The retroviral RNase H is part of the reverse transcriptase (RT). The RT/RNase H is located inside the virus particles and is carried into the cell during infection. During RT-mediated reverse transcription RNA–DNA hybrids are generated. The RNase H removes the RNA in the hybrids, but leaves the polypurine tract (PPT) RNA intact. The PPT is required as primer for initiation of second-strand DNA synthesis . Previous studies with HIV have demonstrated that an oligodeoxynucleotide (ODN) A that specifically binds the PPT of HIV-1, forms a local RNA–DNA hybrid that mimics a natural replication intermediate [2–5]. This hybrid activates RNA hydrolysis by the HIV RT/RNase H  and destroys the viral RNA template before DNA synthesis [3,4]. The partially double-stranded hairpin loop-structured 54 mer ODN A inhibits HIV-1 replication in infected cells, including drug-resistant strains of HIV [2–4]. Furthermore, as the RT-associated RNase H is associated with virus particles , it can be activated outside of the cell and this renders the virus noninfectious [5,8]. The effect of ODN A is sequence-specific , and a partially double-stranded ODN is more strongly inhibitory than a single-stranded antisense (as) ODN [4,5].
An ODN also showed in-vivo efficacy against a murine oncogenic retrovirus . The virus causes splenomegaly in mice and we were able to demonstrate reduction of virus load in the blood and of disease progression. However, a murine oncogenic retrovirus does not reflect the situation of primary blood-borne HIV. We designed this study to be as close to the human situation, it builds on the unique property that virus can be targeted before entering a cell. It therefore intervenes earlier than the inhibitors of RT, which only works inside of a cell. The RNase H can also be activated early during infection of a cell and prevents DNA provirus formation [2,3].
Here we demonstrate that HIV particles present in the plasma of infected individuals from Zurich and HIV field isolates from Africa including two drug-resistant strains can be treated before entering a cell with an ODN for activation of the viral RNase H and self-destruction of the virus. Statistically significant reduction of viral titers was observed in all cases. Premature activation of an enzyme instead of its inhibition may be an approach worth pursuing further. Reduction of viral load ex vivo may be a model for the in-vivo situation.
Patients and HIV isolates
We obtained HIV-infected blood from 19 patients from the Zurich University Hospital with a broad range of history, HAART, ART, or no therapies, high or low viral loads, and CD4 cell counts (see Fig. 3a). Furthermore, we obtained HIV-1 field isolates from Africa Uganda-92 (Ug92), Uganda-93 (Ug93), Rwanda (Rw), Malawi (Mw), HIV-1 ELI, HIV-1 LAI/BRU, HIV-1 BaL, HIV-1 IIIB as well as strains resistant to the nucleoside analogue AZT (Azt) and the protease inhibitor Saquinavir (Saqi) from the National Institute of Health (NIH) AIDS Research and Reference Reagent Program (ARRRP) .
Plasma, primary peripheral blood mononuclear cells, and HIV
Plasma was extracted from 10 ml of blood of HIV-1-infected patients (available through the Department of Infectious Diseases of the University Hospital, University of Zurich, Zurich, Switzerland) using the BD Vacutainer CPT method (Becton Dickinson, Franklin Lakes, New Jersey, USA). Primary peripheral blood mononuclear cells (pPBMCs) and plasma were extracted from fresh blood obtained from healthy donors (available through the Blood Donation Center, Zurich, Switzerland) using Ficoll Paque Plus methods (GE Health Technologies, USA). Sucrose-purified virions from patient 8898 used for cleavage assays were a kind gift from Dr J. Boeni from the Swiss National Center for Retroviruses, Zurich, Switzerland.
The ODN A  consists of a 25 mer antisense and a 25 mer passenger strand connected by four thymidines (Fig. 1a). ODN Sc , which has the same length and nucleotide composition as ODN A but a randomized sequence of both strands, served as a control for nonspecific activity of ODN A (Fig. 1a). The ODNs were phosphorothioated at each end (three bases) and in the T4 linker. The ODNs were purchased from Operon (Germany) or Integrated DNA Technology (USA).
Viral RNA was isolated from plasma  or cell supernatants using the QIAamp Viral RNA Mini Kit (Qiagen, Germany) according to the manufacturer's directions.
Determination of HIV RNA by qRT-PCR
HIV-1 RNA levels were determined by quantitative reverse transcription polymerase chain reaction (qRT-PCR) using RNA extracted from plasma or cell culture supernatants as indicated in the legends. Isolated viral RNA was reverse transcribed using High-Capacity cDNA Archive Kit (Applied Biosystems, Foster City, California, USA) and qRT-PCR was then performed using the Taqman Universal PCR master mix and the ABI 7300 instrument from Applied Biosystems. Location of the PCR primers and probes on the HIV-1 genome are schematically indicated in Fig. 1b. Coordinates for the primers and probes refer to HIV Gene Bank accession number 9629357. Sequences were gag forward primer, 5′-GCAGCCATGCAAATGTTAAAAGAG-3′; gag reverse primer, TCCCCTTGGTTCTCTCATCTGG ; FAM-TAMRA-gag probe, TCTATCCCATTCTGCAGCTTCCTCATT; UTR forward primer, CAATGACTTACAAGGCAGCTGTAGA; UTR reverse primer, TTAGCAGAACTACACACCAGGGC; FAM-TAMRA-UTR probe, TTCACTCCCAAAGAAGACAAGATATCCTTGAT. Sequences for nef-readout and cycling conditions were published previously [4,5]. All primers and the gag probe were purchased from Microsynth (Switzerland), the nef and UTR probes from Applied Biosystems. The results are presented in the figures as HIV RNA corresponding to absolute numbers of copies per assay unless otherwise stated.
Determination of HIV RNA in virions
HIV-1 virions, derived from 20 μl of patient plasma, were incubated with different concentrations of ODNs. At indicated time points, viral RNA was purified and the amount of undigested RNA was quantified by qRT-PCR using two sets of primers covering the PPT region of the HIV genome, either nef or UTR. Samples (3 μl) obtained from the NIH were diluted to 20 μl in RPMI containing 20% FBS and treated similarly. qRT-PCR was performed using nef primers.
Test for infectivity of cell-free HIV on primary peripheral blood mononuclear cells
One hundred microliters of HIV-containing plasma from patients (Zurich isolates) or 5 μl of the African isolates were treated with either 1 or 5 μmol/l of ODNs for 1 or 4 h. One million phytohemagglutinin (PHA)-activated PBMCs  were infected with pretreated HIV-1 virions and grown in culture at 37°C for 3 days in a total volume of 5 ml of RPMI-1640 with 20% FBS in the presence of interleukin 2 (IL-2, 20 U/ml; Zeptometrix, USA). At day 3, 3 ml of RPMI-1640, 20% FBS, and IL-2 were added. At day 7 HIV RNA levels were analyzed in the supernatants using gag-specific qRT-PCR (Fig. 1b). For the experiments with passaged materials, the equivalents of approximately 2 × 104 HIV RNA copies were used.
RNA isolated from HIV-1-infected patients and cell culture supernatants was reverse transcribed using random primers and High-Capacity cDNA Archive Kit (Applied Biosystems) and the cDNA was sequenced in the HIV-1 PPT-flanking genomic regions using forward primer CCTCAGGTACCTTTAAGACCAATGAC and reverse primer CCCCTGGCCCTGGTGTGTA. Thermal cycle sequencing was performed on a thermocycler (Touchgene from Genius, USA) for 25 cycles with either the forward or reverse primer using 10 ng of DNA templates. Purified sequencing products were run on an automated ABI 3100 genetic analyzer (Applied Biosystems). Sequences of both strands were analyzed and edited individually using the software package Sequencer 4.7 (Gene-Codes, USA), assembled as contigs with Sequence Assembler (Applied Biosystems) and subsequently aligned through Clustal-W algorithm of MacVector 7.1 software program (Accerlerys, USA).
HIV-PPT-RNA was transcribed in vitro, dephosphorylated, and 5′-labeled as described . Cleavage with virus-associated RT/RNase H in virions was performed with a patient isolate, purified with a sucrose-step gradient, and permeabilized. HIV virions were from patient 8898 (3 × 107 copies/ml). Ten nanomol/l HIV-PPT-RNA and 50 nmol/l oligonucleotide in RT buffer  containing 100 ng/ml carrier RNA and 0.1% NP-40 were annealed (2 min 90°C, 10 min 37°C), supplemented with 0.2 U/μl RNasin and incubated for the indicated times at 37°C. The control cleavage was performed with 0.1 U/μl RT/RNase H (GE Healthcare) for 30 min. The samples were proteinase K-treated, phenolized, precipitated, and subjected to electrophoresis in 10% polyacrylamide containing 8 mol/l urea, together with a marker, HIV-PPT-RNA partially digested with RNase T1 (Ambion, USA) as described by the manufacturer.
Samples from several patients were excluded from the analysis for individual experiments, when HIV RNA levels were not detectable in the presence of phosphate-buffered saline (PBS) (four samples) or unusually high (one sample). In the analysis of Zurich virions, some samples were not tested for ODN Sc control, but only for PBS control (six samples). In case of undetectable HIV RNA levels upon treatment with ODN A or ODN Sc, the values were set to the detection limit of 1 copy per assay. Viral RNA levels after treatment were compared using the Wilcoxon signed-rank test using nontransformed HIV RNA levels in copies per assay. To address the problem of multiple comparisons, a Bonferroni correction was performed and thus P values smaller than 0.017 are considered statistically significant. SPSS 15.0 (SPSS Inc., Chicago, Illinois, USA) was used for statistical analyses.
Experimental design and targeting the RNase H in virions
We are using ODNs to target the RNase H in HIV particles. The ODNs are designed to bind to the PPT of HIV-1 IIIB. As controls, we used a scrambled ODN Sc  or buffer PBS. The sequences of the PPT and ODNs are shown in Fig. 1a. ODN A effects on HIV were analyzed by measuring HIV RNA levels by quantitative RT-PCR (qRT-PCR) using primers flanking the PPT, designated as nef or UTR, close to the cleavage site to avoid other effects on the RNA. Because this is a highly variable region, we used two primer pairs. The more conserved gag (gag) primers are similar to those used in routine diagnostics tests (Fig. 1b).
In order to verify whether virus-associated RT/RNase H can be activated by an oligonucleotide to degrade PPT-containing RNA, in-vitro transcribed radioactively end-labeled HIV-PPT RNA together with the oligonucleotide was allowed to form a local hybrid as substrate for the RNase H and incubated with permeabilized virions from a sucrose gradient-purified patient-derived HIV preparation. The cleavage site of the viral RT/RNase H was verified with recombinant RT/RNase H and proven to correspond to the natural cleavage site, 5′ of the ACU sequence (Fig. 1c). Furthermore, in a separate study using lysate supplementation assays, we have shown that overexpression of only wild-type RT/RNaseH but not of a mutant deficient in its RNase H activity leads to enhanced antiviral activity of ODN A (Kwok et al., unpublished results).
In order to establish the conditions required for treatment of HIV in blood, we isolated plasma from the blood of healthy donors, spiked it with the cell culture-adapted strain HIV-1 IIIB, and treated it with ODNs for up to 24 h in vitro. The amount of nonhydrolyzed RNA was determined by quantitative RT-PCR using primers flanking the PPT and shown to be significantly reduced with time and dose (Fig. 1d and e).
We used HIV-containing plasma from patient 13 and one field isolate from Rwanda, for which sufficient material was available, to establish the conditions with two concentrations of ODNs and various times (4, 8, 12, 16, and 24 h; Fig. 2a and b). We chose 8 h and 5 μmol/l ODNs for further experiments, as these conditions were sufficient to cause degradation of HIV RNA.
We then tested the samples for their infectivity as a second independent read-out, which correlates with viral RNA loads , by treating a sample from Zurich patient 3 and HIV-1 BaL strain with ODN A for 1 and 4 h, infecting normal human pPBMCs and culturing them for 7 days (Fig. 2c and d). The amount of HIV-1 RNA in the supernatants was reduced 4 h postincubation with 5 μmol/l ODN A as determined by qRT-PCR using the conserved gag primers. These conditions were further used.
Reduction of viral loads in plasma
HIV-1-containing blood samples of patients from Zurich with different histories were analyzed in this study immediately after isolation. The samples were supplied with information on plasma viral loads and CD4+ cell counts determined prior to this study (Fig. 3a, top). We also analyzed well characterized HIV-1 African primary field isolates and strains resistant against the protease inhibitor Saquinavir (Saqi) or the nucleoside analogue AZT (Azt) from the AIDS Reagent Program (Fig. 3a, bottom). These samples differed from Zurich isolates, as they were supplied after passaging in primary PBMCs. Therefore, we present them as a separate group. The sequences of the HIV-1 PPT regions of few selected patients and field strains were determined and are presented in Fig. 3b.
We tested for ODN-mediated degradation of HIV RNA in plasma samples using qRT-PCR with nef or UTR primers. The amount of nonhydrolyzed viral RNA in blood samples from Zurich and African field isolates of this and the following experiments are shown as median in a box plot (Fig. 3c). Treatment of Zurich samples with ODN A led to reduction of viral RNA levels more than two-fold in 94% and more than 10-fold in 25% of the cases in comparison to PBS. Similar results were obtained with African field isolates and drug-resistant strains (Tables 1 and 2).
Inhibition of viral infectivity of plasma
We then tested the samples for infectivity of ODN-treated HIV-1 virions by measuring HIV RNA levels using the conserved gag primers. The infectivity of all samples of the plasma from patients from Zurich, African field isolates, and drug-resistant strains was analyzed and showed a strong antiviral effect, ranging from two-fold to more than 1000-fold (Fig. 3c and Tables 1 and 2).
In order to examine the antiviral effect of ODNs in the absence of blood factors and differences in viral titers, we used the supernatants from these infected cell cultures (first passage), matched the viral inputs to identical RNA copy numbers, treated them with ODNs and infected primary human pPBMCs. The supernatants (second passage) were analyzed after 7 days, showing even more pronounced the ODN A-mediated inhibition (Fig. 3c and Tables 1 and 2).
We demonstrate that the viral load in HIV-infected plasma samples from patients can be reduced by activation of the viral RNase H by an ODN. This leads to self-destruction of the virus particles. All ODN A-treated samples had a statistically significant decrease of the mean value of viral RNA compared to the control ODN Sc-treated and PBS-treated groups. The infectivity of treated virions was reduced on average 52–63-fold (Fig. 3c and Table 1). Seventy-five percentage of all isolates showed a more than two-fold reduction of infectivity, 54% more than 10-fold, and 21% more than 1000-fold reduction (Table 2). A combination of all experiments analyzed by a nonparametric test with Bonferroni post-hoc test correction (Fig. 3c) revealed a significant decrease not only in viral load but also in variability. The variations may in part be due to the presence of blood-borne components, the status of the patient, the sequences and fitness of the viruses, and so on. All virus samples of patients with or without a history of antiviral therapy were susceptible to ODN A treatment. A second passage with standardized viral inputs strongly increased the specificity and confirmed the antiviral activity of ODN A to be statistically significant (Fig. 3c and Tables 1 and 2).
We have previously shown that multidrug-resistant mutants in the RT can be inhibited by the ODN A-assisted activation of the RNase H , which we confirm here with two natural drug-resistant isolates. We noticed the mutations in the 5′ part of the PPT (Fig. 2b) and do not know their contribution to the inhibitory effect of ODN A, yet the isolates were still susceptible to inhibition by ODN A. We showed previously that three mismatches out of 25 between PPT and ODN sequences reduced the inhibitory effect . Some mismatches were determined here (Fig. 3b). Matched ODNs may therefore enhance the antiviral efficiencies and combinations of ODNs are feasible against a population of mutant viruses.
Furthermore, our unpublished data indicate that in tissue culture, serial passage of HIV-1 with ODN A did not result in mutations in the PPT during 11 passages, and in the RNase H-coding regions of the HIV genome, whereas escape mutants appeared after four passages when control drugs such as forscarnet were used (data not shown).
We previously demonstrated in-vivo efficacy of ODN against a murine retrovirus, the spleen focus-forming virus (SFFV), in mice . The ODN was modified to match the PPT sequence of SFFV and resembles ODN A. Chronically infected mice exhibited either transient or long-term reductions of virus titer depending on the regimen of treatment. Treatment prior to, during, or shortly after infection reduced the viral load in the blood and can delay disease progression, increase survival rates, and prevent viral infection . However, SFFV is a cancer virus with significantly different biological properties than HIV; yet, in both cases, reduction in viral load was observed.
The ODN A used here to target the PPT is directed against the extended PPT covering the ACU site, the natural cleavage site of the RNAse H during replication, and 3′ sequences (see Fig. 1a). The sequencing results demonstrate that the G tract and the sequences 3′ of the ACU site are highly conserved, whereas the 5′ A rich region was more variable (Fig. 3b).
Although the antiviral potential of different forms of ODNs have been extensively studied (reviewed in [14,15]), the ODN A utilized in our experiments may have special properties, as the target strand as well as the second strand of the ODN A are rich in Gs, which may form G tetrads . The second strand may confer higher stability to the ODN A compared to single-stranded antisense strands [4,5]. Furthermore, phosphorothioated modifications present at the termini and in the linker region of the ODNs may help protect against nuclease digestion and increase binding to serum proteins. Other modifications might further improve the antiviral activity of ODN A by changing its stability, uptake, transport, and so on. The properties of ODN A and virus in the blood cannot easily be extrapolated from in-vitro conditions. A better understanding of the mechanism of the antiviral activity of ODN A regarding different aspects of its action would be helpful for further development. Furthermore, the dose, regimen, and toxicity have to be determined. We consider these ex-vivo studies as a model for in-vivo effects in the blood stream.
The fact that the ODN can induce retroviral RT/RNase H-mediated cleavage of HIV outside of cells rendering the virions noninfectious suggests some possibilities for future applications, such as prevention of mother-to-child or sexual transmission, the two major routes of spread of HIV in the Third World. Also, a bridging therapy, allowing multidrug-resistant viruses to become sensitive again, could be envisaged. Our approach suggests that it is feasible to activate a viral enzyme for self-destruction of the virus rather than designing inhibitors.
We are very grateful to Professor Dr Rainer Weber for his interest, support, and cooperation in this project and for providing us with Zurich HIV patient samples. We thank Mrs Christina Gruber for her excellent technical assistance and her careful organization. We also acknowledge Professor Dr Burkhardt Seifert for performing statistical analysis. We thank Dr Juerg Boeni for providing sucrose-purified patient HIV isolate 8898, Dr Cyril Shah for assistance with sequencing, and Mrs Lucia Bertodetto for help with preparation of primary PBMCs.
1. Wohrl BM, Moelling K. Interaction of HIV-1 ribonuclease H with polypurine tract containing RNA-DNA hybrids. Biochemistry 1990; 29:10141–10147.
2. Jendis J, Strack B, Volkmann S, Boni J, Moelling K. Inhibition of replication of fresh HIV type 1 patient isolates by a polypurine tract-specific self-complementary oligodeoxynucleotide. AIDS Res Hum Retroviruses 1996; 12:1161–1168.
3. Jendis J, Strack B, Moelling K. Inhibition of replication of drug-resistant HIV type 1 isolates by polypurine tract-specific oligodeoxynucleotide TFO A. AIDS Res Hum Retroviruses 1998; 14:999–1005.
4. Moelling K, Abels S, Jendis J, Matskevich A, Heinrich J. Silencing of HIV by hairpin-loop-structured DNA oligonucleotide. FEBS Lett 2006; 580:3545–3550.
5. Matskevich AA, Ziogas A, Heinrich J, Quast S, Moelling K. Short partially double-stranded oligodeoxynucleotide induces reverse transcriptase-mediated cleavage of HIV RNA and abrogates infectivity of virions. AIDS Res Hum Retroviruses 2006; 22:1220–1230.
6. Hansen J, Schulze T, Mellert W, Moelling K. Identification and characterization of HIV-specific RNase H by monoclonal antibody. EMBO J 1988; 7:239–243.
7. Moelling K, Bolognesi DP, Bauer H, Busen W, Plassmann HW, Hausen P. Association of viral reverse transcriptase with an enzyme degrading the RNA moiety of RNA-DNA hybrids. Nat New Biol 1971; 234:240–243.
8. Matzen K, Elzaouk L, Matskevich A, Nitzsche A, Heinrich J, Moelling K. RNase H-induced suicide of a retrovirus by oligodeoxynucleotides in a mouse model. Nat Biotechnol 2007; 25:669–674.
9. Brown BK, Darden JM, Tovanabutra S, Oblander T, Frost J, Sanders-Buell E, et al
. Biologic and genetic characterization of a panel of 60 human immunodeficiency virus type 1 isolates, representing clades A, B, C, D, CRF01_AE, and CRF02_AG, for the development and assessment of candidate vaccines. J Virol 2005; 79:6089–6101.
10. Volkmann S, Jendis J, Frauendorf A, Moelling K. Inhibition of HIV-1 reverse transcription by triple-helix forming oligonucleotides with viral RNA. Nucleic Acids Res 1995; 23:1204–1212.
11. Fransen K, Mortier D, Heyndrickx L, Verhofstede C, Janssens W, van der Groen G. Isolation of HIV-1 RNA from plasma: evaluation of seven different methods for extraction (part two). J Virol Methods 1998; 76:153–157.
12. Bisset LR, Bosbach S, Tomasik Z, Lutz H, Schupbach J, Boni J. Quantification of in vitro retroviral replication using a one-tube real-time RT-PCR system incorporating direct RNA preparation. J Virol Methods 2001; 91:149–155.
13. Gallo RC, Wong-Staal F. A human T-lymphotropic retrovirus (HTLV-III) as the cause of the acquired immunodeficiency syndrome. Ann Intern Med 1985; 103:679–689.
14. Becker Y. CpG ODNs treatments of HIV-1 infected patients may cause the decline of transmission in high risk populations: a review hypothesis and implications. Virus Genes 2005; 30:251–266.
15. Van Aerschot A. Oligonucleotides as antivirals: dream or realistic perspective? Antiviral Res 2006; 71:307–316.
16. Burge S, Parkinson GN, Hazel P, Todd AK, Neidle S. Quadruplex DNA: sequence, topology and structure. Nucleic Acids Res 2006; 34:5402–5415.