Fitting data reveals the complexities of NADP as a Ca²⁺ ATPase inhibitor

Wang et al. (2019) propose that NADP inhibits SERCA pump and that this explains the inhibitory effect of NADP on Ca²⁺ uptake by rat liver microsomes. The analysis of these data extracted from figures 2 and 10 is shown in Figure 1, the data on Ca²⁺ uptake from Figure 2 by Wang et al. are shown as red squares and data on Ca²⁺/Mg²⁺ ATPase activity from figure 10 are shown as blue diamonds. The fitting of these data using one single inverse saturation function ($100\ast {\text{IC}}_{50}^n/({\text{IC}}_{50}^n+{[\text{NADP}]}^n)$) with only two parameters (IC₅₀ and n, Hill coefficient) is shown in Figure 1. It is evident that the inhibitory effect of NADP on microsomal ATPase activity (blue line) has a different concentration dependence that the effect of NADP on the accumulation of Ca²⁺ by rat liver microsomes (red line). The former has an IC₅₀ of 0.11 mmol/L and a small Hill coefficient (0.32) while the latter has an IC₅₀ of 0.6 mmol/L and a Hill coefficient of 3.0 (Fig. 1). The authors argue that SERCA pump accounts for only 14% of the total ATPase activity (Wang et al. 2019). However, this does not seem to explain that the effect of NADP on ATPase activity and microsomal Ca²⁺ uptake occurs in a different concentration range resulting from different Hill coefficients. NAADP was discovered as a Ca²⁺ releasing agent that came as a contaminant in NADP (Lee and Aarhus 1995). NAADP not only is highly potent in releasing Ca²⁺ from microsomes derived from different tissues, but also from rat liver microsomes that were Ca²⁺ loaded by the activity of SERCA pump (Mándi et al. 2006). NAADP also induces a substantial inactivation of its Ca²⁺ releasing activity (Lee 1997), this situation has also been shown in NAADP‐responding channels (Fig. 2) isolated from liver lysosomes (Zhang and Li 2007) or for Ca²⁺ released from rat liver microsomes (Mándi et al. 2006). Wang et al. did experiments to discard the participation of NAADP. However, to achieve this goal it is essential to consider that NAADP induces a time‐ and concentration‐dependent self‐inactivation (Lee 1997; Mándi et al. 2006). It is clear then that there is an optimal concentration of NAADP to induce Ca²⁺ release from microsomes that is near 1 μmol/L and that this inactivation is not complete (Fig. 2).

Figure 1.

Fitting data extracted from figures 2 and 10 of Wang et al. The microsomal Ca²⁺ uptake activities at different NADP concentrations were taken from Figure 2 and they are shown as red squares. The fitting of these data (red line) was achieved using a single inverse saturation curve with an IC₅₀ = 0.6 mmol/L and a Hill coefficient, n = 3.0 with an R ² = 0.994. The Ca²⁺/Mg²⁺ ATPase activities at different NADP concentrations were taken from Figure 10 and shown as blue diamonds. The fitting of these data (blue line) with a single inverse saturation curve resulted in an IC₅₀ = 0.11 mmol/L and a Hill coefficient, n = 0.32 with an R ² = 0.996.

Figure 2.

NAADP‐induced ion channel activity data extracted from Figure 3 in Zhang and Li (2007) are shown as blue circles. The fitting of the data (blue line) was obtained with the difference between two saturation curves where the maximal activation was considered to be 100% activity and the maximal inactivation was 88% of maximal activity. The activation curve has an EC₅₀ = 0.27 μmol/L and a Hill coefficient, n = 0.472 while the inactivation function has an EC₅₀ = 6.8 μmol/L and n = 2.17 with an R ² = 0.993. Due to the presence of these two opposing effects, the maximal ion channel activity induced by NAADP is not higher than 70% and the inactivation by NAADP is not complete.

Nevertheless, Wang et al. used NAADP at 1 mmol/L, a concentration that induces a significant inactivation of NAADP‐triggered Ca²⁺ release activity. Accordingly, they saw that this high concentration of NAADP decreased Ca²⁺ uptake and more importantly, fully inhibited the effect of NADP (figure 5 of Wang et al.). This result is compatible with the idea that the effect of NADP depends on the activity of NAADP. Figure 6 of Wang et al. shows that NADP is not increasing Ca²⁺ efflux from microsomes, but the conditions used here are different to those used to assess Ca²⁺ uptake, for instance, the presence of oxalate, ATP, and a higher concentration of Mg²⁺ in the latter compared with the former. So it could be that NAADP activity is different in these two different assay conditions. Finally, the use of trans‐Ned‐19 to discard participation of NAADP resulted in unexpected observations. All the inhibitors tested (2‐APB, ruthenium red, and dantrolene) increased the inhibitory effect of NADP on Ca²⁺ uptake, except for trans‐Ned‐19. There is no clear explanation for these results. In any event, the data shown in table 1 of Wang et al. is not strong enough to discard the participation of NAADP that might be present in their NADP solution. In conclusion, the work by Wang et al. has not entirely ruled out the possibility that traces of NAADP in their NADP solution are responsible for activating a Ca²⁺ release channel resulting in the apparent NADP inhibitory effect on Ca²⁺ uptake by liver microsomes.