APOBEC3A catabolism of electroporated plasmid DNA in mouse muscle.

The mouse is widely used as a model for DNA therapy and vaccination even though the efficiency of DNA delivery in higher mammals and humans is much less. The human APOBEC3 (A3) enzymes impact viral genomes by cytidine deamination, which introduces multiple uridine residues into single-stranded DNA, a process known as genetic editing. This initiates rapid DNA catabolism via a uracil DNA glycosylase dependent pathway. In tissue culture, A3A, A3C and A3B can hyperedit transfected plasmid DNA. We explored plasmid catabolism in vivo initiated by A3A, the most efficient of the human enzymes and one that is functionally conserved across most mammals. As rodents do not encode an A3A enzyme, it was possible to explore DNA degradation in the mouse model. Human A3A genetically edits co-electroporated luciferase plasmid DNA in mouse skeletal muscle that initiates DNA degradation resulting in approximately fourfold decrease in bioluminescence. Part of the degradation occurs in the nucleus as indicated by complex hyperedited DNA molecules. As human A3A is strongly upregulated by interferon α and DNA sensing pathways, it is a strong candidate enzyme for restricting plasmid DNA in higher mammals.

Erroneous identification of APOBEC3-edited chromosomal DNA in cancer genomics.

BACKGROUND: The revolution in cancer genomics shows that the dominant mutations are CG->TA transitions. The sources of these mutations are probably two host cell cytidine deaminases APOBEC3A and APOBEC3B. The former in particular can access nuclear DNA and monotonously introduce phenomenal numbers of C->T mutations in the signature 5'TpC context. These can be copied as G->A transitions in the 5'GpA context. METHODS: DNA hypermutated by an APOBEC3 enzyme can be recovered by a technique called 3DPCR, which stands for differential DNA denaturation PCR. This method exploits the fact that APOBEC3-edited DNA is richer in A+T compared with the reference. We explore explicitly 3DPCR error using cloned DNA. RESULTS: Here we show that the technique has a higher error rate compared with standard PCR and can generate DNA strands containing both C->T and G->A mutations in a 5'GpCpR context. Sequences with similar traits have been recovered from human tumour DNA using 3DPCR. CONCLUSIONS: Differential DNA denaturation PCR cannot be used to identify fixed C->T transitions in cancer genomes. Presently, the overall mutation frequency is ∼10(4)-10(5) base substitutions per cancer genome, or 0.003-0.03 kb(-1). By contrast, the 3DPCR error rate is of the order of 4-20 kb(-1) owing to constant selection for AT DNA and PCR-mediated recombination. Accordingly, sequences recovered by 3DPCR harbouring mixed C->T and G->A mutations associated with the 5'GpC represent artefacts.

Polymorphic APOBEC3 modulates chronic hepatitis B in Moroccan population.

The cytidine deaminase apolipoprotein B mRNA editing catalytic subunit-3 (APOBEC3) induces G-to-A hypermutation in hepatitis B virus (HBV) genomes and operates as part of the innate antiviral immune system. We investigated the associations between the presence of APOBEC3 variants and HBV carriage in a case-control study in the Moroccan population. A polymorphic deletion affecting the APOBEC3B gene and the H186R variant of APOBEC3G were genotyped in 179 HBV chronic carriers and 216 healthy control subjects. In addition, to assess the overall impact of APOBEC3 deaminases on circulating HBV, we looked for hyperedited forms of the viral genome using the 3DPCR technique and analysed editing context. Data analysis showed that there was no significant difference in the frequencies of deleted APOBEC3B alleles (P = 0.261) or genotypes (P = 0.333) between patients with chronic hepatitis B and control subjects. By contrast, subjects bearing deleted genotype had a faster progression of liver disease than those with the insertion genotype (adjusted OR, 3.72; 95% CI, 0.38-36.12). The analysis of the APOBEC3G H186R polymorphism revealed that R/R genotype frequencies were not significantly different in HBV infected patients and in healthy subjects. 3DPCR was positive in 26 samples (14%) among 179. Amplified viral segments displayed monomorphic G>A transitions highly reminiscent of APOBEC3G activity. Most intriguingly, hemi/homozygous carriers of the APOBEC3B deletion had significantly lower virus loads than patients with the wild type (median 539 vs. 2213 IU/mL, P = 0.0023). This result suggests that genetic variations in APOBEC3 cytidine deaminases do not predispose to chronicity but may modulate the course of persistent HBV infection.

Simulating pseudogene evolution in vitro: determining the true number of mutations in a lineage.

Hypermutagenic PCR has been used to simulate pseudogene evolution of the Escherichia coli R67 dihydrofolate reductase gene. Each time the most divergent clone was used as template for another round of hypermutagenesis. After six rounds, with an average mutation rate of 0.05 per base per round, up to a 46% nucleic acid sequence variation was achieved. For a few clones the protein information content could be annihilated. As the intermediates were cloned and sequenced, it was possible to establish the real lineage and compute the true number of mutations. Not surprisingly the true number of forward and back mutations as well as variable sites exceeded those based on comparing any single intermediate to the initial sequence. However, the true number of forward and backward mutations, as well as the number of variable sites, increased linearly with sequence divergence from the original sequence, suggesting an empirical means to correct for branch lengths.

AMD-3100 (AnorMED).

AnorMED is developing AMD-3100, a CXCR4 chemokine receptor antagonist discovered by a collaboration between AnorMED and the Rega Institute, as a potential treatment for HIV infection [337657]. AMD-3100 is a bis-tetraazamacrocycle, which is a potent inhibitor of HIV-1 and HIV-2 viruses. It interacts with the virus-associated uncoating process [172189]. It is currently being evaluated in phase II clinical trials at The John's Hopkins University School of Medicine and two other US sites. The phase II trial will include 20 to 30 patients and is expected to be completed by the middle of 2000 [326272,326832]. In January 2000, analysts at CIBC World Markets predicted a 2003 launch date for AMD-3100, with sales of CAN 43.5 million dollars, rising to CAN 111 million dollars in 2004 [370903].

HIV genetic variation is directed and restricted by DNA precursor availability.

The effects of deoxynucleoside triphosphate (dNTP) imbalances on the fidelity of human immunodeficiency virus type 1 (HIV-1) replication were investigated. Using detergent permeabilized virions and biased dNTP concentrations different types of hypermutants were readily produced. However, the mutant spectrum was different from naturally occurring hypermutants demonstrating that the host cell may restrict variation. Using a genetic screen based on the blue/white beta-galactosidase complementation assay, G --> A hypermutants were recovered from HIV-infected thymidine treated U937 cells. Furthermore, hypermutants were recovered from 1 to 2% of resting or activated peripheral blood mononuclear cells indicating that small proportions of primary cells had distorted intracellular [dTTP] and [dCTP]. Such imbalances may underlie a proportion of somatic and germline point mutations and shape to some extent the evolution of mammalian and viral genomes.

Hypermutagenic PCR involving all four transitions and a sizeable proportion of transversions.

Very complex mutant libraries of the dihydrofolate reductase (DHFR) gene encoded by the Escherichia coli plasmid R67 were created using hypermutagenic PCR with biased deoxynucleotide triphosphate (dNTP) concentrations. Exploiting the particular stability of the G:T mismatch, the DHFR gene could be enriched in A+T by employing biased deoxypyrimidine triphosphate concentrations, i.e. [dTTP] > [dCTP]. A sizeable fraction of hypermutants were functional. A combination of [dTTP] > [dCTP] and [dGTP] > [dATP] biases generated mutations at unexpectedly low frequencies. This could be overcome by the addition of Mn2+ cations. Overall mutation frequencies of 10% per amplification (range 4-18% per clone) could be attained. All four transitions and a smaller number of transversions were produced throughout the gene. PCR mutagenesis could be so extensive as to inactivate all amplified versions of the gene.

Reverse transcriptase and substrate dependence of the RNA hypermutagenesis reaction.

G-->A hypermutation is a remarkable phenomenon resulting from retroviral reverse transcription in the presence of highly biased dNTP concentrations. Of the three reverse transcriptases (RTases) available, those of human immunodeficiency virus type 1 (HIV-1), avian myeloblastosis virus (AMV) and Moloney murine leukemia virus (MoMLV), the HIV-1 enzyme showed the greatest sensitivity to biased [dCTP]/[dTTP] ratios. The HIV-1 RTase was able to discriminate between dUTP, dITP and the four DNA precursors and was insensitive to pH. There was little preference for nucleotide contexts. A few exceptionally modified sequences were found presumably resulting from G-->A hypermutation and multiple strand transfer. This particular predilection of the HIV-1 and, by extrapolation, the lentiviral RTases towards G-->A hypermutation suggests that the phenomenon may have contributed to the remarkably elevated A content of these retroviral genomes.

Clonal expansion of human T-cell leukemia virus type I-infected cells in asymptomatic and symptomatic carriers without malignancy.

Human T-cell leukemia virus type I (HTLV-I) is associated with adult T-cell leukemia/lymphoma and tropical spastic paraparesis/HTLV-associated myelopathy. Both diseases are usually preceded by a long clinically asymptomatic period. PCR amplification of the HTLV-I proviral integration sites shows that clonal expansion of HTLV-I-bearing T cells, rather than being an occasional phenomenon in nonmalignant disease, is the norm for both symptomatic and asymptomatic carriers. Sequencing of 100 molecular clones derived by PCR amplification of part of the envelope gene from two asymptomatic carriers revealed almost no genetic variation. Viral amplification via clonal expansion, rather than by reverse transcription, would explain this remarkable genetic stability.

Hypermutagenesis of RNA using human immunodeficiency virus type 1 reverse transcriptase and biased dNTP concentrations.

The finding of G-->A hypermutated retroviral genomes in which up to 40% of guanines may be substituted by adenines was proposed to result from the depletion of the intracellular dCTP concentration and suggested a means to hypermutagenize nucleic acids. Using a RNA/reverse transcriptase ratio of approximately 1:30, comparable to that within the retroviral replication complex, G-->A hypermutants were produced in a simple in vitro reaction using highly biased dNTP concentrations--i.e., a low ratio of [dCTP]/[dTTP]. Up to 38% of G residues could be substituted, the proportion being inversely proportional to the concentration of dCTP. As G-->A hypermutation resulted from elongation beyond multiple rG.dT mismatches, U-->C hypermutants resulting from multiple rU.dG mismatches were sought, and found, during cDNA synthesis using low [dATP] and high [dGTP]. Mixed G-->A and U-->C hypermutants could also be produced under conditions of low [dCTP] plus low [dATP] and high [dTTP] plus high [dGTP]. Hypermutagenesis should allow jumping through, and subsequent exploration of, sequence space to a greater degree than heretofore and, in conjunction with genetic screening, might be of use in the search of proteins or ribozymes with novel or enhanced properties.

Spatial discontinuities in human immunodeficiency virus type 1 quasispecies derived from epidermal Langerhans cells of a patient with AIDS and evidence for double infection.

A nonhomogeneous spatial distribution of human immunodeficiency virus type 1 quasispecies was observed for epidermal Langerhans cells purified from skin patches taken from a patient with AIDS soon after death. Each patch presented a unique collection of sequences, distinct from those of juxtaposed patches or those derived from the other leg. Infection of Langerhans cells by virus from underlying T cells in the dermis might explain this partition. The analysis revealed the presence of two distinct cocirculating viral strains, indicating double infection.

G-->A hypermutation of the human immunodeficiency virus type 1 genome: evidence for dCTP pool imbalance during reverse transcription.

The quasispecies model for RNA viruses predicts the existence of a replication error threshold beyond which there is a melting or total loss of sequence information. Retroviral G-->A hypermutation is probably an example. Here it is shown that G-->A transitions may occur in both GpG and GpA dinucleotide contexts. Transitions in GpG preferentially occur via base mispairing at the ends of runs of G residues, whereas G-->A transitions within GpA may result from temporary dislocation of the primer and template strands by a single base. The two circumstances may be related by the local dCTP substrate concentration. An in vitro elongation assay shows that primer/template dislocation is more frequent for the human immunodeficiency virus type 1 reverse transcriptase than for murine or avian retroviral enzymes. Taken together these data suggest that G-->A hypermutation is an example of induced mutation whereby the viral reverse transcriptase is forced into making errors by imbalances in the intracellular dCTP concentration.

Restriction and enhancement of human immunodeficiency virus type 1 replication by modulation of intracellular deoxynucleoside triphosphate pools.

Human immunodeficiency virus type 1 (HIV-1) replication is shown to be sensitive to the intracellular concentration of deoxynucleoside triphosphate substrates. Addition of thymidine to established cell lines resulted in a dramatic reduction of virus production. The effect could be substantially alleviated by addition of deoxycytidine, which, alone, enhanced viral titers by a factor of 2 to 3. Hydroxyurea treatment abolished HIV-1 replication in peripheral blood mononuclear cells and could be reversed by deoxyadenosine. These data show that HIV-1 replication occurs under suboptimal DNA precursor conditions.

High-resolution structure of an HIV-1 quasispecies: identification of novel coding sequences.

OBJECTIVE: To characterize an HIV-1 quasispecies in vivo at high resolution (1%) in order to determine its genetic structure. METHODS: The first coding exon of tat was amplified by polymerase chain reaction from uncultured peripheral blood mononuclear cells of an HIV-1-infected patient. The products were cloned into M13mp18 RF, and 106 clones were sequenced. RESULTS: Thirty-one different Tat protein variants were found. Amongst these, five major forms with frequencies of 44, 11, 8 and 5% were identified. All of the remaining 26 sequences were unique, 15 of which were defective. Within the variant spectrum a small number of genomes encoded novel open reading frames, for example, a tat-vpu fusion product. CONCLUSION: Some of the myriad proviruses present in an individual harbour novel coding sequences. While these are probably of little importance for AIDS pathogenesis they emphasize the ability of HIV to explore a huge range of genetic configurations.

Strand specific PCR amplification of low copy number DNA.

A method for the amplification of a single DNA strand at low copy number is described. It is a wholly PCR based approach which involves an initial linear amplification of the target using a tagged strand specific primer. This is followed by classical PCR amplification of the progeny using a pair of primers, one specific for the sequence tagged onto the 5' end of the first round primer, the second specific for the target sequence. Given the protocol used the ratio of the two strands in the final amplification product was 50:1.

In vivo persistence of a HIV-1-encoded HLA-B27-restricted cytotoxic T lymphocyte epitope despite specific in vitro reactivity.

A large number of human immunodeficiency virus type 1 (HIV-1) specific HLA-restricted cytotoxic T cell (CTL) epitopes have been mapped, including an HLA-B27-restricted immunodominant epitope within p25gag. Accordingly, this segment of the HIV-1 provirus was amplified by the polymerase chain reaction from DNA derived from fresh uncultured peripheral blood mononuclear cells (PBMC) of four HLA-B27 HIV-1-infected individuals. In all cases the majority of infected PBMC bore sequences encoding the HLA-B27-restricted peptide. CTL escape mutants had not accumulated in vivo 8 and 14 months later despite demonstrable CTL activity in vitro. These data emphasize the importance of silently infected lymphocytes in evading immune surveillance.

LAV revisited: origins of the early HIV-1 isolates from Institut Pasteur.

Two of the first human immunodeficiency virus type-1 (HIV-1) strains isolated were authenticated by reanalyzing original cultured samples stored at the Collection Nationale de Culture des Microorganismes as well as uncultured primary material. Cloned polymerase chain reaction products were used to analyze coding sequences of the V3 loop in the gp120 glycoprotein. The original isolate HIV-1 Bru, formerly called LAV, was derived from patient BRU. HIV-1 Lai was derived from patient LAI and contaminated a HIV-1 Bru culture between 20 July and 3 August 1983. The culture became, in effect, HIV-1 Lai, identifiable by a unique motif in the V3 loop. Because of this contamination two, rather than one, HIV-1 isolates were sent to the Laboratory of Tumor Cell Biology at the National Cancer Institute on 23 September 1983. Original HIV-1 Bru was indeed present in the sample marked JBB/LAV. However the M2T-/B sample harbored HIV-1 Lai, a strain capable of growing on established cell lines. The striking similarity between HIV-1 Lai (formerly LAV-Bru) and HTLV-3B sequences remains.