The histone deacetylase inhibitor chidamide induces intermittent viraemia in HIV-infected patients on suppressive antiretroviral therapy

OBJECTIVES

To evaluate the safety and efficacy of chidamide to reverse HIV-1 latency in vivo and to compare the effects of four clinically tested histone deacetylase (HDAC) inhibitors on non-histone proteins in vitro.

METHODS

Participants received chidamide orally at 10 mg twice weekly for 4 weeks while maintaining baseline antiretroviral therapy. The primary outcome was plasma viral rebound during chidamide dosing and the secondary outcomes were safety, pharmacokinetic and pharmacodynamic profiles, changes in cell-associated HIV-1 RNA and HIV-1 DNA, and immune parameters. Western blotting was used to compare the in vitro effects of the four HDAC inhibitors on HSP90, NF-κB and AP-1.

RESULTS

Seven aviraemic participants completed eight oral doses of chidamide, and only grade 1 adverse events were observed. Cyclic increases in histone acetylation were also detected. All participants showed robust and repeated plasma viral rebound (peak viraemia 147-3850 copies/mL), as well as increased cell-associated HIV-1 RNA, during chidamide treatment. Furthermore, we identified an enhanced HIV-1-specific cellular immune response and a modest 37.7% (95% CI: 12.7-62.8%, P = 0.028) reduction in cell-associated HIV-1 DNA. Compared with the other three HDAC inhibitors, chidamide had minimal cytotoxicity in vitro at clinically relevant concentrations and showed mechanistically superior effects on non-histone proteins, including HSP90, NF-κB and AP-1.

CONCLUSIONS

Chidamide safely and vigorously disrupts HIV-1 latency in vivo, which makes it a promising latency-reversing agent.

Inhibition of ryanodine binding to sarcoplasmic reticulum vesicles of cardiac muscle by Zn(2+) ions

Using the assay of [(3)H]ryanodine binding to the sarcoplasmic reticulum, the effect of Zn(2+) on ryanodine receptors (RyRs) of cardiac muscle was investigated. There was no obvious change in the binding at [Zn(2+)](f) of less than 0.2 microM. However, a decrease of the binding became significant with raising [Zn(2+)](f) to 0.5 microM. The inhibitory effect of Zn(2+) was [Zn(2+)](f)-dependent, with IC(50/ZnI) of 2.1+/-0.4 microM (mean+/-S.D.). Scatchard analysis indicates that both an increase of K(d) and a decrease of B(max) were responsible for Zn(2+)-induced decrease of the binding. The Hill coefficient for this inhibitory effect of Zn(2+) was between 0.8 and 1.2. The interactions of the effects of Zn(2+) and various modulators of RyR indicate that the inhibitory effect of Zn(2+) was mostly mediated through inhibiting Ca(2+) activation sites (CaA) on RyR. Since the [Zn(2+)](f) dependence was not clearly changed by [Ca(2+)](f), the inhibitory effect of Zn(2+) may not be due to competition of Zn(2+) with Ca(2+) for CaA and probably is indirect. The inhibitory effect of Zn(2+) could not be antagonized by 2 mM dithiothreitol, a thiol-reducing agent, suggesting that the binding of Zn(2+) ions to RyRs of cardiac muscle is not accompanied by obvious change of redox state of the RyRs. In comparison with that seen in skeletal muscle [3], the effects of Zn(2+) on ryanodine binding to the sarcoplasmic reticulum of cardiac muscle show several distinct differences. It is indicated that the effect of Zn(2+) on RyRs may be isoform-dependent. The physiological significance of the effects of Zn(2+) is discussed.

Biphasic modulation of ryanodine binding to sarcoplasmic reticulum vesicles of skeletal muscle by Zn2+ ions

With the use of a [(3)H]ryanodine binding assay, the modulation of skeletal muscle ryanodine receptor (RyR1) by Zn(2+) was investigated. In the presence of 100 microM free Ca(2+) concentration ([Ca(2+)](f)) as activator, the equilibrium [(3)H]ryanodine binding to heavy sarcoplasmic reticulum vesicles was biphasically modulated by Zn(2+). The binding was increased by a free Zn(2+) concentration ([Zn(2+)](f)) of less than 1 microM; a peak binding, approx. 140% of the control (without added Zn(2+)) was obtained at 0.3 microM [Zn(2+)](f). An inhibitory effect of Zn(2+) became obvious with a [Zn(2+)](f) of more than 1 microM; the [Zn(2+)](f) for producing half inhibition was 2.7+/-0.5 microM (mean+/-S.D.). Scatchard analysis indicated that the increase in the binding induced by low [Zn(2+)](f) was due to a decrease in K(d), whereas both an increase in K(d) and a possible decrease in B(max) were responsible for the decrease in binding induced by high [Zn(2+)](f). The binding in the presence of micromolar [Zn(2+)](f) showed a biphasic time course. In the presence of 3 microM [Zn(2+)](f), after reaching a peak with an increased rate of initial binding, the binding gradually declined. The decline phase could be prevented by decreasing [Zn(2+)](f) to 0.5 microM or by adding 2 mM dithiothreitol, a thiol-reducing agent. The [Ca(2+)](f) dependence of binding was changed significantly by Zn(2+), whereas Ca(2+) had no clear effect on the [Zn(2+)](f) dependence of binding. Moreover, some interactions were found in the effects between Zn(2+) and other RyR1 modulators. It is indicated that Zn(2+) can modulate the activation sites and inactivation sites for Ca(2+) on RyR1. The physiological significance of the effects of Zn(2+) on ryanodine binding is discussed.