We screened for inhibitory activities of 25 wild vegetables and fruits for human immunodeficiency virus type 1 (HIV-1) reverse transcriptase (RT). Among them, ethanol-and water-extracts ofBrasenia schreberi (Junsai) and water-extract of Petasites japonicus (Fuki) strongly inhibited the HIV-1 RT activity to incorporate dTTP into poly(rA)-p(dT)15. We tested HIV-1 RT inhibitory activities of 15 polyphenols, isolated from Brasenia schreberi. Among them, gossypetin and hypolaetin 7-O-glucoside inhibited the activity.

MMLV RT expressed in the cell-free system is indistinguishable from that expressed in E. coli.

We have previously used site-directed mutagenesis to introduce basic residues (i.e., Arg; Lys) in the nucleic acid binding cleft of the Moloney murine leukemia virus reverse transcriptase (MMLV RT) in order to increase its template-primer (T/P) binding affinity. Three stabilizing mutations (i.e., E286R, E302K, and L435R) were identified (Yasukawa et al., 2010). Now, we studied the mechanism by which those mutations increase the thermal stability of the RT. The three single-mutants (E286R, E302K, and L435R), an RNase H-deficient MMLV RT (carrying the RNase H-inactivating mutation D524A), a quadruple mutant (E286R/E302K/L435R/D524A, designated as MM4) and the wild-type enzyme (WT) were produced in Escherichia coli. All RTs exhibited similar dissociation constants (Kd) for heteropolymeric DNA/DNA (2.9-6.5 nM) and RNA/DNA complexes (1.2-2.9 nM). Unlike the WT, mutant enzymes (E286R, E302K, L435R, D524A, and MM4) were devoid of RNase H activity, and were not able to degrade RNA in RNA/DNA complexes. These results suggest that the mutations, E286R, E302K, and L435R increase the thermostability of MMLV RT not by increasing its affinity for T/P but by abolishing its RNase H activity.

Human immunodeficiency virus type 1 (HIV-1) reverse transcriptase (RT) possesses two distinct enzymatic activities: those of RNA- and DNA-dependent DNA polymerases and RNase H. In the current HIV-1 therapy, all HIV-1 RT inhibitors inhibit the activity of DNA polymerase, but not that of RNase H. We previously reported that ethanol and water extracts of Brasenia schreberi (Junsai) inhibited the DNA polymerase activity of HIV-1 RT [Hisayoshi et al. (2014) J Biol Macromol 14:59-65]. In this study, we screened 43 edible plants and found that ethanol and water extracts of Brasenia schreberi and water extract of Petasites japonicus strongly inhibit not only the activity of DNA polymerase to incorporate dTTP into poly(rA)-p(dT)15 but also the activity of RNase H to hydrolyze the RNA strand of an RNA/DNA hybrid. In addition, these three extracts inhibit HIV-1 replication in human cells, with EC50 values of 1-2 µg/ml. These results suggest that Brasenia schreberi and Petasites japonicus contain substances that block HIV-1 replication by inhibiting the DNA polymerase activity and/or RNase H activity of HIV-1 RT.

Moloney murine leukemia virus reverse transcriptase (MMLV RT) contains fingers, palm, thumb, and connection subdomains as well as an RNase H domain. The DNA polymerase active site resides in the palm subdomain, and the RNase H active site is located in the RNase H domain. The RNase H domain contains a positively charged α-helix called the C helix (H(594)GEIYRRR(601)), that is thought to be involved in substrate recognition. In this study, we expressed three versions of the RNase H domain in Escherichia coli, the wild-type domain (WT) (residues Ile498-Leu671) and two variants that lack the regions containing the C helix (Ile593-Leu603 and Gly595-Thr605, which we called ΔC1 and ΔC2, respectively) with a strep-tag at the N-terminus and a deca-histidine tag at the C-terminus. These peptides were purified from the cells by anion-exchange, Ni(2+) affinity, and Strep-Tactin affinity column chromatography, and then the tags were removed by proteolysis. In an RNase H assay using a 25-bp RNA-DNA heteroduplex, WT, ΔC1, and ΔC2 produced RNA fragments ranging from 7 to 16 nucleotides (nt) whereas the full-length MMLV RT (Thr24-Leu671) produced 14-20-nt RNA fragments, suggesting that elimination of the fingers, palm, thumb, and connection subdomains affects the binding of the RNase H domain to the RNA-DNA heteroduplex. The activity levels of WT, ΔC1, and ΔC2 were estimated to be 1%, 0.01%, and 0.01% of full-length MMLV RT activity, indicating that the C helix is important, but not critical, for the activity of the isolated RNase H domain.