Abstract
Human T-cell leukemia virus type 1 (HTLV-1) is the etiologic agent of adult T-cell leukemia/lymphoma (ATLL), HTLV-1-associated myelopathy/tropical spastic paraparesis (HAM/TSP), and other diseases. The mechanisms of virus pathogenesis are still obscure. The occurrence of defective proviruses in HTLV-1-infected cell lines and the peripheral blood mononuclear cells (PBMC) of infected individuals is a frequent feature of virus infection. We detected defective proviruses with large internal deletions in PBMC from ATLL and HAM/TSP patients and in asymptomatic HTLV-1 carriers. Seventeen PCR-amplified defective proviruses were sequenced, and three types of deletions were found. Besides truncated MA and the 5′ end of the genome, truncated CA, truncated SU, and more frequently truncated TM linked to the pX region were detected. Reverse transcription-PCR analysis of PBMC from ATLL patients and asymptomatic carriers also revealed RNA transcripts with large internal deletions. Analysis of two RT-PCR cDNA clones confirmed a Gag-TM-pX structure of the transcripts. Most defective proviruses contained numerous internal stop codons, but some were capable of coding for the truncated MA linked to a variable out-of-frame peptide. Cloned defective proviruses with long open reading frames were subjected to in vitro transcription-translation followed by radioimmunoprecipitation, which showed expression of chimeric proteins between 8 and 12 kDa. Possible roles of defective proviruses and chimeric proteins are discussed, although there is no firm association with pathogenesis.
The human T-cell leukemia virus type 1 (HTLV-1) is a retrovirus which causes two distinct pathologies: adult T-cell leukemia/lymphoma (ATLL) (10, 27) and chronic progressive myelopathy-tropical spastic paraparesis (TSP), also called HTLV-1-associated myelopathy (HAM) (5, 26). Recent studies have suggested that HTLV-1 is also associated with other human diseases (23).
Previous comparative studies of different HTLV-1 isolates indicated the following: the protein patterns of the viruses from cells of individuals with ATL and HAM/TSP were identical (6), DNA blotting (11, 33) and sequence comparisons of proviruses from ATLL and HAM/TSP revealed nearly 97% homology (4, 21), and restriction endonuclease analysis did not reveal any selected integration sites for the proviruses in the DNA of infected individuals (33). Occasional cases of ATLL associated with TSP-HAM pathology have been reported (14, 24), and infection of a patient with blood from a HAM/TSP donor resulted in HAM/TSP in the recipient (8). Thus, different diseases in association with one virus remains a paradox of HTLV-1. The mechanisms by which the virus induces different diseases and the cofactors, either virus associated or host related, that contribute to different forms of disease manifestations or progression remain to be determined.
Defective proviruses have been observed in cells from ATLL patients and could be important elements in pathogenesis (9, 17, 32). We have therefore compared the presence of defective proviruses in HTLV-1 asymptomatic carriers and in patients with ATLL or HAM/TSP.
Detailed analysis of defective proviruses were previously carried out on the MT-2 cell line (15, 16). In this cell line, besides a complete provirus, seven defective proviruses were detected by Southern hybridization, and four of them were found to be identical with a large internal deletion resulting in the 5′ portion of the gag gene being linked to a pX region (15, 16). These proviruses contained an open reading frame (ORF) coding for a chimeric protein (p28) composed of Gag (MA [matrix] and truncated CA [capsid]) and a short pX fragment (12). RNA transcripts for MT-2 cell defective provirus are packaged in virions and can be translated into a p28 chimeric protein (25), and they establish new provirus in culture (1). HTLV-1 proviruses with internal deletions have also been found in other cell lines (9, 33) and in peripheral blood mononuclear cells (PBMC) of ATLL patients (18, 31), as well as defective proviruses in human immunodeficiency virus infection (29).
Using the p28 provirus of MT-2 cells as a prototype, we designed a set of PCR primers corresponding to the Gag and pX regions (25), which allowed us to amplify proviruses with large internal deletions provided they contained the corresponding gag and pX sequences but not the gag-pol-env-pX sequence of the completed provirus. The aim of our research was to detect and analyze possible defective proviruses with large internal deletions in HTLV-1-infected individuals with different pathologies: HAM/TSP, ATLL, and asymptomatic carriers. We report the identification and nucleotide sequence analysis of different proviruses with large internal deletions in DNA of patients with HTLV-1-associated pathologies and show that some are transcribed and can be translated into proteins.
MATERIALS AND METHODS
Patients.
Patients with ATLL or HAM/TSP and asymptomatic HTLV-1 carriers were studied. HTLV-1 infection of each individual was confirmed by Western blotting of serum and by PCR of PBMC DNA using pol and tax pairs of primers. HAM/TSP patients were also tested for the presence of complete provirus by Southern hybridization with corresponding subgenomic probes. Blood was collected by venipuncture, and PBMC were separated on a Ficoll-Hypaque gradient, washed twice in phosphate-buffered saline (PBS), and stored at −80°C until use. Some PBMC samples and HTLV-1 immortalized cell lines from HAM/TSP patients were kindly provided by G. Peries (Hôpital St. Louis, Paris, France) and have been previously described (6, 7). Most PBMC samples from patients with ATLL, asymptomatic carriers, and some patients with HAM/TSP were from individuals studied at the Royal Marsden Hospital and St. Mary's Hospital, London, United Kingdom, with ethical approval.
Cells.
MT-2 and CEM cells were cultivated in RPMI 1640 medium supplemented with 10% fetal calf serum and antibiotics as previously described (3). Cells were passaged every 4 days.
Plasmids.
pMT-2-42 containing defective HTLV-1 provirus (12) was a gift from M. Hatanaka (Institute of Virus Research, Kyoto University, Kyoto, Japan). The expression vector pTargeT was purchased from Promega, Madison, Wis.
Sera and monoclonal antibodies.
Serum samples (verified by different immunological tests) were obtained from two HTLV-1-infected individuals with HAM/TSP. Monoclonal antibodies against p28, p19, p24, and p21E were obtained from Chemicon Int. (Temacula, Calif.). Anti-human immunoglobulin G-alkaline phosphatase (IgG-AP) and anti-mouse IgG-AP were obtained from Boehringer-Mannheim, Mannheim, Germany.
Isolation of DNA and RNA.
Total genomic DNA was extracted from 3 × 106 to 5 × 106 PBMC using a DNA preparation kit (Promega) and was analyzed by agarose gel electrophoresis. From each cell preparation, 107 cells were used for DNA extraction by phenol-chloroform. Total RNA was isolated from 3 × 106 to 5 × 106 PBMCs using RNAzol B (Biogenetics, Poole, United Kingdom) according to the manufacturer's protocol.
PCR and reverse transcription-PCR (RT-PCR).
Sets of primers for nested PCR corresponding to the long terminal repeat (LTR), Gag, and pX regions of HTLV-1 were designed based on the relevant viral sequences (30) and were synthesized by Oswel Labs (Southampton, United Kingdom). HTLV-1 proviruses are available under accession no. U19949 and D13784. Their sequences are as follows: 5′LTR (forward), 5′ GAC AAT GAC CAT GAG CCC CAA (positions 2 to 22); ATG gag (forward), 5′ TAG GCT ATG GGC CAA ATC TT (positions 798 to 817); A1 p28 (forward), 5′ CAA ATC TTT TCC CGT AGC GCT AGC (positions 809 to 832); A2 p28 (forward), 5′ TCC AGT TAC GAT TTC CAC CAG TTG (positions 918 to 940); pX (reverse), 5′ AGG AGG ATT TGA TGG GAG AGG TTA (ATK positions 6727 to 6704); B1 p28 (reverse), 5′ GGA GGC GAT GTA GTT GCA ATA (ATK positions 6683 to 6663); B2 p28 (reverse), 5′ ATG TGC TTG GTT TAC AGG GAT (ATK positions 6640 to 6659); Stop p28 (reverse), 5′ GGT TAA TTA TTG GCA GGG GAG (positions in ATK 6708 to 6687); HTLV-1 pol and tax, as published by Kwok et al. (19); HTLV-1 pol (forward), 5′ AGA TAC AGG AGC AGA CAT GAC (ATK positions 2242 to 2262); HTLV-1 pol (reverse), 5′ GGA CTG GAA AAC ACT ACA GTA (AKT positions 3547 to 3528); HTLV-1 tax forward, 5′ TTT CGG ATA CCC AGT CTA CG (ATK positions 7335 to 7354); and HTLV-1 tax reverse, 5′ GAT AAC GCG TCC ATC GAT GG (ATK positions 7472 to 7491).
We used in each reaction 1 μg of DNA (equivalent to approximately 150,000 cells). Hot-start PCR amplification was performed as instructed by the supplier of Taq polymerase (Perkin-Elmer/Roche, Nutley, N.J.). After initial denaturation at 95°C for 5 min, DNA was subjected to 32 to 40 cycles of amplification (denaturation at 95°C for 1 min, annealing at 48 to 55°C [dependent on primers] for 45 s, extension at 72°C for 1 min). After the last cycle, extension was carried out at 72°C for 10 min, and a 10-μl aliquot of each PCR mixture was electrophoresed in a 1.8% agarose gel containing ethidium bromide. Adequate conditions and amounts of DNAs were confirmed by PCR with β-globin primers, upstream (5′-ACA CAA CTG TGT TCA CTA GC-3′) and downstream (5′-CCA CTT CAT CCA CGT TCA CC-3′). Amplimers were analyzed by agarose (1.8%) electrophoresis with ethidium bromide. High-fidelity PCR was performed with the 5′LTR-pX pair of primers and amplification conditions suggested by the kit supplier (Boehringer-Mannheim). Long PCR fragments were analyzed on 1% agarose gel with ethidium bromide. RT-PCR was performed with a Titan one-tube RT-PCR kit according to the protocol of the supplier (Boehringer-Mannheim).
Southern blot hybridization.
After overnight digestion of 10 μg of each genomic DNA sample with SacI (Promega), gel electrophoresis, and overnight transfer to a Hybond N membrane (Amersham, Little Chalfont, United Kingdom), Southern hybridization was done by using a digoxigenin (DIG) DNA labeling and detection kit (Boehringer-Mannheim). For Southern blots of PCR products, 10 μl of the PCR product was loaded per track. Conditions of prehybridization, hybridization, washing, and color development were as proposed by the manufacturer. The SacI-PstI fragment from pMT-2-42 was used as a probe. DIG-labeled DNA molecular weight markers III and VII (Boehringer-Mannheim) were used for gel calibration. Conditions of DNA transfer and Southern hybridization of PCR-amplified fragments were the same as for the genomic DNA.
Protein electrophoresis, Western blotting, and RIP.
Electrophoresis was performed in a precast sodium dodecyl sulfate-Tricine 10 to 20% polyacrylamide gel (Novex, San Diego, Calif.). Rainbow markers (Amersham) were used for gel calibration. After electrophoresis, proteins were blotted for 2 h at 4°C on 0.22-μm-pore-size nitrocellulose membrane (BA83; Schleicher & Schuell, Dassell, Germany). Membranes were blocked with 8% dry milk in PBS with 0.1% Tween 20 overnight at 4°C. After 2 h of incubation with a corresponding serum or monoclonal antibody, anti-human or anti-mouse IgG-AP (1:2,000 dilution; Boehringer-Mannheim) were used as the secondary antibody. After five washes (5 min each) in PBS-Tween 20, color was developed with 4-chloro-1-naphthol (Bio-Rad, Hercules, Calif.) in PBS with 0.01% H2O2 or BM purple substrate (Boehringer-Mannheim). Virus-specific proteins synthesized in the TNT in vitro transcription-translation system (Promega) were precipitated with monoclonal antibodies or with polyclonal sera from HTLV-1-infected individuals. One hour later, protein G (Sigma, St. Louis, Mo.) was added to increase immune complexes, and incubation was continued for 16 h. Precipitates were washed five times in buffer with 0.5% NP-40, disrupted in radioimmunoprecipitation (RIP) assay buffer at 100°C for 5 min, and applied to the gel.
Cloning and sequencing.
PCR-amplified fragments were purified on 1.5% low-melting-point agarose (Bethesda Research Laboratories, Gaithersburg, Md.) and cloned in the pTargeT vector (Promega). Recombinant clones were analyzed by restriction with EcoRI (Promega) to confirm correct fragment insertion. Inserts were sequenced using a ABI Prism dye terminator cycle sequencing kit with AmpliTaq DNA polymerase (Applied Biosystems/Perkin-Elmer) and analyzed using an Applied Biosystems model 373 automatic DNA sequencer.
In vitro expression of cloned sequences.
Expression of cloned amplimers was carried out in the TNT in vitro transcription-translation system (Promega) with l-[U-14C]leucine (Amersham) as specified by the manufacturer. Synthesized proteins were analyzed by electrophoresis or immunoprecipitated with HAM/TSP patient sera, followed by electrophoresis of precipitated proteins on a 10% NuPAGE gel (Novex).
RESULTS
Defective proviruses revealed by genomic Southern blotting and long-range PCR.
To estimate complete provirus in patient DNA, Southern hybridization analyses of total genomic DNAs obtained from ATLL- and HAM/TSP-derived cell lines were performed with SacI, known to have cleavage sites in both LTRs of the provirus (30, 33). Ten micrograms of DNA extracted from each cell line was subjected to restriction endonuclease digestion followed by electrophoresis and hybridization with the DIG-labeled subgenomic probe. DNA from the human lymphoid cell line CEM was used as a negative control. Southern blot analysis revealed that five of six HAM/TSP-derived cell lines and one ATLL-derived cell line contained both complete and defective proviruses; one ATLL-derived cell line was found to contain only complete provirus (Fig. 1A). One to four defective proviruses ranging from 7 to 2 kb were detected in these DNA samples. Similar results were obtained following EcoRI digestion and Southern hybridization (data not shown).
FIG. 1.
Complete and deleted HTLV-1 proviruses detected by Southern hybridization in the DNA from TSP/HAM- and ATLL-derived cell lines. (A) Total genomic DNA was cleaved with SacI and, after electrophoresis and transfer, hybridized to DIG-labeled subgenomic probe. Lanes: 1, DNA from MT-2 cells; 2 and 3, DNA from ATLL cell lines; 4 to 9, DNA from TSP/HAM-derived cell lines; M, DIG-labeled Boehringer III markers. Positions of full-size proviruses are indicated by the arrow on the left. Most of deleted proviruses are localized between 2 to 3.5 kb. (B) The same DNA samples tested by PCR and Southern hybridization for the defective proviruses with large internal deletions (A1-B1 pair of primers). M, DIG-labeled Boehringer VI markers. Position of the p28 provirus is indicated by an arrow.
PCR analysis performed with the A1-B1 primer pair on the same DNA samples followed by Southern blot hybridization revealed numerous defective proviruses (Fig. 1B). Complete provirus was analyzed in 20 asymptomatic carriers and 7 ATLL patients and was detected in 8 and 6, respectively. However, because of low proviral load especially in asymptomatic carriers, we cannot exclude that more samples contained complete HTLV-1 genomes. Analysis of all samples for the presence of the HTLV-1 genome was confirmed by PCR with the pol-tax pair of primers and by long-range PCR with the 5′LTR-pX pair of primers (data not shown).
Defective proviruses with Gag-pX structure revealed by nested PCR.
To detect defective proviruses with a Gag-pX0 structure, further analysis of the defective proviruses was carried out by nested PCR with two sets of primers, followed by Southern hybridization. In the first round of PCR, we used a pair of primers that corresponds to the 5′ LTR (2 to 22) and pX (6727 to 6704) nucleotides; nested PCR was done with a second pair of primers, ATG (798 to 817) and stop p28 (6708 to 6687). Using these primers, we were able to detect proviruses with internal deletions that contain both the 5′ LTR and the beginning of pX0 region. Another two primer pairs for nested PCR, representing the N-terminal part of MA (A1; 809 to 832) and pX0 (B1; 6684 to 6664), followed by internal MA (A2; 918 to 941) and pX0 (B2; 6639 to 6659), allowed us to detect only the internal part of defective proviruses, with or without a 5′ LTR. Use of both pairs of primers allows efficient amplification of deleted proviruses (Fig. 2).
FIG. 2.
Detection of HTLV-1 proviruses with large internal deletions by nested PCR-Southern blotting in the DNA from PBMCs of the ATLL and TSP/HAM patients. PCR was performed with the A1-B1 and A2-B2 pairs of primers. Lanes: 1, DNA from MT-2 cells; 2, DNA from noninfected CEM cells; 3 to 8, DNAs from ATLL patients; 9 to 12, DNAs from TSP/HAM patients; 13, water control. Position of the p28 provirus from MT-2 cells is marked by an arrow.
To ensure that the defective proviruses are genuine and did not result either from carryover contamination or from PCR artifacts, the specificity of the detection was further analyzed by testing DNA from MT-2 cells, known to contain seven defective proviruses, and by spiking DNA from HTLV-1-free CEM cells with plasmid DNA containing a full-length HTLV-1 genome. As shown in Fig. 3, deleted proviral genomes were amplified only from MT-2 cells, confirming the specificity and validity of our analysis of clinical specimens.
FIG. 3.
Estimation of the specificity of defective provirus amplification by PCR and nested PCR. (A) Amplification of defective proviruses using the LTR-B1 pair of primers (35 cycles). Lanes: M, markers; 1, DNA from MT-2 cells (200 ng); 2, DNA from noninfected CEM cells (500 ng) mixed with 100 ng of plasmid expressing full-size HTLV-1; 3, DNA from CEM cells; 4, no-DNA control. (B) Amplification of defective proviruses using the A1-B1 pair of primers (35 cycles). Lanes: M, markers; 1, DNA from CEM cells (500 ng); 2, DNA from CEM cells mixed with 50 ng of pMT-2-42 (cloned p28 provirus); 3, DNA from CEM cells (500 ng) mixed with 100 ng of pHTLV-1; 4, DNA from MT-2 cells (200 ng). (C) Amplification of defective proviruses by nested PCR using the A2-B2 pair of primers (30 cycles). One microliter from the first reaction (B) was taken for nested PCR. Positions of amplimers are given on the right. Five microliters of each PCR mixture was run on a 2% agarose gel with ethidium bromide.
Overall, 51 DNA samples, comprising 13 cell lines and 38 extracted directly from PBMC of infected individuals, were examined for the defective proviruses (Table 1). After the first round of PCR especially with the 5′LTR-pX pair of primers, we detected amplimers only in a few samples, indicating a relatively low load of defective proviruses containing both the 5′ LTR and corresponding pX region. However, after either Southern hybridization or a second PCR round with the 5′-Gag-pX pair of primers, numerous defective proviruses were detected in the majority of samples.
TABLE 1.
Defective proviruses with large internal deletions between gag and pX detected by nested PCR
Origin of samples | No. tested | No. positive |
---|---|---|
Cell lines from: | ||
ATCC | ||
MT-2 | 1 | 1 |
Hut 102 | 1 | 1 |
ATLL patients | 5 | 3 |
TSP/HAM patients | 6 | 6 |
PBMC from: | ||
ATLL patients | 10 | 8 |
TSP patients | 6 | 6 |
Patient with uveitis | 1 | 0 |
Patient with HIV, HTLV-1 indeterminate by Western blotting, negative PCR for tax and pol | 1 | 1 |
HTLV-1 carriers | 20 | 18 |
Total | 51 | 44 |
The sizes of the majority of Gag-pX0 amplimers varied from nearly 200 to 1,200 bp, signifying a spectrum of internal deletions. Notably, comparison of DNA from PBMC and from cultivated lymphocytes of the same patient with HAM/TSP indicated that the quantity of the deleted proviruses increased during in vitro culture, possibly through mitosis, although not all of the defective proviruses proliferated equally (data not shown). More efficient amplification was observed with the MA-pX pair of primers, indicating that many of the analyzed proviruses either lack a 5′ LTR or have a mutated 5′ LTR.
Nucleotide sequence of the amplimers.
To analyze the internal structure of the PCR-amplified proviral sequences, 17 PCR amplimers from different patients (11 HAM/TSP, 3 ATLL, and 3 carriers) were purified on low-melting-point agarose and sequenced directly in both directions using internal primers. Most but not all deleted proviruses contained stop codons within the sequenced fragments.
As expected from the primers used, the internal parts of all defective proviruses were flanked at the 5′ by the Gag (MA) fragment and at the 3′ by pX sequences, whereas the internal parts (between MA and pX) of these proviruses were quite different. Two proviruses contained the 3′-terminal part of SU (surface) linked to the 5′-terminal part of TM (transmembrane), and 12 of 14 deleted proviruses contained 3′-terminal fragments of TM. Two proviruses contained only 5′-terminal fragments of CA. However, pol-related sequences were not revealed in any of these proviruses. Based on genomic structure, we grouped defective proviruses into three principal classes (Table 2). The MA stretch in all three classes of proviruses was relatively long, from 70 to 360 bp. Differences in size were observed between TM stretches in proviruses from classes 1 and 2, from 6 to nearly 530 bp. CA fragments in both CA-containing proviruses (class 3) were nearly 120 bp. Class 3 strongly resembles the p28 provirus present and expressed in MT-2 cells (15, 25). No clear association with pathology was observed for any of the identified groups of deleted proviruses, although class 2 was more frequent in HAM/TSP patients.
TABLE 2.
Simplified nucleotide structures of different classes of defective HTLV-1 proviruses
Class | Structure | No. detected | Disease (no. affected) |
---|---|---|---|
1 | δMA----------δSU-δTM-pX | 3 | HAM/TSP (2), ATLL (1) |
2 | δMA----------------δTM-pX | 12 | HAM/TSP (8), asymptomatic (3), ATLL (1) |
3 | δMA-δCA-----------------pX | 2 | ATLL (1), HAM/TSP (1) |
Four proviruses with a 5′ LTR and long ORFs and the δMA-δTM-pX structure from HAM/TSP patients were sequenced from the first ATG (position 804) to the pX region (position 6670). Sequence alignments are shown in Fig. 4. The size of MA in these proviruses varied from 126 bp (case 2) to 330 bp (case 4). TM varied from 6 bp (case 4) to 363 bp (case 3). There was a high degree of similarity to the prototype HTLV-1 clone pATK, with identical mutations in all MA, TM, and pX fragments between sequenced proviruses. For example in MA of all four proviruses, G was substituted for A (position 889), in TM of all three (containing the corresponding sequence), T was substituted for C (position 6532) and A was substituted for C (position 6631), and in pX of all four proviruses, G was substituted for A (position 6659) and T was substituted for C (position 6664). It is important to point out that these sequences do not represent cross-contamination in PCRs and that all five base substitutions were exactly the same as in a sequenced HTLV-1 provirus from a HAM/TSP patient (4).
FIG. 4.
Nucleotide alignments of the four deleted HTLV-1 proviruses with δMA-δTM-pX structure and complete HTLV-1 (ATK). All proviruses were detected in TSP/HAM patients. The upper line is the nucleotide sequence of ATK; nucleotide positions are given in parentheses. Only differences in the defective proviruses are indicated. Aligned identical bases are indicated (---). Termination of codon of the env gene is underlined. Numbers 1 to 4 correspond to the proviruses detected in different individuals.
Analysis of proviruses 1 to 3 indicated that they are capable of coding for 8 to 12-kDa proteins. They were initiated at the first ATG (position 804) followed by an MA fragment at the N-terminus and a chimeric sequence at the C-terminal part of the molecule. The chimeric C-terminal part was derived as a consequence of single nucleotide deletions or insertions resulting in a change of the reading frame as follows: in the first provirus, insertion of C (position 133); in the second provirus, deletion of G (position 6503); and in the third provirus, insertion of G (between 1133 and 1134). These mutations occurred in the regions of MA-TM junctions, and provirus 3 was detected in PBMC of a patient who rapidly developed HAM/TSP after blood transfusion (8). Provirus 4 contained deletion of G (position 807) in the second codon, so that the putative synthesized protein is totally out of frame, yet the coding capacity of this unusual ORF is nearly 4.2 kDa. Structures of the ORFs of these proviruses are shown in Fig. 5.
FIG. 5.
Deduced protein sequences coded by the ORFs of defective HTLV-1 proviruses. Virus-specific sequences are marked in bold; mutated in-frame amino acids (aa) are given in small letters; out-of-frame amino acids are underlined.
Detection of chimeric transcripts in PBMC by RT-PCR.
To confirm that expression of defective proviruses with large internal deletions could take place in vivo, we isolated total RNA from fresh PBMC of 6 ATLL and 14 asymptomatic carriers for RT-PCR or RT-PCR plus Southern hybridization with the A1-B1 pair of primers and corresponding probes (Fig. 6). The cDNAs ranging from nearly 280 to 950 bp were revealed in 11 of 14 asymptomatic carriers and in 5 of 6 ATLL patients. Short transcripts (nearly 280 bp) were more frequently detected. No DNA contamination was found in the same samples examined by direct PCR with the same pair of primers. Some of these amplimers were sequenced using the A2-B2 pair of primers; alignment of the two cDNAs and complete HTLV-1 provirus (Fig. 7) shows high homology between the two cDNAs. Seven substitutions were detected in a 192-bp MA fragment of both amplimers, resulting in five amino acid changes. Replacements found were Ala for Val (positions 976 and 977), Arg for Trp (position 978), Ser for Leu (position 1114), Asp for Ala (position 1129), and Pro for Ala (positions 1134 and 1136). In both cDNAs, the first putative (since the 5′ end of the proviruses was not sequenced) stop codon (TAA) was detected between MA and TM as a result of a G substitution for T (position 1137). Partial sequence analysis of two other transcripts also demonstrated a δMA-δTM-pX structure.
FIG. 6.
Transcripts of deleted HTLV-1 proviruses detected in PBMCs of ATLL and asymptomatic carriers by RT-PCR and Southern hybridization. Lanes: M, size markers; −, no RNA; +, RNA from MT-2 cells; 1 to 3, RNA from asymptomatic HTLV-1 carriers; 4 to 8, RNA from ATLL patients.
FIG. 7.
Sequence alignments of the RT-PCR-amplified transcripts from PBMC of the asymptomatic carries (A and B) and ATK. Only the sequence differences between ATK and amplimers are indicated.
Expression of defective proviruses in vitro.
To confirm the possible expression and immunoreactivity of the chimeric proteins, we cloned defective proviruses 1 to 3 into the pTargeT expression vector and analyzed protein expression in the TNT in vitro transcription-translation system. After incubation of plasmids in TNT reaction mixture, the expressed proteins were immunoprecipitated with HAM/TSP patient serum or monoclonal antibodies to p19/p28, electrophoresed, and analyzed by fluorography (Fig. 8). Protein expression was detected in all three reaction tubes, and all three synthesized proteins were recognized by patient sera and by monoclonal antibody against p19 and p28. The sizes of the synthesized proteins varied from nearly 8 to 12 kDa and corresponded well to the coding capacities of the ORFs deduced from the nucleotide sequences of cloned proviruses. These results were confirmed by Western blotting (data not shown). Reactivity with patient serum as well as with anti-MA monoclonal antibody clearly indicated that the N-terminal parts of these proteins are intact and immunogenic; if expressed in vivo, these proteins might thus be myristylated and associated with the plasma membrane.
FIG. 8.
RIP analysis of the proteins coded by three HTLV-1 defective proviruses with long ORFs that were expressed in vitro. Lanes: M, 14C-rainbow markers (Amersham); 1, positive control of in vitro transcription-translation reaction; 2 to 4, RIP with TSP/HAM patient serum; 5 to 7, RIP with monoclonal antibodies to p17/p28. Sizes of the proteins are given in kilodaltons on the left.
DISCUSSION
HTLV-1 is associated with several independent pathologies in humans, although over 90% of infected individuals remain asymptomatic throughout life. The two main diseases are ATLL and HAM/TSP. Analysis of virus variants associated with these pathologies did not reveal principal differences, indicating that there are no pathology-specific sequences or mutations. Thus, other elements or factors might be linked to pathology progression.
The formation of defective proviruses is known to take place in cells infected with replication-competent retroviruses. These deletions could occur at different steps of the virus replication cycle: reverse transcription, proviral integration, or DNA replication. In evolutionary terms, one might consider exogenous deleted proviruses as intermediate forms between complete exogenous and silent endogenous viruses. Thus, analysis of these proviruses could shed light not only on virus-associated pathogenesis but also on the fundamental processes of long-term virus-cell coexistence.
This study shows that the majority of HTLV-1-infected individuals studied have defective integrated proviral genomes in white blood cells and that several of these genomes have potential polypeptide coding capacity. The goal of our research was to detect and characterize defective proviruses with large internal deletions and a Gag-pX structure in ATLL, HAM/TSP, and asymptomatic HTLV-1 carriers. Complete proviruses were found in 45 of the 51 DNA samples analyzed. Since genomic Southern hybridization and/or long-range PCR do not detect small genomic changes and point mutations, it is likely that many of these full-length proviruses are not replication competent.
We detected numerous defective proviruses with δgag-δenv-pX0 structure in most DNA samples from ATLL patients, HAM/TSP patients, asymptomatic HTLV-1 carriers, and PBMC-derived cell lines from these individuals. Proviruses without a 5′ LTR were frequent, indicating a likely block to transcription of viral genes. The amplimers with a 5′ LTR ranged in size from nearly 200 to 1,300 bp. Evidently, amplimers without the 5′ LTR were shorter, ranging from 180 to nearly 800 bp. Direct sequence analysis of 17 deleted proviruses amplified by nested PCR indicated that most of these proviruses, besides the truncated MA and pX0 sequence, contain fragments of CA, SU, and TM. While most of the sequenced proviruses contained numerous stop codons, several proviruses had relatively long ORFs, starting from the ATG of MA (position 804) and capable of coding for proteins of up to 12 kDa. While the N-terminal parts of these proteins belong to MA (from 41 to 109 amino acids), the C-terminal parts (because of deletions) were usually out of frame and did not correspond to any known protein. One provirus with the ATG of MA has an ORF that because of a mutation in the second codon gives rise to a completely out of frame 4.2-kDa protein. Thus, these ORFs have a chimeric structure and represent a potential new class of HTLV-1-related proteins composed of truncated MA and variable out-of-frame polypeptides. We examined three such proviruses in an in vitro transcription-translation system and detected proteins with N-terminal p19 (MA) epitopes.
When DNA samples from asymptomatic HTLV-1 carriers were examined by PCR, 18 of 20 DNA samples were found to have defective proviruses with large internal deletions and a δMA-δTM-pX structure. By RT-PCR analysis, corresponding deleted transcripts were detected. Direct nucleotide sequencing of two RT-PCR-amplified transcripts demonstrated that both amplimers contained exactly the same mutations as detected in the proviral DNA.
The detection of deleted transcripts in cells of HTLV-1-infected individuals raises not only a possible role of chimeric proteins in infected cells but also the question of whether they are incorporated into particles, and transmitted, as shown for the defective proviruses in MT-2 cells (1, 25). The RNAs detected in asymptomatic carriers are likely to contain a slip site (UUAAAAU) with a potential pseudoknot structure between gag and env. Since we did not estimate the frameshift efficiency of this sequence, it is difficult to make any firm conclusions, although the UUUAAAC sequence present in coronavirus MHV-A59 yielded a frameshift efficiency of nearly 40% when tested in a rabbit reticulocyte lysate (2). Several further points need to be investigated. First, how frequently and how efficiently is deleted RNA (if Ψ is intact) packaged inside the virion? Second, is there any competition between deleted and complete RNA for packaging? Third, if copackaging is possible, are heterozygous particles infectious? Are chimeric MA-pX proteins expressed in vivo? If so, do they interface with cellular processes, with HTLV-1 replication, or with host immune responses?
Taken together, our results indicate that HTLV-1 proviruses with large internal deletions are present in more than 80% of HTLV-1-infected individuals, with no special association of these proviruses with one of the HTLV-1-associated pathologies. However, defective proviruses with long ORFs were detected only in cells (or cell lines) from TSP/HAM patients. It may be informative to test further HAM/TSP patients and examine whether defective proviral load and the ratio of complete and defective proviruses differ between categories of patients. Expression of chimeric proteins from defective proviruses might contribute to the autoimmune features of HAM/TSP.
ACKNOWLEDGMENTS
We thank C. Patience, D. Griffiths, and H. King for fruitful discussions and help.
V.A.M. was supported by a UICC Yamagiwa-Yoshida Research Fellowship. This work was funded in part by Medical Research Council and was supported by the European Union HTLV European Research Network.
REFERENCES
- 1.Benovic S, Kok T, Stephenson A, McInnes J, Burrell C, Li P. De novo reverse transcription of HTLV-1 following cell-to-cell transmission of infection. Virology. 1998;244:294–301. doi: 10.1006/viro.1998.9111. [DOI] [PubMed] [Google Scholar]
- 2.Bredenbeek P J, Pachuk C J, Noten A F H, Charite J, Weiss S R, Spaan W J M. The primary structure and expression of a second open reading frame of a polymerase gene of a coronavirus MHV-A59; a highly conserved polymerase is expressed by an efficient ribosomal frameshifting mechanism. Nucleic Acids Res. 1990;18:1825–1832. doi: 10.1093/nar/18.7.1825. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Clapham P, Nagy K, Weiss R A. Pseudotypes of human T-cell leukemia virus types 1 and 2: neutralization by patients' sera. Proc Natl Acad Sci USA. 1984;81:2886–2889. doi: 10.1073/pnas.81.9.2886. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Evangelista A, Maroushek S, Minnigan H, Larson A, Retzel E, Haase A, Gonzalez-Dunia D, McFarlin D, Mingioli E, Jacobson S, Osame M, Sonoda S. Nucleotide sequence analysis of a provirus derived from an individual with tropical spastic paraparesis. Lancet. 1990;ii:407–410. doi: 10.1016/0882-4010(90)90052-r. [DOI] [PubMed] [Google Scholar]
- 5.Gessain A, Barin F, Vernant J C, Gout O, Maurs L, Calender A, de Thé G. Antibodies to human T-lymphotropic virus type-1 in patients with tropical spastic paraparesis. Lancet. 1985;ii:407–410. doi: 10.1016/s0140-6736(85)92734-5. [DOI] [PubMed] [Google Scholar]
- 6.Gessain A, Saal F, Morozov V, Lasneret J, Vilette D, Gout O, Emanoil-Ravier R, Sigaux F, de Thé G, Peries G. Characterisation of HTLV-1 isolates and T lymphoid cell lines derived from French West Indian patients with tropical spastic paraparesis. Int J Cancer. 1989;43:327–333. doi: 10.1002/ijc.2910430227. [DOI] [PubMed] [Google Scholar]
- 7.Gessain A, Saal F, Giron M-L, Lasneret J, Lagaye S, Gout O, de Thé G, Sigaux F, Peries G. Cell surface phenotype and human T lymphotropic virus type 1 antigen expression in 12 T cell lines derived from peripheral blood and cerebrospinal fluid of West Indian, Guyanese and African patients with tropical spastic paraparesis. J Gen Virol. 1990;71:333–341. doi: 10.1099/0022-1317-71-2-333. [DOI] [PubMed] [Google Scholar]
- 8.Gout O, Baulac M, Gessain A, Semah F, Saal F, Peries G, Cabrol C, Foucault-Fretz C, Laplane D, Sigaux F, de Thé G. Rapid development of myelopathy after HTLV-1 infection acquired by transfusion during cardiac transplantation. N Engl J Med. 1990;322:383–388. doi: 10.1056/NEJM199002083220607. [DOI] [PubMed] [Google Scholar]
- 9.Hill S A, Shuh M, Derse D. Comparisons of defective HTLV-1 proviruses predict the mode of origin and coding potential of internally deleted genomes. Virology. 1999;263:273–281. doi: 10.1006/viro.1999.9922. [DOI] [PubMed] [Google Scholar]
- 10.Hinuma Y, Nagata K, Hanaoka M, Nakai M, Matsumoto T, Kinoshita K, Shirakawa S, Miyoshi I. Adult T-cell leukemia antigen in an ATLL cell line and detection of antibodies to the antigen in human sera. Proc Natl Acad Sci USA. 1981;78:6476–6480. doi: 10.1073/pnas.78.10.6476. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Hiramatsu K, Yoshikura H. Frequent partial deletion of human adult T-cell leukemia virus type 1 proviruses in experimental transmission: pattern and possible implication. J Virol. 1986;58:508–512. doi: 10.1128/jvi.58.2.508-512.1986. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Iino T, Takeuchi K, Nam S H, Siomi H, Sabe H, Kobayashi N, Hatanaka M. Structural analysis of p28 adult T-cell leukaemia-associated antigen. J Gen Virol. 1986;67:1373–1379. doi: 10.1099/0022-1317-67-7-1373. [DOI] [PubMed] [Google Scholar]
- 13.Imamura J, Tsujimoto A, Ohta Y, Hirose S, Shimotohno K, Miwa K, Miyoshi I. DNA blotting analysis of human retroviruses in cerebrospinal fluid of spastic paraparesis patients: the viruses are identical to human T-cell leukemia virus type 1 (HTLV-1) Int J Cancer. 1988;42:221–224. doi: 10.1002/ijc.2910420213. [DOI] [PubMed] [Google Scholar]
- 14.Kawai H, Nishida Y, Takagi M, Nakamura K, Masuda K, Saito S, Shirakami A. HTLV-1 associated myelopathy with adult T-cell leukemia. Neurology. 1989;39:1129–1131. doi: 10.1212/wnl.39.8.1129. [DOI] [PubMed] [Google Scholar]
- 15.Kobayashi N, Konishi H, Sabe H, Sigesada H K, Noma T, Honjo T, Hatanaka M. Genomic Structure of HTLV (human T-cell leukemia virus): detection of defective genome and its amplification in MT-2 cells. EMBO J. 1984;3:1339–1343. doi: 10.1002/j.1460-2075.1984.tb01974.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Kobayashi N, Hatanaka M. Molecular aspects of human T-cell leukemia virus type 1 (HTLV-1) Cancer Rev. 1986;1:64–95. [Google Scholar]
- 17.Korber B, Okayama A, Donnelly R, Tachibana N, Essex M. Polymerase chain reaction analysis of defective human T-cell leukemia virus type I proviral genomes in leukemic cells of patients with adult T-cell leukemia. J Virol. 1991;65:5471–5476. doi: 10.1128/jvi.65.10.5471-5476.1991. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Kubota S, Furuta R A, Siomi H, Maki M, Hatanaka M. Analysis of a novel defective HTLV-1 provirus and detection of a new HTLV-1-induced cellular transcript. FEBS Lett. 1995;375:31–36. doi: 10.1016/0014-5793(95)01166-c. [DOI] [PubMed] [Google Scholar]
- 19.Kwok S, Ehrlick G, Poiesz B, Kalish R, Sninsky J J. Enzymatic amplification of HTLV-1 viral sequences from peripheral blood mononuclear cells and infected tissues. Blood. 1988;72:1117–1123. [PubMed] [Google Scholar]
- 20.Leis J, Baltimore D, Bishop J M, Coffin J, Fleissner E, Goff S P, Oroszlan S, Robinson H, Skalka A M, Temin H M, Vogt V. Standardized and simplified nomenclature from proteins common to all retroviruses. J Virol. 1988;62:1808–1809. doi: 10.1128/jvi.62.5.1808-1809.1988. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Malik K, Even J, Karpas A. Molecular cloning and complete nucleotide sequence of an adult T-cell leukemia virus/human T-cell leukemia virus type 1 (ATLV/HTLV-1) isolate of Caribbean origin: relationship to other members of the HTLV-1 subgroup. J Gen Virol. 1988;69:1694–1670. doi: 10.1099/0022-1317-69-7-1695. [DOI] [PubMed] [Google Scholar]
- 22.Miyoshi I, Kubonishi I, Yoshimoto S, Shiraishi Y. A T-cell line derived from normal human cord leukocytes by co-culturing with human leukemic T-cell. Gann. 1981;72:978–981. [PubMed] [Google Scholar]
- 23.Mochizuki M, Ono A, Ikeda E, Hikita N, Watanabe T, Yamaguchi K, Sagawa K, Ito K. HTLV-1 uveitis. J AIDS Hum Retrovirol. 1996;13(Suppl.):S50–S56. doi: 10.1097/00042560-199600001-00010. [DOI] [PubMed] [Google Scholar]
- 24.Morozov V A, Lagaye S, Saal F, Bazarbachi A, Gout O, Lyon-Caen O, Peries G. High level of HTLV-1 specific protein expression in a patient with ATLL, chronic progressive myelopathy and Kaposi's sarcoma. Leukemia. 1992;6:746–750. [PubMed] [Google Scholar]
- 25.Morozov V A, Weiss R A. Two types of HTLV-1 particles are released from MT-2 cells. Virology. 1999;255:279–284. doi: 10.1006/viro.1998.9578. [DOI] [PubMed] [Google Scholar]
- 26.Osame M, Usuku K, Izumo S, Ijichi N, Amitini H, Igata A. HTLV-1 associated myelopathy: a new clinical entity. Lancet. 1986;i:1031–1032. doi: 10.1016/s0140-6736(86)91298-5. [DOI] [PubMed] [Google Scholar]
- 27.Poiesz B J, Ruscetti F W, Gazdar A F, Bunn P A, Minna J D, Gallo R C. Detection and isolation of type-C retrovirus particles from fresh and cultured lymphocytes of a patient with cutaneous T-cell lymphoma. Proc Natl Acad Sci USA. 1980;77:7415–7419. doi: 10.1073/pnas.77.12.7415. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Ratner L, Josephs S, Starchich B, Hahn B, Shaw G, Gallo R C, Wong-Staal F. Nucleotide sequence analysis of a variant of human T-cell leukemia virus (HTLV-1b) provirus with a deletion in pX-1. J Virol. 1985;54:781–790. doi: 10.1128/jvi.54.3.781-790.1985. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Sanchez G, Xu X, Chermann J C, Hirsch I. Accumulation of defective viral genomes in peripheral blood mononuclear cells of human immunodeficiency virus type 1-infected individuals. J Virol. 1997;71:2233–2240. doi: 10.1128/jvi.71.3.2233-2240.1997. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Seiki M, Hattori S, Hirayama Y, Yoshida M. Human adult T-cell leukemia virus: complete nucleotide sequence of the provirus genome integrated leukemia cell DNA. Proc Natl Acad Sci USA. 1983;80:3618–3623. doi: 10.1073/pnas.80.12.3618. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Shuh M, Hill S A, Derse D. Defective and wild-type human T-cell leukemia virus type I proviruses: characterization of gene products and trans-interactions between proviruses. Virology. 1999;262:442–451. doi: 10.1006/viro.1999.9909. [DOI] [PubMed] [Google Scholar]
- 32.Tamaiya S, Matsuoka M, Etoh K-I, Watanabe T, Kamihira S, Yamaguchi K, Takatsuki K. Two types of defective human T-lymphotropic virus type I provirus in adult T-cell leukemia. Blood. 1996;88:3065–3073. [PubMed] [Google Scholar]
- 33.Yoshida M, Osame M, Usuhku K, Matsumoto M, Igata A. Viruses detected in HTLV-1-associated myelopathy and adult T-cell leukemia are identical on DNA blotting. Lancet. 1987;i:1085–1086. doi: 10.1016/s0140-6736(87)90506-x. [DOI] [PubMed] [Google Scholar]