ADAP1 limits neonatal cardiomyocyte hypertrophy by reducing integrin cell surface expression

Abstract

The ArfGAP with dual PH domains 1 (ADAP1) regulates the activation of the hypertrophic mitogen-activated protein kinase ERK1/2 pathway in non-cardiomyocytes. However, its role in cardiomyocytes is unknown. Our aim was to characterize the role of ADAP1 in the hypertrophic process of cardiomyocytes. We assessed the expression of ADAP1 in the hearts of adult and neonatal rats by RT-qPCR and Western blotting and showed that it is preferentially expressed in cardiomyocytes. Adenoviral-mediated ADAP1 overexpression in cultured rat neonatal ventricular cardiomyocytes limited their serum-induced hypertrophic response as measured by immunofluorescence microscopy. Furthermore, ADAP1 overexpression completely blocked phenylephrine- and Mek1 constitutively active (Mek1ca) mutant-induced hypertrophy in these cells. The anti-hypertrophic effect of ADAP1 was not caused by a reduction in protein synthesis, interference with the Erk1/2 pathway, or disruption of the fetal gene program activation, as assessed by nascent protein labeling, Western blotting, and RT-qPCR, respectively. An analysis of cultured cardiomyocytes by confocal microscopy revealed that ADAP1 partially re-organizes α-actinin into dense puncta, a phenomenon that is synergized by Mek1ca overexpression. Biotin labeling of cell surface proteins from cardiomyocytes overexpressing ADAP1 revealed that it reduces the surface expression of β1-integrin, an effect that is strongly potentiated by Mek1ca overexpression. Our findings provide insights into the anti-hypertrophic function of ADAP1 in cardiomyocytes.

Figure 1

Adap1 expression in cardiac cells. (A) Relative levels of Adap1 mRNA expression in the rat brain (adult) and heart (adult and 2-day-old neonate) were measured by RT-qPCR and were normalized to the Rpl30 reporter gene (n = 4 independent tissues). (B) Representative Western blots of Adap1 and Gapdh (loading control) detected from whole brain and heart extracts. (C) The histogram represents the relative expression level of Adap1 normalized to Gapdh in the respective tissues (n = 4 independent tissues). (D) Representative Western blot of Adap1, α-Actinin (cardiomyocyte specific marker), and Gapdh (loading control) detected in the protein lysates of enriched rat neonatal ventricular cardiomyocytes (RNVC) and non-cardiomyocytes (Non-CM). (E) The histogram represents the relative expression level of Adap1 normalized to Gapdh in the respective cell lysates (n = 4 independent cell isolations). Full-length blots are presented in Supplementary Fig. 1.

Figure 2

ADAP1 restrains the serum-induced increase in cell size of cultured cardiomyocytes. (A) Western blot detection of adenovirus-mediated 3xFLAG-hADAP1 overexpression (MOI of 50) in rat neonatal ventricular cardiomyocytes (RNVC) cultured for 72 h post-infection. (B) RNVC were infected with either β-Gal- (negative control) or ADAP1-overexpressing adenovirus and were cultured for 72 h in the absence (0%) or presence (10%) of serum. Representative images of α-Actinin-immunostained RNVC (left) and corresponding segmented images were acquired using the Operetta High-Content Imaging System (Perkin Elmer). The scale bar represents 50 µm. (C) The histogram represents the cell surface areas of RNVC overexpressing either β-Gal or ADAP1 and cultured for 72 h with increasing concentrations of serum (n = 3 independent experiments). *P < 0.05 and **P < 0.005 vs. β-Gal; ***P < 0.0005 and ****P < 0.0001 vs. β-Gal at 0% [serum]. ^####P < 0.0001 vs. ADAP1 at 0% [serum]; ns, not significant; Two-way ANOVA with Sidak’s multiple comparison test. Full-length blots are presented in Supplementary Fig. 2.

Figure 3

ADAP1 blocks phenylephrine- and Mek1ca-induced hypertrophy. Cell surface area measurements of rat neonatal ventricular cardiomyocytes (RNVC) overexpressing ADAP1 in the absence or presence of 50 µM phenylephrine (A) or Mek1ca-overexpressing adenovirus (B) compared with an adβ-Gal-infected control. Quantification of at least 3 independent experiments expressed as means ± SD. *P < 0.05; **P < 0.005; ****P < 0.0001 vs. unstimulated β-Gal. ^####P < 0,0001 vs. stimulated β-Gal; ns, not significant; One-way ANOVA with Tukey’s multiple comparison test. (C) Representative Western blots of protein lysates from adβ-Gal-, adADAP1-, or adMek1ca-infected RNVC metabolically labeled for 2 h with L-azidohomoalanine (AHA) and detected with biotin-alkyne and streptavidin-HRP. (D) The histogram represents the rates of nascent protein synthesis relative to the β-Gal control, which were measured using the same experimental conditions as in C (n = 3 independent experiments). *P < 0.05 vs. β-Gal; ns, not significant; One-way ANOVA with Tukey’s multiple comparison test. (E) Representative Western blots of Erk1/2 phosphorylation status, ADAP1 and Mek1ca overexpression levels, and Gapdh (loading control) from protein lysates of RNVC cultured for 72 h following the indicated adenoviral infection. (F) The histogram represents the phospho/total Erk1/2 ratios relative to the β-Gal control and which were measured using the same experimental conditions as in E (n = 4 independent experiments). ****P < 0.0001 vs. β-Gal; ns, not significant; One-way ANOVA with Tukey’s multiple comparison test. Full-length blots are presented in Supplementary Fig. 3.

Figure 4

ADAP1 does not interfere with Mek1ca-induced fetal gene program activation. (A–I) Analysis by RT-qPCR of different mRNA expressed in rat neonatal ventricular cardiomyocytes (RNVC) that are representative of the fetal gene program. The RNVC were infected with an β-Gal- (negative control), ADAP1-, or Mek1ca-overexpressing adenovirus, individually or in combination as indicated, and were cultured for 72 h post-infection. The histograms represent mRNA expression levels relative to the β-Gal control and normalized to the Rpl30 reporter gene (n = 3 independent experiments). *P < 0.05; **P < 0.005; ***P < 0.0005; ****P < 0.0001 vs. β-Gal; ns, not significant; One-way ANOVA with Tukey’s multiple comparison test.

Figure 5

ADAP1 relocalizes cytoskeletal α-Actinin. (A) Representative confocal images (Olympus FluoView FV1000 microscope) of α-Actinin-immunostained rat neonatal ventricular cardiomyocytes (RNVC) infected with a β-Gal- (negative control), ADAP1-, or Mek1ca-overexpressing adenovirus, individually or in combination as indicated, and cultured for 72 h post-infection. Arrows point to α-Actinin dense puncta. The scale bar represents 12 µm. (B) Number of α-Actinin puncta per cell measured with the Operetta High-Content Imaging System (Perkin Elmer) using the same experimental conditions as in A (n = 4 independent experiments). ****P < 0.0001 vs. β-Gal; One-way ANOVA with Tukey’s multiple comparison test.

Figure 6

ADAP1 reduces cell surface β1-integrin expression in cardiomyocytes. (A) Representative Western blot of biotin-labeled cell surface β1-integrin expression compared to total β1-integrin expression in protein lysates of rat neonatal ventricular cardiomyocytes (RNVC) infected with a β-Gal- (negative control), ADAP1-, or Mek1ca-overexpressing adenovirus, individually or in combination as indicated, and cultured for 72 h post-infection. The expression levels of ADAP1, Mek1ca, and Gapdh (loading control) were also verified. (B) The histogram represents the level of cell surface β1-integrin expression normalized to total β1-integrin expression relative to the β-Gal control and measured using the same experimental conditions as in A (n = 3 independent experiments). *P < 0.05; **P < 0.005 vs. β-Gal; One-way ANOVA with Tukey’s multiple comparison test. (C) Representative Western blot of Hic-5, ADAP1, Mek1ca, and Gapdh (loading control) in protein lysates obtained from RNVC cultured for 72 h following the indicated adenoviral infection. (D) The histogram represents the relative expression levels of Hic-5 normalized to Gapdh, relative to the β-Gal control, and measured under the same experimental conditions as in C (n = 4 independent experiments). *P < 0.05; **P < 0.01 vs. β-Gal; ns, not significant; One-way ANOVA with Fisher’s LSD test. Full-length blots are presented in Supplementary Fig. 6.