Human T-cell leukemia virus type 1 reverse transcriptase (RT) originates from the pro and pol open reading frames and requires the presence of RT-RNase H (RH) and RT-RH-integrase proteins for its activity

Abstract

The first description of an active form of a recombinant human T-cell leukemia virus type 1 (HTLV-1) reverse transcriptase (RT) and subsequent predictions of its amino acid sequence and quaternary structure are reported here. By using amino acid alignment methods, the NH2 and COOH termini of the RT, RNase H (RH), and integrase (IN) domains of the Pol polyprotein were determined. The HTLV-1 RT seems to be unique since its NH2 terminus is probably encoded by the pro open reading frame (ORF) fused downstream, via a transframe peptide, to the polypeptide encoded by the pol ORF. The HTLV-1 Pol amino acid sequence was revealed to be highly similar to that of Rous sarcoma virus (RSV), particularly at the RT-RH hinge region. These two domains remain linked for RSV; this may also be the case for HTLV-1. In light of these results, RT, RT-RH, and RT-RH-IN genes were constructed and introduced into His-tagged protein expression vectors. The corresponding proteins were synthesized in vitro, and the DNA polymerase activities of different protein combinations were tested. Solely the RT-RH-RT-RH-IN combination was found to have a significant activity level. Velocity sedimentation analysis suggested that the HTLV-1 RT-RH and RT-RH-IN monomers are likely associated in an oligomeric structure, probably of the alpha3/beta type.

FIG. 1

Strategy for expression of the HTLV-1 Pol polyprotein and localization of encoding domains. (A) Schematic presentation of gag, pro, and pol ORFs. Gray box, mature PR; black box, part of pro ORF which would encode the RT NH₂ terminus; open arrowhead, frameshift site; hatched box, domains that encode part of RT, RH, and IN in the pol ORF; closed and open inverted triangles, described and hypothesized PR maturation sites, respectively. (B) Alignments of different NH₂ and COOH termini of HIV-1 and RSV RT, RH, and IN domains with amino acids encoded by the HTLV-1 pro ORF’s 3′ end and pol ORFs. Amino acids in lowercase letters in the HTLV-1 pol ORF do not encode the RT protein. Amino acids in capital letters in the HTLV-1 pro and pol ORFs encode the NH₂-terminal segment of RT. Alignments were generated with the CLUSTAL and HCA programs. Identification of perfectly conserved (∗) or well-conserved ( · ) amino acids was made with MDM-78 (5). I and L are considered to be identical. The arrow indicates the shift from the pro to the pol HTLV-1 ORF. Double-boxed amino acids are previously described PR cleavage sites leading to mature (i) PR, RT, RH, and IN of HIV-1; (ii) RT-RH and IN of RSV; or (iii) PR of HTLV-1. Single-boxed amino acids are the deduced junctions between the HTLV-1 RT-RH and RH-IN domains.

FIG. 2

Comparative alignments of HTLV-BLV pol ORF NH₂-terminal domains and HTLV-BLV pro ORF COOH-terminal domains with previously characterized HIV-1, MuLV, and RSV RT NH₂ termini. Residues in single gray boxes are those conserved in HTLV-1, HTLV-2, and BLV. Double-boxed amino acids are HTLV-1 pro or pol ORF residues shared at least once with either HIV-1, MuLV, or RSV. Consensus sequence 1 was obtained by comparing the HTLV-1 pol ORF-encoded amino acids with those of HIV-1, MuLV, and RSV RTs, and consensus sequence 2 was obtained by comparing the HTLV-1 pro ORF-encoded amino acids with HIV-1, MuLV, and RSV RT residues. Perfectly conserved (capital letters), hydrophobic (h), charged (c), and aromatic (a) residues are noted in the consensus sequences when they are present in the HTLV-1 sequence and in at least two of the previously described RTs. #, stop codon; -, amino acid insertion; /, previously described cleavage site.

FIG. 3

SDS-PAGE of in vitro-translated Pol proteins. DNA matrices were simultaneously transcribed by either T3 (A) or T7 (B) RNA polymerase and translated in the presence of [³⁵S]methionine. Proteins were analyzed by SDS-PAGE and fluorography. The positions of molecular mass markers positions are indicated on the left. (A) Translation products synthesized in the presence of either luciferase (lanes 3 and 4), pPOL (lanes 1 and 5), pRT-RH (lane 6), or pRT (lane 7) DNA template or no DNA template (lane 2). In vitro-synthesized, [³⁵S]methionine-labeled POL polyprotein precursor was incubated for 4 h at 37°C with BLV protease (lane 8). (B) Translation products synthesized in the presence of either luciferase (lane 2), pHPOL (lane 3), pHRT-RH (lane 4), or pHRT (lane 5) DNA template or no DNA template (lane 1). (C) Purification of [³⁵S]methionine-labeled, His-tagged RT protein (lane 1) on nickel resin. Lanes: 2, bound His-RT; 3, unbound His-RT; 4 to 7, fractions obtained at each resin washing step; 8 to 11, fractions of imidazole-eluted pure His-RT.

FIG. 4

Sucrose gradient analysis of a mixture of RT-RH and RT-RH-IN proteins. (A) Both proteins were centrifuged together through a 5 to 20% sucrose gradient. Fractions were collected and subjected to trichloroacetic acid precipitation, and precipitated proteins were analyzed by SDS-PAGE followed by autoradiography. The top and the bottom of the gradient are indicated. The positions of molecular mass markers are indicated on the right in kilodaltons. Fractions in which protein standards sedimented under the same conditions are indicated at the top of the panel. These were (from the bottom of the gradient) α₂-macroglobulin (340 kDa, dimer), β-amylase (200 kDa, tetramer), alcohol dehydrogenase (150 kDa, tetramer), and BSA (69 kDa, monomer). (B) [³⁵S]methionine-labeled proteins of each fraction were quantitated. The amounts of RT-RH-IN protein (closed bars) and of RT-RH protein (open bars) are represented for each fraction. The amount of protein in the 24th fraction is represented on a 10-fold-reduced scale.