000] for RSV). Homology modeling was performed using the Swiss Model server (Arnold et al., 2006). Analysis of interacting interfaces was performed using ePISA (Krissinel and Henrick, 2007).

REFERENCES Al-Mawsawi, L.Q., Fikkert, V., Dayam, R., Witvrouw, M., Burke, T.R., Jr., Borchers, C.H., and Neamati, N. (2006). Discovery of a small-molecule HIV-1 integrase inhibitor-binding site. Proc. Natl. Acad. Sci. USA 103, 10080–10085. Alian, A., Griner, S.L., Chiang, V., Tsiang, M., Jones, G., Birkus, G., Geleziunas, R., Leavitt, A.D., and Stroud, R.M. (2009). Catalytically-active complex of HIV-1 integrase with a viral DNA substrate binds anti-integrase drugs. Proc. Natl. Acad. Sci. USA 106, 8192–8197. Arnold, K., Bordoli, L., Kopp, J., and Schwede, T. (2006). The SWISS-MODEL workspace: a web-based environment for protein structure homology modelling. Bioinformatics 22, 195–201. Ballandras, A., Moreau, K., Robert, X., Confort, M.P., Merceron, R., Haser, R., Ronfort, C., and Gouet, P. (2011). A crystal structure of the catalytic core domain of an avian sarcoma and leukemia virus integrase suggests an alternate dimeric assembly. PLoS ONE 6, e23032. Barsov, E.V., Huber, W.E., Marcotrigiano, J., Clark, P.K., Clark, A.D., Arnold, E., and Hughes, S.H. (1996). Inhibition of human immunodeficiency virus type 1 integrase by the Fab fragment of a specific monoclonal antibody suggests that different multimerization states are required for different enzymatic functions. J. Virol. 70, 4484–4494. Berthoux, L., Sebastian, S., Muesing, M.A., and Luban, J. (2007). The role of lysine 186 in HIV-1 integrase multimerization. Virology 364, 227–236. Bienzle, D. (2014). FIV in cats—a useful model of HIV in people? Vet. Immunol. Immunopathol. 159, 171–179. Bojja, R.S., Andrake, M.D., Merkel, G., Weigand, S., Dunbrack, R.L., Jr., and Skalka, A.M. (2013). Architecture and assembly of HIV integrase multimers in the absence of DNA substrates. J. Biol. Chem. 288, 7373–7386. Cellier, C., Moreau, K., Gallay, K., Ballandras, A., Gouet, P., and Ronfort, C. (2013). In vitro functional analyses of the human immunodeficiency virus type 1 (HIV-1) integrase mutants give new insights into the intasome assembly. Virology 439, 97–104.

ACCESSION NUMBERS

Emsley, P., Lohkamp, B., Scott, W.G., and Cowtan, K. (2010). Features and development of Coot. Acta Crystallogr. D Biol. Crystallogr. 66, 486–501.

The Protein Data Bank accession numbers for the atomic coordinates and structure factors for the FIV IN core domain structure in the monomeric and dimeric forms are 4MQ3 and 4PA1.

Engelman, A., Bushman, F.D., and Craigie, R. (1993). Identification of discrete functional domains of HIV-1 integrase and their organization within an active multimeric complex. EMBO J. 12, 3269–3275.

SUPPLEMENTAL INFORMATION

Faure, A., Calmels, C., Desjobert, C., Castroviejo, M., Caumont-Sarcos, A., Tarrago-Litvak, L., Litvak, S., and Parissi, V. (2005). HIV-1 integrase crosslinked oligomers are active in vitro. Nucleic Acids Res. 33, 977–986.

Supplemental Information includes one slider image and one 3D molecular model and can be found with this article online at http://dx.doi.org/10.1016/ j.str.2014.08.001.

Fletcher, T.M., 3rd, Soares, M.A., McPhearson, S., Hui, H., Wiskerchen, M., Muesing, M.A., Shaw, G.M., Leavitt, A.D., Boeke, J.D., and Hahn, B.H. (1997). Complementation of integrase function in HIV-1 virions. EMBO J. 16, 5123–5138.

AUTHOR CONTRIBUTIONS

Goldgur, Y., Dyda, F., Hickman, A.B., Jenkins, T.M., Craigie, R., and Davies, D.R. (1998). Three new structures of the core domain of HIV-1 integrase: an active site that binds magnesium. Proc. Natl. Acad. Sci. USA 95, 9150–9154.

M.G. and A.A. designed and conducted the experiments, analyzed the data, and prepared the manuscript. ACKNOWLEDGMENTS We thank HZB for the allocation of synchrotron radiation beam time. The research leading to these results has received funding from the European Community’s Seventh Framework Program (FP7/2007-2013) under BioStruct-X (grant agreement no. 283570), and from the Rubin Scientific and Medical Research Fund, and benefited from use of the Technion Center for Structural Biology facility of the Lorry I. Lokey Center for Life Sciences and Engineering and the Russell Berrie Nanotechnology Institute. We thank Dr. Orit Goldshmidt for assistance in subcloning. Received: May 31, 2014 Revised: July 14, 2014 Accepted: August 2, 2014 Published: September 4, 2014

Gouet, P., Robert, X., and Courcelle, E. (2003). ESPript/ENDscript: Extracting and rendering sequence and 3D information from atomic structures of proteins. Nucleic Acids Res. 31, 3320–3323. Hare, S., Di Nunzio, F., Labeja, A., Wang, J., Engelman, A., and Cherepanov, P. (2009a). Structural basis for functional tetramerization of lentiviral integrase. PLoS Pathog. 5, e1000515. Hare, S., Shun, M.C., Gupta, S.S., Valkov, E., Engelman, A., and Cherepanov, P. (2009b). A novel co-crystal structure affords the design of gain-of-function lentiviral integrase mutants in the presence of modified PSIP1/LEDGF/p75. PLoS Pathog. 5, e1000259. Hare, S., Gupta, S.S., Valkov, E., Engelman, A., and Cherepanov, P. (2010). Retroviral intasome assembly and inhibition of DNA strand transfer. Nature 464, 232–236. Jaskolski, M., Alexandratos, J.N., Bujacz, G., and Wlodawer, A. (2009). Piecing together the structure of retroviral integrase, an important target in AIDS therapy. FEBS J. 276, 2926–2946.



Johnson, B.C., Me´tifiot, M., Ferris, A., Pommier, Y., and Hughes, S.H. (2013). A homology model of HIV-1 integrase and analysis of mutations designed to test the model. J. Mol. Biol. 425, 2133–2146. Krishnan, L., and Engelman, A. (2012). Retroviral integrase proteins and HIV-1 DNA integration. J. Biol. Chem. 287, 40858–40866. Krishnan, L., Li, X., Naraharisetty, H.L., Hare, S., Cherepanov, P., and Engelman, A. (2010). Structure-based modeling of the functional HIV-1 intasome and its inhibition. Proc. Natl. Acad. Sci. USA 107, 15910–15915. Krissinel, E., and Henrick, K. (2007). Inference of macromolecular assemblies from crystalline state. J. Mol. Biol. 372, 774–797. Krug, M., Weiss, M.S., Heinemann, U., and Mueller, U. (2012). XDSAPP: a graphical user interface for the convenient processing of diffraction data using XDS. J. Appl. Cryst. 45, 568–572. Lesbats, P., Me´tifiot, M., Calmels, C., Baranova, S., Nevinsky, G., Andreola, M.L., and Parissi, V. (2008). In vitro initial attachment of HIV-1 integrase to viral ends: control of the DNA specific interaction by the oligomerization state. Nucleic Acids Res. 36, 7043–7058. Li, M., Mizuuchi, M., Burke, T.R., Jr., and Craigie, R. (2006). Retroviral DNA integration: reaction pathway and critical intermediates. EMBO J. 25, 1295– 1304. Li, X., Krishnan, L., Cherepanov, P., and Engelman, A. (2011). Structural biology of retroviral DNA integration. Virology 411, 194–205. Long, F., Vagin, A.A., Young, P., and Murshudov, G.N. (2008). BALBES: a molecular-replacement pipeline. Acta Crystallogr. D Biol. Crystallogr. 64, 125–132. Maertens, G.N., Hare, S., and Cherepanov, P. (2010). The mechanism of retroviral integration from X-ray structures of its key intermediates. Nature 468, 326–329. Maignan, S., Guilloteau, J.P., Zhou-Liu, Q., Cle´ment-Mella, C., and Mikol, V. (1998). Crystal structures of the catalytic domain of HIV-1 integrase free and complexed with its metal cofactor: high level of similarity of the active site with other viral integrases. J. Mol. Biol. 282, 359–368. Maillot, B., Le´vy, N., Eiler, S., Crucifix, C., Granger, F., Richert, L., Didier, P., Godet, J., Pradeau-Aubreton, K., Emiliani, S., et al. (2013). Structural and functional role of INI1 and LEDGF in the HIV-1 preintegration complex. PLoS ONE 8, e60734.

Murshudov, G.N., Skuba´k, P., Lebedev, A.A., Pannu, N.S., Steiner, R.A., Nicholls, R.A., Winn, M.D., Long, F., and Vagin, A.A. (2011). REFMAC5 for the refinement of macromolecular crystal structures. Acta Crystallogr. D Biol. Crystallogr. 67, 355–367. Pandey, K.K., Bera, S., and Grandgenett, D.P. (2011). The HIV-1 integrase monomer induces a specific interaction with LTR DNA for concerted integration. Biochemistry 50, 9788–9796. Pei, J., Kim, B.H., and Grishin, N.V. (2008). PROMALS3D: a tool for multiple protein sequence and structure alignments. Nucleic Acids Res. 36, 2295– 2300. Savarino, A., Pistello, M., D’Ostilio, D., Zabogli, E., Taglia, F., Mancini, F., Ferro, S., Matteucci, D., De Luca, L., Barreca, M.L., et al. (2007). Human immunodeficiency virus integrase inhibitors efficiently suppress feline immunodeficiency virus replication in vitro and provide a rationale to redesign antiretroviral treatment for feline AIDS. Retrovirology 4, 79. Serrao, E., Thys, W., Demeulemeester, J., Al-Mawsawi, L.Q., Christ, F., Debyser, Z., and Neamati, N. (2012). A symmetric region of the HIV-1 integrase dimerization interface is essential for viral replication. PLoS ONE 7, e45177. Shibagaki, Y., Holmes, M.L., Appa, R.S., and Chow, S.A. (1997). Characterization of feline immunodeficiency virus integrase and analysis of functional domains. Virology 230, 1–10. Sippel, M., and Sotriffer, C.A. (2010). Molecular dynamics simulations of the HIV-1 integrase dimerization interface: guidelines for the design of a novel class of integrase inhibitors. J. Chem. Inf. Model. 50, 604–614. Tintori, C., Demeulemeester, J., Franchi, L., Massa, S., Debyser, Z., Christ, F., and Botta, M. (2012). Discovery of small molecule HIV-1 integrase dimerization inhibitors. Bioorg. Med. Chem. Lett. 22, 3109–3114. Toogood, P.L. (2002). Inhibition of protein-protein association by small molecules: approaches and progress. J. Med. Chem. 45, 1543–1558. Vink, C., van der Linden, K.H., and Plasterk, R.H. (1994). Activities of the feline immunodeficiency virus integrase protein produced in Escherichia coli. J. Virol. 68, 1468–1474. Yang, Z.N., Mueser, T.C., Bushman, F.D., and Hyde, C.C. (2000). Crystal structure of an active two-domain derivative of Rous sarcoma virus integrase. J. Mol. Biol. 296, 535–548.

Merkel, G., Andrake, M.D., Ramcharan, J., and Skalka, A.M. (2009). Oligonucleotide-based assays for integrase activity. Methods 47, 243–248.

Yin, Z., and Craigie, R. (2010). Modeling the HIV-1 Intasome: A Prototype View of the Target of Integrase Inhibitors. Viruses 2, 2777–2781.

Minor, W., Cymborowski, M., Otwinowski, Z., and Chruszcz, M. (2006). HKL3000: the integration of data reduction and structure solution—from diffraction images to an initial model in minutes. Acta Crystallogr. D Biol. Crystallogr. 62, 859–866.

Zhao, Z., McKee, C.J., Kessl, J.J., Santos, W.L., Daigle, J.E., Engelman, A., Verdine, G., and Kvaratskhelia, M. (2008). Subunit-specific protein footprinting reveals significant structural rearrangements and a role for N-terminal Lys-14 of HIV-1 Integrase during viral DNA binding. J. Biol. Chem. 283, 5632–5641.
