The B30.2(SPRY) Domain of the Retroviral Restricti
http://www.100md.com
病菌学杂志 2005年第10期
Department of Cancer Immunology and AIDS, Dana-Farber Cancer Institute, Department of Pathology, Division of AIDS, Harvard Medical School, Boston, Massachusetts 02115
Laboratory of Genomic Diversity, National Cancer Institute, Frederick, Maryland 21702-1201
SAIC-Frederick, Frederick, Maryland 21702-1201
Department of Immunology and Infectious Diseases, Harvard School of Public Health, Boston, Massachusetts 02115
ABSTRACT
Tripartite motif (TRIM) proteins are composed of RING, B-box 2, and coiled coil domains. Some TRIM proteins, such as TRIM5, also possess a carboxy-terminal B30.2(SPRY) domain and localize to cytoplasmic bodies. TRIM5 has recently been shown to mediate innate intracellular resistance to retroviruses, an activity dependent on the integrity of the B30.2 domain, in particular primate species. An examination of the sequences of several TRIM proteins related to TRIM5 revealed the existence of four variable regions (v1, v2, v3, and v4) in the B30.2 domain. Species-specific variation in TRIM5 was analyzed by amplifying, cloning, and sequencing nonhuman primate TRIM5 orthologs. Lineage-specific expansion and sequential duplication occurred in the TRIM5 B30.2 v1 region in Old World primates and in v3 in New World monkeys. We observed substitution patterns indicative of selection bordering these particular B30.2 domain variable elements. These results suggest that occasional, complex changes were incorporated into the TRIM5 B30.2 domain at discrete time points during the evolution of primates. Some of these time points correspond to periods during which primates were exposed to retroviral infections, based on the appearance of particular endogenous retroviruses in primate genomes. The results are consistent with a role for TRIM5 in innate immunity against retroviruses.
INTRODUCTION
Following entry, retroviruses must negotiate a series of processes to establish a permanent infection of the host cell. These include uncoating of the viral core, reverse transcription, nuclear access, and integration of the viral DNA into the host genome (1, 11, 60). Early postentry restrictions to retrovirus infection can determine tropism at the species level. Infection by N-tropic murine leukemia virus, for example, is inefficient in most human cells and in certain cell lines from African green monkeys (5, 53). Human immunodeficiency virus type 1 (HIV-1) encounters a postentry block in Old World monkeys, whereas simian immunodeficiency virus (SIVmac) is blocked in most New World monkey cells (17, 18, 42). These species-specific restrictions share several features. First, the block occurs prior to or concurrent with reverse transcription, which occurs in the cytoplasm of the host cell. At most, low levels of early reverse transcripts are made in restricted cells (9, 17, 29, 42). Second, the viral determinant of susceptibility to the block is the capsid protein (9, 14, 22, 32, 33, 53). Other capsid-binding proteins, such as cyclophilin A in the case of HIV-1, can modify the degree of restriction (33, 34, 49, 55). Third, restriction is mediated by dominant host factors, the activity of which can be titrated by the introduction of virus-like particles containing proteolytically processed capsid proteins of the restricted viruses (4, 6, 9, 13, 29, 33, 46, 54).
These observations suggested a model in which host restriction factors interact, directly or indirectly, with the viral capsid and prevent its progression along the infection pathway. A genetic screen identified a major restriction factor in monkey cells that acts on HIV-1, and to a lesser extent, on SIVmac (47). The factor, TRIM5rh, was selected from a cDNA library prepared from primary rhesus monkey lung fibroblasts. TRIM5rh was shown to be sufficient to confer potent resistance to HIV-1 infection on otherwise susceptible cells. Moreover, TRIM5rh was necessary for maintenance of the block to the early phase of HIV-1 infection in Old World monkey cells, as demonstrated by interference with the expression of the endogenous TRIM5 ortholog in these cells (47). HIV-1 infection in cells expressing TRIM5rh was blocked at the earliest stage of reverse transcription (47). Cells expressing TRIM5rh exhibited a partial inhibition of SIVmac infection but were as susceptible as control cells to infection by Moloney murine leukemia virus vectors. Humans express a protein, TRIM5hu, that is 87% identical in amino acid sequence to the rhesus monkey protein TRIM5rh (47). Even when expressed at comparable levels, TRIM5hu was less potent in suppressing HIV-1 and SIVmac infections than TRIM5rh (47). Recently, TRIM5hu was shown to be responsible for the postentry restriction of N-tropic murine leukemia virus (N-MLV) in human cells (15, 20, 34, 63). TRIM5rh was much less effective than TRIM5hu at blocking this murine leukemia virus. Thus, TRIM5hu potently restricts N-MLV, specifies an intermediate level of resistance to HIV-1, and does not affect SIVmac infection. In contrast, TRIM5rh potently restricts HIV-1 and exhibits a modest inhibition of SIVmac and N-MLV infections. The TRIM5 protein from African green monkeys, TRIM5agm, has recently been shown to inhibit N-MLV, HIV-1, and SIVmac infections (15, 20, 43, 63). These observations underscore the importance of species-specific variations in TRIM5 orthologs for their ability to restrict infections by particular retroviruses.
More than 50 genes encode tripartite motif (TRIM) proteins (36). The tripartite motif includes a RING domain, a B-box 2 domain, and a coiled coil (cc) domain; TRIM proteins have also been called RBCC proteins. Some TRIM proteins, including TRIM5, contain a C-terminal B30.2 or SPRY domain (16, 36). Differential splicing of the TRIM5 primary transcript gives rise to the expression of several isoforms of the protein product (36). The TRIM5 isoform is the largest product (493 amino acid residues in humans) and contains the B30.2/SPRY domain. The other TRIM5 isoforms ( and are the best substantiated of these) lack an intact B30.2/SPRY domain. The TRIM5rh isoform does not inhibit HIV-1 or SIVmac infection (47). In fact, TRIM5rh has been shown to exhibit a weak dominant-negative activity, repressing the ability of wild-type TRIM5rh to inhibit HIV-1 infection (47). Thus, the B30.2 domain is critical for the ability of TRIM5 to mediate antiretroviral effects.
A common feature of TRIM proteins is the ability to assemble into large complexes in the cytoplasm or nucleus of the cell (36). The function of these cytoplasmic and nuclear bodies is largely unknown. The demonstration of a potent, specific antiretroviral activity for TRIM5 raised the possibility that the major function of some TRIM proteins is the establishment of innate immunity to infectious agents. The localization of TRIM5 to cytoplasmic bodies (36, 62) is consistent with its ability to block retroviral infection shortly after entry of the viral capsid into the cytoplasm of the host cell.
Studies of endogenous retroviral sequences have indicated that vertebrates, including humans, have been exposed to retroviruses for many millions of years (2, 12, 19, 21, 25, 26, 28, 35, 44, 45, 56, 58, 59, 61). This long history of host-retrovirus coevolution might favor the selection of particular TRIM5 proteins that effectively suppress lethal retrovirus infections. Since many viruses preferentially infect particular host species, the selection of molecules involved in antiviral immunity often occurs in a lineage-specific manner. In this report, we characterize the TRIM5 proteins of several primate species and provide evidence of lineage-specific variation in particular B30.2 domain elements.
MATERIALS AND METHODS
Cells for TRIM5 cDNA synthesis. SQMK-FP (Pindak) cells from a Bolivian squirrel monkey (Saimiri boliviensis boliviensis), Vero cells from an African green monkey (Cercopithecus aethiops pygerythrus), OMK cells from an owl monkey (Aotus trivirgatus), and human HeLa cells were obtained from the American Type Culture Collection. Primary rhesus lung fibroblasts were derived from a rhesus macaque (Macaca mulatta) (18). The Coriell Institute for Medical Research (Camden, N.J.) supplied the following cells: AG05352 cells from a black-handed spider monkey (Ateles geoffroyi), AG05308 cells from a red-chested mustached tamarin (Saguinus labiatus), AG06116 cells from a patas monkey (red guenon) (Erythrocebus patas), AG06209 cells from a Sumatran orangutan (Pongo pygmaeus abelii), AG05251 cells from a Western lowland gorilla (Gorilla gorilla), and GM03448 cells from a chimpanzee (Pan troglodytes verus).
Samples for DNA analysis. Nonhuman primate DNAs from a previously described collection of samples (2, 3) included two independent samples from squirrel monkeys (B263 [female] and B26) (Saimiri sciureus), samples from two baboon species (B21 [Papio cynocephalus] and B856 and B1542 [Papio anubis]), a patas monkey sample (B1530) (Erythrocebus patas), a grivet sample (B1532 [female]) (Cercopithecus sabaeus), two sooty mangabey samples (B400 [female] and B859) (Cercocebus atys), a gelada sample (B853) (Theropithecus gelada), two colobus samples (B130 [Colobus guereza] and B1527 [Colobus guereza caudatus]), samples from two species of langur (B131 [female] [Presbytis obscurus] and B401 [male] [Presbytis senex]), samples from two species of gibbon (B23, B1533, and B837 [Hylobates lar] and B128 [Hylobates concolor]), and samples from gorillas (B27 [female], B461, and B1535 [female]) (Gorilla gorilla). Additional samples of patas monkey (Erythrocebus patas), white-cheeked gibbon (Hylobates concolor), and gorilla (Gorilla gorilla) DNA were obtained from the core collection of the Laboratory of Genomic Diversity, National Cancer Institute.
TRIM5 cDNA cloning and sequencing. First- and second-strand cDNA synthesis was performed by use of a cDNA synthesis kit (Clontech), using RNAs prepared from cells derived from an African green monkey, a rhesus macaque, a Bolivian squirrel monkey, an owl monkey, a patas monkey, a tamarin, a spider monkey, an orangutan, a gorilla, and a chimpanzee (see above). Human and primate TRIM5 cDNAs encoding TRIM5 were amplified with the primers TRIMf2 (5'-GCGGAATTCGCCATGGCTTCTGGAATCCTGGTT-3') and TRIMr2 (5'-GCGATCGATGCCTCAAGAGCTTGGTGAGCACAG-3').
Amplification was carried out by use of a Clontech Advantage PCR kit, with thermocycling of the reactions at 95°C for 30 s, followed by 30 cycles of 95°C for 30 s, 55°C for 1 min, and 68°C for 3 min. Amplified cDNAs were inserted into the pCR-BluntII-TOPO plasmid (Invitrogen) and sequenced by use of the following primers: M13f (–20), GTAAAACGACGGCCAGT; M13r (–21), AACAGCTATGACCATG; TRIMf3, GGAAGCTGACATCAGAGA; TRIMf4, GATAAGAGACAAGTGAGC; TRIMr3, TCTACCTCCCAGTAATG; and TRIMr4, TCCTTCTCCAGGTTTTGC.
PCR amplification of human TRIM5 exon 8 from genomic DNA. Thirty nanograms of individual DNAs from the primates described above was amplified with 5 U of AmpliTaq Gold enzyme in 3.5 mM MgCl2 and 330 μM nucleoside triphosphates, using 200 nanomoles each of the following primers containing an M13 forward or reverse tag: TRIM5Ex8SeqF, 5'-GTAAAACGACGGCCAGTTCCCTTAGCTGACCTGTTAATTT-3'; and TRIM5Ex8SeqR, 5'-GGAAACAGCTATGACCATGGCTGTACAGAAGGGGCTGAG-3'. The following cycling conditions were used: an initial 94°C, 5-min activation of AmpliTaq Gold, followed by 20 cycles of 94°C for 20 seconds, 54°C for 20 seconds, and 68°C for 1 min and then an additional 20 cycles of 94°C for 20 seconds, 65°C for 20 seconds, and 68°C for 1 min, a final extension of 68°C for 7 min, and a hold overnight, if necessary, at 4°C.
DNA sequencing. Amplicons were treated with exonuclease I and shrimp alkaline phosphatase and subjected to Big Dye cycle sequencing in the presence of either an M13 forward or reverse primer tag (m13 forward, 5'-GTAAAACGACGGCCAGT-3'; m13 reverse, 5'-GGAAACAGCTATGACCATG-3'). Sequenced reactions were cleaned with Sephadex G-50, dried on a thermocycler (95°C without caps for 20 min), and resuspended in undiluted formamide (Applied Biosystems) in preparation for capillary electrophoresis on an ABI 3730 HT apparatus. Subsequent to sequencing, forward and trace files were examined with Mutation Explorer software (Softgenetics) and sequences were visually determined.
Phylogenetic analysis. The data sets for the complete TRIM5 protein and the B30.2(SPRY) domain were examined separately. Predicted amino acid sequences were compiled and aligned for subsequent phylogenetic analyses by ClustalX (51) and were verified visually. Phylogenetic reconstruction was performed with MEGA, version 2.0, software (23) using the following methods: minimum evolution, neighbor joining, and maximum parsimony. A bootstrap analysis using 1,000 iterations was performed with each method (38).
Nonsynonymous/synonymous variation. The nonsynonymous/synonymous (Ka/Ks) ratios at various codon positions for pairwise comparisons of TRIM5 cDNAs were calculated by the method of Li (24). The Ka/Ks ratio was estimated as a rolling average for a window of 150 codons, with the center of the window being moved codon-by-codon to produce a plot showing local variations in the degree of sequence conservation.
Nucleotide sequence accession numbers. The complete primate TRIM5 coding sequences have been deposited in the GenBank database under accession numbers AY740612 to AY740621. The primate TRIM5 nucleotide sequences have been deposited in the GenBank database under accession numbers AY710295 to AY710304.
RESULTS
Variation among TRIM proteins related to TRIM5. TRIM proteins first appeared with the metazoans and dramatically expanded in number during vertebrate evolution (36). Thus, current TRIM genes presumably arose from a small number of ancestral genes. To gain insight into the conserved and variable elements of TRIM proteins, we aligned the primary amino acid sequences of several TRIM proteins. The TRIM proteins most closely related to TRIM5hu (as ascertained by BLAST analysis) were selected for study. Nearly all of these TRIM proteins contain a B30.2 domain, and many have been shown to localize, at least in part, to the cytoplasm of the cell (36). Several of these TRIM proteins are found in cytoplasmic bodies, as is TRIM5 (36).
The predicted amino acid sequences of TRIM proteins related to TRIM5hu are aligned in Fig. 1. The sequences of several TRIM5 orthologs from nonhuman primates that were determined in this study were included in the alignment (see below). A moderate degree of variation was observed in all of the domains of these selected TRIM proteins (Fig. 1). The existence of four regions within the B30.2 domain that exhibited substantial variation in length as well as extensive amino acid differences among the TRIM proteins was particularly noteworthy. We designated these variable regions v1, v2, v3, and v4 (Fig. 1). Because of the modest level of amino acid variation among the TRIM B30.2 domains, the boundaries of these variable regions are somewhat arbitrary. We chose reasonably conserved sets of boundary residues to eliminate any ambiguity in defining the length of a particular variable region. This allowed us to examine the lengths of the B30.2 domain variable regions in different TRIM proteins, including those which are less closely related to TRIM5 (Table 1). Based on length variations in the B30.2 v1, v2, and v3 regions, two groups of TRIM proteins can be discerned. For one group, consisting of TRIM21, BIA2, mouse LOC216781, rat LOC303167, RNF137, TRIM17, TRIM11, TRIM10, TRIM26, TRIM27, TRIM20, TRIM39, and TRIM50A, each B30.2 variable region conforms to a narrow range of lengths. The v1 regions of these TRIM proteins are 19 to 21 residues long, the v2 regions are 11 to 13 residues long, and the v3 regions are 20 to 23 residues long. An examination of the available sequences of TRIM10, TRIM11, TRIM20, TRIM26, TRIM27, TRIM39, and TRIM50A from different mammalian species indicated that the B30.2 domain sequences and the variable region lengths are relatively well conserved (data not shown). The B30.2 domain v1, v2, and v3 regions of TRIM21 and the v3 region of TRIM17 exhibit more amino acid variability between humans and rodents (data not shown). Nonetheless, the lengths of the TRIM17 and TRIM21 variable regions are similar in humans and rodents (Table 1 and data not shown). Thus, one group of TRIM proteins exhibits a narrow range of B30.2 variable region lengths and, in many cases, minimal variation between human and rodent orthologs. Based on the sequences of the TRIM proteins in this group, consensus TRIM B30.2 domain variable region lengths can be defined as 19 to 21 residues for v1, 11 to 13 residues for v2, and 20 to 23 residues for v3.
For the second group of TRIM proteins, the lengths of the B30.2 variable regions differ from those found for the group described above. This second group includes TRIM4, TRIM5, TRIM6, TRIM22, TRIM34, rat LOC308906, and mouse 9230105E10RiK. Compared with the consensus B30.2 variable loop lengths, TRIM4 exhibits a long v1 region, and TRIM6, TRIM22, TRIM30, TRIM34, rat LOC308906, and mouse 9230105E10RiK exhibit long v2 and v3 regions. Relative to the consensus, the primate TRIM5 proteins exhibit differences in the lengths of v1, v2, and v3. It is noteworthy that a group of related TRIM proteins (TRIM5, TRIM6, TRIM22, TRIM30, TRIM34, rat LOC308906, and mouse 9230105E10RiK) all share the property of increased B30.2 variable region lengths compared to the consensus. The human TRIM5, TRIM6, TRIM22, and TRIM34 genes are located in a gene cluster at 11p15.4. In rodents, TRIM6 and TRIM34 are adjacent to the rat LOC308903 and mouse 9230105E10RiK genes. This clustering may indicate recent paralogous relationships among these TRIM genes. In the available TRIM6 and TRIM34 sequences from other mammals, the B30.2 variable region lengths are similar to those found for the human proteins (Table 1), even though considerable amino acid variation is observed in these regions (data not shown). The human TRIM4 and TRIM22 genes do not have orthologs in the rodent genome, and conversely, TRIM30 is not found in the human genome (data not shown). Thus, in most cases for which data are available, the B30.2 variable region lengths are preserved in TRIM proteins from different species. TRIM5 and the proteins encoded by the rat LOC308906 and mouse 9230105E10RiK genes are exceptions (Table 1; see below).
The v4 variable regions of the B30.2 domains of the TRIM proteins examined are short, varying from 4 to 10 residues in length. The v4 variable region is five residues long in all of the TRIM proteins closely related to TRIM5 (Table 1).
Interspecies variation in the TRIM5 protein. Variations among TRIM5 proteins in primates affect the efficacy of restriction of particular retroviruses (15, 20, 34, 43, 47, 63). Ancient retroviral infections encountered by the members of specific primate lineages may have exerted selective pressure on the TRIM5 structure. To investigate this possibility, we cloned and sequenced the TRIM5 cDNAs encoding the complete TRIM5 proteins of several primate species. These included apes (chimpanzee, gorilla, and orangutan), Old World monkeys (rhesus macaque, African green monkey, and patas monkey), and New World monkeys (squirrel monkey, tamarin, and spider monkey). To ensure that these sequences were representative, we determined the sequences of several TRIM5 cDNAs from individual humans, rhesus macaques, African green monkeys, and squirrel monkeys. The variation among the predicted TRIM5 proteins of individuals within a species was <2.5% (data not shown).
We also examined the mouse and rat genomes for potential TRIM5 orthologs. The rodent proteins most closely related to TRIM5 are encoded by TRIM12 (mouse), 9230105E10RiK (mouse), and LOC308906 (rat). Reciprocally, TRIM5 is the human protein most closely related to the protein products of these rodent genes (data not shown). TRIM12 and 9230105E10RiK are essentially identical in their 5' portions and have been assembled at adjacent loci on mouse chromosome 7. It is possible that the current assembly will be revised and that TRIM12 and 9230105E10RiK are in fact the same gene. It is also possible that these genes are paralogs that arose from a duplication event after the divergence of mice and rats. An examination of current drafts of the chicken and dog genomes failed to identify candidate TRIM5 orthologs.
An alignment of the predicted amino acid sequences of the TRIM5-related proteins of rodents and several primates is shown in Fig. 1. The degree of variation among the TRIM5-related proteins of these different species was larger than that observed within the individuals of a species (data not shown). The full-length protein product of the 9230105E10RiK gene is only 49% identical to human TRIM5. This is lower than the 80 and 67% identities seen when the human and rodent versions of the TRIM6 and TRIM34 proteins, respectively, are compared. In fact, the degree of sequence identity between the human TRIM5 and rodent 9230105E10RiK proteins is not higher than the similarity among TRIM5 and other TRIM proteins encoded by the human TRIM cluster located at 11p15.4. Whether the mouse 9230105E10 and rat LOC308906 genes represent TRIM5 orthologs will be addressed elsewhere. At a minimum, the rodent TRIM5-related proteins exhibit substantial differences from TRIM5hu. Thus, our analysis of interspecies variation among TRIM5 proteins will focus mainly on primate sequences.
Full-length TRIM5 cDNA sequences from 10 primate species were used to construct a phylogenetic tree (Fig. 2A). Bootstrap simulation using neighbor joining, minimum evolution, and maximum parsimony algorithms all resolved the New Word/Old World dichotomy and the great ape/Old World monkey dichotomy with a high confidence. Patas monkeys were differentiable from African green monkeys and rhesus monkeys in comparisons of the full-length TRIM5 sequences; comparisons of the complete TRIM5 sequences for this particular branch differed somewhat from comparisons between rhesus macaques and related primates when only the B30.2 regions were compared (see below).
Interspecies variation in TRIM5-like proteins was observed in all four of the defined TRIM5 domains, with the most variation occurring in the B30.2 domain (Table 2). To assess B30.2 domain variation more thoroughly, we amplified DNA fragments corresponding to TRIM5 exon 8, which encodes the TRIM5 B30.2 domain, from several nonhuman primates and then sequenced the fragments. Attempts to amplify TRIM5 exon 8 sequences from two prosimian species, lemurs and galagoes, were unsuccessful (data not shown). As expected from our analysis of TRIM proteins, the observed variation within the B30.2 domains of primate TRIM5 proteins was not uniform, being concentrated within the v1, v2, and v3 variable regions defined above (Fig. 1). In the TRIM5 proteins of the primates examined, all three of these variable B30.2 regions differed in length from the consensus TRIM variable region lengths (Fig. 2B). With the sole exception of the spider monkey TRIM5 protein, in which v2 is 15 residues long, the primate TRIM5 v2 regions are 17 residues in length. Thus, although the v2 region exhibits a high level of substitution of individual amino acids among primate TRIM5 proteins, the length of this region is well conserved. In contrast, a striking lineage-specific variation in the lengths of the v1 and v3 variable regions of the primate TRIM5 proteins was observed (Fig. 2B). The v1 regions of the New World monkeys are 17 residues long, slightly shorter than the consensus v1 length. In contrast, the v1 regions of all of the Old World (catarrhine) primates are longer than that of the consensus. Since a v1 length of 26 residues is found in all hominoids and in the Colobinae subfamily of Old World monkeys, this may represent the v1 length of the TRIM5 proteins of the shared catarrhine ancestor. In the Cercopithecinae subfamily of Old World monkeys, the v1 length is expanded to 28 residues, and in the African green monkey/grivet/patas monkey lineage, it is expanded to 46 residues. The 46-residue v1 region of TRIM5 from these monkeys contains two tandem 20-residue repeats (Fig. 3, top). These are perfect repeats in some individual African green monkeys, but not in others (data not shown). Thus, the TRIM5 B30.2 v1 region appears to have undergone some expansion in all Old World primates. In certain lineages, this expansion is dramatic and involves tandem sequence duplication.
The B30.2 v3 region is 32 residues long in the TRIM5 proteins of all Old World (catarrhine) primates. This is longer than the consensus v3 length and likely represents the v3 length of the TRIM5 protein of the shared catarrhine ancestor. The TRIM5 B30.2 v3 region is longer in the New World (platyrrhine) monkeys than in the Old World monkeys. The v3 region is 41 residues long in all of the New World monkey TRIM5 sequences examined, except for that of the spider monkey. The B30.2 v3 region in spider monkeys has expanded to 96 residues in length. This remarkable expansion results from the presence of three imperfect tandem repeats, of 30, 31, and 24 residues (Fig. 3, bottom). Thus, the TRIM5 B30.2 v3 region appears to have expanded in New World (platyrrhine) monkeys. In at least one lineage, that of spider monkeys, the expansion is dramatic and involves tandem sequence duplication.
Analysis of nonsynonymous/synonymous variation in TRIM5 genes from different species. Analyses of nonsynonymous/synonymous variations (Ka/Ks ratios) can provide insight into selection for or against a change in the coding capacity of a gene (24). However, this approach cannot accommodate sequences that exhibit large insertions or deletions (indels) relative to one another. Therefore, for our comparison of TRIM5 sequences from different primates, the indels involving the v1 and v3 regions of the B30.2 domain were excluded from the analysis. The Ka/Ks ratios at various codon positions were calculated for pairwise comparisons of TRIM5 cDNAs from a hominoid (human), an Old World monkey (patas monkey), and two New World monkeys (tamarin and spider monkey), providing several phylogenetically independent paired comparisons (Fig. 4). For the sequences encoding the N-terminal half of TRIM5 (the RING, B-box 2, and coiled coil domains), the Ka/Ks ratio was generally <1, indicating selection against amino acid changes. For the sequences encoding the B30.2 domain, the Ka/Ks ratio increased dramatically, reaching values above 5 in some cases. Such high Ka/Ks ratios are strongly suggestive of selectively driven diversity in the B30.2 domain.
DISCUSSION
The functions of most TRIM proteins are not understood (36). The TRIM genes arose in bilaterans and expanded in number during vertebrate evolution in response to as yet unknown selective forces. Several observations are consistent with the possibility that some of these proteins contribute to the establishment of innate intracellular defense systems against infectious agents. The TRIM5 proteins of rhesus monkeys and humans block infections by HIV-1 and murine leukemia viruses, respectively (15, 20, 34, 47, 63). The African green monkey TRIM5 protein exhibits activity against a diverse group of retroviruses (15, 20, 43, 63). Some antiviral activity has also been reported for other TRIM family members (7, 8, 27, 57), although this activity lacks the potency and degree of specificity observed for TRIM5 against particular retroviruses. Finally, some TRIM genes are inducible by interferon (8, 31, 37, 52), suggesting a link to other elements of the innate immune system.
The above considerations raise the possibility that infectious agents influenced the evolution of certain TRIM proteins. Length variation in particular segments of different TRIM B30.2 domains were examined as one potential indicator of such selection. Our observations are consistent with a parsimonious model in which ancestral TRIM proteins possessed B30.2 domains with relatively short variable regions. In such a model, the ancestral TRIM B30.2 domain possessed a v1 region of 19 to 21 amino acids, a v2 region of 11 to 13 residues, and a v3 region of 20 to 23 residues, the variable region lengths found in many examples of extant TRIM proteins. For some cytoplasmic TRIM proteins such as TRIM21, amino acid variation, but not length variation, in the B30.2 variable regions occurred during the evolution of rodents and primates. For TRIM4 and the TRIM proteins encoded by the human 11p15.4 cluster (TRIM5, TRIM6, TRIM22, and TRIM34), expansion of the length of one or more B30.2 variable regions apparently occurred at various times in mammalian evolution. For TRIM6 and TRIM34, the mutational events that caused the increased lengths of the v2 and v3 regions likely occurred before the divergence of rodents and primates. Once fixed, these variable region lengths were apparently maintained within narrow limits for at least 90 million years. In contrast, the evolution of different B30.2 variable region lengths in TRIM5 continued during the diversification of primates (see below). Since TRIM4 and TRIM22 are found in humans but not rodents, the emergence or loss of these genes must have occurred after the rodent-primate divergence.
The TRIM5 B30.2 domain exhibits length variations in particular primate lineages. The shorter versions of the TRIM5 B30.2 domain v1, v2, and v3 regions among the primate species examined exhibit lengths of 17 residues, 15 to 17 residues, and 30 to 32 residues, respectively. These v2 and v3 regions are longer than the corresponding regions of the hypothetical ancestral TRIM protein, suggesting that some expansion of these regions may have occurred in a TRIM5 ancestor prior to the diversification of primates. Notably, expansion of the v2 and v3 regions characterizes all of the TRIM proteins encoded by the human 11p15.4 cluster; it seems plausible that these related TRIM proteins share a common ancestor that exhibited v2 and v3 regions which were longer than those of the putative ancestral TRIM protein. Concomitant with the evolution of primates, the TRIM5 B30.2 domain apparently underwent episodic expansions of either the v1 or v3 variable region in a lineage-specific fashion. These v1 or v3 expansions occurred at discrete points in the TRIM5 phylogeny (Fig. 2B, arrows). The evolution of a v1 length of 26 residues apparently occurred at the root of the Old World (catarrhine) branch of primates. A further expansion of v1 to 28 residues occurred with the evolution of the subfamily Cercopithecinae, beginning 9 to 13 million years ago. A final v1 expansion to 46 residues evolved with the African green and patas monkeys.
In the New World (platyrrhine) branch of monkeys, the TRIM5 B30.2 domains evolved long v3 regions. At the root of this lineage, a v3 expansion to 41 residues occurred. A further expansion to 96 residues occurred in spider monkeys. Remarkably, the longest B30.2 variable regions (v1 in African green monkeys and patas monkeys and v3 in spider monkeys) contain tandem duplications. Doublets, defined as short duplications of between 25 and 100 bp, are common in the genomes of mammals and other eukaryotes (50). Tandem doublets, such as those that encode the B30.2 variable region duplications observed in our study, tend to have arisen recently in evolution (50). It has been speculated that double-stranded breaks that are filled in and nonhomologously recombined might be the origins of adjacent doublets (50). Whatever the mechanism by which they arose, the resulting long variable regions in TRIM5 were preserved and became prevalent in the respective monkey species. TRIM5 proteins with long B30.2 variable regions possibly provide better protection against retroviral infections. Indeed, the TRIM5 proteins of African green monkeys and spider monkeys exhibit antiviral activities against a diverse group of retroviruses (15, 20, 43, 63); it is possible that the tandem duplications in the B30.2 variable regions contribute to this breadth.
Standard approaches to assess the selection of genes (e.g., comparisons of the ratios of nonsynonymous/synonymous changes) typically ignore insertions and deletions that result in length variations. An analysis of Ka/Ks ratios was useful for assessing the evolution of TRIM5 sequences outside the B30.2 v1 and v3 variable regions. Pairwise comparisons of TRIM5 cDNAs from four species, representing hominoids, Old World monkeys, and New World monkeys, revealed a similar pattern. For the 5' portion of the TRIM5 cDNA, which encodes the RING, B-box 2, and coiled coil domains of TRIM5, the Ka/Ks ratios were generally <1. This indicates that purifying selection has operated on these TRIM5 domains to preserve the amino acid sequences and, presumably, function. This conclusion is consistent with the conservation of amino acid residues known to be important for the integrity of these domains in the TRIM5 proteins from the different primate species examined (Fig. 1). The demonstrated functional activity of TRIM5 proteins derived from different primate species in restricting particular retroviruses (15, 20, 43, 47, 63) also supports the conservation of TRIM5's function during primate evolution. Even in owl monkeys, in which a retrotransposition has resulted in a loss of the ability to encode a complete TRIM5 protein and instead a TRIM5-cyclophilin A fusion protein is produced, the capacity to restrict some retroviruses is preserved (30, 39).
In contrast to the results for the 5' portion of TRIM5, the Ka/Ks ratios were >1 for the 3' end of TRIM5, which encodes the B30.2 domain. Thus, even when the v1 and v3 variable regions, which exhibit significant variations in individual amino acid residues as well as in their lengths, were excluded from consideration, the analysis strongly suggested that diversity within primate TRIM5 B30.2 domains has been driven by selection. In summary, evolutionary forces have operated to preserve the amino acids in the N-terminal portion and to diversify the amino acids in the B30.2 domain of primate TRIM5 proteins.
Although our analysis of Ka/Ks ratios suggests that selection has influenced the TRIM5 gene segments flanking the indels, the selective nature of the inserts themselves (i.e., whether an increased length is deleterious, neutral, or positively selected) requires further interpretation. The probability of fixation of an insert will be less than, equal to, or more than its creation rate through mutation when selection of the insert is negative, absent, or positive, respectively. The frequency of indels is known to be considerably lower than that of point substitutions (20A, 59A). Thus, the coincident fixation of several TRIM5 inserts during a short period of primate evolution is indicative of either an unusually high rate of indel creation or the presence of positive selection favoring fixation of the insert. Because of evidence of selection in flanking regions and the relative stability of the indel regions in TRIM5 genes over long evolutionary periods, we consider positive selection to be a more likely explanation for the fixation of the inserts.
Beginning at discrete time points spaced over millions of years, TRIM5 B30.2 variable regions of particular lengths, and presumably with particular amino acid sequences as well, were sufficiently advantageous that the corresponding TRIM5 genes became prevalent within the species. Episodic waves of lethal retrovirus infections may have contributed to this evolutionary pattern. The timing of some ancient retroviral epidemics has been deduced from the presence of endogenous retroviruses in the germ lines of mammalian species (2, 12, 19, 21, 25, 26, 28, 35, 44, 45, 58, 59, 61). Notably, several endogenous retroviruses appeared in primate germ lines during the same time periods in which two of the expansions of v1 length in the catarrhine TRIM5 proteins seem to have occurred. Between 25 and 40 million years ago, when Old World primates diverged from New World monkeys, several endogenous retroviruses (human endogenous retrovirus type E [HERV-E], HERV-W, HERV-K, and ERV-9) were introduced into primate genomes (21, 28, 35, 44, 45, 61). This corresponds to the period in which the TRIM5 B30.2 v1 region length apparently expanded to 26 residues (Fig. 2B). About 9 to 13 million years ago, the simian endogenous retrovirus (SERV) and Papio cynocephalus endogenous retrovirus were introduced into the genomes of cercopithecine monkeys (26, 58, 59). The TRIM5 B30.2 v1 expansion to 28 residues occurred during this same period (Fig. 2B). Later TRIM5 v1 expansions to 46 residues in the African green monkey/patas monkey lineage may also be related to retroviral infections. SIV infections are currently prevalent in these monkeys (10, 41), and SIV represents a candidate retrovirus that could have applied selective pressure on TRIM5. However, given the absence of endogenous SIV proviruses, the extent of such infections in ancient times is unknown.
TRIM5 v3 expansion in New World monkeys could also involve ancient retroviral infections. The HERV-S and HERV-F endogenous retroviruses were introduced into primate germ lines between 32 and 56 million years ago, encompassing the period of divergence of the New World and Old World monkeys (61). Further studies may reveal additional candidate retroviruses that potentially influenced platyrrhine TRIM5 proteins. Moreover, infectious agents beyond the retrovirus group may conceivably have influenced TRIM protein evolution.
The v1 region of the TRIM5 B30.2 domain has undergone extensive changes during the evolution of Old World primates. In this light, it is noteworthy that the major determinant of anti-HIV-1 potency distinguishing the human and rhesus monkey TRIM5 proteins was recently mapped to the B30.2 v1 region (48). Further studies will be needed to understand how the B30.2 variable regions might contribute to TRIM5 antiretroviral activity. Length expansions of the degree observed in our survey of primate TRIM5 sequences must be compatible with the fold of the B30.2 domain and thus are probably surface elements. The structure of the B30.2 domain is not known, although it has been speculated that it might be immunoglobulin-like (40). Surface-exposed variable elements of the B30.2 domain are reminiscent of the complementarity-determining region loops of immunoglobulins. Although we were not able to align the B30.2 variable segments and immunoglobulin CDRs in an instructive manner (data not shown), the TRIM B30.2 variable regions may play roles in the recognition of foreign ligands such as viral capsids.
The evolutionary relationships among TRIM genes in diverse vertebrate species require additional investigation and may indicate functional interactions among the members of this interesting family of proteins. In parallel, the ability of TRIM proteins to modulate infections by viruses associated with these vertebrates should be investigated.
ACKNOWLEDGMENTS
We thank Yvette McLaughlin and Sheri Farnum for manuscript preparation and William Murphy and Jill Slattery for helpful discussions.
This work was supported by a grant from the National Institutes of Health (HL54785) and by a Center for AIDS Research award (P30 AI28691). We also acknowledge the support of the Bristol-Myers Squibb Foundation, the International AIDS Vaccine Initiative, and the late William F. McCarty Cooper. This work was funded in part with federal funds from the Center for Cancer Research of the National Cancer Institute and the National Institutes of Health under contract no. NO1-CO-12400.
The content of this publication does not necessarily reflect the views of the Department of Health and Human Services, nor does mention of trade names, commercial products, or organizations imply endorsement by the U.S. Government.
REFERENCES
Artes, E. J., and M. A. Wainberg. 1996. Human immunodeficiency type 1 reverse transcriptase and early events in reverse transcription. Adv. Virus Res. 46:97-163.
Benveniste, R. E., R. Heinemann, G. L. Wilson, R. Callahan, and G. J. Todaro. 1974. Detection of baboon type C viral sequences in various primate tissues by molecular hybridization. J. Virol. 14:56-67.
Benveniste, R. E., L. Kuller, S. T. Roodman, S. L. Hu, and W. R. Morton. 1993. Long-term protection of macaques against high-dose type D retrovirus challenge after immunization with recombinant vaccinia virus expressing envelope glycoproteins. J. Med. Primatol. 22:74-79.
Besnier, C., Y. Takeuchi, and G. Towers. 2002. Restriction of lentivirus in monkeys. Proc. Natl. Acad. Sci. USA 99:11920-11925.
Besnier, C., L. Ylinen, B. Strang, A. Lister, Y. Takeuchi, S. P. Goff, and G. J. Towers. 2003. Characterization of murine leukemia virus restriction in mammals. J. Virol. 77:13403-13406.
Bieniasz, P. D. 2003. Restriction factors: a defense against retroviral infection. Trends Microbiol. 11:286-291.
Bonilla, W. V., D. D. Pinschewer, P. Klenerman, V. Rousson, M. Gamboli, P. P. Pandolfi, R. M. Zinkernagel, M. S. Salvato, and H. Hengartner. 2002. Effects of promyelocytic leukemia protein on virus-host balance. J. Virol. 76:3810-3818.
Chee, A. V., P. Lopez, P. P. Pandolfi, and B. Roizman. 2003. Promyelocytic leukemia protein mediates interferon-based anti-herpes simplex virus 1 effects. J. Virol. 77:7101-7105.
Cowan, S., T. Hatziioannou, T. Cunningham, M. A. Muesing, H. G. Gottlinger, and P. D. Bieniasz. 2002. Cellular inhibitors with Fv1-like activity restrict human and simian immunodeficiency virus tropism. Proc. Natl. Acad. Sci. USA 99:11914-11919.
Daniel, M. D., N. L. Letvin, N. W. King, M. Kannagi, P. K. Sehgal, R. D. Hunt, P. J. Kanki, M. Essex, and R. C. Desrosiers. 1985. Isolation of T-cell tropic HTLV-II-like retrovirus from macaques. Science 228:1201-1204.
Freed, E. O. 1998. HIV-1 gag proteins: diverse functions in the virus life cycle. Virology 251:1-15.
Goodchild, N. L., D. A. Wilkinson, and D. L. Mager. 1993. Recent evolutionary expansions of a subfamily of RTVL-H human endogenous retrovirus-like elements. Virology 196:778-788.
Hatziioannou, T., S. Cowan, S. P. Goff, P. D. Bieniasz, and G. J. Towers. 2003. Restriction of multiple divergent retroviruses by Lv1 and Refl. EMBO J. 22:385-394.
Hatziioannou, T., S. Cowan, U. K. von Schwedler, W. I. Sundquist, and P. Bieniasz. 2004. Species-specific tropism determinants in the human immunodeficiency virus type 1 capsid. J. Virol. 78:6005-6012.
Hatziioannou, T., D. Perez-Caballero, A. Yang, S. Cowan, and P. D. Bieniasz. 2004. Retrovirus resistance factors Ref1 and Lv1 are species-specific variants of TRIM5. Proc. Natl. Acad. Sci. USA 101:10774-10779.
Henry, J., I. H. Mather, M. F. McDermott, and P. Pontarotti. 1998. B30.2-like domain proteins: update and new insights into a rapidly expanding family of proteins. Mol. Biol. Evol. 15:1696-1705.
Himathongkham, S., and P. A. Luciw. 1996. Restriction of HIV-1 (subtype B) replication at the entry step in rhesus macaque cells. Virology 219:485-488.
Hofmann, W., D. Schubert, J. LaBonte, L. Munson, S. Gibson, J. Scammell, P. Ferrigno, and J. Sodroski. 1999. Species-specific, postentry barriers to primate immunodeficiency virus infection. J. Virol. 73:10020-10028.
Johnson, W. E., and J. M. Coffin. 1999. Constructing primate phylogenies from ancient retrovirus sequences. Proc. Natl. Acad. Sci. USA 96:10254-10260.
Keckesova, Z., L. M. J. Ylinen, and G. J. Towers. 2004. The human and African green monkey TRIM genes encode Ref1 and Lv1 retroviral restriction factor activities. Proc. Natl. Acad. Sci. USA 101:10780-10785.
Keightley, P. D., and D. J. Gaffney. 2003. Functional constraints and frequency of deleterious mutations in noncoding DNA of rodents. Proc. Natl. Acad. Sci. USA 100:13402-13406.
Kim, H. S., O. Takenaka, and T. J. Crow. 1999. Isolation and phylogeny of endogenous retrovirus sequences belonging to the HERV-W family in primates. J. Gen. Virol. 80:2613-2619.
Kootstra, N. A., C. Munk, N. Tonnu, N. R. Landau, and I. M. Verma. 2003. Abrogation of postentry restriction of HIV-1-based lentiviral vector transduction in simian cells. Proc. Natl. Acad. Sci. USA 100:1298-1303.
Kumar, S., K. Tamura, and M. Nei. 1994. MEGA: molecular evolutionary genetics analysis software for microcomputers. Comput. Appl. Biosci. 10:189-191.
Li, W. H. 1993. Unbiased estimation of the rates of synonymous and nonsynonymous substitution. J. Mol. Evol. 36:96-99.
Mager, D. L., and J. D. Freeman. 1995. HERV.H endogenous retroviruses—presence in the New-World branch but amplification in the Old-World primate lineage. Virology 213:395-404.
Mang, R., J. Maas, A. C. van der Kuyl, and J. Goudsmit. 2000. Papio cynocephalus endogenous retrovirus among Old World monkeys: evidence for coevolution and ancient cross-species transmission. J. Virol. 74:1578-1586.
Marcello, A., A. Ferrari, V. Pellegrini, G. Pegoraro, M. Lusic, F. Beltram, and M. Giacca. 2003. Recruitment of human cyclin T1 to nuclear bodies through direct interaction with the PML protein. EMBO J. 22:2156-2166.
Mariana-Costantini, R., T. M. Horn, and R. Callahan. 1989. Ancestry of a human endogenous retrovirus family. J. Virol. 63:4982-4985.
Munk, C., S. M. Brandt, G. Luccero, and N. R. Landau. 2002. A dominant block to HIV-1 replication at reverse transcription in simian cells. Proc. Natl. Acad. Sci. USA 99:13843-13848.
Nisole, S., C. Lynch, J. P. Stoye, and M. W. Yap. 2004. A Trim5-cyclophilin A fusion protein found in owl monkey kidney cells can restrict HIV-1. Proc. Natl. Acad. Sci. USA 101:13324-13328.
Orimo, A., N. Tominaga, K. Yoshimura, Y. Yamauchi, M. Nomura, M. Sato, Y. Nogi, M. Suzuki, H. Suzuki, K. Ikeda, S. Inoue, and M. Mauramatsu. 2000. Molecular cloning of ring finger protein 21 (RNF21)/interferon-responsive finger protein (ifp1), which possesses two RING-B box-coiled coil domains in tandem. Genomics 69:143-149.
Owens, C. M., P. C. Yang, H. Gottlinger, and J. Sodroski. 2003. Human and simian immunodeficiency virus capsid proteins are major viral determinants of early, postentry replication blocks in simian cells. J. Virol. 77:726-731.
Owens, C. M., B. Song, M. J. Perron, P. C. Yang, M. Stremlau, and J. Sodroski. 2004. Binding and susceptibility to post-entry restriction factors in monkey cells are specified by distinct regions of the human immunodeficiency virus type 1 capsid. J. Virol. 78:5423-5437.
Perron, M., M. Stremlau, B. Song, W. Ulm, R. Mulligan, and J. Sodroski. 2004. TRIM5 mediates the postentry block to N-tropic murine leukemia viruses in human cells. Proc. Natl. Acad. Sci. USA 101:11827-11832.
Reus, K., J. Mayer, M. Sauter, H. Zischler, N. Müller-Lantzsch, and E. Meese. 2001. HERV-K (OLD): ancestor sequences of the human endogenous retrovirus family HERV-K (HML-2). J. Virol. 75:8917-8926.
Reymond, A., G. Meroni, A. Fantozzi, G. Merla, S. Cairo, L. Luzi, D. Riganelli, E. Zanaria, S. Messali, S. Cainarca, A. Guffanti, S. Minucci, P. G. Pelicci, and A. Ballabio. 2001. The tripartite motif family identifies cell compartments. EMBO J. 20:2140-2151.
Rhodes, D. A., G. Ihrke, A. T. Reinicke, G. Malcherek, M. Towey, D. A. Isenberg, and J. Trowsdale. 2002. The 52,000 MW Ro/SS-A autoantigen in Sjogren's syndrome/systemic lupus erythematosus (Ro52) is an interferon-gamma inducible tripartite motif protein associated with membrane proximal structures. Immunology 106:246-256.
Rogers, J. S., and D. I. Swofford. 1998. A fast method for approximating maximum likelihoods of phylogenetic trees from nucleotide sequences. Syst. Biol. 47:77-89.
Sayah, D. M., E. Sokolskaja, L. Berthoux, and J. Luban. 2004. Cyclophilin A retrotransposition into TRIM5 explains owl monkey resistance to HIV-1. Nature 430:569-573.
Seto, M. H., H.-L. C. Liu, D. Zajchowski, and M. Whitlow. 1999. Protein fold analysis of the B30.2-like domain. Proteins 35:235-249.
Sharp, P. M., E. Bailes, R. R. Chaudhuri, C. M. Rodenburg, M. O. Santiago, and B. H. Hahn. 2001. The origins of acquired immune deficiency syndrome viruses: where and when? Philos. Trans. R. Soc. Lond. B Biol. Sci. 356:867-876.
Shibata, R., H. Sakai, M. Kawamura, K. Tokunaga, and A. Adachi. 1995. Early replication block of human immunodeficiency virus type 1 in monkey cells. J. Gen. Virol. 76:2723-2730.
Song, B., H. Javanbakht, M. Perron, D. Park, M. Stremlau, and J. Sodroski. 2005. Retrovirus restriction by TRIM5 variants from Old World and New World primates. J. Virol. 79:3930-3937.
Steele, P. E., M. A. Martin, A. B. Rabson, T. Bryan, and S. J. O'Brien. 1986. Amplification and chromosomal dispersion of human endogenous retroviral sequences. J. Virol. 59:545-550.
Steinhuber, S., M. Brack, G. Hunsmann, H. Schwelberger, M. P. Dierich, and W. Vogetseder. 1995. Distribution of human endogenous retrovirus HERV-K genomes in humans and different primates. Hum. Genet. 96:188-192.
Stoye, J. P. 2002. An intracellular block to primate lentivirus replication. Proc. Natl. Acad. Sci. USA 99:11549-11551.
Stremlau, M., C. M. Owens, M. J. Perron, M. Kiessling, P. Autissier, and J. Sodroski. 2004. The cytoplasmic body component TRIM5 restricts human immunodeficiency virus (HIV-1) infection in Old World monkeys. Nature 427:848-853.
Stremlau, M., M. Perron, S. Welikala, and J. Sodroski. 2005. Species-specific variation in the B30.2(SPRY) domain of TRIM5 determines the potency of human immunodeficiency virus type 1 restriction. J. Virol. 79:3139-3145.
Thali, M., A. Bukovsky, E. Kondo, B. Rosenwirth, C. T. Walsh, J. Sodroski, and H. G. Gottlinger. 1994. Functional association of cyclophilin A with HIV-1 virions. Nature 372:363-365.
Thomas, E. E., N. Srebro, J. Sebat, N. Navin, J. Healy, B. Mishra, and M. Wigler. 2004. Distribution of short paired duplications in mammalian genomes. Proc. Natl. Acad. Sci. USA 101:10349-10354.
Thompson, J. D., T. J. Gibson, F. Plewniak, F. Jeanmougin, and D. G. Higgins. 1997. The CLUSTAL_X windows interface: flexible strategies for multiple sequence alignment aided by quality analysis tools. Nucleic Acids Res. 25:4876-4882.
Toniato, E., X. P. Chen, J. Losman, V. Flati, L. Donahue, and P. Rothman. 2002. TRIM8/GERP RING finger protein interacts with SOCS-1. J. Biol. Chem. 277:37315-37322.
Towers, G., M. Bock, S. Martin, Y. Takeuchi, J. P. Stoye, and O. Danos. 2000. A conserved mechanism of retrovirus restriction in mammals. Proc. Natl. Acad. Sci. USA 97:12295-12299.
Towers, G., M. Collins, and Y. Takeuchi. 2002. Abrogation of Ref1 retrovirus restriction in human cells. J. Virol. 76:2548-2550.
Towers, G. J., T. Hatziioannou, S. Cowan, S. P. Goff, J. Luban, and P. D. Bieniasz. 2003. Cyclophilin A modulates the sensitivity of HIV-1 to host restriction factors. Nat. Med. 9:1138-1143.
Tristem, M. 2000. Identification and characterization of novel human endogenous retrovirus families by phylogenetic screening of the human genome mapping project database. J. Virol. 74:3715-3730.
Turelli, P., V. Doucas, E. Craig, B. Mangeat, N. Klages, R. Evans, G. Kalpana, and D. Trono. 2001. Cytoplasmic recruitment of INI1 and PML on incoming HIV preintegration complexes: interference with early steps of viral replication. Mol. Cell 7:1245-1254.
van der Kuyl, A. C., J. T. Dekker, and J. Goudsmit. 1995. Distribution of baboon endogenous virus among species of African monkeys suggests multiple ancient cross-species transmissions in shared habitats. J. Virol. 69:7877-7887.
van der Kuyl, A. C., R. Mang, J. T. Dekker, and J. Goudsmit. 1997. Complete nucleotide sequence of simian endogenous type D retrovirus with intact genome organization: evidence for ancestry to simian retrovirus and baboon endogenous virus. J. Virol. 71:3666-3676.
Watanabe, H., A. Fujiyama, M. Hattori, T. D. Taylor, A. Toyoda, Y. Kuroki, H. Noguchi, A. BenKahla, H. Lehrach, R. Sudbrak, M. Kube, S. Taenzer, P. Galgoczy, M. Platzer, M. Scharfe, G. Nordsiek, H. Blocker, I. Hellmann, P. Khaitovich, S. Paabo, R. Reinhardt, H. J. Zheng, X. L. Zhang, G. F. Zhu, B. F. Wang, G. Fu, S. X. Ren, G. P. Zhao, Z. Chen, Y. S. Lee, J. E. Cheong, S. H. Choi, K. M. Wu, T. T. Liu, K. J. Hsiao, S. F. Tsai, C. G. Kim, S. Oota, T. Kitano, Y. Kohara, N. Saitou, H. S. Park, S. Y. Wang, M. L. Yaspo, and Y. Sakaki. 2004. DNA sequence and comparative analysis of chimpanzee chromosome 22. Nature 429:382-388.
Whitcomb, J. M., and S. H. Hughes. 1992. Retroviral reverse transcription and integration: progress and problems. Annu. Rev. Cell Biol. 8:275-306.
Wilkinson, D. A., D. L. Mager, and J.-A. C. Leong. 1994. Endogenous human retroviruses, p. 465-535. In J. A. Levy (ed.), The Retroviridae, vol. III. Plenum Press, New York, N.Y.
Xu, L., L. Yang, P. K. Moitra, K. Hashimoto, P. Rallabhandi, S. Kaul, G. Meroni, J. P. Jensen, A. M. Weissman, and P. D'Arpa. 2003. BTBD1 and BTBD2 colocalize to cytoplasmic bodies with the RBCC/tripartite motif protein, TRIM5delta. Exp. Cell Res. 288:84-93.
Yap, M. W., S. Nisole, C. Lynch, and J. P. Stoye. 2004. Trim5 protein restricts both HIV-1 and murine leukemia virus. Proc. Natl. Acad. Sci. USA 101:10786-10791.(Byeongwoon Song, Bert Gol)
Laboratory of Genomic Diversity, National Cancer Institute, Frederick, Maryland 21702-1201
SAIC-Frederick, Frederick, Maryland 21702-1201
Department of Immunology and Infectious Diseases, Harvard School of Public Health, Boston, Massachusetts 02115
ABSTRACT
Tripartite motif (TRIM) proteins are composed of RING, B-box 2, and coiled coil domains. Some TRIM proteins, such as TRIM5, also possess a carboxy-terminal B30.2(SPRY) domain and localize to cytoplasmic bodies. TRIM5 has recently been shown to mediate innate intracellular resistance to retroviruses, an activity dependent on the integrity of the B30.2 domain, in particular primate species. An examination of the sequences of several TRIM proteins related to TRIM5 revealed the existence of four variable regions (v1, v2, v3, and v4) in the B30.2 domain. Species-specific variation in TRIM5 was analyzed by amplifying, cloning, and sequencing nonhuman primate TRIM5 orthologs. Lineage-specific expansion and sequential duplication occurred in the TRIM5 B30.2 v1 region in Old World primates and in v3 in New World monkeys. We observed substitution patterns indicative of selection bordering these particular B30.2 domain variable elements. These results suggest that occasional, complex changes were incorporated into the TRIM5 B30.2 domain at discrete time points during the evolution of primates. Some of these time points correspond to periods during which primates were exposed to retroviral infections, based on the appearance of particular endogenous retroviruses in primate genomes. The results are consistent with a role for TRIM5 in innate immunity against retroviruses.
INTRODUCTION
Following entry, retroviruses must negotiate a series of processes to establish a permanent infection of the host cell. These include uncoating of the viral core, reverse transcription, nuclear access, and integration of the viral DNA into the host genome (1, 11, 60). Early postentry restrictions to retrovirus infection can determine tropism at the species level. Infection by N-tropic murine leukemia virus, for example, is inefficient in most human cells and in certain cell lines from African green monkeys (5, 53). Human immunodeficiency virus type 1 (HIV-1) encounters a postentry block in Old World monkeys, whereas simian immunodeficiency virus (SIVmac) is blocked in most New World monkey cells (17, 18, 42). These species-specific restrictions share several features. First, the block occurs prior to or concurrent with reverse transcription, which occurs in the cytoplasm of the host cell. At most, low levels of early reverse transcripts are made in restricted cells (9, 17, 29, 42). Second, the viral determinant of susceptibility to the block is the capsid protein (9, 14, 22, 32, 33, 53). Other capsid-binding proteins, such as cyclophilin A in the case of HIV-1, can modify the degree of restriction (33, 34, 49, 55). Third, restriction is mediated by dominant host factors, the activity of which can be titrated by the introduction of virus-like particles containing proteolytically processed capsid proteins of the restricted viruses (4, 6, 9, 13, 29, 33, 46, 54).
These observations suggested a model in which host restriction factors interact, directly or indirectly, with the viral capsid and prevent its progression along the infection pathway. A genetic screen identified a major restriction factor in monkey cells that acts on HIV-1, and to a lesser extent, on SIVmac (47). The factor, TRIM5rh, was selected from a cDNA library prepared from primary rhesus monkey lung fibroblasts. TRIM5rh was shown to be sufficient to confer potent resistance to HIV-1 infection on otherwise susceptible cells. Moreover, TRIM5rh was necessary for maintenance of the block to the early phase of HIV-1 infection in Old World monkey cells, as demonstrated by interference with the expression of the endogenous TRIM5 ortholog in these cells (47). HIV-1 infection in cells expressing TRIM5rh was blocked at the earliest stage of reverse transcription (47). Cells expressing TRIM5rh exhibited a partial inhibition of SIVmac infection but were as susceptible as control cells to infection by Moloney murine leukemia virus vectors. Humans express a protein, TRIM5hu, that is 87% identical in amino acid sequence to the rhesus monkey protein TRIM5rh (47). Even when expressed at comparable levels, TRIM5hu was less potent in suppressing HIV-1 and SIVmac infections than TRIM5rh (47). Recently, TRIM5hu was shown to be responsible for the postentry restriction of N-tropic murine leukemia virus (N-MLV) in human cells (15, 20, 34, 63). TRIM5rh was much less effective than TRIM5hu at blocking this murine leukemia virus. Thus, TRIM5hu potently restricts N-MLV, specifies an intermediate level of resistance to HIV-1, and does not affect SIVmac infection. In contrast, TRIM5rh potently restricts HIV-1 and exhibits a modest inhibition of SIVmac and N-MLV infections. The TRIM5 protein from African green monkeys, TRIM5agm, has recently been shown to inhibit N-MLV, HIV-1, and SIVmac infections (15, 20, 43, 63). These observations underscore the importance of species-specific variations in TRIM5 orthologs for their ability to restrict infections by particular retroviruses.
More than 50 genes encode tripartite motif (TRIM) proteins (36). The tripartite motif includes a RING domain, a B-box 2 domain, and a coiled coil (cc) domain; TRIM proteins have also been called RBCC proteins. Some TRIM proteins, including TRIM5, contain a C-terminal B30.2 or SPRY domain (16, 36). Differential splicing of the TRIM5 primary transcript gives rise to the expression of several isoforms of the protein product (36). The TRIM5 isoform is the largest product (493 amino acid residues in humans) and contains the B30.2/SPRY domain. The other TRIM5 isoforms ( and are the best substantiated of these) lack an intact B30.2/SPRY domain. The TRIM5rh isoform does not inhibit HIV-1 or SIVmac infection (47). In fact, TRIM5rh has been shown to exhibit a weak dominant-negative activity, repressing the ability of wild-type TRIM5rh to inhibit HIV-1 infection (47). Thus, the B30.2 domain is critical for the ability of TRIM5 to mediate antiretroviral effects.
A common feature of TRIM proteins is the ability to assemble into large complexes in the cytoplasm or nucleus of the cell (36). The function of these cytoplasmic and nuclear bodies is largely unknown. The demonstration of a potent, specific antiretroviral activity for TRIM5 raised the possibility that the major function of some TRIM proteins is the establishment of innate immunity to infectious agents. The localization of TRIM5 to cytoplasmic bodies (36, 62) is consistent with its ability to block retroviral infection shortly after entry of the viral capsid into the cytoplasm of the host cell.
Studies of endogenous retroviral sequences have indicated that vertebrates, including humans, have been exposed to retroviruses for many millions of years (2, 12, 19, 21, 25, 26, 28, 35, 44, 45, 56, 58, 59, 61). This long history of host-retrovirus coevolution might favor the selection of particular TRIM5 proteins that effectively suppress lethal retrovirus infections. Since many viruses preferentially infect particular host species, the selection of molecules involved in antiviral immunity often occurs in a lineage-specific manner. In this report, we characterize the TRIM5 proteins of several primate species and provide evidence of lineage-specific variation in particular B30.2 domain elements.
MATERIALS AND METHODS
Cells for TRIM5 cDNA synthesis. SQMK-FP (Pindak) cells from a Bolivian squirrel monkey (Saimiri boliviensis boliviensis), Vero cells from an African green monkey (Cercopithecus aethiops pygerythrus), OMK cells from an owl monkey (Aotus trivirgatus), and human HeLa cells were obtained from the American Type Culture Collection. Primary rhesus lung fibroblasts were derived from a rhesus macaque (Macaca mulatta) (18). The Coriell Institute for Medical Research (Camden, N.J.) supplied the following cells: AG05352 cells from a black-handed spider monkey (Ateles geoffroyi), AG05308 cells from a red-chested mustached tamarin (Saguinus labiatus), AG06116 cells from a patas monkey (red guenon) (Erythrocebus patas), AG06209 cells from a Sumatran orangutan (Pongo pygmaeus abelii), AG05251 cells from a Western lowland gorilla (Gorilla gorilla), and GM03448 cells from a chimpanzee (Pan troglodytes verus).
Samples for DNA analysis. Nonhuman primate DNAs from a previously described collection of samples (2, 3) included two independent samples from squirrel monkeys (B263 [female] and B26) (Saimiri sciureus), samples from two baboon species (B21 [Papio cynocephalus] and B856 and B1542 [Papio anubis]), a patas monkey sample (B1530) (Erythrocebus patas), a grivet sample (B1532 [female]) (Cercopithecus sabaeus), two sooty mangabey samples (B400 [female] and B859) (Cercocebus atys), a gelada sample (B853) (Theropithecus gelada), two colobus samples (B130 [Colobus guereza] and B1527 [Colobus guereza caudatus]), samples from two species of langur (B131 [female] [Presbytis obscurus] and B401 [male] [Presbytis senex]), samples from two species of gibbon (B23, B1533, and B837 [Hylobates lar] and B128 [Hylobates concolor]), and samples from gorillas (B27 [female], B461, and B1535 [female]) (Gorilla gorilla). Additional samples of patas monkey (Erythrocebus patas), white-cheeked gibbon (Hylobates concolor), and gorilla (Gorilla gorilla) DNA were obtained from the core collection of the Laboratory of Genomic Diversity, National Cancer Institute.
TRIM5 cDNA cloning and sequencing. First- and second-strand cDNA synthesis was performed by use of a cDNA synthesis kit (Clontech), using RNAs prepared from cells derived from an African green monkey, a rhesus macaque, a Bolivian squirrel monkey, an owl monkey, a patas monkey, a tamarin, a spider monkey, an orangutan, a gorilla, and a chimpanzee (see above). Human and primate TRIM5 cDNAs encoding TRIM5 were amplified with the primers TRIMf2 (5'-GCGGAATTCGCCATGGCTTCTGGAATCCTGGTT-3') and TRIMr2 (5'-GCGATCGATGCCTCAAGAGCTTGGTGAGCACAG-3').
Amplification was carried out by use of a Clontech Advantage PCR kit, with thermocycling of the reactions at 95°C for 30 s, followed by 30 cycles of 95°C for 30 s, 55°C for 1 min, and 68°C for 3 min. Amplified cDNAs were inserted into the pCR-BluntII-TOPO plasmid (Invitrogen) and sequenced by use of the following primers: M13f (–20), GTAAAACGACGGCCAGT; M13r (–21), AACAGCTATGACCATG; TRIMf3, GGAAGCTGACATCAGAGA; TRIMf4, GATAAGAGACAAGTGAGC; TRIMr3, TCTACCTCCCAGTAATG; and TRIMr4, TCCTTCTCCAGGTTTTGC.
PCR amplification of human TRIM5 exon 8 from genomic DNA. Thirty nanograms of individual DNAs from the primates described above was amplified with 5 U of AmpliTaq Gold enzyme in 3.5 mM MgCl2 and 330 μM nucleoside triphosphates, using 200 nanomoles each of the following primers containing an M13 forward or reverse tag: TRIM5Ex8SeqF, 5'-GTAAAACGACGGCCAGTTCCCTTAGCTGACCTGTTAATTT-3'; and TRIM5Ex8SeqR, 5'-GGAAACAGCTATGACCATGGCTGTACAGAAGGGGCTGAG-3'. The following cycling conditions were used: an initial 94°C, 5-min activation of AmpliTaq Gold, followed by 20 cycles of 94°C for 20 seconds, 54°C for 20 seconds, and 68°C for 1 min and then an additional 20 cycles of 94°C for 20 seconds, 65°C for 20 seconds, and 68°C for 1 min, a final extension of 68°C for 7 min, and a hold overnight, if necessary, at 4°C.
DNA sequencing. Amplicons were treated with exonuclease I and shrimp alkaline phosphatase and subjected to Big Dye cycle sequencing in the presence of either an M13 forward or reverse primer tag (m13 forward, 5'-GTAAAACGACGGCCAGT-3'; m13 reverse, 5'-GGAAACAGCTATGACCATG-3'). Sequenced reactions were cleaned with Sephadex G-50, dried on a thermocycler (95°C without caps for 20 min), and resuspended in undiluted formamide (Applied Biosystems) in preparation for capillary electrophoresis on an ABI 3730 HT apparatus. Subsequent to sequencing, forward and trace files were examined with Mutation Explorer software (Softgenetics) and sequences were visually determined.
Phylogenetic analysis. The data sets for the complete TRIM5 protein and the B30.2(SPRY) domain were examined separately. Predicted amino acid sequences were compiled and aligned for subsequent phylogenetic analyses by ClustalX (51) and were verified visually. Phylogenetic reconstruction was performed with MEGA, version 2.0, software (23) using the following methods: minimum evolution, neighbor joining, and maximum parsimony. A bootstrap analysis using 1,000 iterations was performed with each method (38).
Nonsynonymous/synonymous variation. The nonsynonymous/synonymous (Ka/Ks) ratios at various codon positions for pairwise comparisons of TRIM5 cDNAs were calculated by the method of Li (24). The Ka/Ks ratio was estimated as a rolling average for a window of 150 codons, with the center of the window being moved codon-by-codon to produce a plot showing local variations in the degree of sequence conservation.
Nucleotide sequence accession numbers. The complete primate TRIM5 coding sequences have been deposited in the GenBank database under accession numbers AY740612 to AY740621. The primate TRIM5 nucleotide sequences have been deposited in the GenBank database under accession numbers AY710295 to AY710304.
RESULTS
Variation among TRIM proteins related to TRIM5. TRIM proteins first appeared with the metazoans and dramatically expanded in number during vertebrate evolution (36). Thus, current TRIM genes presumably arose from a small number of ancestral genes. To gain insight into the conserved and variable elements of TRIM proteins, we aligned the primary amino acid sequences of several TRIM proteins. The TRIM proteins most closely related to TRIM5hu (as ascertained by BLAST analysis) were selected for study. Nearly all of these TRIM proteins contain a B30.2 domain, and many have been shown to localize, at least in part, to the cytoplasm of the cell (36). Several of these TRIM proteins are found in cytoplasmic bodies, as is TRIM5 (36).
The predicted amino acid sequences of TRIM proteins related to TRIM5hu are aligned in Fig. 1. The sequences of several TRIM5 orthologs from nonhuman primates that were determined in this study were included in the alignment (see below). A moderate degree of variation was observed in all of the domains of these selected TRIM proteins (Fig. 1). The existence of four regions within the B30.2 domain that exhibited substantial variation in length as well as extensive amino acid differences among the TRIM proteins was particularly noteworthy. We designated these variable regions v1, v2, v3, and v4 (Fig. 1). Because of the modest level of amino acid variation among the TRIM B30.2 domains, the boundaries of these variable regions are somewhat arbitrary. We chose reasonably conserved sets of boundary residues to eliminate any ambiguity in defining the length of a particular variable region. This allowed us to examine the lengths of the B30.2 domain variable regions in different TRIM proteins, including those which are less closely related to TRIM5 (Table 1). Based on length variations in the B30.2 v1, v2, and v3 regions, two groups of TRIM proteins can be discerned. For one group, consisting of TRIM21, BIA2, mouse LOC216781, rat LOC303167, RNF137, TRIM17, TRIM11, TRIM10, TRIM26, TRIM27, TRIM20, TRIM39, and TRIM50A, each B30.2 variable region conforms to a narrow range of lengths. The v1 regions of these TRIM proteins are 19 to 21 residues long, the v2 regions are 11 to 13 residues long, and the v3 regions are 20 to 23 residues long. An examination of the available sequences of TRIM10, TRIM11, TRIM20, TRIM26, TRIM27, TRIM39, and TRIM50A from different mammalian species indicated that the B30.2 domain sequences and the variable region lengths are relatively well conserved (data not shown). The B30.2 domain v1, v2, and v3 regions of TRIM21 and the v3 region of TRIM17 exhibit more amino acid variability between humans and rodents (data not shown). Nonetheless, the lengths of the TRIM17 and TRIM21 variable regions are similar in humans and rodents (Table 1 and data not shown). Thus, one group of TRIM proteins exhibits a narrow range of B30.2 variable region lengths and, in many cases, minimal variation between human and rodent orthologs. Based on the sequences of the TRIM proteins in this group, consensus TRIM B30.2 domain variable region lengths can be defined as 19 to 21 residues for v1, 11 to 13 residues for v2, and 20 to 23 residues for v3.
For the second group of TRIM proteins, the lengths of the B30.2 variable regions differ from those found for the group described above. This second group includes TRIM4, TRIM5, TRIM6, TRIM22, TRIM34, rat LOC308906, and mouse 9230105E10RiK. Compared with the consensus B30.2 variable loop lengths, TRIM4 exhibits a long v1 region, and TRIM6, TRIM22, TRIM30, TRIM34, rat LOC308906, and mouse 9230105E10RiK exhibit long v2 and v3 regions. Relative to the consensus, the primate TRIM5 proteins exhibit differences in the lengths of v1, v2, and v3. It is noteworthy that a group of related TRIM proteins (TRIM5, TRIM6, TRIM22, TRIM30, TRIM34, rat LOC308906, and mouse 9230105E10RiK) all share the property of increased B30.2 variable region lengths compared to the consensus. The human TRIM5, TRIM6, TRIM22, and TRIM34 genes are located in a gene cluster at 11p15.4. In rodents, TRIM6 and TRIM34 are adjacent to the rat LOC308903 and mouse 9230105E10RiK genes. This clustering may indicate recent paralogous relationships among these TRIM genes. In the available TRIM6 and TRIM34 sequences from other mammals, the B30.2 variable region lengths are similar to those found for the human proteins (Table 1), even though considerable amino acid variation is observed in these regions (data not shown). The human TRIM4 and TRIM22 genes do not have orthologs in the rodent genome, and conversely, TRIM30 is not found in the human genome (data not shown). Thus, in most cases for which data are available, the B30.2 variable region lengths are preserved in TRIM proteins from different species. TRIM5 and the proteins encoded by the rat LOC308906 and mouse 9230105E10RiK genes are exceptions (Table 1; see below).
The v4 variable regions of the B30.2 domains of the TRIM proteins examined are short, varying from 4 to 10 residues in length. The v4 variable region is five residues long in all of the TRIM proteins closely related to TRIM5 (Table 1).
Interspecies variation in the TRIM5 protein. Variations among TRIM5 proteins in primates affect the efficacy of restriction of particular retroviruses (15, 20, 34, 43, 47, 63). Ancient retroviral infections encountered by the members of specific primate lineages may have exerted selective pressure on the TRIM5 structure. To investigate this possibility, we cloned and sequenced the TRIM5 cDNAs encoding the complete TRIM5 proteins of several primate species. These included apes (chimpanzee, gorilla, and orangutan), Old World monkeys (rhesus macaque, African green monkey, and patas monkey), and New World monkeys (squirrel monkey, tamarin, and spider monkey). To ensure that these sequences were representative, we determined the sequences of several TRIM5 cDNAs from individual humans, rhesus macaques, African green monkeys, and squirrel monkeys. The variation among the predicted TRIM5 proteins of individuals within a species was <2.5% (data not shown).
We also examined the mouse and rat genomes for potential TRIM5 orthologs. The rodent proteins most closely related to TRIM5 are encoded by TRIM12 (mouse), 9230105E10RiK (mouse), and LOC308906 (rat). Reciprocally, TRIM5 is the human protein most closely related to the protein products of these rodent genes (data not shown). TRIM12 and 9230105E10RiK are essentially identical in their 5' portions and have been assembled at adjacent loci on mouse chromosome 7. It is possible that the current assembly will be revised and that TRIM12 and 9230105E10RiK are in fact the same gene. It is also possible that these genes are paralogs that arose from a duplication event after the divergence of mice and rats. An examination of current drafts of the chicken and dog genomes failed to identify candidate TRIM5 orthologs.
An alignment of the predicted amino acid sequences of the TRIM5-related proteins of rodents and several primates is shown in Fig. 1. The degree of variation among the TRIM5-related proteins of these different species was larger than that observed within the individuals of a species (data not shown). The full-length protein product of the 9230105E10RiK gene is only 49% identical to human TRIM5. This is lower than the 80 and 67% identities seen when the human and rodent versions of the TRIM6 and TRIM34 proteins, respectively, are compared. In fact, the degree of sequence identity between the human TRIM5 and rodent 9230105E10RiK proteins is not higher than the similarity among TRIM5 and other TRIM proteins encoded by the human TRIM cluster located at 11p15.4. Whether the mouse 9230105E10 and rat LOC308906 genes represent TRIM5 orthologs will be addressed elsewhere. At a minimum, the rodent TRIM5-related proteins exhibit substantial differences from TRIM5hu. Thus, our analysis of interspecies variation among TRIM5 proteins will focus mainly on primate sequences.
Full-length TRIM5 cDNA sequences from 10 primate species were used to construct a phylogenetic tree (Fig. 2A). Bootstrap simulation using neighbor joining, minimum evolution, and maximum parsimony algorithms all resolved the New Word/Old World dichotomy and the great ape/Old World monkey dichotomy with a high confidence. Patas monkeys were differentiable from African green monkeys and rhesus monkeys in comparisons of the full-length TRIM5 sequences; comparisons of the complete TRIM5 sequences for this particular branch differed somewhat from comparisons between rhesus macaques and related primates when only the B30.2 regions were compared (see below).
Interspecies variation in TRIM5-like proteins was observed in all four of the defined TRIM5 domains, with the most variation occurring in the B30.2 domain (Table 2). To assess B30.2 domain variation more thoroughly, we amplified DNA fragments corresponding to TRIM5 exon 8, which encodes the TRIM5 B30.2 domain, from several nonhuman primates and then sequenced the fragments. Attempts to amplify TRIM5 exon 8 sequences from two prosimian species, lemurs and galagoes, were unsuccessful (data not shown). As expected from our analysis of TRIM proteins, the observed variation within the B30.2 domains of primate TRIM5 proteins was not uniform, being concentrated within the v1, v2, and v3 variable regions defined above (Fig. 1). In the TRIM5 proteins of the primates examined, all three of these variable B30.2 regions differed in length from the consensus TRIM variable region lengths (Fig. 2B). With the sole exception of the spider monkey TRIM5 protein, in which v2 is 15 residues long, the primate TRIM5 v2 regions are 17 residues in length. Thus, although the v2 region exhibits a high level of substitution of individual amino acids among primate TRIM5 proteins, the length of this region is well conserved. In contrast, a striking lineage-specific variation in the lengths of the v1 and v3 variable regions of the primate TRIM5 proteins was observed (Fig. 2B). The v1 regions of the New World monkeys are 17 residues long, slightly shorter than the consensus v1 length. In contrast, the v1 regions of all of the Old World (catarrhine) primates are longer than that of the consensus. Since a v1 length of 26 residues is found in all hominoids and in the Colobinae subfamily of Old World monkeys, this may represent the v1 length of the TRIM5 proteins of the shared catarrhine ancestor. In the Cercopithecinae subfamily of Old World monkeys, the v1 length is expanded to 28 residues, and in the African green monkey/grivet/patas monkey lineage, it is expanded to 46 residues. The 46-residue v1 region of TRIM5 from these monkeys contains two tandem 20-residue repeats (Fig. 3, top). These are perfect repeats in some individual African green monkeys, but not in others (data not shown). Thus, the TRIM5 B30.2 v1 region appears to have undergone some expansion in all Old World primates. In certain lineages, this expansion is dramatic and involves tandem sequence duplication.
The B30.2 v3 region is 32 residues long in the TRIM5 proteins of all Old World (catarrhine) primates. This is longer than the consensus v3 length and likely represents the v3 length of the TRIM5 protein of the shared catarrhine ancestor. The TRIM5 B30.2 v3 region is longer in the New World (platyrrhine) monkeys than in the Old World monkeys. The v3 region is 41 residues long in all of the New World monkey TRIM5 sequences examined, except for that of the spider monkey. The B30.2 v3 region in spider monkeys has expanded to 96 residues in length. This remarkable expansion results from the presence of three imperfect tandem repeats, of 30, 31, and 24 residues (Fig. 3, bottom). Thus, the TRIM5 B30.2 v3 region appears to have expanded in New World (platyrrhine) monkeys. In at least one lineage, that of spider monkeys, the expansion is dramatic and involves tandem sequence duplication.
Analysis of nonsynonymous/synonymous variation in TRIM5 genes from different species. Analyses of nonsynonymous/synonymous variations (Ka/Ks ratios) can provide insight into selection for or against a change in the coding capacity of a gene (24). However, this approach cannot accommodate sequences that exhibit large insertions or deletions (indels) relative to one another. Therefore, for our comparison of TRIM5 sequences from different primates, the indels involving the v1 and v3 regions of the B30.2 domain were excluded from the analysis. The Ka/Ks ratios at various codon positions were calculated for pairwise comparisons of TRIM5 cDNAs from a hominoid (human), an Old World monkey (patas monkey), and two New World monkeys (tamarin and spider monkey), providing several phylogenetically independent paired comparisons (Fig. 4). For the sequences encoding the N-terminal half of TRIM5 (the RING, B-box 2, and coiled coil domains), the Ka/Ks ratio was generally <1, indicating selection against amino acid changes. For the sequences encoding the B30.2 domain, the Ka/Ks ratio increased dramatically, reaching values above 5 in some cases. Such high Ka/Ks ratios are strongly suggestive of selectively driven diversity in the B30.2 domain.
DISCUSSION
The functions of most TRIM proteins are not understood (36). The TRIM genes arose in bilaterans and expanded in number during vertebrate evolution in response to as yet unknown selective forces. Several observations are consistent with the possibility that some of these proteins contribute to the establishment of innate intracellular defense systems against infectious agents. The TRIM5 proteins of rhesus monkeys and humans block infections by HIV-1 and murine leukemia viruses, respectively (15, 20, 34, 47, 63). The African green monkey TRIM5 protein exhibits activity against a diverse group of retroviruses (15, 20, 43, 63). Some antiviral activity has also been reported for other TRIM family members (7, 8, 27, 57), although this activity lacks the potency and degree of specificity observed for TRIM5 against particular retroviruses. Finally, some TRIM genes are inducible by interferon (8, 31, 37, 52), suggesting a link to other elements of the innate immune system.
The above considerations raise the possibility that infectious agents influenced the evolution of certain TRIM proteins. Length variation in particular segments of different TRIM B30.2 domains were examined as one potential indicator of such selection. Our observations are consistent with a parsimonious model in which ancestral TRIM proteins possessed B30.2 domains with relatively short variable regions. In such a model, the ancestral TRIM B30.2 domain possessed a v1 region of 19 to 21 amino acids, a v2 region of 11 to 13 residues, and a v3 region of 20 to 23 residues, the variable region lengths found in many examples of extant TRIM proteins. For some cytoplasmic TRIM proteins such as TRIM21, amino acid variation, but not length variation, in the B30.2 variable regions occurred during the evolution of rodents and primates. For TRIM4 and the TRIM proteins encoded by the human 11p15.4 cluster (TRIM5, TRIM6, TRIM22, and TRIM34), expansion of the length of one or more B30.2 variable regions apparently occurred at various times in mammalian evolution. For TRIM6 and TRIM34, the mutational events that caused the increased lengths of the v2 and v3 regions likely occurred before the divergence of rodents and primates. Once fixed, these variable region lengths were apparently maintained within narrow limits for at least 90 million years. In contrast, the evolution of different B30.2 variable region lengths in TRIM5 continued during the diversification of primates (see below). Since TRIM4 and TRIM22 are found in humans but not rodents, the emergence or loss of these genes must have occurred after the rodent-primate divergence.
The TRIM5 B30.2 domain exhibits length variations in particular primate lineages. The shorter versions of the TRIM5 B30.2 domain v1, v2, and v3 regions among the primate species examined exhibit lengths of 17 residues, 15 to 17 residues, and 30 to 32 residues, respectively. These v2 and v3 regions are longer than the corresponding regions of the hypothetical ancestral TRIM protein, suggesting that some expansion of these regions may have occurred in a TRIM5 ancestor prior to the diversification of primates. Notably, expansion of the v2 and v3 regions characterizes all of the TRIM proteins encoded by the human 11p15.4 cluster; it seems plausible that these related TRIM proteins share a common ancestor that exhibited v2 and v3 regions which were longer than those of the putative ancestral TRIM protein. Concomitant with the evolution of primates, the TRIM5 B30.2 domain apparently underwent episodic expansions of either the v1 or v3 variable region in a lineage-specific fashion. These v1 or v3 expansions occurred at discrete points in the TRIM5 phylogeny (Fig. 2B, arrows). The evolution of a v1 length of 26 residues apparently occurred at the root of the Old World (catarrhine) branch of primates. A further expansion of v1 to 28 residues occurred with the evolution of the subfamily Cercopithecinae, beginning 9 to 13 million years ago. A final v1 expansion to 46 residues evolved with the African green and patas monkeys.
In the New World (platyrrhine) branch of monkeys, the TRIM5 B30.2 domains evolved long v3 regions. At the root of this lineage, a v3 expansion to 41 residues occurred. A further expansion to 96 residues occurred in spider monkeys. Remarkably, the longest B30.2 variable regions (v1 in African green monkeys and patas monkeys and v3 in spider monkeys) contain tandem duplications. Doublets, defined as short duplications of between 25 and 100 bp, are common in the genomes of mammals and other eukaryotes (50). Tandem doublets, such as those that encode the B30.2 variable region duplications observed in our study, tend to have arisen recently in evolution (50). It has been speculated that double-stranded breaks that are filled in and nonhomologously recombined might be the origins of adjacent doublets (50). Whatever the mechanism by which they arose, the resulting long variable regions in TRIM5 were preserved and became prevalent in the respective monkey species. TRIM5 proteins with long B30.2 variable regions possibly provide better protection against retroviral infections. Indeed, the TRIM5 proteins of African green monkeys and spider monkeys exhibit antiviral activities against a diverse group of retroviruses (15, 20, 43, 63); it is possible that the tandem duplications in the B30.2 variable regions contribute to this breadth.
Standard approaches to assess the selection of genes (e.g., comparisons of the ratios of nonsynonymous/synonymous changes) typically ignore insertions and deletions that result in length variations. An analysis of Ka/Ks ratios was useful for assessing the evolution of TRIM5 sequences outside the B30.2 v1 and v3 variable regions. Pairwise comparisons of TRIM5 cDNAs from four species, representing hominoids, Old World monkeys, and New World monkeys, revealed a similar pattern. For the 5' portion of the TRIM5 cDNA, which encodes the RING, B-box 2, and coiled coil domains of TRIM5, the Ka/Ks ratios were generally <1. This indicates that purifying selection has operated on these TRIM5 domains to preserve the amino acid sequences and, presumably, function. This conclusion is consistent with the conservation of amino acid residues known to be important for the integrity of these domains in the TRIM5 proteins from the different primate species examined (Fig. 1). The demonstrated functional activity of TRIM5 proteins derived from different primate species in restricting particular retroviruses (15, 20, 43, 47, 63) also supports the conservation of TRIM5's function during primate evolution. Even in owl monkeys, in which a retrotransposition has resulted in a loss of the ability to encode a complete TRIM5 protein and instead a TRIM5-cyclophilin A fusion protein is produced, the capacity to restrict some retroviruses is preserved (30, 39).
In contrast to the results for the 5' portion of TRIM5, the Ka/Ks ratios were >1 for the 3' end of TRIM5, which encodes the B30.2 domain. Thus, even when the v1 and v3 variable regions, which exhibit significant variations in individual amino acid residues as well as in their lengths, were excluded from consideration, the analysis strongly suggested that diversity within primate TRIM5 B30.2 domains has been driven by selection. In summary, evolutionary forces have operated to preserve the amino acids in the N-terminal portion and to diversify the amino acids in the B30.2 domain of primate TRIM5 proteins.
Although our analysis of Ka/Ks ratios suggests that selection has influenced the TRIM5 gene segments flanking the indels, the selective nature of the inserts themselves (i.e., whether an increased length is deleterious, neutral, or positively selected) requires further interpretation. The probability of fixation of an insert will be less than, equal to, or more than its creation rate through mutation when selection of the insert is negative, absent, or positive, respectively. The frequency of indels is known to be considerably lower than that of point substitutions (20A, 59A). Thus, the coincident fixation of several TRIM5 inserts during a short period of primate evolution is indicative of either an unusually high rate of indel creation or the presence of positive selection favoring fixation of the insert. Because of evidence of selection in flanking regions and the relative stability of the indel regions in TRIM5 genes over long evolutionary periods, we consider positive selection to be a more likely explanation for the fixation of the inserts.
Beginning at discrete time points spaced over millions of years, TRIM5 B30.2 variable regions of particular lengths, and presumably with particular amino acid sequences as well, were sufficiently advantageous that the corresponding TRIM5 genes became prevalent within the species. Episodic waves of lethal retrovirus infections may have contributed to this evolutionary pattern. The timing of some ancient retroviral epidemics has been deduced from the presence of endogenous retroviruses in the germ lines of mammalian species (2, 12, 19, 21, 25, 26, 28, 35, 44, 45, 58, 59, 61). Notably, several endogenous retroviruses appeared in primate germ lines during the same time periods in which two of the expansions of v1 length in the catarrhine TRIM5 proteins seem to have occurred. Between 25 and 40 million years ago, when Old World primates diverged from New World monkeys, several endogenous retroviruses (human endogenous retrovirus type E [HERV-E], HERV-W, HERV-K, and ERV-9) were introduced into primate genomes (21, 28, 35, 44, 45, 61). This corresponds to the period in which the TRIM5 B30.2 v1 region length apparently expanded to 26 residues (Fig. 2B). About 9 to 13 million years ago, the simian endogenous retrovirus (SERV) and Papio cynocephalus endogenous retrovirus were introduced into the genomes of cercopithecine monkeys (26, 58, 59). The TRIM5 B30.2 v1 expansion to 28 residues occurred during this same period (Fig. 2B). Later TRIM5 v1 expansions to 46 residues in the African green monkey/patas monkey lineage may also be related to retroviral infections. SIV infections are currently prevalent in these monkeys (10, 41), and SIV represents a candidate retrovirus that could have applied selective pressure on TRIM5. However, given the absence of endogenous SIV proviruses, the extent of such infections in ancient times is unknown.
TRIM5 v3 expansion in New World monkeys could also involve ancient retroviral infections. The HERV-S and HERV-F endogenous retroviruses were introduced into primate germ lines between 32 and 56 million years ago, encompassing the period of divergence of the New World and Old World monkeys (61). Further studies may reveal additional candidate retroviruses that potentially influenced platyrrhine TRIM5 proteins. Moreover, infectious agents beyond the retrovirus group may conceivably have influenced TRIM protein evolution.
The v1 region of the TRIM5 B30.2 domain has undergone extensive changes during the evolution of Old World primates. In this light, it is noteworthy that the major determinant of anti-HIV-1 potency distinguishing the human and rhesus monkey TRIM5 proteins was recently mapped to the B30.2 v1 region (48). Further studies will be needed to understand how the B30.2 variable regions might contribute to TRIM5 antiretroviral activity. Length expansions of the degree observed in our survey of primate TRIM5 sequences must be compatible with the fold of the B30.2 domain and thus are probably surface elements. The structure of the B30.2 domain is not known, although it has been speculated that it might be immunoglobulin-like (40). Surface-exposed variable elements of the B30.2 domain are reminiscent of the complementarity-determining region loops of immunoglobulins. Although we were not able to align the B30.2 variable segments and immunoglobulin CDRs in an instructive manner (data not shown), the TRIM B30.2 variable regions may play roles in the recognition of foreign ligands such as viral capsids.
The evolutionary relationships among TRIM genes in diverse vertebrate species require additional investigation and may indicate functional interactions among the members of this interesting family of proteins. In parallel, the ability of TRIM proteins to modulate infections by viruses associated with these vertebrates should be investigated.
ACKNOWLEDGMENTS
We thank Yvette McLaughlin and Sheri Farnum for manuscript preparation and William Murphy and Jill Slattery for helpful discussions.
This work was supported by a grant from the National Institutes of Health (HL54785) and by a Center for AIDS Research award (P30 AI28691). We also acknowledge the support of the Bristol-Myers Squibb Foundation, the International AIDS Vaccine Initiative, and the late William F. McCarty Cooper. This work was funded in part with federal funds from the Center for Cancer Research of the National Cancer Institute and the National Institutes of Health under contract no. NO1-CO-12400.
The content of this publication does not necessarily reflect the views of the Department of Health and Human Services, nor does mention of trade names, commercial products, or organizations imply endorsement by the U.S. Government.
REFERENCES
Artes, E. J., and M. A. Wainberg. 1996. Human immunodeficiency type 1 reverse transcriptase and early events in reverse transcription. Adv. Virus Res. 46:97-163.
Benveniste, R. E., R. Heinemann, G. L. Wilson, R. Callahan, and G. J. Todaro. 1974. Detection of baboon type C viral sequences in various primate tissues by molecular hybridization. J. Virol. 14:56-67.
Benveniste, R. E., L. Kuller, S. T. Roodman, S. L. Hu, and W. R. Morton. 1993. Long-term protection of macaques against high-dose type D retrovirus challenge after immunization with recombinant vaccinia virus expressing envelope glycoproteins. J. Med. Primatol. 22:74-79.
Besnier, C., Y. Takeuchi, and G. Towers. 2002. Restriction of lentivirus in monkeys. Proc. Natl. Acad. Sci. USA 99:11920-11925.
Besnier, C., L. Ylinen, B. Strang, A. Lister, Y. Takeuchi, S. P. Goff, and G. J. Towers. 2003. Characterization of murine leukemia virus restriction in mammals. J. Virol. 77:13403-13406.
Bieniasz, P. D. 2003. Restriction factors: a defense against retroviral infection. Trends Microbiol. 11:286-291.
Bonilla, W. V., D. D. Pinschewer, P. Klenerman, V. Rousson, M. Gamboli, P. P. Pandolfi, R. M. Zinkernagel, M. S. Salvato, and H. Hengartner. 2002. Effects of promyelocytic leukemia protein on virus-host balance. J. Virol. 76:3810-3818.
Chee, A. V., P. Lopez, P. P. Pandolfi, and B. Roizman. 2003. Promyelocytic leukemia protein mediates interferon-based anti-herpes simplex virus 1 effects. J. Virol. 77:7101-7105.
Cowan, S., T. Hatziioannou, T. Cunningham, M. A. Muesing, H. G. Gottlinger, and P. D. Bieniasz. 2002. Cellular inhibitors with Fv1-like activity restrict human and simian immunodeficiency virus tropism. Proc. Natl. Acad. Sci. USA 99:11914-11919.
Daniel, M. D., N. L. Letvin, N. W. King, M. Kannagi, P. K. Sehgal, R. D. Hunt, P. J. Kanki, M. Essex, and R. C. Desrosiers. 1985. Isolation of T-cell tropic HTLV-II-like retrovirus from macaques. Science 228:1201-1204.
Freed, E. O. 1998. HIV-1 gag proteins: diverse functions in the virus life cycle. Virology 251:1-15.
Goodchild, N. L., D. A. Wilkinson, and D. L. Mager. 1993. Recent evolutionary expansions of a subfamily of RTVL-H human endogenous retrovirus-like elements. Virology 196:778-788.
Hatziioannou, T., S. Cowan, S. P. Goff, P. D. Bieniasz, and G. J. Towers. 2003. Restriction of multiple divergent retroviruses by Lv1 and Refl. EMBO J. 22:385-394.
Hatziioannou, T., S. Cowan, U. K. von Schwedler, W. I. Sundquist, and P. Bieniasz. 2004. Species-specific tropism determinants in the human immunodeficiency virus type 1 capsid. J. Virol. 78:6005-6012.
Hatziioannou, T., D. Perez-Caballero, A. Yang, S. Cowan, and P. D. Bieniasz. 2004. Retrovirus resistance factors Ref1 and Lv1 are species-specific variants of TRIM5. Proc. Natl. Acad. Sci. USA 101:10774-10779.
Henry, J., I. H. Mather, M. F. McDermott, and P. Pontarotti. 1998. B30.2-like domain proteins: update and new insights into a rapidly expanding family of proteins. Mol. Biol. Evol. 15:1696-1705.
Himathongkham, S., and P. A. Luciw. 1996. Restriction of HIV-1 (subtype B) replication at the entry step in rhesus macaque cells. Virology 219:485-488.
Hofmann, W., D. Schubert, J. LaBonte, L. Munson, S. Gibson, J. Scammell, P. Ferrigno, and J. Sodroski. 1999. Species-specific, postentry barriers to primate immunodeficiency virus infection. J. Virol. 73:10020-10028.
Johnson, W. E., and J. M. Coffin. 1999. Constructing primate phylogenies from ancient retrovirus sequences. Proc. Natl. Acad. Sci. USA 96:10254-10260.
Keckesova, Z., L. M. J. Ylinen, and G. J. Towers. 2004. The human and African green monkey TRIM genes encode Ref1 and Lv1 retroviral restriction factor activities. Proc. Natl. Acad. Sci. USA 101:10780-10785.
Keightley, P. D., and D. J. Gaffney. 2003. Functional constraints and frequency of deleterious mutations in noncoding DNA of rodents. Proc. Natl. Acad. Sci. USA 100:13402-13406.
Kim, H. S., O. Takenaka, and T. J. Crow. 1999. Isolation and phylogeny of endogenous retrovirus sequences belonging to the HERV-W family in primates. J. Gen. Virol. 80:2613-2619.
Kootstra, N. A., C. Munk, N. Tonnu, N. R. Landau, and I. M. Verma. 2003. Abrogation of postentry restriction of HIV-1-based lentiviral vector transduction in simian cells. Proc. Natl. Acad. Sci. USA 100:1298-1303.
Kumar, S., K. Tamura, and M. Nei. 1994. MEGA: molecular evolutionary genetics analysis software for microcomputers. Comput. Appl. Biosci. 10:189-191.
Li, W. H. 1993. Unbiased estimation of the rates of synonymous and nonsynonymous substitution. J. Mol. Evol. 36:96-99.
Mager, D. L., and J. D. Freeman. 1995. HERV.H endogenous retroviruses—presence in the New-World branch but amplification in the Old-World primate lineage. Virology 213:395-404.
Mang, R., J. Maas, A. C. van der Kuyl, and J. Goudsmit. 2000. Papio cynocephalus endogenous retrovirus among Old World monkeys: evidence for coevolution and ancient cross-species transmission. J. Virol. 74:1578-1586.
Marcello, A., A. Ferrari, V. Pellegrini, G. Pegoraro, M. Lusic, F. Beltram, and M. Giacca. 2003. Recruitment of human cyclin T1 to nuclear bodies through direct interaction with the PML protein. EMBO J. 22:2156-2166.
Mariana-Costantini, R., T. M. Horn, and R. Callahan. 1989. Ancestry of a human endogenous retrovirus family. J. Virol. 63:4982-4985.
Munk, C., S. M. Brandt, G. Luccero, and N. R. Landau. 2002. A dominant block to HIV-1 replication at reverse transcription in simian cells. Proc. Natl. Acad. Sci. USA 99:13843-13848.
Nisole, S., C. Lynch, J. P. Stoye, and M. W. Yap. 2004. A Trim5-cyclophilin A fusion protein found in owl monkey kidney cells can restrict HIV-1. Proc. Natl. Acad. Sci. USA 101:13324-13328.
Orimo, A., N. Tominaga, K. Yoshimura, Y. Yamauchi, M. Nomura, M. Sato, Y. Nogi, M. Suzuki, H. Suzuki, K. Ikeda, S. Inoue, and M. Mauramatsu. 2000. Molecular cloning of ring finger protein 21 (RNF21)/interferon-responsive finger protein (ifp1), which possesses two RING-B box-coiled coil domains in tandem. Genomics 69:143-149.
Owens, C. M., P. C. Yang, H. Gottlinger, and J. Sodroski. 2003. Human and simian immunodeficiency virus capsid proteins are major viral determinants of early, postentry replication blocks in simian cells. J. Virol. 77:726-731.
Owens, C. M., B. Song, M. J. Perron, P. C. Yang, M. Stremlau, and J. Sodroski. 2004. Binding and susceptibility to post-entry restriction factors in monkey cells are specified by distinct regions of the human immunodeficiency virus type 1 capsid. J. Virol. 78:5423-5437.
Perron, M., M. Stremlau, B. Song, W. Ulm, R. Mulligan, and J. Sodroski. 2004. TRIM5 mediates the postentry block to N-tropic murine leukemia viruses in human cells. Proc. Natl. Acad. Sci. USA 101:11827-11832.
Reus, K., J. Mayer, M. Sauter, H. Zischler, N. Müller-Lantzsch, and E. Meese. 2001. HERV-K (OLD): ancestor sequences of the human endogenous retrovirus family HERV-K (HML-2). J. Virol. 75:8917-8926.
Reymond, A., G. Meroni, A. Fantozzi, G. Merla, S. Cairo, L. Luzi, D. Riganelli, E. Zanaria, S. Messali, S. Cainarca, A. Guffanti, S. Minucci, P. G. Pelicci, and A. Ballabio. 2001. The tripartite motif family identifies cell compartments. EMBO J. 20:2140-2151.
Rhodes, D. A., G. Ihrke, A. T. Reinicke, G. Malcherek, M. Towey, D. A. Isenberg, and J. Trowsdale. 2002. The 52,000 MW Ro/SS-A autoantigen in Sjogren's syndrome/systemic lupus erythematosus (Ro52) is an interferon-gamma inducible tripartite motif protein associated with membrane proximal structures. Immunology 106:246-256.
Rogers, J. S., and D. I. Swofford. 1998. A fast method for approximating maximum likelihoods of phylogenetic trees from nucleotide sequences. Syst. Biol. 47:77-89.
Sayah, D. M., E. Sokolskaja, L. Berthoux, and J. Luban. 2004. Cyclophilin A retrotransposition into TRIM5 explains owl monkey resistance to HIV-1. Nature 430:569-573.
Seto, M. H., H.-L. C. Liu, D. Zajchowski, and M. Whitlow. 1999. Protein fold analysis of the B30.2-like domain. Proteins 35:235-249.
Sharp, P. M., E. Bailes, R. R. Chaudhuri, C. M. Rodenburg, M. O. Santiago, and B. H. Hahn. 2001. The origins of acquired immune deficiency syndrome viruses: where and when? Philos. Trans. R. Soc. Lond. B Biol. Sci. 356:867-876.
Shibata, R., H. Sakai, M. Kawamura, K. Tokunaga, and A. Adachi. 1995. Early replication block of human immunodeficiency virus type 1 in monkey cells. J. Gen. Virol. 76:2723-2730.
Song, B., H. Javanbakht, M. Perron, D. Park, M. Stremlau, and J. Sodroski. 2005. Retrovirus restriction by TRIM5 variants from Old World and New World primates. J. Virol. 79:3930-3937.
Steele, P. E., M. A. Martin, A. B. Rabson, T. Bryan, and S. J. O'Brien. 1986. Amplification and chromosomal dispersion of human endogenous retroviral sequences. J. Virol. 59:545-550.
Steinhuber, S., M. Brack, G. Hunsmann, H. Schwelberger, M. P. Dierich, and W. Vogetseder. 1995. Distribution of human endogenous retrovirus HERV-K genomes in humans and different primates. Hum. Genet. 96:188-192.
Stoye, J. P. 2002. An intracellular block to primate lentivirus replication. Proc. Natl. Acad. Sci. USA 99:11549-11551.
Stremlau, M., C. M. Owens, M. J. Perron, M. Kiessling, P. Autissier, and J. Sodroski. 2004. The cytoplasmic body component TRIM5 restricts human immunodeficiency virus (HIV-1) infection in Old World monkeys. Nature 427:848-853.
Stremlau, M., M. Perron, S. Welikala, and J. Sodroski. 2005. Species-specific variation in the B30.2(SPRY) domain of TRIM5 determines the potency of human immunodeficiency virus type 1 restriction. J. Virol. 79:3139-3145.
Thali, M., A. Bukovsky, E. Kondo, B. Rosenwirth, C. T. Walsh, J. Sodroski, and H. G. Gottlinger. 1994. Functional association of cyclophilin A with HIV-1 virions. Nature 372:363-365.
Thomas, E. E., N. Srebro, J. Sebat, N. Navin, J. Healy, B. Mishra, and M. Wigler. 2004. Distribution of short paired duplications in mammalian genomes. Proc. Natl. Acad. Sci. USA 101:10349-10354.
Thompson, J. D., T. J. Gibson, F. Plewniak, F. Jeanmougin, and D. G. Higgins. 1997. The CLUSTAL_X windows interface: flexible strategies for multiple sequence alignment aided by quality analysis tools. Nucleic Acids Res. 25:4876-4882.
Toniato, E., X. P. Chen, J. Losman, V. Flati, L. Donahue, and P. Rothman. 2002. TRIM8/GERP RING finger protein interacts with SOCS-1. J. Biol. Chem. 277:37315-37322.
Towers, G., M. Bock, S. Martin, Y. Takeuchi, J. P. Stoye, and O. Danos. 2000. A conserved mechanism of retrovirus restriction in mammals. Proc. Natl. Acad. Sci. USA 97:12295-12299.
Towers, G., M. Collins, and Y. Takeuchi. 2002. Abrogation of Ref1 retrovirus restriction in human cells. J. Virol. 76:2548-2550.
Towers, G. J., T. Hatziioannou, S. Cowan, S. P. Goff, J. Luban, and P. D. Bieniasz. 2003. Cyclophilin A modulates the sensitivity of HIV-1 to host restriction factors. Nat. Med. 9:1138-1143.
Tristem, M. 2000. Identification and characterization of novel human endogenous retrovirus families by phylogenetic screening of the human genome mapping project database. J. Virol. 74:3715-3730.
Turelli, P., V. Doucas, E. Craig, B. Mangeat, N. Klages, R. Evans, G. Kalpana, and D. Trono. 2001. Cytoplasmic recruitment of INI1 and PML on incoming HIV preintegration complexes: interference with early steps of viral replication. Mol. Cell 7:1245-1254.
van der Kuyl, A. C., J. T. Dekker, and J. Goudsmit. 1995. Distribution of baboon endogenous virus among species of African monkeys suggests multiple ancient cross-species transmissions in shared habitats. J. Virol. 69:7877-7887.
van der Kuyl, A. C., R. Mang, J. T. Dekker, and J. Goudsmit. 1997. Complete nucleotide sequence of simian endogenous type D retrovirus with intact genome organization: evidence for ancestry to simian retrovirus and baboon endogenous virus. J. Virol. 71:3666-3676.
Watanabe, H., A. Fujiyama, M. Hattori, T. D. Taylor, A. Toyoda, Y. Kuroki, H. Noguchi, A. BenKahla, H. Lehrach, R. Sudbrak, M. Kube, S. Taenzer, P. Galgoczy, M. Platzer, M. Scharfe, G. Nordsiek, H. Blocker, I. Hellmann, P. Khaitovich, S. Paabo, R. Reinhardt, H. J. Zheng, X. L. Zhang, G. F. Zhu, B. F. Wang, G. Fu, S. X. Ren, G. P. Zhao, Z. Chen, Y. S. Lee, J. E. Cheong, S. H. Choi, K. M. Wu, T. T. Liu, K. J. Hsiao, S. F. Tsai, C. G. Kim, S. Oota, T. Kitano, Y. Kohara, N. Saitou, H. S. Park, S. Y. Wang, M. L. Yaspo, and Y. Sakaki. 2004. DNA sequence and comparative analysis of chimpanzee chromosome 22. Nature 429:382-388.
Whitcomb, J. M., and S. H. Hughes. 1992. Retroviral reverse transcription and integration: progress and problems. Annu. Rev. Cell Biol. 8:275-306.
Wilkinson, D. A., D. L. Mager, and J.-A. C. Leong. 1994. Endogenous human retroviruses, p. 465-535. In J. A. Levy (ed.), The Retroviridae, vol. III. Plenum Press, New York, N.Y.
Xu, L., L. Yang, P. K. Moitra, K. Hashimoto, P. Rallabhandi, S. Kaul, G. Meroni, J. P. Jensen, A. M. Weissman, and P. D'Arpa. 2003. BTBD1 and BTBD2 colocalize to cytoplasmic bodies with the RBCC/tripartite motif protein, TRIM5delta. Exp. Cell Res. 288:84-93.
Yap, M. W., S. Nisole, C. Lynch, and J. P. Stoye. 2004. Trim5 protein restricts both HIV-1 and murine leukemia virus. Proc. Natl. Acad. Sci. USA 101:10786-10791.(Byeongwoon Song, Bert Gol)