3.1 Patch-clamp studies of single channel characteristics
The first series of experiments assessed the effects of HCQ (Fig. 1) and AZM alone (Fig. 2) and in combination (Fig. 3) on a range of human cardiac ion channels. The experiments on HCQ and AZM themselves adopted concentrations fully encompassing plasma levels associated with their clinical use. The same HCQ concentration range was then combined with 10 µM AZM. These examined inward, cardiac channel currents from hNav1.5 (INa, A ), Cav1.2 (ICa,L,B ), mediating or maintaining membrane action potential depolarization. They also studied outward Kv4.3 (Ito,C ), hERG (IKr, C ), KCNQ1/E1 (IKs, E ) and Kir2.1 (IK1,F ) normally mediating action potential recovery. This encompassed the range of ion channels recommended by the CiPA protocol.
Figure1 (left panels, i) exemplifies observed ionic currents in the presence of 0, 1, 10 and 100 µM HCQ. The insets summarize the related voltage clamping protocols, fully described in extended Methods in Supplementary data. Five concentrations ranging from 10-8 to 10-4 M were tested for each current examined in 3 to 6 cells at each concentration. Mean concentration-response plots (right panels, ii) used the maximum magnitudes of each current. Where applicable, the consequent alterations in such current magnitudes following pharmacological challenge were used to derive IC50 values.
Figure1 (right panels, ii) plots mean (±SEM) fractional reductions in current magnitude against HCQ concentration for INa (A), ICaL (B), Ito (C), IKs(D), IKr (E) and IK1 (F) . µM-HCQ inhibited both IKr tail and IK1 pulse currents with IC50s of 10±0.6 µM (D) and 34±5 µM (F), respectively, with more marked effects on IKr, where the IC50 fell close to expected therapeutic levels in current Covid-19 treatment regimes (Borba et al., 2020), than IK1. In contrast, µM-HCQ has less effect on INa and ICaL with IC50s of 113 ± 78 µM and 209 ± 94 µM respectively, Nevertheless, HCQ caused graded INa reductions through the entire HCQ concentration range typified in a 22±1.8% reduction at 1 µM. No significant inhibition of Ito and IKs by µM-HCQ was observed.
Figure 2 presents corresponding results obtained with AZM applied by itself similarly encompassing clinical therapeutic levels. A previously reported dose regime administering 500, 250 and 250 mg/day on each of three successive days gave a day 2 serum concentration of 0.22 μM(Lai, Ho, Jain & Walters, 2011)). However, the entire, 0-100 μM explored AZM concentration range yielded only minimal inhibitory effects. Even 10 µM and 100 µM AZM reduced INa (A), ICaL,(B) IKs (E) and IKr (D) by <17% and <35% respectively and produced negligible effects on IK1 (F) and no effect on Ito(C).
Finally, Figure 3 summarizes results of applying the same range of five HCQ concentrations displayed in Fig. 1 now combined with 10 µM AZM. Adding AZM did not significantly alter the effects of HCQ on INa (A), ICaL (B), Ito(C), and IKs (E), but reduced the IC50 for IKr (IC50 = 7.7 ± 0.8 µM) and IK1 (30.4 ± 3.0 µM).