Rapamycin reverses hypertrophic cardiomyopathy in a mouse model of LEOPARD syndrome-associated PTPN11 mutation

Abstract

LEOPARD syndrome (LS) is an autosomal dominant "RASopathy" that manifests with congenital heart disease. Nearly all cases of LS are caused by catalytically inactivating mutations in the protein tyrosine phosphatase (PTP), non-receptor type 11 (PTPN11) gene that encodes the SH2 domain-containing PTP-2 (SHP2). RASopathies typically affect components of the RAS/MAPK pathway, yet it remains unclear how PTPN11 mutations alter cellular signaling to produce LS phenotypes. We therefore generated knockin mice harboring the Ptpn11 mutation Y279C, one of the most common LS alleles. Ptpn11(Y279C/+) (LS/+) mice recapitulated the human disorder, with short stature, craniofacial dysmorphia, and morphologic, histologic, echocardiographic, and molecular evidence of hypertrophic cardiomyopathy (HCM). Heart and/or cardiomyocyte lysates from LS/+ mice showed enhanced binding of Shp2 to Irs1, decreased Shp2 catalytic activity, and abrogated agonist-evoked Erk/Mapk signaling. LS/+ mice also exhibited increased basal and agonist-induced Akt and mTor activity. The cardiac defects in LS/+ mice were completely reversed by treatment with rapamycin, an inhibitor of mTOR. Our results demonstrate that LS mutations have dominant-negative effects in vivo, identify enhanced mTOR activity as critical for causing LS-associated HCM, and suggest that TOR inhibitors be considered for treatment of HCM in LS patients.

Figure 1. Generation of mice expressing an inducible Ptpn11 Y279C allele.

(A) Structure of the Ptpn11 targeting locus and targeting construct region. Protein coding regions, LoxP sites, and Frt sites are shown. The arrows highlight the target positions of PCR primers used to identify homologous recombinants. Note that primers PLF and PRR are outside the region of homology. The expected amplicons are indicated with brackets and are specific to homologously recombined templates. LoxP sites are shown as black triangles; mutant LoxP511 sites are shown as white triangles. Their relative orientation is shown by the orientation of the triangles. FRT sites are shown as white ovals. Exons are indicated by rectangles. (B) Structure of locus after Flp-mediated recombination. (C) Structure of locus after Cre-mediated recombination. (D) Confirmation of correct targeting of ES cell clone 1F10 by PCR, revealing a 4.6-kb fragment on each correctly targeted side of the allele. M indicates molecular weight markers. (E) Germline transmission in F1 mice, as depicted by PCR, for the presence of the neo gene. (F) PCR detection strategy using LSGenofwd and LSR primers for detection of WT and LS/+ mice. (G) Restriction digest (HpyCH4V) of Ptpn11 exon 7 RT-PCR–derived fragment confirming expression of the mutant allele (full-length = 185 bp; fragment sizes = 100, 75 bp, 58 bp, 42 bp, and 10 bp [LS]; 100 bp, 75 bp, and 10 bp [WT]). Note that the 10-bp fragment probably comigrates with primer alone and/or contaminating RNA. PCR primers used are 721 forward/900 reverse. See Supplemental Table 1 for details on primers used throughout this figure. Fwd, forward; Rev, reverse.

Figure 2. LS/+ mice demonstrate phenotypic abnormalities similar to those in human LS.

(A) Representative photographs of en face and nose bridge profile (arrow) (image was brightened to enable a clear view of the facial phenotype). (B) Skeletal/chest deformity in LS/+ mice, with pectus carinatum superiorly (arrow) and pectus excavatum inferiorly (arrow). (C) Overall smaller body size in LS/+ mice as compared with that of WT mice. (D) Body length and weight are significantly less in LS/+ mice than in WT mice; n = 5–10 mice/group at each age measured. Data represent mean ± SEM; *P < 0.05, 2-tailed Student’s t test.

Figure 3. LS/+ mice have pathological cardiac hypertrophy.

(A) Representative hearts from 16-week-old WT and LS/+ mice. (B) Transverse sections of WT and LS/+ hearts stained with H&E (original magnification, ×100). (C) Heart weight to body weight (HW/BW) ratios of WT and LS/+ mice at indicated ages (n = 5–10 mice/group at each age measured). Data represent mean ± SEM; *P < 0.05, 2-tailed Student’s t test. (D) Paraffin-embedded heart sections from 16-week-old WT and LS/+ mice stained with reticulin (original magnification, ×400; scale bar: 100 μm). The average area of WT cardiomyocytes is 2,038 ± 98.5 μm²/cardiomyocyte and that of LS/+ cardiomyocytes is 2,578 ± 115.9 μm²/cardiomyocyte (P < 0.001, 2-tailed Student’s t test). Data represent mean ± SEM (200–500 cardiomyocytes/group). (E) Cardiomyocytes from 8- to 10-week-old WT and LS/+ hearts (original magnification, ×200). Cardiomyocyte length for WT hearts is 108.4 ± 2.6 μm and for LS/+ hearts is 113.5 ± 2.9 μm; cardiomyocyte width for WT hearts is 18.9 ± 0.8 μm and for LS/+ hearts is 22.4 ± 0.7 μm (P < 0.05, 2-tailed Student’s t test). Data represent mean ± SEM (200–500 cardiomyocytes; n = 4 mice each.). (F) H&E staining (original magnification, ×400; white bar: 100 μm) of paraffin-embedded longitudinal heart sections from 16-week-old WT and LS/+ mice. Note the myofiber disarray and increased inflammatory cells (indicated by white arrows and magnified in inset) in the interstitial spaces. (G) Scanning electron microscopy of cardiomyocytes (original magnification, ×4,800) from 12-week-old mice. Double arrowhead: 1 μm.

Figure 4. Decreased survival in LS/+ mice.

(A) Kaplan-Meier curve of WT versus LS/+ mice (n = 5–8 mice per group at each age point). Note significantly decreased survival in LS/+ mice at 12 months. *P < 0.05, log-rank test. (B) H&E-stained sections (original magnification, ×400; scale bar: 100 μm) depicting myofiber disarray and enlarged nuclei in 52-week-old LS/+ hearts compared with WT hearts. (C) Masson-Trichrome stain of paraffin-embedded longitudinal heart sections from WT and LS/+ mice reveals marked fibrosis in LS/+ hearts at 52 weeks of age (original magnification, ×400). Fibrous tissue is stained in blue and magnified in inset. (D) Reticulin stain of paraffin-embedded heart sections from 52-week-old WT and LS/+ mice (original magnification, ×400), showing enlarged LS/+ cardiomyocytes.

Figure 5. LS/+ mice have HCM that progresses to chamber dilation.

Representative echocardiographs of WT and LS/+ mice at 12 to 14 weeks of age and 52 weeks of age, respectively. Two-headed arrows indicate left ventricle chamber size. Note that at 12 weeks, LS/+ hearts have small chamber dimensions compared with those of WT controls; at 52 weeks, LS/+ hearts have increased chamber dimensions and have dilated/decompensated compared with those of WT controls.

Figure 6. LS/+ mice upregulate expression of cardiac fetal genes.

Total RNA from WT and LS/+ mice (n = 4–6 of each genotype) of the indicated ages was used to perform quantitative RT-PCR (each sample in triplicate). The ratio of ΔDC_T was analyzed using GAPDH as a control. Data represent mean ± SEM; *P < 0.01, 2-tailed Student’s t test. Note the significant difference in cardiac fetal gene expression in LS/+ and WT hearts of the same age.

Figure 7. LS mutants are catalytically impaired and show increased binding to IRS-1.

(A) WT, DG/+, and LS/+ 8- to 10-week-old mice were treated for 10 minutes with PBS or insulin (10 mU/g body weight). Hearts were isolated and lysed, and Shp2 immune complex PTP assays were conducted using pNPP as a substrate. The bottom panel shows comparable recovery of immunoprecipitated Shp2. Experiments were performed in triplicate (n = 4–6 animals per group). Data represent mean ± SEM; *P < 0.05, ^†P < 0.001, ^‡P < 0.01. All P values were derived from ANOVA with Bonferroni post-test when ANOVA was significant. (B) Primary cardiomyocytes isolated from 8- to 10-week-old WT or LS/+ mice were either left unstimulated or were stimulated for 15 minutes with IGF-1 (10 nM). Lysates from these cells were immunoprecipitated with anti–IRS-1 antibodies and immunoblotted with anti-Shp2 or anti–IRS-1 antibodies, as indicated, to detect complex formation. The bottom panel represents quantification of immunoprecipitated Shp2 normalized to total IRS-1 levels from the above experiment. Note increased IRS-1/Shp2 complex formation in LS/+ cardiomyocytes as compared with that in WT cardiomyocytes, even in the absence of IGF-1 stimulation.

Figure 8. Agonist-evoked Erk MAPK activation is impaired in LS/+ mice.

(A) Whole heart lysates from either unstimulated or 10-minute insulin-stimulated (10 mU/g body weight) WT and LS/+ mice (8- to 10-week-old) were harvested, lysed, and immunoblotted with anti-phospho–Erk1/2 antibodies. The blot was reprobed with anti-Erk1/2 antibodies as a loading control. Each lane represents an individual animal. (B) Quantification of data collected from 4–6 animals for each group, each derived from 3 different experiments. *P < 0.01, ANOVA plus Bonferroni post-test when ANOVA was significant. Note the significant difference between basal and insulin-stimulated Erk activity of WT heart lysates, but the lack of significant differences between Erk activity in either basal WT and LS/+ hearts or in basal and insulin-stimulated LS/+ hearts, respectively. (C) Primary cardiomyocytes isolated from 8-week-old WT and LS/+ mice were collected, cultured overnight, and then either left unstimulated or stimulated for the indicated times with IGF-1 (10 nM), EGF (25 ng/ml), angiotensin II (100 nmol/l), or IL-6 (10 ng/ml). Data are representative of n = 3 independent experiments. Cell lysates were immunoblotted with anti-phospho–Erk1/2 antibodies, and then membranes were reprobed with anti-Erk1 antibodies to control for loading.

Figure 9. Activation of the Akt/mTOR pathway is enhanced in LS/+ hearts and cardiomyocytes.

(A) Whole heart lysates from either unstimulated or 10-minute insulin-stimulated (10 mU/g body weight) WT and LS/+ mice (8 weeks old) were harvested, lysed, and immunoblotted with anti-phospho–Akt⁴⁷³ antibodies, followed by anti-Akt antibodies to control for loading. Each lane represents an individual animal. The dividing line represents grouping of images from the same gel. (B) Quantification of data collected from 4–6 animals from each group, derived from 3 independent experiments. *P < 0.001, ^†P < 0.05, ^‡P < 0.05. All P values were derived from ANOVA plus Bonferroni post-test when ANOVA was significant. (C) Primary cardiomyocytes from either unstimulated or IGF-1 stimulated (10 nM) WT and LS/+ mice (8 to 10 weeks old) were lysed and immunoblotted with anti-phospho–Akt⁴⁷³, Tsc2¹⁴⁶², or p70S6K antibodies, as indicated, followed by anti-Akt antibodies to control for loading. (D) Quantification of Akt phosphorylation, derived from 3 independent experiments. *P < 0.05, ^†P < 0.001, ^‡P < 0.001. All P values were derived from ANOVA plus Bonferroni post-test when ANOVA was significant.

Figure 10. Multiple cardiac signaling pathways are affected in LS/+ mice.

(A) Whole heart lysates from WT and LS/+ 8-week-old mice were harvested, lysed, and immunoblotted with anti-phospho-FAK³⁹⁷ antibodies. The blot was reprobed with total anti-Fak antibodies as a loading control. Each lane represents an individual animal. (B) Quantification of data collected from 4–6 animals from each group, derived from 3 independent experiments. *P < 0.05, 2-tailed Student’s t test. (C and D) Primary cardiomyocytes isolated from 8- to 10-week-old WT and LS/+ mice were collected, cultured overnight, and then either left unstimulated or stimulated for the indicated times with IGF-1 (10 nM). Cells were lysed and immunoblotted with (C) anti-phospho-JNK1/2 or (D) anti-phospho-Stat3 antibodies. Blots were reprobed with anti-JNK or anti-Stat3 antibodies, as indicated, to control for loading. Data are representative of n = 3 independent experiments.

Figure 11. LS/+ cardiomyocytes respond to rapamycin treatment.

(A) Photomicrograph of primary cardiomyocytes isolated from vehicle-injected WT, vehicle-injected LS/+, or rapamycin-injected LS/+ mice after 4 weeks of daily injections. Original magnification, ×200. (B) Quantification of the differences in area between vehicle- or rapamycin-treated WT and LS/+ cardiomyocytes (200–500 cells per group). Note the significant difference in total area between WT and LS/+ cardiomyocytes. Rapamycin had little/no effect on WT cardiomyocyte cell size. Results are the mean ± SEM. *P < 0.001, 2-tailed Student’s t test. (C) Inhibiting mTorc1 activity rescues aberrant LS/+ signaling. Whole heart lysates from vehicle-treated WT, vehicle-treated LS/+, and rapamycin-treated LS/+ mice after 4 weeks of daily injections were harvested, lysed, and immunoblotted with anti-phospho–Akt⁴⁷³, –GSK-3α (Ser21), –Tsc2¹⁴⁶², and –p70S6K antibodies, followed by anti-Akt antibodies to control for loading, respectively. Note that rapamycin treatment reversed the Akt/mTor pathway signaling defects associated with LS/+ mutation. rapa, rapamycin.

Figure 12. Rapamycin normalizes HCM in LS/+ mice.

(A) H&E-stained longitudinal sections of hearts from WT and LS/+ mice. Note normalization of hypertrophy in LS/+ hearts after rapamycin treatment (original magnification, ×100). (B) Reticulin stain of paraffin-embedded heart sections from 16-week-old WT and LS/+ mice (original magnification, ×400). (C) Quantification of average area (in μm²) of cardiomyocytes (200–500 cells counted/group) from WT or LS/+ cardiomyocytes isolated from mice that were either vehicle- or rapamycin-treated (2 mg/kg body weight) daily by i.p. injection for 4 weeks, then weekly for 4 weeks; see Results and Methods for details. Results are shown as the mean ± SEM. *P < 0.05, ^†P < 0.05. (D) Heart weight to body weight ratios of WT and LS/+ mice with vehicle- or rapamycin-treatment, as indicated. *P < 0.05, ^†P < 0.001. (E) Representative echocardiographs from 16-week-old vehicle-treated WT, vehicle-treated LS/+, or rapamycin-treated LS/+ mice. Two-headed arrows indicate left ventricle chamber size. (F) Anatomic and functional parameters of 12- and 16-week-old WT, LS/+, and LS/+ rapamycin-treated mice, as assessed by echocardiography, either after 4 weeks (measured at 12 weeks of age) or after 4 weeks of daily i.p. injections, followed by 4 weeks of weekly (measured at 16 weeks of age) i.p. injections. LVID-d, left ventricular chamber dimension in diastole. *P < 0.001 denotes significance between the vehicle-treated 16-week-old LS/+ mice and the LS/+ rapamycin-treated mice. All P values in C, D, and F were derived from ANOVA and Bonferroni post-test when ANOVA was significant.

Figure 13. Rapamycin reverses HCM in LS/+ mice.

(A) Representative echocardiographs from 16-week-old vehicle-treated WT, rapamycin-treated WT, vehicle-treated LS/+, or rapamycin-treated LS/+ mice. Two-headed arrows indicate left ventricle chamber size. (B) Anatomic and functional parameters of 12-, 16-, and 20-week-old WT, rapamycin-treated WT, LS/+, and rapamycin-treated LS/+ mice, as assessed by echocardiography, either after 4 weeks (measured at 16 weeks of age) or after 4 weeks of daily i.p. injections, followed by 4 weeks of weekly (measured at 20 weeks of age) i.p. injections. *P < 0.001, ^†P < 0.001, ^‡P < 0.001. All P values were derived from ANOVA and Bonferroni post-test when ANOVA was significant.

Figure 14. Rapamycin treatment normalizes cardiac fetal gene expression in LS/+ mice.

Total RNA from vehicle- or rapamycin-treated WT and LS/+ mice (n = 4–6 of each genotype) of the indicated ages was used to perform quantitative RT-PCR (each sample in triplicate). The ratio of ΔδC_T was analyzed using Gapdh as a control. Note that rapamycin-treated LS/+ heart lysates have normalized fetal gene expression profiles as compared with those of WT heart lysates. Data represent mean ± SEM. *P < 0.01, ^†P < 0.001, 2-tailed Student’s t test.