Isradipine therapy in Cacna1dIle772Met/+ mice ameliorates primary aldosteronism and neurologic abnormalities

Abstract

Somatic gain-of-function mutations in the L-type calcium channel CaV1.3 (CACNA1D gene) cause adrenal aldosterone-producing adenomas and micronodules. De novo germline mutations are found in a syndrome of primary aldosteronism, seizures, and neurologic abnormalities (PASNA) as well as in autism spectrum disorder. Using CRISPR/Cas9, we here generated mice with a Cacna1d gain-of-function mutation found in both adenomas and PASNA syndrome (Cacna1dIle772Met/+). These mice show reduced body weight and increased mortality from weaning to approximately 100 days of age. Male mice do not breed, likely due to neuromotor impairment, and the offspring of female mice die perinatally, likely due to lack of maternal care. Mice generated by in vitro fertilization showed elevated intracellular calcium in the aldosterone-producing zona glomerulosa, an elevated aldosterone/renin ratio, and persistently elevated serum aldosterone on a high-salt diet as signs of primary aldosteronism. Anesthesia with ketamine and xylazine induced tonic-clonic seizures. Neurologic abnormalities included hyperlocomotion, impaired performance in the rotarod test, impaired nest building, and slight changes in social behavior. Intracellular calcium in the zona glomerulosa, aldosterone levels, and rotarod performance responded to treatment with the calcium channel blocker isradipine, with implications for the therapy of patients with aldosterone-producing lesions and with PASNA syndrome.

Keywords: Endocrinology; Hypertension; Mouse models; Neurological disorders; Neuroscience.

Figure 1. Generation of Cacna1d^Ile772Met/+ mice.

(A) Topology of the Cacna1d gene (isoform ENSMUST00000238504.1) with Exons as blue boxes and intronic segments as thin lines. Introns longer than 2,000 bp are truncated. The sequence encoding the first 18 amino acids of exon 16 (encoding Ile772) is shown below. Substitution of ATG (Met; orange) for ATC (Ile; blue) was performed using CRISPR/Cas9 editing. The PAM site (gray background) as well as the gRNA (black box) used for mutagenesis are highlighted. (B) Sequencing of genomic DNA obtained from ear biopsies shows the heterozygous mutation. (C) Sequencing of cDNA from adrenal gland demonstrates expression of the mutant RNA.

Figure 2. Cacna1d^Ile772Met/+ mice show higher mortality and gain less weight than WT mice.

(A) Kaplan-Meier curve showing survival probability of Cacna1d^Ile772Met/+ (Het, orange) and WT (blue) mice relative to the number of mice weaned at 21 days. The 95% CI is plotted in a lighter shade around the mean. Mortality is higher in Cacna1d^Ile772Met/+ than in WT mice from weaning to approximately 100 days of age (log rank test; n_WT = 103, n_Het = 114; P < 0.005; χ²= 24.39, df = 1). (B) Body weight determined during the SHIRPA assessment prior to behavioral phenotyping experiments (age 13–18 weeks). Cacna1d^Ile772Met/+ mice are lighter than WT littermates regardless of sex (Mann-Whitney U; n_{WT, female} = 26, n_{Het, female} = 18, P = 0.0004, U = 329.5; n_{WT, male} = 22, n_{Het, male} = 20, P = 6.2×10^–5, U = 333). (C) Body length measured during the SHIRPA assessment show that Cacna1d^Ile772Met/+ mice are shorter than WT littermates regardless of sex (Mann-Whitney U; n_{WT, female} = 22, n_{Het, female} = 17, P = 7.0×10^–5, U = 56; n_{WT, male} = 18, n_{Het, male} = 17, P = 2.3×10^–5, U = 33). ***P < 0.001.

Figure 3. Adrenal histology and Cyp11b2 ISH of Cacna1d^Ile772Met/+ and WT.

(A and D) H&E staining of FFPE adrenal sections from 14-week-old female WT (A) or Cacna1d^Ile772Met/+ mice (Het, D). F, fat; C, capsule; ZG, zona glomerulosa; ZF, zona fasciculata; XZ, x-zone, M, medulla. (B and E) ISH of WT or Het mice. A probe for the detection of Cyp11b2 mRNA was used. Staining is confined to the ZG. (C and F) Controls for ISH without any probes are shown. The stainings shown in this figure are representative for 8 WT (6 female, 2 male) and 6 Het (3 female, 3 male). Scale bars: 500 μm (overview, top) or 100 μm (magnification, bottom).

Figure 4. Serum aldosterone and aldosterone/renin ratio are increased in 18-week-old Cacna1d^Ile772Met/+ mice.

(A) Serum aldosterone levels as determined by ELISA are elevated in Cacna1d^Ile772Met/+ (Het) versus WT mice (2-tailed t test; n_WT = 19, n_Het=12, P = 0.005, t = –3.06). (B) Renin concentrations as determined by ELISA exhibit no significant difference between WT and Het mice (Mann-Whitney U test; n_WT = 19, n_Het = 13, P = 0.94, U = 121). (C) The aldosterone/renin ratio is increased in Het mice (2-tailed t test; n_WT = 19, n_Het=12, P = 0.02, t = –2.47). (D) A 2-week high-salt diet (4% Na⁺ in chow, 1% NaCl in drinking water) resulted in lower aldosterone levels in both genotypes. However, Het mice still exhibited significantly higher serum aldosterone levels (Mann-Whitney U test; n_WT = 31, n_Het = 28, P = 1.21 × 10^–9, U = 33). *P < 0.05, **P < 0.01, ***P < 0.001.

Figure 5. Calcium imaging in acute adrenal slices to determine the response of the zona glomerulosa to potassium and angiotensin II.

(A and B) Representative image of a WT (A) or Ile772Met/+ (B) adrenal slice stained with Fura-2 AM (top). On the bottom, representative traces are shown. Scale bars: 10 μm. (C) Mean calcium concentrations in zona glomerulosa cells determined by imaging of Fura-2 AM–stained slices. The extracellular concentration of angiotensin II was varied between 2 and 500 pM in the presence of either 3 (left) or 5 mM K⁺ (right) (likelihood ratio test of linear mixed models; exact statistical values are given in Table 2; *P <0.05, **P < 0.01).

Figure 6. Cacna1d^Ile772Met/+ mice show motor abnormalities as well as deficits in nest construction.

(A) WT mice stay significantly longer on the rotating rod in rotarod experiments than Cacna1d^Ile772Met/+ (Het) mice (Mann-Whitney U; n_WT = 19, n_Het = 19, P = 3.48 × 10^–12, U = 2852.5). (B) The total distance covered by Het mice in 10 minutes in the open-field experiment is significantly higher than for WT as a sign of hyperlocomotion (2-tailed t test; n_WT = 19, n_Het = 19, P = 2.6 × 10^–7, t = -6.32). (C) The percentage of the open-field observation time spent in the center of the box is not different between genotypes (Mann-Whitney U; n_WT = 19, n_Het = 19, P = 0.08, U = 119). (D) Het mice are largely unable to build well-formed nests over 18 hours, as indicated by the lower nest scores (Mann-Whitney U; n_WT = 19, n_Het = 19, P = 0.0006, U = 281.5). ***P < 0.001.

Figure 7. Cacna1d^Ile772Met/+ mice show hyperlocomotion in response to changes in ambient light but are unable to maintain a hanging position at the top of the cage.

(A–C) Course of 3 groups of home-cage behaviors over time for 24 hours (see Supplemental Figure 5 for all other groups). Cacna1d^Ile772Met/+ mice show hyperlocomotion as indicated by longer distances traveled (A; likelihood ratio test of linear mixed models; n_WT = 19, n_Het = 17; P_adj = 0.02, z = 0.085) together with increased periods spent moving around (B; likelihood ratio test of linear mixed models; n_WT = 19, n_Het = 17; P_adj = 0.02, z = 0.00208). Hanging from the lid of the cage is significantly reduced in Cacna1d^Ile772Met/+ mice compared with WT (C; likelihood ratio test of linear mixed models; n_WT = 19, n_Het = 17; P_adj = 0.04, z = 2.631). For all statistical models, genotype was used as fixed and recording as well as time as random factors. All P values are adjusted for 10 comparisons (including the remaining behaviors shown in Supplemental Figure 5) using the FDR method. Data are shown as mean ± 95% CI.

Figure 8. Cacna1d^Ile772Met/+ mice tolerate free interaction but show avoidance behavior in forced interactions.

(A) In the second session of the 3-chamber test, Cacna1d^Ile772Met/+ (Het) mice show a slight preference for social interaction as seen by longer total durations spent near the unfamiliar stranger (Mann-Whitney U test; n_WT = 19, P = 0.22, U = 201; n_Het = 18, P = 0.04, U = 251). (B) During the third session of the 3-chamber test, Cacna1d ^Ile772Met/+ — but not WT mice — show a preference for the unfamiliar stranger (Mann-Whitney U test; n_WT = 19, P = 0.60, U = 162; n_Het = 18, P = 0.001, U = 70) (C–I) Parameters determined in the social proximity test. The increase in nose-to-anogenital, crawl-over, crawl-under, and jump escape events are considered indicative of increased anxiety and avoidance behavior (Mann-Whitney U test; n_WT = 10, n_Het = 8). P = 0.66, U = 34.5 (C); P = 0.17, U = 24 (D); P = 0.008, U = 9.5 (E);P = 0.002, U = 10 (F); P = 0.1, U = 21 (G); P = 0.003, U = 6 (H); P = 0.002, U = 5.5 (I). *P < 0.05, **P < 0.01.

Figure 9. Treatment with the L-type calcium channel blocker isradipine partially reverses the effects of the Ile772Met mutation.

(A) Isradipine lowers mean intracellular calcium concentrations (likelihood ratio test of linear mixed models; 50 nM versus 0 nM isradipine: n_{WT, cells} = 143, n_{WT, slices} = 12, P_adj = 0.028, χ² = 6.08, df = 1; n_{Het, cells} = 85, n_{Het, slices} = 9, P_adj = 2.16 × 10^–7, χ²(1) = 28.36; 300 nM versus 0 nM isradipine: n_{WT, cells} = 143, n_{WT, slices} = 12, P_adj = 1.0, χ²(1) = 0.22; n_{Het, cells} = 85, n_{Het, slices} = 9, P_adj = 2.56 × 10^–7, χ²(1) = 27.89; 4 mM K⁺ and 100 pM Ang II). (B) Isradipine lowers baseline intracellular calcium concentrations in Het mice (cells/slices from A; mean ± 95% CI; likelihood ratio test of linear mixed models; 50 nM isradipine: WT: P_adj = 0.8, χ²(1) = 0.72; Het:, P_adj = 0.0016, χ²(1) = 11.15; 300 nM isradipine: WT: P_adj = 0.11, χ²(1) = 3.72; Het: P_adj = 4.48×10^–8, χ²(1) = 32.82). P values in A and B were adjusted for multiple comparisons using the Bonferroni procedure. (C) Four hours after the last isradipine dose, treated (+) Het show a significant improvement in rotarod performance compared with untreated controls (–) (2-tailed t test; n_{WT, treated} = 10, n_{WT, control} = 11, P_wt = 0.63, t_wt = -0.50; n_{Het, treated} = 9, n_{Het, control} =9, P_Het = 0.03, t_Het = 2.17). Values were normalized to the mean of untreated mice of the same genotype. (D) In Het mice, serum aldosterone is lower relative to controls (95% CI, 92.1–327.1, Het: 27.7–99.0; mean ± 95% CI, black circles ± whiskers; 100% of controls is indicated with dashed line, and individual values are shown as colored circles). The break in the y axis extends from 340 to 750%. See Supplemental Figure 8A for absolute values.