Kaempferol derivatives as antiviral drugs against the 3a channel protein of coronavirus

Abstract

The protein coded by the open-reading-frame 3a of SARS coronavirus has been demonstrated to form a cation-selective channel that may become expressed in the infected cell. The activity of the channel is involved in the mechanism of virus release. Drugs that inhibit the ion channel can, therefore, inhibit virus release, and they could be a source for development of novel therapeutic antiviral agents. Various drugs found in Chinese herbs that are well known as anticancer agents also have an antiviral potency. Here we tested the flavonols kaempferol, kaempferol glycosides, and acylated kaempferol glucoside derivatives with respect to their potency to block the 3a channel. We used the Xenopus oocyte with a heterologously expressed 3a protein as a model system to test the efficacy of the flavonols. Some of these drugs turned out to be potent inhibitors of the 3a channel. The most effective one was the glycoside juglanin (carrying an arabinose residue) with an IC50 value of 2.3 µM for inhibition of the 3a-mediated current. Kaempferol derivatives with rhamnose residue also seem to be quite effective. We suggest that viral ion channels, in general, may be a good target for the development of antiviral agents, and that, in particular, kaempferol glycosides are good candidates for 3a channel proteins of coronaviruses.

Fig. 1

Structure of the flavonoids tested with respect to their effects on the 3a-mediated current. The flavonol kaempferol (1) and the kaempferol glycosides afzelin (2), juglanine (3), and tiliroside (4), as well as two tiliroside derivatives (5–6), the kaempferol triglycoside (7), the flavonol quercetin (8), the flavanone naringenin (9), and the isoflavone genistein (10). Rha stands for rhamnose and Araf for arabinofuranose. For their respective effects, compare Table 1.

Fig. 2

Effect of kaempferol on current-voltage (IV) curves of Ba²⁺-sensitive current. Open squares describe the current-voltage dependencies in the absence and filled squares in the presence of 20 µM kaempferol. a Endogenous currents, b currents of cells with heterologously expressed 3a protein, and c the 3a-protein-mediated current component (endogenous current subtracted). Data represent averages of n = 4 to 7 experiments ± SEM.

Fig. 3

Effect of kaempferol glycosides on Ba²⁺-sensitive and 3a-mediated current. Results in the absence of drug are given by open squares and in the presence of drug by filled symbols. a Effect of 10 µM juglanin on current-voltage dependencies of Ba²⁺-sensitve current in control oocytes without a 3a protein. b Effect of 2.5 (filled squares) and 10 µM (filled circles) juglanin on 3a-mediated current that was determined as the difference of Ba²⁺-sensitive current in a 3a-expressing cells. Data points represent averages from 4–7 experiments ± SEM. c Inhibition of a 3-mediated current at − 100 mV by juglanin. The data points represent averages ± SEM of 5–17 measurements. The dashed line is an approximation of equ. 1 to the data points with n = 1.2 and K_1/2 = 2.3 µM. d Effect of 20 µM tiliroside (squares) and 10 µM afzelin (circles) on voltage dependence of a Ba²⁺-sensitive current in 3a-expressing cell. Data points represent averages from 4–7 experiments ± SEM.