A new, expressed multigene family containing a hot spot for insertion of retroelements is associated with polymorphic subtelomeric regions of Trypanosoma brucei

Abstract

We describe a novel gene family that forms clusters in subtelomeric regions of Trypanosoma brucei chromosomes and partially accounts for the observed clustering of retrotransposons. The ingi and ribosomal inserted mobile element (RIME) non-LTR retrotransposons share 250 bp at both extremities and are the most abundant putatively mobile elements, with about 500 copies per haploid genome. From cDNA clones and subsequently in the T. brucei genomic DNA databases, we identified 52 homologous gene and pseudogene sequences, 16 of which contain a RIME and/or ingi retrotransposon inserted at exactly the same relative position. Here these genes are called the RHS family, for retrotransposon hot spot. Comparison of the protein sequences encoded by RHS genes (21 copies) and pseudogenes (24 copies) revealed a conserved central region containing an ATP/GTP-binding motif and the RIME/ingi insertion site. The RHS proteins share between 13 and 96% identity, and six subfamilies, RHS1 to RHS6, can be defined on the basis of their divergent C-terminal domains. Immunofluorescence and Western blot analyses using RHS subfamily-specific immune sera show that RHS proteins are constitutively expressed and occur mainly in the nucleus. Analysis of Genome Survey Sequence databases indicated that the Trypanosoma brucei diploid genome contains about 280 RHS (pseudo)genes. Among the 52 identified RHS (pseudo)genes, 48 copies are in three RHS clusters located in subtelomeric regions of chromosomes Ia and II and adjacent to the active bloodstream form expression site in T. brucei strain TREU927/4 GUTat10.1. RHS genes comprise the remaining sequence of the size-polymorphic "repetitive region" described for T. brucei chromosome I, and a homologous gene family is present in the Trypanosoma cruzi genome.

FIG. 1.

New multigene family containing a hot spot for retroelement insertion. (A) Map of homologous cDNAs (cDNA-004, cDNA-005, cDNA-040, and cDNA-132 with accession numbers AF403385, AF403388, AF403386, and AF403387, respectively) isolated from a T. brucei AnTat1 library. The large and small black boxes represent coding and noncoding sequences, respectively, and the white box shows the 5" end of a non-LTR retrotransposon ingi. The 12-bp sequence upstream of the retrotransposon is shown below the cDNA-005 map. (B) Schematic map of the RHS genes and pseudogenes sequenced from five different cosmid clones of AnTat1 genomic DNA (Cos-02, -03, -12, -17, and -23 with accession numbers AY046893, AY046894, AY046896, AY046895, and AY046897S1, and AY046897S2, respectively). Coding and noncoding sequences are represented by large and small boxes, respectively. The black boxes correspond to sequences presenting at least 80% identity, while the hatched boxes correspond to unrelated sequences. Premature stop codons (S) and retroelement RIME (○) or ingi (•) insertion are indicated above the maps. The RIME and ingi retroelements inserted in the RHS pseudogene of Cos-12 and Cos-23 clones are shown below the maps. The 12-bp repetitive duplication sequences flanking the retroelements are shown below the RIME/ingi map.

FIG. 2.

Schematic representation of RHS (pseudo)genes present in T. brucei (TREU927/4) database. The name of each gene or pseudogene is on the left; those in a grey box are potentially functional, while the unboxed names correspond to nonfunctional pseudogenes. Arrows, chimeric sequences; ∗, sequences (RHS genes and conserved flanking regions) present in the database under accession numbers AY046887 (RHS1a), AY046888 (RHS2a), AY046889 (RHS3a), AY046890 (RHS4a), AY046891 (RHS5a), and AY046892 (RHS6a). The amino acid sequences of box 1 encoded by the RHS1a to RHS6a genes which are labeled by an asterisk are compared in Fig. 3. Chromosome and/or BAC clones containing the sequence are indicated on the right. The color code for each subfamily is as follows: RHS1 (), RHS2 (), RHS3 (▭), RHS4 (), RHS5 (), and RHS6 (). In the coding sequences, the positions of frame shifts (F), premature stop codons (S), and RIME (○) and/or ingi (•) insertions are indicated. Multiple retroelement insertions are shown by a corresponding number of open and filled circles. Horizontal lines in the middle of ϕRHS1h and ϕRHS2f coding sequences represent a deletion of a part of the RHS coding sequence. The most conserved (box 1) and most divergent (box 2) coding regions between RHS subfamilies are shaded.

FIG. 3.

Amino acid alignment of box 1 domain from RHS1a-6a proteins and related T. cruzi protein. The aligned box 1 sequences indicated by an asterisk in the left margin of Fig. 2 are representative of each RHS subfamily. Dashes were introduced to maximize the alignment. Identical amino acids are shaded and in bold. The positions of the ATP/GTP-binding motif and the duplicated insertion site sequence associated with non-LTR retroelement (RIME and/or ingi) insertion are indicated above the alignment. The last row (T.cr.) represents the chimeric RHS-related protein identified in the T. cruzi database, as shown in Fig. 10. All boxed and bold residues of the T. cruzi sequence are identical to residues located at the same relative position in the T. brucei RHS proteins.

FIG. 4.

Phylogenetic analysis of RHS proteins and RHS (pseudo) genes. (A) The phylogenetic tree was constructed with three full-length RHS amino acid sequences (a, b, and c) specific for each RHS subfamily except RHS6, for which only a single full-length sequence has been identified. Premature stop codons and frame shifts present in three RHS pseudogenes (ϕRHS1c, ϕRHS3b, and ϕRHS3c) were corrected to obtain a chimeric full-length RHS protein. The two other phylogenetic trees were constructed using the full-length RHS2 (B) or RHS1 (C) (pseudo)gene sequences located downstream of the RIME/ingi insertion site. The grey boxes include RHS (pseudo)genes with RIME (○) and/or ingi (•) retroelement(s). In each panel, the scale bar represents a genetic distance of 0.1 or 0.01 amino acid or nucleotide substitutions per site, and numbers given beside the nodes represent the percentage of bootstrap replicas yielding these trees.

FIG. 5.

Comparison of 12-bp sequences flanking RIME or ingi elements inserted into RHS pseudogenes. The RIME and ingi retroelements (□), the RHS flanking pseudogenes (░⃞), and the unknown flanking region (▨) are schematically represented. The 12 bp located at the junction between the RIME/ingi retroelements and the RHS/unknown flanking regions, and also between retroelements, are indicated by numbered black boxes, and the corresponding sequence is indicated on the right. The nucleotides in bold correspond to the duplicated region associated with the RIME/ingi insertion, and the boxes define the conserved region. The AAAAAA and CCCTGG sequences correspond to the end and the beginning of the RIME/ingi elements, respectively, and dots represent the RHS or unknown sequences. The crosses (✖) in the middle of retroelements indicate in which RIME or ingi element homologous recombination probably occurred. The accession numbers of ϕRHS2j, ϕRHS4d, and ϕRHS1 in Cos-12 and Cos-23, ϕRHS1e, ϕRHS1g, ϕRHS1f, and ϕRHS2g are AL359782 (ChrIa), AC008146 (BAC-30P15), AY046896, AY046897S1 and AY046897S2, AC079606 (BAC-3B10), AC087701 (BAC-26P8), AL359782 (ChrIa), and AC087701 (BAC-26P8), respectively.

FIG. 6.

Western blot analysis of RHS proteins. Lysates (4 × 10⁷ cells) of T. brucei procyclic form EATRO1125 (PF) and bloodstream form AnTat1 (BF) were analyzed by Western blotting with the immune sera specific for tubulin, RHS1, RHS2, RHS4, RHS5, and RHS6. The positions of the molecular mass markers (in kilodaltons) are indicated on the left and right, and the names of the immune sera is given under each blot.

FIG. 7.

Immunolocalization of RHS proteins. T. brucei procyclic cells (EATRO1125) were stained with anti-RHS1 (A), anti-RHS2 (B), anti-RHS4 (C), anti-RHS5 (D), and anti-RHS6 (E) immune sera (first column) and with DAPI (second column). Respective phase contrast (phase) images are shown in the third column. Bar = 5 μm.

FIG. 8.

Southern blot analysis of RHS (pseudo)genes. Genomic DNA from T. brucei TREU927/4 (lanes 1) and 427 (lanes 2) was digested to completion with HincII (RHS1 and RHS5), ClaI (RHS2 and RHS6), AseI (RHS3), or HpaII and KpnI (RHS4) and was analyzed by hybridization with the RHS1- to RHS6-specific probes to a Southern blot. The names of the probes and molecular size markers (in kilobases) are indicated below and on the left of each panel, respectively.

FIG. 9.

Gene organization of the BAC-26P8 clone which contains the B-ES of VSG 10.1. (A) Map of the B-ES of VSG 10.1 and upstream regions in TREU927/4 GUTat10.1, as previously described (36). The locations of the genes and retrotransposons (RIME and ingi) in BAC-26P8 are shown (□). Genes shown above the line are oriented towards the telomere, whereas those shown below the line are oriented away from the telomere. ▪, 50-bp, 70-bp, and telomere repeats. Expressions site-associated genes (ESAGs) are numbered 1 to 8 and the black flag indicates the position of the B-ES promoter. The B-ES, starting at the promoter, is highlighted by a large grey box. The region containing uncharacterized repeated sequences located upstream of the B-ES is indicated. (B) Detailed analysis of the region containing RHS pseudogenes in BAC-26P8. Sequences encoding RHS1 to RHS4 pseudogenes (large boxes) are shown using the same color code as in Fig. 2. The name and orientation of each RHS pseudogene are indicated above or below the boxes. The upstream (5") and downstream (3") sequences conserved between the different RHS (pseudo)genes of the same or different subfamilies are indicated by intermediate size boxes containing horizontal lines. All the RIME and ingi retroelements that are inserted into RHS genes are indicated by small white boxes.

FIG. 10.

Characterization of an RHS-related multigene family in T. cruzi. (A) Map of the pBAC52 and f1 genomic clones from T. cruzi CL-Brener and Maracay strains, respectively, as previously reported by Olivares et al. (49). Grey, hatched, and white boxes represent the L1Tc retrotransposons (5 kb) and the RS1Tc (1.4 kb) and Seq3Tc (1.9 kb) repetitive elements, respectively. The conserved duplicated 9-bp sequences (TGCAGACAT) flanking the L1Tc elements and the corresponding L1Tc insertion site in the f1 sequences are shown under the map. (B) Chimeric RS1Tc/Seq3Tc sequence coding for an RHS-related protein. To generate an RS1Tc/Seq3Tc-related sequence containing a large open reading frame (black box on the map), a selection of nine representative T. cruzi GSS were assembled. The relative positions and accession numbers of the selected GSSs are indicated under the map. Box 1 indicates the amino acid sequence, aligned with the T. brucei RHS proteins in Fig. 3.