Evolutionary and functional analyses of the interaction between the myeloid restriction factor SAMHD1 and the lentiviral Vpx protein

Abstract

SAMHD1 has recently been identified as an HIV-1 restriction factor operating in myeloid cells. As a countermeasure, the Vpx accessory protein from HIV-2 and certain lineages of SIV have evolved to antagonize SAMHD1 by inducing its ubiquitin-proteasome-dependent degradation. Here, we show that SAMHD1 experienced strong positive selection episodes during primate evolution that occurred in the Catarrhini ancestral branch prior to the separation between hominoids (gibbons and great apes) and Old World monkeys. The identification of SAMHD1 residues under positive selection led to mapping the Vpx-interaction domain of SAMHD1 to its C-terminal region. Importantly, we found that while SAMHD1 restriction activity toward HIV-1 is evolutionarily maintained, antagonism of SAMHD1 by Vpx is species-specific. The distinct evolutionary signature of SAMHD1 sheds light on the development of its antiviral specificity.

Figure 1. Characterization of the Atypical Catarrhini Ancestral Branch Length in the SAMHD1 Gene Tree

(A) Amino acid branch lengths inferred for SAMHD1 sequences available for 34 mammals in the OrthoMaM database (Ranwez et al., 2007) as inferred by maximum likelihood under the LG+G+F model implemented in RAxML (Stamatakis, 2006). Only the Scandentia+primates sub-tree is presented. Note the extremely long ancestral branch (0.1032 substitutions per site) leading to Old World monkeys (OWMs, Catarrhini) represented in red. (B) Amino acid branch lengths inferred from a concatenation of 786 1:1 orthologous CDSs available for the same 34 taxa in the OrthoMaM database. Note the much shorter Catarrhini ancestral branch (0.0058 substitutions per site) represented in red. (C) Distribution of the ancestral Catarrhini amino-acid branch length inferred across the 786 1:1 orthologous gene trees showing that SAMHD1 is an extreme outlier with a branch length about 17 times longer than the one inferred from the concatenation. See also Figure S1 and Table S1.

Figure 2. Evolutionary Conservation of SAMHD1 Restriction Activity against HIV-1

(A) SAMHD1 phylogenetic tree for 31 primate species rooted by the treeshrew Tupaia belangeri. The topology was inferred by maximum likelihood from the nucleotide sequences under a GTR+G model using RAxML. This topology is fully congruent with the most recent multigene phylogeny obtained by Perelman et al. (2011). Branch lengths are those inferred by Codeml under the best fitting branch model (Yang and Nielsen, 1998). Note the episode of positive selection along the Catarrhini ancestral branch (ω = 1.30), followed by a decrease in OWMs and hominids (ω = 0.59), whereas New World monkeys (NWMs, Platyrrhini) and other primates show a dN/dS characteristic of purifying selection (ω = 0.26). * indicates sequenced SAMHD1 in this study. Scale is in number of substitutions per codon. (B) The variation of ω in SAMHD1 along the primate phylogeny was jointly reconstructed with divergence times while controlling the effect of three life-history traits (body mass, longevity, and maturity) using the Bayesian framework recently proposed by Lartillot and Poujol (2011). The Bayesian chronogram obtained shows that the positive selection episode experienced by SAMHD1 during primate evolution occurred along the ancestral Catarrihini approximately between 45 and 25 million years ago. Time scale is expressed in million years with vertical lines delimitating the main geological periods (Pli., Pliocene; P., Pleistocene). (C) SAMHD1 expression was induced through retroviral transduction of SAMHD1-silenced THP-1 myeloid cells. Forty-eight hours after transduction, cells were differentiated and infected with a HIV-LUC-G. Luciferase activity was measured 24 hr after infection, normalized for protein levels and represented as relative infection compared to untransduced cells. Mock, untransduced; H1.4R, shRNA-resistant huSAMHD1; Gib, gibbon; Mng, mangabey; Rh, rhesus macaque; Agm, African green monkey; Tam, tamarin; Owl, owl monkey; Gml, gray mouse lemur; OWM, Old World monkeys; NWM, New World monkeys. (D) THP-1 and THP-1 stably expressing Vpx were differentiated on coverslips and infected with a SIVmac239-IRES-eGFP molecular clone with an IRES-eGFP sequence as a reporter (SIVmac239) or with the same virus with a stop codon in the vpx ORF (SIVmac239ΔVpx). Cells were analyzed by immunofluorescence 48 hr after infection for SAMHD1 expression. Nuclei were stained with DAPI. Cells were further analyzed by flow cytometry for the expression of GFP. Results are presented as the percent of GFP-positive cells, with the percent of GFP+ THP-1 cells infected with SIVmac239 set to 1. Error bars represent the standard deviation from the mean. See also Figure S2.

Figure 3. Site-Specific Positive Selection Profile of SAMHD1 in 31 Primates

(A) The dN/dS ratio (ω) of the 626 codons of the human SAMHD1 sequence was inferred under the M8 site model (Nielsen and Yang, 1998) implemented in Codeml (Yang, 2007) from an alignment of 31 primates species with Tupaia belangeri as an outgroup. The dash line represents the dN/dS = 1 limit, above which codons are inferred to be under positive selection. The 17 codons shown in red are those identified as being under positive selection with posterior probability PP > 0.95 under M8 model. The four codons indicated by red stars are those identified using the more conservative M2a site model with PP > 0.95 (Zhang et al., 2005). The position of the two functional SAM and HD domains is indicated. (B) Amino acids in the C-terminal domain of SAMHD1 that are subject to strong positive selection and their distribution among primates is represented. See also Figure S3.

Figure 4. Interaction between Vpx_mac251 and huSAMHD1 through the C-Terminal Domain of SAMHD1

(A) Schematic representation of truncations and point mutants of huSAMHD1 that were engineered in the pOZ retroviral vector. All constructs were C-terminally FLAG-tagged. The limits of SAM and HD domains are indicated. (B) Amino acids 575 to 626 of SAMHD1 are required for Vpx_mac251-induced hSAMHD1 degradation. FLAG-tagged WT-SAMHD1, SAMHD1-HD/AA, SAMHD1-ΔSAM, and SAMHD1-F575 were transfected in 293T cells together or not with HA-tagged Vpx_mac251. Whole-cell extracts were analyzed 48 hr later by immunoblot using anti-FLAG, anti-HA, and anti-tubulin antibodies. (C) Amino acids 595 to 626 of huSAMHD1 are involved in the interaction between Vpx_mac251 and huSAMHD1. Whole-cell extracts from 293T cells transfected to express HA-Vpx_mac251 together with WT-SAMHD1, SAMHD1-F575, or SAMHD1-F595 were FLAG immunoprecipitated. After peptide elution, immunoprecipitates and whole-cell extracts were analyzed by immunoblot with anti-FLAG, anti-HA, and anti-DDB1 antibodies. (D) FLAG-tagged WT-SAMHD1, SAMHD1-F611, SAMHD1-S614A, SAMHD1-V618A, and SAMHD1-M626A were FLAG immunoprecipitated from 293T cells expressing HA-Vpx_mac251. Peptide-eluted immunoprecipitates were analyzed by western blot with anti-HA, anti-FLAG, and anti-tubulin antibodies. See also Figure S4.

Figure 5. Ability of Vpx Alleles from SIV and HIV-2 Strains to Degrade huSAMHD1

(A) THP-1 cells were transduced with retroviral vectors allowing expression of FLAG- and HA-tagged Vpx_mac251, Vpx_mac239, Vpx_Smm, Vpx_Rcm-Ng, Vpx_mnd2, Vpx_drl1, Vpx_ROD, Vpx_2A, and Vpx_2B. Forty-eight hours later, transduced cells were differentiated overnight on polylysine treated coverslips prior to immunostaining with anti-SAMHD1 and anti-HA antibodies. Nuclei are stained in mounting media with DAPI. Arrows point at examples of cells displaying a representative staining. (B) Vpx-positive cells from (A) were counted. Results are expressed as the percent of Vpx-positive cells where Vpx/SAMHD1 costaining, Vpx/SAMHD1 exclusion and cytoplasmic SAMHD1 is observed. (C) THP-1 cells differentiated on coverslips were treated with VLP-Vpx. Cells were fixed and stained with anti-SAMHD1 antibody at 0, 2, and 6 hr after VLP-Vpx exposure. Nuclei are stained in mounting media with DAPI. Arrows point at examples of cells displaying a representative staining. (D) Cells treated as in (A) were infected with HIV-LUC-G. Luciferase activity was measured 24 hr after infection and is expressed as fold increase luciferase activity over untransduced parental THP-1 cells. Error bars represent the standard deviation from the mean. (E) FLAG- and HA-tagged Vpx alleles from selected SIV strains were coexpressed with FLAG-tagged huSAMHD1 or empty vector in 293T cells. Whole-cell extracts were subjected to HA immunoprecipitation prior to peptide elution and analysis by immunoblot with anti-FLAG, anti-HA, and anti-tubulin antibodies. (F) FLAG- and HA-tagged Vpx alleles from HIV-2_ROD and HIV-2B strains were coexpressed with FLAG-tagged huSAMHD1 or empty vector in 293T cells. Whole-cell extracts were subjected to HA immunoprecipitation prior to peptide elution and analysis by immunoblot using anti-FLAG, anti-HA and anti-tubulin antibodies. See also Figure S5.

Figure 6. Vpx-Induced SAMHD1 Degradation Is Species Specific

(A) SAMHD1 expression patterns in the presence of Vpx from HIV-2 and SIV strains. FLAG- and HA-tagged SAMHD1 proteins from a selected panel of primates were coexpressed with various FLAG- and HA-tagged Vpx proteins in HeLa cells. Whole-cell extracts were analyzed 30 hr after transfection by immunoblot with anti-HA and anti-tubulin antibodies. Hu, human; Gib, gibbon; Rh, rhesus macaque; Mng, mangabey; OWM, Old World monkeys; Owl, owl monkey; NWM, New World monkeys; Gml, gray mouse lemur. (B) Vpx_Rcm-Ng and Vpx_mac251 interact with rhesus SAMHD1. FLAG- and HA-tagged Vpx alleles were coexpressed with FLAG-tagged rhesus SAMHD1 or empty vector in 293T cells. Whole-cell extracts were subjected to HA immunoprecipitation prior to peptide elution and analysis by immunoblot with anti-FLAG, anti-HA, and anti-tubulin antibodies. (C) Schematic representation summarizing the data obtained showing that Vpx targeting of SAMHD1 is species specific. HIV-2 Clade B Vpx (discontinuous arrow) degrades huSAMHD1 but with slower kinetics compared to HIV-2_ROD Vpx.