Human adenylyl cyclase 9 is auto-stimulated by its isoform-specific C-terminal domain

The increase of basal levels is dependent on the presence of the C2b domain. (A) Basal intracellular levels of cAMP reported by GloSensor22F bioluminescence in HEK293FT cells transiently transfected with 2 μg GloSensor22F plasmid and 1 (1:2) or 2 (2:2) μg of the plasmids of interest that encoded pcDNA3, E326D_AC9, or E326D_AC9C2a, respectively. Data are the average of the apparent plateau reached at 25–30 min of incubation (see Fig 3 of the main text) standardized in each well by the peak of the response elicited by 5 μM forskolin and 100 μM rolipram administered at the end of the experiment. Data are means ± S.D. n = 14/group. As the variances of the six groups were significantly different, the statistical analysis was carried out after log transformation of the data. One-way ANOVA, F(5,78) = 1,554, P < 0.0001, followed by Tukey’s post hoc multiple comparison test: +++P < 0.0001 significantly different from the respective pcDNA3 group, and ***P < 0.0001 when compared to the respective E326D_AC9C2A group. (B) Increment over the respective basal levels induced by 100 nM isoproterenol calculated by averaging three consecutive time-points once the peak level of bioluminescence was reached (see Fig 3 of the main text). The relative light units were standardized in each well by the peak of the response elicited by 5 μM forskolin and 100 μM rolipram administered at the end of the experiment. All points are shown, and the lines represent the means. (C) Increment over the respective basal levels induced by 1,000 nM prostaglandin E₁. Other details as in (B). (D) Concentration–response to isoproterenol. All measured points are shown. Non-linear four-parameter regression was used to fit the curves in GraphPad Prism v.6; the respective basal values were set as the bottom of the curves. The calculated EC50s (range 28–55 nM) and the maximal responses were not different at α = 0.05. (E) Concentration–response to prostaglandin E₁. All measured points are shown. As the maximum of the curve could not be estimated, EC50 values were not calculated.

(A) Basal intracellular cAMP levels reported by GloSensorF22 bioluminescence in HEK293FT cells transiently expressing pcDNA3, AC9, AC9C2a, E326D_AC9, or E326D_AC9C2a. Data are the average of the apparent plateau reached at 25–30 min of incubation (see Fig 3 of the main text) standardized in each well by the peak of the response elicited by 5 μM forskolin and 100 μM rolipram administered at the end of the experiment. Data are means ± S.D. n = 6/group. As the variances of the five groups were significantly different, statistical analysis was carried out after log transformation of the data. One-way ANOVA, F(4,25) = 320, P < 0.0001; Tukey’s post hoc multiple comparison test: ***P < 0.001 versus respective variant lacking the C2b domain, and +++P < 0.0001 versus the pcDNA3 group. (A, B) Concentration–response to isoproterenol from the cells described in (A). All measurements of the standardized peak responses are shown. Curves were fitted by non-linear four-parameter regression with a variable slope in GraphPad Prism v.6; the respective basal values were set as the bottom of the curves. As the maximum of the curves could not be estimated reliably, no EC50 values are reported. (C) Increment over the respective basal levels induced by 100 nM isoproterenol calculated by averaging three consecutive time-points once the peak level of bioluminescence was reached (see Fig 3 of the main text). The relative light units were standardized in each well by the peak of the response elicited by 5 μM forskolin and 100 μM rolipram administered at the end of the experiment. All points are shown, and the lines represent the means.

(A) Basal cAMP levels, all as in Fig S2 (A) except that data are means ± S.D. n = 14/group. As the variances of the five groups were significantly different, statistical analysis was carried out after log transformation of the data. One-way ANOVA, F(4,65) = 935, P < 0.0001; Tukey’s post hoc multiple comparison test: ***P < 0.001 versus respective variant lacking the C2b domain, and +++P < 0.0001 versus the pcDNA3 group. (A, B) Concentration–response to isoproterenol from the cells described in (A). All measurements of the standardized peak responses are shown. Curves were fitted by non-linear four-parameter regression with a variable slope in GraphPad Prism v.6; the respective basal values were set as the bottom of the curves. The EC50 of the E326D_AC9C2a (21 nM) was significantly lower (95% confidence intervals) than that of the other four groups—range 35–65 nM. The maximal responses were E326D_AC9C2a > E326D_AC9 = AC9C2a > pcDNA3 > AC9, where > denotes statistical significance at P < 0.05, as indicated by the 95% confidence intervals calculated by the non-linear regression algorithm. (C) Increment of cAMP levels above the respective basal levels at 30 nM isoproterenol. Means ± S.D. n = 4/group. Data reduction protocol as described in Fig S1B. One-way ANOVA, F(4,15) = 186.1, P < 0.0001; Tukey’s post hoc multiple comparison test: ***P < 0.001 versus respective full-length variant, and +++P < 0.0001 versus the pcDNA3 group. The results for 10 nM isoproterenol were closely similar.