This Article Abstract Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Sadoshima, J. Articles by Izumo, S. Search for Related Content PubMed PubMed Citation Articles by Sadoshima, J. Articles by Izumo, S. Pubmed/NCBI databases Compound via MeSH Substance via MeSH

(Circulation Research. 1997;80:228-241.)
© 1997 American Heart Association, Inc.

Articles

Angiotensin II and Serum Differentially Regulate Expression of Cyclins, Activity of Cyclin-Dependent Kinases, and Phosphorylation of Retinoblastoma Gene Product in Neonatal Cardiac Myocytes

Junichi Sadoshima, Hiroki Aoki, Seigo Izumo

the Cardiovascular Research Center, Division of Cardiology, University of Michigan Medical Center, Ann Arbor.

Correspondence to Dr Seigo Izumo, Cardiovascular Research Center, University of Michigan Medical Center, Ann Arbor, MI 48109-0644. E-mail sizumo@umich.edu

	Abstract

Top Abstract Introduction Materials and Methods Results Discussion References

 
The hypertrophic response in cardiac myocytes and the mitogenic response in other cell types share various early cellular responses. However, how the subsequent cell growth response, such as cell cycle machinery, is regulated in cardiac hypertrophy is not understood. Using cultured neonatal rat cardiac myocytes, we examined the effect of angiotensin II (Ang II), a hypertrophic stimulus, on mRNA and protein expression of cyclins and cyclin-dependent protein kinases (cdks), activity of cdks, and phosphorylation of retinoblastoma gene product (pRb). The effect of FCS, a stimulus that was previously reported to initiate both protein and DNA synthesis in cardiac myocytes, was also examined for comparison. Ang II activated cdk4 and caused phosphorylation of pRb, peaking at 12 hours, but subsequently downregulated cyclin D₁, D₃, and A expression and cdk2 activity. FCS increased the expression of G₁-S cyclins, caused activation of cdk4, cdk2, and cdc2, and strongly phosphorylated pRb but failed to significantly stimulate DNA synthesis in neonatal cardiac myocytes. These results suggest that Ang II transiently activates but subsequently downregulates cell cycle regulators. Induction of G₁ and G₁-S cyclins and activation of cdks by FCS are not sufficient to drive cardiac myocytes into S phase. The functional role of pRb phosphorylation by Ang II and serum stimulation and, by inference, the subsequent liberation of E2F in terminally differentiated myocytes remain to be elucidated.

Key Words: cell cycle • hypertrophy • angiotensin II • cardiac myocyte • retinoblastoma gene product

	Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

 
Cardiac hypertrophy is an adaptive process used by postmitotic cardiac myocytes in response to various stresses, including hemodynamic overload and ischemia.¹ Increased cell size caused by an increase in protein synthesis is a cardinal feature of cardiac hypertrophy.¹ Because cardiac myocytes lose the ability to synthesize DNA soon after birth,^{2 3 4 5} they increase cell size but not cell number in response to growth stimuli.^{1 2} Accumulating evidence suggests that growth factors, such as Ang II and endothelin-1, may play an important role in the pathogenesis of cardiac hypertrophy.^{6 7 8} Intensive studies on the cellular signal transduction mechanisms of cardiac hypertrophy have indicated that hypertrophic stimuli activate multiple second messenger systems, such as phospholipases C, D, and A2, Ca²⁺, PKC, tyrosine kinase, Ras, Raf, MAP kinase, and 90- and 70-kD S6 kinases.^{9 10 11 12 13 14 15} Hypertrophic stimuli also cause induction of various IE genes, such as c-fos, c-jun, c-myc, and Egr-1.¹⁶ Interestingly, these second messenger systems and IE genes are activated by mitogenic stimuli in various cell types.¹⁷ Thus, hypertrophic stimuli and mitogenic stimuli seem to share the immediate-early cell signaling.

Mitogenic responses in undifferentiated cells are known to be regulated by cell cycle regulators, such as cyclins, cdks, and cdk inhibitors (reviewed in Reference 18). Cyclins are a family of proteins that have oscillating levels of expression during the cell cycle. Cyclins associate with cdks and act as a regulatory subunit of cdks, thus activating them in a cell cycle–specific manner. Cyclin C, cyclins D₁, D₂, and D₃, and cyclin E play important roles in the G₁ phase. Cyclin A is a key molecule in the S and G₂/M phases; cyclin B is essential in the G₂/M phase. The cdks are a family of serine/threonine kinases closely related to the cdc2 gene product of the yeast Schizosaccharomyces pombe and the cdc28 product of Saccharomyces cerevisiae.¹⁸ It has been shown in cultured cell lines that the cyclin D–cdk4/cdk6 complex regulates G₁ progression, the cyclin E/cyclin A–cdk2 complex is essential for G₁/S transition, and cyclin A/cyclin B–cdc2 (cdk1) promotes entry into mitosis. Activity of cdks is regulated not only by binding of cyclins but also by phosphorylation of threonine and tyrosine residues and by binding of cdk inhibitors, such as p21^CIP1/WAF1, p27^KIP1, p57^KIP2, and the INK4 family (p15, p16, p18, and p19).¹⁹

The activated cyclin-cdk complex is thought to phosphorylate various cellular substrates relevant to cell cycle progression. The most intensely studied among cdk substrates is pRb. In the G₁ phase, pRb is hyperphosphorylated by the cyclin D–cdk4 complex,²⁰ which causes release and activation of the E2F/DP transcription factor complex, which is otherwise sequestered and inhibited by pRb. E2F/DP in turn activates expression of genes required for S-phase progression.²¹ It should be noted, however, that in vivo substrates of cdks largely remain to be characterized. This raises the possibility that cdks may have some roles even in the growth response of nondividing cells, if they are activated.

Although the cell cycle regulation has been extensively studied in various mitogenic responses, it has not been well characterized in postmitotic cardiac myocytes (for review see References 4 and 5). Because hypertrophic stimuli and mitogenic stimuli share the early cell signaling, it is important to examine to what extent two different growth stimuli share the further downstream cell signaling. Therefore, we examined the effects of hypertrophic stimuli on cell cycle regulators in cardiac myocytes and compared them with those of mitogenic stimuli. We have previously shown that treatment with Ang II causes a hypertrophic response in neonatal rat cardiac myocytes.²² In the same preparation, serum stimulation has been shown not only to increase total protein content¹³ but also to initiate DNA synthesis without an increase in cell number during the early period of cell culture.³ Using these two different types of cardiac hypertrophic stimuli, we systematically examined how Ang II and serum affect the cell cycle machinery, such as expression of cyclins and cdks, activity of cdks, and phosphorylation of pRb in the neonatal rat cardiac myocyte culture system.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

 
Materials
Anti–cyclin D₁ (monoclonal, 72-13G), anti–cyclin D₃ (rabbit polyclonal, C-16), anti–cyclin E (rabbit polyclonal, M-20), anti–cyclin A (rabbit polyclonal, H-432), anti-cdk2 (rabbit polyclonal, M-2), anti-cdk4 (rabbit polyclonal, C-22), anti-cdk6 (rabbit polyclonal, C-21), and anti-cdc2 (monoclonal, 17) antibodies were purchased from Santa Cruz. Anti-pRb (monoclonal, G3-245) was from Pharmingen. Rabbit anti-mouse IgG antibody and normal rabbit serum (nonimmune serum) were from Jackson Immuno Research. Histone H₁ was from Boehringer-Mannheim. Anti-BrdU antibody (FITC-conjugated Bu 5.1) was from IBL Research Products.

Cell Culture
Primary culture of the neonatal rat cardiac myocyte was prepared as described previously.^{22 23} After enzymatic dissociation, the cells were preplated for 1 hour to selectively enrich for cardiac myocytes. The resultant suspension of myocytes was plated onto gelatin-coated 35- or 60-mm culture dishes at a density of 1.5x10⁵ cells/cm² and cultured as described previously.^{22 23} The use of the preplating procedure, mitogen-poor serum for plating (5% horse serum), BrdU, and a prompt switch to the serum-free medium enabled us to routinely obtain myocyte cultures in which >90% are myocytes, as assessed by immunofluorescence staining with a monoclonal antibody against sarcomeric myosin heavy chain (MF20). Cardiac fibroblast culture was prepared as described previously.^{22 23} All experiments were performed 48 hours after changing to the serum-free medium. Comparison of time-zero samples obtained from multiple experiments in the same Northern blot indicated that the difference in the levels of cdc2 mRNA expression was ±20%, suggesting that the expression levels of cell cycle regulators were almost comparable among multiple experiments. To suppress proliferation of "contaminating" cardiac fibroblasts in myocyte culture in response to FCS stimulation, most of the cardiac myocyte experiments were performed in the presence of BrdU (100 µmol/L) unless otherwise described. Neither Ang II nor FCS caused a significant increase in total cell number in the myocytes culture over 48 hours (data not shown), confirming that the BrdU treatment effectively suppressed proliferation of "contaminating" cardiac fibroblasts even after FCS stimulation. Ang II was applied every 12 hours to compensate for its degradation by endogenous angiotensinase. We also used [Sar¹]Ang II, which is more resistant to angiotensinase, in some experiments. The results of the experiments using [Sar¹]Ang II were indistinguishable from those using Ang II. The Ang II concentration in FCS does not seem significant because the induction of cyclin D₁ protein expression by FCS was not significantly affected by losartan (data not shown).

Northern Blot Analysis
Isolation of total cellular RNA and Northern blot analyses were performed as described previously.^{22 23} The probes for cyclin A (human), cyclin D₁ (murine CYL1/cyclin D1), cyclin D₂ (murine CYL2/cyclin D2), cyclin D₃ (murine CYL3/cyclin D3), cyclin E (human), cdc2 (murine), cdk2 (human), and GAPDH were used as described previously.^{23 24} The hybridization signals of specific mRNAs were normalized to those of GAPDH mRNA to correct for differences in loading and/or transfer. Densitometric analyses were performed using Eagle Eye II (Stratagene) and NIH Image (1.57 version).

Immunoprecipitation and Immunoblotting
Cell lysates were prepared with ice-cold lysis buffer containing 50 mmol/L Tris-HCl (pH 7.4), 250 mmol/L NaCl, 0.1% NP-40, 5 mmol/L EDTA, 10 mmol/L NaF, 0.1 mmol/L dithiothreitol, 1 mmol/L AEBSF, 10 µg/mL aprotinin, and 0.1 mmol/L leupeptin. Total protein content in the cell lysate was determined by the Lowry method (Sigma Chemical Co). Immunoprecipitation was performed by incubating the cell lysates obtained from one 60-mm dish with 1 µg of specific antibodies against cyclins or cdks. We used the entire cell lysate (except 10 µL for determination of protein content) from one dish for immunoprecipitation, because neither Ang II nor FCS significantly changed the cell number over 24 hours. To precipitate immune complexes, protein A–Sepharose was used for rabbit polyclonal antibodies. When mouse monoclonal antibodies were used, rabbit anti-mouse IgG antibody was added before protein A–Sepharose. Immunoprecipitates were separated by SDS-PAGE and transferred to an Immobilon-P membrane (Millipore). Immunoblotting was performed with antibodies at 1 µg/mL. HRP-conjugated protein A at 1:10 000 dilution was used as a secondary antibody for rabbit polyclonal antibodies; HRP-conjugated anti-mouse IgG at 1:10 000 was used for mouse monoclonal antibodies. ECL (Amersham) was used as a detection system. Linearity of signals obtained by ECL and subsequent fluorography was confirmed as follows: The myocyte lysate containing different amounts of total protein (0, 100, 250, 500, 750, 1000, 1500, and 2000 µg) was aliquoted into eight different tubes, and lysis buffer was added so that each tube had equal volume. These tubes were subjected to immunoprecipitation/blotting, and the signals were visualized by the ECL system. The intensity of the bands was quantified by densitometric analysis and plotted against the amount of total protein subjected for immunoprecipitation/blotting. A linear relationship was observed between them (Fig 1). The expression of the cyclin and cdk protein was determined per dish or after it was normalized to the amount of total protein in the cell lysates.

View larger version (25K): [in this window] [in a new window]   Figure 1. Linearity of the immunoprecipitation (IP)/immunoblotting/ECL system. Cardiac myocytes were stimulated with FCS for 24 hours. The cell lysates containing different amounts of total protein were subjected to IP with anti-cdk2 antibody (1 µg per sample). The immunoprecipitates were immunoblotted with the same antibody (1 µg/mL). Signals were visualized by the ECL system. A, A representative autoradiogram is shown. In lanes 1 through 8, increasing amounts of total protein (shown in panel B) were subjected to IP. The band at 33 kD (indicated by an arrow) corresponds to cdk2. In this experiment, a slightly faster migrating faint band was also observed when large amounts of protein were subjected to IP. This band may represent an isoform of cdk2 or a posttranslationally modified cdk2. B, The intensity of the upper band was quantified by densitometric analysis (NIH image). The intensity of each band (arbitrary units) was plotted against the amount of total protein subjected to IP. A similar result was obtained when a faster migrating faint band was analyzed together with the main 33-kD band. The result shown is representative of two experiments.

cdk Assays
The immune complex cdk assays were performed as described previously.^{22 24} The kinase reaction was performed in 50 µL kinase buffer supplemented with 2.5 µg histone H₁ or 3 µg GST-pRb (amino acids 769 to 921 of mouse pRb; see below), 10 µmol/L cold ATP, and 5 µCi [-³²P]ATP (6000 Ci/mmol) for 20 minutes at 37°C with shaking. The reaction was stopped by adding &fraq14; vol of 5x sample buffer. Phosphorylated proteins were separated on 12.5% SDS-polyacrylamide gels and then subjected to autoradiography. The migration of histone H₁ or GST-pRb on SDS-polyacrylamide gel was determined by Coomassie blue staining.

A DNA fragment encoding the carboxy-terminal region of mouse pRb (amino acids 769 to 921) was obtained by reverse transcription followed by polymerase chain reaction. The fragment was cloned into pGEX4T-3 (Pharmacia), in frame with the GST gene. The resulting fusion protein was purified from bacterial lysates with the aid of glutathione agarose beads (Pharmacia) and used as a substrate for the pRb kinase assay.

Immunostaining
Immunofluorescence cell staining was performed as described previously.²³ Anti–cyclin D₁ (2 µg/mL), anti–cyclin E (2 µg/mL), anti-cdk2 (2 µg/mL), anti-cdk4 (2 µg/mL), or anti–sarcomeric myosin (MF20, 1:2 dilution) antibody was used as a primary antibody.

BrdU Incorporation
BrdU incorporation and immunostaining of the BrdU-incorporated cells were performed as described previously.²² Briefly, cells were plated on gelatin-coated glass coverslips after preplating. Cells were cultured in myocyte culture medium with 5% horse serum but without BrdU for 24 hours. Cells were then kept in serum-free medium for 24 or 48 hours and then stimulated with Ang II (100 nmol/L) or FCS (20%) for 48 hours. Control cultures were prepared without stimulation with Ang II or FCS. BrdU (100 µmol/L) was added together with Ang II or FCS, and cells were maintained with BrdU for the entire experiment. Cells were stained with antibodies against BrdU and sarcomeric myosin as described previously.²²

Statistics
Data are given as mean±SEM. Statistical analysis was performed using ANOVA as appropriate. Posttest multiple comparisons were performed by the method of Bonferroni. Significance was accepted at the P<.05 level.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

 
Ang II and FCS Differentially Regulate mRNA Expression of G₁-S Cyclins
We first examined the effect of Ang II and FCS on mRNA expression of G₁-S cyclins in neonatal rat cardiac myocytes. Cardiac myocytes were cultured in serum-free conditions for 48 hours and then stimulated with Ang II or FCS for various times. Results of Northern blot analyses are shown in Figs 2 and 3. Hybridization of each specific probe was observed at the expected size for each probe: cyclin D₁, 4.5 kb²⁵ ; cyclin D₂, 6.0 kb²⁵ ; cyclin D₃, 2.1 kb²⁵ ; cyclin E, 2 to 2.5 kb²⁶ ; and cyclin A, 1.8 and 2.7 kb.²⁷ An additional band smaller in size was observed in cyclin D₂ blots (Fig 2B, asterisk). Although we do not know the identity of this band, it was regulated in a manner similar to that for the upper 6.0-kb band. Low levels of basal expression of cyclins were observed at time zero, but they did not significantly change over the next 24 hours in unstimulated cardiac myocytes (Fig 3, open squares). Ang II treatment slightly but significantly increased the expression of cyclin D₁ at 4 to 8 hours but gradually downregulated it thereafter (Figs 2A, 3A, and 3B). At 20 hours, expression of cyclin D₁ was significantly lower than that of control (Fig 3B). On the other hand, FCS treatment significantly increased the expression of cyclin D₁ over 24 hours, peaking at 8 and 24 hours (Figs 2A, 3A, and 3B). Both Ang II and FCS treatment significantly upregulated the expression of cyclin D₂ (a 6.0-kb form) over 24 hours (Figs 2B, 3C, and 3D). Ang II treatment gradually downregulated the expression of cyclin D₃, whereas FCS treatment upregulated it over 24 hours (Figs 2C and 3E). Neither Ang II nor FCS significantly affected the expression of cyclin E (Figs 2D and 3F). Ang II significantly increased the expression of cyclin A by 1.5- to 2.0-fold at 4 hours but gradually downregulated it thereafter (Figs 2E, 3G, and 3H). FCS treatment significantly upregulated the expression of cyclin A over 24 hours (Figs 2E, 3G, and 3H). These results indicate that Ang II and FCS differentially regulate mRNA expression of cyclins D₁, D₃, and A, whereas they both increase mRNA expression of cyclin D₂ and do not significantly affect that of cyclin E.

View larger version (78K): [in this window] [in a new window]   Figure 2. The effect of Ang II and FCS on mRNA expression of G1-S cyclins. Cardiac myocytes were treated with Ang II (100 nmol/L) or FCS (20%) for the indicated times. Total RNA was isolated, and Northern blot analyses were performed. Total RNA (15 µg) was loaded for each lane. Representative Northern blots are shown: A, cyclin D1; B, cyclin D2; C, cyclin D3; D, cyclin E; and E, cyclin A. The message for each cyclin is indicated by an arrow. Ethidium bromide staining of 18S and 28S RNA showed equal amounts of RNA in each lane. Similar results were obtained from two other experiments. *Smaller-sized mRNA was observed in cyclin D2 blots.

View larger version (29K): [in this window] [in a new window]   Figure 3. Time course of mRNA expression of G1-S cyclins by Ang II and FCS. Cardiac myocytes were treated with or without Ang II (100 nmol/L) or FCS (20%) for the indicated times. Northern blot analyses of cyclin D1 (A and B), cyclin D2 (C and D), cyclin D3 (E), cyclin E (F), and cyclin A (G and H) were performed as in Fig 2. Densitometric analyses of Northern blots are shown. For cyclins D2 and A, densitometric analyses of 6.0- and 2.7-kb transcripts are shown, respectively. Values were normalized to those at time zero. For each analysis, the signal was normalized to that obtained with a GAPDH probe. Results shown in panels A, C, E, F, and G are from representative experiments. In panel F, cyclin E mRNA increased by 55% of control after a 20-hour treatment with FCS. However, it was not statistically significant when it was analyzed with data from two other experiments. Results in panels B, D, and H are mean±SEM obtained from two to four separate experiments. In panels B and H, figures on FCS columns indicate fold increase of cyclins D1 and A, respectively. *P<.05 and **P<.005 vs control. The number of experiments is indicated in parentheses.

We also obtained total RNA from cardiac fibroblasts, which had been cultured in conditions identical to those for myocytes, and performed Northern blot analyses in parallel with myocytes. Signals obtained from the cardiac fibroblast culture were not stronger than those observed in the myocyte culture (data not shown), which contains a small population of fibroblasts (<10% of total cells), suggesting that the mRNA signals observed in the myocyte blots originated mostly, if not exclusively, from myocytes.

Ang II and FCS Differentially Regulate Protein Expression of G₁-S Cyclins
Expression of some cyclins is known to be regulated by a posttranslational mechanism. Therefore, mRNA levels of cyclins may not necessarily correspond to those of the proteins. Accordingly, we next examined protein expression of G₁-S cyclins. To determine the existence of cyclin proteins, we immunoprecipitated cyclins using specific antibodies, except for cyclin D₂, where an antibody that does not cross-react with other cyclins was not available. The immunoprecipitates were immunoblotted with the same antibody after SDS-PAGE. As shown in Fig 4, a band was observed at 34 kD, and a weaker band was observed at 36 kD in the cyclin D₁ immunoblot. For the cyclin D₃ blot, a band was observed at 34 kD. In the cyclin E immunoblot, a doublet was observed at 50/55 kD. For the cyclin A blot, a band was observed at 58 kD, just above the IgG heavy chain band. These bands were not observed when immunoprecipitation was performed without primary antibody, with nonimmune serum, or with the antibody preabsorbed with 10-fold excess antigen peptide. The molecular masses of these bands are similar to those reported in other cell systems (cyclin D₁, 34 to 36 kD²⁵ ; cyclin D₃, 34 kD²⁹ ; cyclin E, 50 and 55 kD³⁰ ; and cyclin A, 58 kD³¹ ), suggesting that they represent specific cyclin isoforms.

View larger version (43K): [in this window] [in a new window]   Figure 4. Effects of Ang II and FCS on protein expression of G1-S cyclins. Cardiac myocytes were treated with or without Ang II (100 nmol/L) or FCS (20%) for the indicated times. Cyclins were immunoprecipitated with specific antibodies. In some experiments, primary antibody was omitted in the immunoprecipitation, or nonimmune serum or antibody preabsorbed with antigen peptide was used as a primary antibody (indicated by -). The immunoprecipitates were immunoblotted with the same antibody. A, Representative immunoblot analyses. Each cyclin is indicated by an arrow. Note that cyclin D1 has been shown to migrate as doublets during mid-G1 to early S phase in colony stimulating factor-1–stimulated macrophages, probably because of posttranslational modifications.25 Cyclin E also has been shown to undergo cell cycle–dependent changes in electrophoretic mobility, and a 50-kD form has been suggested to correlate with cyclin E–associated cdk activity.28 B, Densitometric analysis of levels of protein expression. For cyclin E, densitometric analysis of 50 kD is shown. Values were normalized to those at time zero. Results were further normalized with total protein content of the cell lysates. The results shown are representative of three to six experiments for each cyclin. The levels of cyclin expression obtained from multiple experiments are as follows (mean±SEM): cyclin D1 (34 kD), Ang II at 9 hours, 1.23±0.09 (P<.05 vs control, n=4), Ang II at 24 hours, 0.71±0.02 (P<.001 vs control, n=4), FCS at 24 hours, 4.01±0.88 (P<.05 vs control, n=4); cyclin D3, Ang II at 24 hours, 0.50±0.07 (P<.005 vs control, n=5), FCS at 24 hours, 1.16±0.14 (n=6); cyclin E (50 kD), Ang II at 24 hours, 2.09±0.25 (P<.05 vs control, n=4), FCS at 24 hours, 2.39±0.54 (P=.08 vs control, n=4); and cyclin A, Ang II at 4 hours, 3.52±0.25 (P<.05 vs control, n=3), FCS at 24 hours, 4.36±1.10 (P<.05 vs control, n=6).

Ang II treatment did not largely affect protein expression of cyclin D₁, whereas FCS treatment upregulated it by 8-fold, peaking at 9 hours (Fig 4; see figure legend for statistics). Ang II downregulated protein expression of cyclin D₃ by 50% over 24 hours (Fig 4). Although FCS increased cyclin D₃ protein by 2-fold per cell over 30 hours (Fig 4A), upregulation was not observed when the signal was normalized by total protein content in the cell lysates (Fig 4B). Both Ang II and FCS upregulated the expression of cyclin E (a 50-kD form) protein by 2- to 2.5-fold over 18 to 24 hours (Fig 4). Ang II transiently upregulated cyclin A protein by 3.5-fold, peaking at 4 hours, and the level of expression returned to the control level at 12 to 18 hours. FCS upregulated the expression of cyclin A protein, peaking at 18 hours. Thus, protein expression of G₁-S cyclins seems to be regulated by Ang II and FCS almost in parallel with mRNA expression, except for cyclin E. Expression of cyclin E protein showed upregulation by both Ang II and FCS despite the fact that the mRNA level did not change significantly with these agents, suggesting that posttranslational mechanisms may regulate cyclin E protein expression.

Existence of Cyclin Proteins in Cardiac Myocytes
To confirm that cardiac myocytes express cyclins, immunofluorescent cell staining was performed. Fig 5 shows the expression of cyclin D₁ in primary cardiac myocytes and fibroblasts. In this experiment, we could not perform double staining with anti–sarcomeric myosin antibody because the anti–cyclin D₁ and anti-myosin antibodies available were both mouse antibodies. In unstimulated cells, a diffuse cytoplasmic staining slightly higher than background level was observed. After 6 hours of stimulation with Ang II or FCS, bright nuclear staining of cyclin D₁, along with weaker cytoplasmic staining compared with unstimulated cells, was observed in cells having dark granular cytoplasm or two nuclei, most likely cardiac myocytes, and in cells having lighter cytoplasm, most likely cardiac fibroblasts (Figs 5C and 5I). The pattern of cyclin D₁ staining reverted to the control state after 24 hours of stimulation with Ang II (Fig 5C), whereas nuclear staining was still observed after 24 hours of stimulation with FCS (Fig 5K). The nuclear staining and diffuse cytoplasmic staining were not observed when control mouse IgG was used as a primary antibody (not shown). It is not clear at present why nuclear translocation of cyclin D₁ was observed after Ang II stimulation, when upregulation of cyclin D₁ protein expression is less dramatic. Subcellular localization of cyclin D₁ may be regulated by a mechanism similar to that for cdks, which translocate from the cytosol to the nucleus in late G₁ and S phases without de novo synthesis of cdks' protein.³² Alternatively, the cell lysis condition we used might have failed to completely solubilize nuclear cyclin D₁, and immunoprecipitation and subsequent immunoblotting underestimated the protein expression of cyclin D₁ (Fig 4).

View larger version (273K): [in this window] [in a new window]   Figure 5. Immunofluorescence staining of cardiac myocytes and nonmyocytes with anti–cyclin D1 antibody. Cardiac myocytes were treated with Ang II (100 nmol/L) or FCS (20%) for the indicated times and then subjected to immunostaining with anti–cyclin D1 antibody (72-13G). FITC-conjugated anti-mouse IgG antibody was used as a secondary antibody. A, B, G, and H, Control cells. C, D, E, and F, Ang II–treated cells. I, J, K, and L, FCS-treated cells; A, C, E, G, I, and K, Cyclin D1 staining. B, D, F, H, J, and L, Phase-contrast micrographs of the corresponding fields. Cardiac myocytes have a darker and granular cytoplasm (arrows) compared with cardiac fibroblasts (arrowheads). Some myocytes are binuclear. Bar=10 µm.

Fig 6 shows double immunostaining of cardiac myocytes with anti–cyclin E and anti–sarcomeric myosin heavy chain antibodies. In unstimulated myocytes stained with anti–cyclin E antibody, a filamentous staining and faint perinuclear staining were observed (Fig 6A). After 18 hours of stimulation with Ang II, bright nuclear staining of cyclin E was observed in some myosin-positive cells (cardiac myocytes) (Fig 6B) and myosin-negative cells (fibroblasts, not shown). Similar results were obtained after stimulation with FCS for 18 hours (not shown). Neither nuclear nor filamentous cytoplasmic staining was observed when nonimmune control rabbit serum was used as a primary antibody (not shown).

View larger version (129K): [in this window] [in a new window]   Figure 6. Immunofluorescence staining of cardiac myocytes and nonmyocytes with anti–cyclin E and anti-myosin antibodies. Cardiac myocytes were treated with or without Ang II (100 nmol/L) for 18 hours and then subjected to double immunostaining with anti–cyclin E (M-20) and anti-myosin (MF20) antibodies. FITC-conjugated anti-rabbit IgG and Texas red–conjugated anti-mouse IgG antibodies were used as secondary antibodies. A, C, and E, Control cells. B, D, and F, Ang II–treated cells. A and B, Cyclin E staining. C and D, Phase-contrast micrographs of the corresponding fields. E and F, Myosin staining. The arrows indicate nuclear staining of cyclin E. Bar=10 µm.

mRNA and Protein Expression of cdk in Cardiac Myocytes
We next examined the expression of cdks in cardiac myocytes. Fig 7 (panels A and B) shows the Northern blot analysis of cdc2 (also termed as cdk1) in cardiac myocytes. Ang II increased mRNA expression of cdc2 by 70% during the initial 6 hours but downregulated it thereafter, whereas FCS gradually upregulated it by 2.5-fold, peaking at 16 to 20 hours. The level of cdc2 expression did not change over 24 to 36 hours in unstimulated myocytes (Fig 7A and 7B). We obtained only a very faint band for cdk2 mRNA with the probe we used (data not shown). Appropriate probes to perform the Northern blot analyses for cdk4 and cdk6 mRNA were not available to us.

View larger version (88K): [in this window] [in a new window]   Figure 7. Effects of Ang II and FCS on mRNA and protein expression of cdks. Cardiac myocytes were treated with or without Ang II (100 nmol/L) or FCS (20%) for the indicated times. A and B, Total RNA was isolated, and Northern blot analyses were performed. Total RNA (15 µg) was loaded for each lane. Panel A shows representative Northern blots. Similar results were obtained from two other experiments. Panel B shows densitometric analyses of the levels of mRNA expression. Values were normalized to those at time zero. C and D, Cell lysates were subjected to immunoprecipitation with specific antibody against cdks. In some immunoprecipitations, nonimmune serum or antibody preabsorbed with antigen peptide was used as a primary antibody (indicated by - in panel C). The immunoprecipitates were immunoblotted with the same antibody. In panel C, representative immunoblot analyses are shown. Each cdk is indicated by an arrow. In panel D, densitometric analyses of the level of protein expression are shown. Note that cdk4 migrated as two bands. Two forms of cdk4 have been shown in exponentially growing PC12 cell cultures.33 Because these two forms may represent different isoforms, we analyzed their time courses separately. In panel D, densitometric analysis of the 34-kD form is shown. Expression of the 36-kD form of cdk4 was relatively stable after Ang II or FCS stimulation (data not shown). Values were normalized to those at time zero. Results were further normalized to total protein content of the cell lysates. The results shown are representative of three experiments for each cdk.

We next examined protein expression of cdks. cdks, including cdc2 (cdk1), cdk2, cdk4, and cdk6, were immunoprecipitated with specific antibodies, and the immunoprecipitates were immunoblotted with the same antibody after SDS-PAGE. As shown in Fig 7C, bands with the expected molecular size (33 kD for cdk2, 34 and 36 kD for cdk4, 40 kD for cdk6, and 34 kD for cdc2) were detected for each cdk examined.^{29 32 34} These bands were not observed when immunoprecipitation was performed with nonimmune serum or the antibody preabsorbed with 10-fold excess antigen peptide, suggesting that they correspond to the respective cdks.

Treatment of myocytes with Ang II increased cdk4 (a 34-kD form) and cdk6 proteins by 3.1- and 2.2-fold at 18 hours, respectively, whereas it did not significantly change protein expression of cdk2, cdk4 (a 36-kD form), or cdc2 over 36 hours (Fig 7C and 7D). On the other hand, treatment with FCS upregulated the protein expression of cdk4 (a 34-kD form), cdk2, and cdc2 by 7-, 2.2-, and 2.3-fold, respectively, whereas it did not affect cdk4 (a 36-kD form) and cdk6 protein expression (Fig 7C and 7D).

To estimate the amounts of cyclin and cdk proteins from nonmyocytes present in myocyte culture, cell lysates were prepared from subconfluent nonmyocyte cultures treated in a manner identical to that used for myocytes, except that BrdU was not added until the fibroblasts became subconfluent. The cell lysates containing equal amounts of total protein obtained from myocytes and nonmyocytes were subjected to immunoprecipitation and immunoblotting side by side. Table 1 summarizes the relative levels of expression of cyclins and cdks. After FCS stimulation, myocyte cultures had either comparable (cyclin D₁) or higher levels of expression (cyclin D₃, cyclin A, and cdk2) than nonmyocyte cultures. This suggests that signals from contaminating nonmyocytes cannot account for the observed increase in the expression of cyclins and cdks in myocyte culture.

View this table: [in this window] [in a new window]   Table 1. Comparison of Cyclins and cdk2 Protein Expression in Cardiac Myocytes and Nonmyocytes

To further confirm the expression of cdks in cardiac myocytes, immunofluorescent cell staining was performed. Fig 8 shows double staining with anti-cdk2 and anti–sarcomeric myosin antibodies. Bright nuclear staining of cdk2 was observed in both myosin-positive and -negative cells stimulated with FCS for 24 hours, indicating that cardiac myocytes express cdk2 protein.

View larger version (55K): [in this window] [in a new window]   Figure 8. Immunofluorescence staining of cardiac myocytes and nonmyocytes with anti-cdk2 and anti-myosin antibodies. Cardiac myocytes were treated with FCS for 24 hours and then subjected to double immunostaining with anti-cdk2 (A) and anti-myosin antibodies (MF20) (C). A phase-contrast micrograph of the corresponding field is shown (B). Each arrow indicates a cardiac myocyte. Bar=10 µm.

Fig 9 shows double staining with anti-cdk4 and anti–sarcomeric myosin antibodies. In control myosin-positive cells, staining of cdk4 was observed predominantly in the perinuclear region (Fig 9A, small arrows). After 8 hours of stimulation with Ang II, nuclear staining of cdk4 was observed in some myosin-positive cells (Fig 9D, large arrow). Strong nuclear staining of cdk4 was observed in myosin-positive cells 18 hours after stimulation with FCS (Fig 9G, large arrow). Neither nuclear nor perinuclear staining was observed when nonimmune control rabbit serum was used as a primary antibody (not shown).

View larger version (134K): [in this window] [in a new window]   Figure 9. Immunofluorescence staining of cardiac myocytes and nonmyocytes with anti-cdk4 and anti-myosin antibodies. Cardiac myocytes were treated with Ang II (100 nmol/L) for 8 hours or FCS (20%) for 18 hours and then subjected to double immunostaining with anti-cdk4 antibody and anti-myosin antibody (MF20). A through C, Control cells. D through F, Ang II–treated cells. G through I, FCS-treated cells. A, D, and G, Staining with anti-cdk4. B, E, and H, Phase-contrast micrographs of the corresponding fields. C, F, and I, Staining with anti-myosin. The small arrows in panel A indicate perinuclear staining of cdk4; the large arrows in panels D and G indicate nuclear staining of cdk4. Bar=10 µm.

Ang II and FCS Activate cdk4 With a Different Time Course
We next examined whether cdks are activated by Ang II and FCS. Kinase activity of cdks was determined by the immune complex kinase assays. We first determined the activity of the cdks that are activated in the G₁ phase, using pRb as a substrate (Fig 10A). Treatment of myocytes with Ang II did not activate but rather decreased cdk2 activity toward pRb over 24 hours. cdk4 was activated by Ang II treatment by 2.2-fold, with a peak at 6 to 12 hours. cdk6 was not significantly activated by Ang II treatment (Fig 10B; see figure legend for statistics). FCS activated cdk2 and cdk4 activity toward pRb by 7- and 2.2-fold, respectively, at 24 hours. FCS did not significantly activate cdk6 (Fig 10A and 10B). Almost comparable levels of cdk activation were observed in the absence of BrdU (data not shown), suggesting that the presence of BrdU in culture media did not affect Rb kinase activity of cdks in cardiac myocytes.

View larger version (93K): [in this window] [in a new window]   Figure 10. Effects of Ang II and FCS on kinase activity toward pRb. Myocytes were stimulated with Ang II (100 nmol/L) or FCS (20%) for the indicated times. Cell lysates were subjected to immunoprecipitation using antibodies against the cdks or cyclins, indicated on the left, or nonimmune serum (indicated by -), followed by kinase assay using GST-pRb (769-921) as the substrate. A, Representative autoradiograms of phosphorylated pRb. For Rb kinase assay of cyclin D1, pRb phosphorylation by anti-cdk2 immune complex performed in parallel is shown as a positive control (P). B, Densitometric analyses of pRb kinase activity shown in panel A. The results are normalized to total protein content of the cell lysates and are relative to time zero. The levels of pRb kinase activity obtained from multiple experiments are as follows (mean±SEM): cdk2, Ang II at 24 hours, 0.18±0.21 (P=.06 vs control, n=3), FCS at 18 hours, 6.96±1.88 (P=.08 vs control, n=3); cdk4, Ang II at 6 hours, 2.24±0.26 (P<.05 vs control, n=4), FCS at 24 hours, 2.19±0.27 (P<.05 vs control, n=3); cdk6 Ang II at 24 hours, 0.57±0.15 (n=3), FCS at 18 hours, 0.43±0.16 (P=.07 vs control, n=3); cyclin D3, Ang II at 24 hours, 0.68±0.15 (n=3), FCS at 24 hours, 2.11±0.54 (n=3); cyclin E, Ang II at 24 hours, 1.82±0.10 (P<.05 vs control, n=3), FCS at 24 hours, 3.06±0.52 (P=.06 vs control, n=3); and cyclin A, Ang II at 6 hours, 1.61±0.14 (P<.05 vs control, n=3), FCS at 24 hours, 2.24±0.35 (P=.07 vs control, n=3).

Activation of the cdks in part depends on binding to their regulatory subunits, the cyclins. To determine the role of cyclins in the activation of cdks, cyclins were immunoprecipitated and the cdks' activity that associates with each cyclin was determined (Fig 10). In Ang II–treated myocytes, pRb kinase activity in the cyclin D₃ immune complex did not significantly change over 24 hours. pRb kinase activity in cyclin A slightly increased by 1.6-fold after the Ang II treatment, with a peak at 6 hours. pRb kinase activity in the cyclin E complex increased by 1.8-fold after 24 hours of treatment with Ang II. Interestingly, we did not detect any pRb kinase activity in the cyclin D₁ immune complex obtained from Ang II–stimulated or FCS-stimulated myocytes using various cell lysis conditions (see "Discussion").

In serum-stimulated myocytes, pRb kinase activity of cyclin D₃, cyclin E, and cyclin A immunoprecipitates all tended to be upregulated >2-fold over 24 hours (Fig 10), suggesting that the association of these cyclins with cdks may play a role in the FCS-induced increase in pRb kinase activity of cdk4 and cdk2.

Ang II Did Not Activate cdk Activity Toward Histone H₁
We next determined the activities of cdk2 and cdc2, which are activated in S and G₂/M phases, by the immune complex kinase assay using histone H₁ as a substrate. Ang II treatment did not significantly increase the activity of cdk2 toward histone H₁ (lanes 3 to 8, Fig 11A) compared with unstimulated myocytes (lanes 1 and 2 in Fig 11A). Little or no cdc2 kinase activity was detected in control or Ang II–treated myocytes (lanes 3 to 8, Fig 11B), confirming our previous observation.¹³ Ang II stimulation failed to activate histone H₁ kinase activity of cdk2 and cdc2 significantly, even in the absence of BrdU (data not shown). On the other hand, FCS significantly activated histone H₁ kinase activity of cdk2 and cdc2 over 24 to 36 hours (lanes 9 to 14 in Fig 11).

View larger version (44K): [in this window] [in a new window]   Figure 11. Effect of Ang II and FCS on cdk2 (A) and cdc2 (B) activity toward histone H1. Myocytes were stimulated with Ang II (100 nmol/L) or FCS (20%) for the indicated times. Cell lysates were subjected to immunoprecipitation (IP) using antibodies against the cdks, indicated on the left, or nonimmune serum indicated by -, followed by kinase assay using histone H1 as the substrate. Representative autoradiograms of phosphorylated histone H1 are shown. Similar results were obtained from two additional experiments. cont indicates control.

Effects of Ang II and FCS on pRb Phosphorylation
pRb is thought to represent an important substrate of the G₁-cyclin cdks, particularly the cyclin D (cdk4) complex and the cyclin E (cdk2) complex.²⁰ Phosphorylation of pRb suppresses the interaction between pRb and the transcription factor complex DP/E2F, which regulates expression of the genes required for S phase. This releases DP/E2F from an inhibitory constraint by pRb.²¹ Therefore, we examined the effects of Ang II and FCS on the phosphorylation status of pRb in vivo in cardiac myocytes. pRb was immunoprecipitated with a specific antibody, which recognizes only pRb but not other members of the pocket protein family, such as p107 and p130, and the immunoprecipitates were immunoblotted with the same antibody after SDS-PAGE. pRb was detected as multiple bands 110 kD, which were not observed when immunoprecipitation was performed without anti-pRb antibody (Fig 12 A). Expression of pRb was observed in both cardiac myocytes and fibroblasts, and the levels of expression in both cell types are almost comparable (Table 1). Although Ang II did not significantly affect the overall levels of pRb over 24 hours, FCS caused a 2-fold increase in total pRb protein (Fig 12). Phosphorylated pRb is known to migrate slower in SDS-PAGE gel than unphosphorylated pRb.²¹ Only the faster migrating form of pRb was detected in the control state, whereas Ang II caused a band shift of pRb (25% of total Rb), peaking at 12 hours. FCS caused a more sustained and intense band shift of pRb (50% of total pRb) over 24 hours. The time course of the band shift of pRb by Ang II was similar to that of cdk4 (Fig 10B), a kinase known to phosphorylate pRb in vivo. A similar band shift of pRb following Ang II and FCS stimulation was also observed in cardiac fibroblasts (data not shown).

View larger version (48K): [in this window] [in a new window]   Figure 12. Effects of Ang II and FCS on phosphorylation status of pRb. Myocytes were stimulated with Ang II (100 nmol/L) or FCS (20%) for the indicated times. Cell lysates were subjected to immunoprecipitation (IP) using anti-pRb antibody. The immunoprecipitates were subjected to immunoblotting with the same antibody after SDS-PAGE on a 7% gel. A, A representative immunoblot is shown. In lane 12, anti-pRb antibody was not included in the IP step. Note that phosphorylated forms of pRb migrate slower on SDS-polyacrylamide gel. B, Densitometric analyses of the pRb bands shown in panel A. Intensity of the band is indicated by arbitrary unit after normalization to the total protein content of each cell lysate (left vertical axis). Open bars indicate unphosphorylated pRb (lower band); solid bars indicate phosphorylated form of pRb (ppRb, upper band). Note the sum of the open and solid columns indicates total amounts of pRb. ppRb/ppRb+pRb (%) is shown by the line graph (right vertical axis). Similar results were obtained from two additional experiments.

Effects of Ang II and FCS on DNA Synthesis in Neonatal Cardiac Myocytes
We finally examined whether stimulation of cardiac myocytes with Ang II or FCS causes different effects on DNA synthesis in cardiac myocytes in our experimental conditions. For this experiment, BrdU was initially omitted from the plating medium. Seventy-two hours after plating, myocytes were then incubated with BrdU (100 µmol/L) for 48 hours in the presence of Ang II or FCS to maximize the chance to detect nuclear accumulation of BrdU. Myocytes were subsequently subjected to double immunostaining with anti-BrdU antibody and anti–sarcomeric myosin antibody. When Ang II was added 48 hours after myocyte dissociation (72 hours after birth), it did not increase the number of BrdU-positive myocytes. FCS slightly (from 1.8% to 6.2%) increased BrdU-positive cells (Table 2). Using the same experimental protocol, however, 99% of cardiac primary fibroblasts were BrdU positive after FCS stimulation. When Ang II or FCS was added 72 hours after myocyte dissociation (96 hours after birth), BrdU-positive cardiac myocytes were not observed (Table 2). This suggests that cardiac myocytes are already terminally differentiated at this stage in our culture conditions. Similar results were obtained when cells were incubated with BrdU for 24 instead of 48 hours (not shown).

View this table: [in this window] [in a new window]   Table 2. BrdU Incorporation Into Cardiac Myocytes

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

 
We have shown that Ang II and serum differentially regulate the expression of cell cycle regulators, the activity of cdks, and the phosphorylation status of pRb in cultured cardiac myocytes. Ang II transiently activates cdk4 but does not activate cdk2 or cdc2. Serum, on the other hand, activates all of these cdks in neonatal cardiac myocytes. Differential regulation of cell cycle machinery may cause differential gene expression and thus account for the specificity of cell signals.

The cellular signaling mechanisms causing differential regulation of cyclin and cdk expression after Ang II and FCS stimulation remain to be elucidated. Recent evidence suggests that Ras positively regulates the expression of cyclin D₁ in NIH 3T3 cells.³⁵ The expression of cyclin E, cyclin A, and cdc2 is positively regulated by E2F.^{36 37} On the other hand, expression of cyclin A is attenuated by TGF-ß1 in Chinese hamster lung fibroblasts.³⁸ Expression of cyclins E and A is inhibited by PKC in human vascular smooth muscle cells.³⁹ Because Ang II activates PKC and Ras^{10 12} and upregulates the expression of TGF-ß1 in cardiac myocytes,²² it will be interesting to study the role of PKC, Ras, and TGF-ß1 in the regulation of the cell cycle machinery by Ang II.

The functional consequence of differential regulation of cdks by Ang II and FCS is not clear at present. However, considering the diverse functions of the cdks as well as those of pRb,¹⁸ several important questions emerge from the present study. First, what is the functional significance of transient activation of cdk4 and phosphorylation of pRb by the hypertrophic stimulus? Phosphorylation of pRb will release its binding as well as its inhibition of the transcription factor DP/E2F heterodimers. It is possible that genes regulated by DP/E2F may play an important role in hypertrophic responses. It is also possible that cdk4 may phosphorylate other important substrates that promote hypertrophic responses. Second, is active downregulation of cyclin and cdk2 activity essential for hypertrophy? It is known that cyclin D₁ prevents activation of muscle-specific gene transcription by the myogenic basic helix-loop-helix regulator MyoD in differentiating skeletal myoblasts.^{40 41 42 43 44} Although a basic helix-loop-helix regulator functionally equivalent to MyoD has not been found in cardiac myocytes, active downregulation of the cyclin-cdk2 complex by Ang II may create positive signals by stimulating the transcription of genes that play an important role in cardiac hypertrophy. Recently, adenovirus E1A, infecting neonatal cardiac myocytes, has been shown to repress cardiac gene transcription, probably through an interaction with the "pocket" proteins and release of E2F, suggesting that E2F may negatively regulate muscle-specific gene expression, such as expression of skeletal -actin.^{5 43} Analogous to the increase in the cell size of budding yeast caused by Ras/cAMP, which inhibits CLN1 and CLN2,⁴⁴ active downregulation of G₁ cyclins and cdks might be an important process in hypertrophy of the mammalian cell.⁴⁵

In these experiments, we used myocyte-rich primary culture, which contains nonmyocytes. This may raise the possibility that some of the results of cyclin and cdk expression may result from the nonmyocyte fraction. However, this is very unlikely for the following reasons: First, parallel experiments using fibroblast-rich cultures showed that signals of mRNA and protein obtained from nonmyocytes were comparable or, at most, only twofold greater than those from myocyte cultures (Table 1 and data not shown). If the noncardiac myocyte fraction is the sole source of the mRNA and protein signals in the myocyte culture, the contaminating nonmyocytes (<10% of the total cell population) in myocyte culture would not account for the levels of mRNA and protein signals in cardiac myocyte culture. Second, we did not observe a significant increase in cell number even after 48 hours of stimulation with FCS in the presence of BrdU. Thus, cardiac myocytes remained the predominant cell type throughout the experiment. Third, we performed cellular staining for several cyclins and cdks and found their signals in myosin-positive myocytes. Furthermore, nuclear staining of cyclins D₁, E, and cdk4 was observed in cardiac myocytes after stimulation with Ang II and FCS, suggesting that cdks are activated in cardiac myocytes.⁴⁶ It is therefore likely that most, if not all, of the results of biochemical analyses in the myocyte culture originated from cardiac myocytes.

Our results indicate that neonatal cardiac myocytes express mRNA of cyclins D₁, D₂, D₃, E, and A and cdc2. We also detected protein expression of these cyclins and cdks, including cdk2, cdk4, cdk6, and cdc2. Yoshizumi et al³¹ recently reported that mRNA expression of all cyclins could be observed in rat fetal and neonatal heart in vivo. Interestingly, they found that expression of cyclin A decreased quickly after birth and that downregulation of cyclin A temporally correlated with the known time course of the permanent withdrawal of cardiac myocytes from the cell cycle. We found that although FCS strongly stimulates mRNA and protein expression of cyclin A and cdk activity in the cyclin A immune complex, neonatal myocytes do not enter S phase after 3 days in culture. Thus, our results provide evidence of temporal dissociation between cyclin A downregulation and terminal differentiation in neonatal cardiac myocytes.

Although FCS increased cdk4, cdk2, and cdc2 kinase activity in neonatal cardiac myocytes, it failed to stimulate DNA synthesis if cardiac myocytes had been cultured for >72 hours after cell preparation (>96 hours after birth). Thus, activation of cdks and DNA synthesis are dissociated in neonatal cardiac myocytes. Although the reason for this dissociation is unclear, several possibilities can be considered. First, activation of cdks by FCS may not be strong enough to drive the cells into S phase. Second, terminally differentiated myocytes might have lost critical mechanisms (molecules) that act downstream from activated cdks. We have previously shown that in the skeletal myogenic cell line C2C12, where proliferation and differentiation are mutually exclusive, restimulation of terminally differentiated myotubes with serum activates cdk2 despite the fact that they never reenter S phase.²⁴ It is therefore likely that the activity of cdks does not necessarily correlate with DNA synthesis in terminally differentiated cells.

Although cardiac myocytes express both cyclin D₁ and cdk4 proteins and FCS significantly increases protein expression of both cyclin D₁ and cdk4, we did not detect pRb kinase activity in the cyclin D₁ immune complex. Interestingly, immunoblotting of the cyclin D₁ immunoprecipitate with anti-cdk4 antibody indicated that cdk4 was associated with cyclin D₁ (data not shown). Thus, the anti–cyclin D₁ antibody we used does not interfere with the physical association between cyclin D₁ and cdk4. The same anti–cyclin D₁ antibody has been widely used to immunoprecipitate the cyclin D₁–cdk4 complex without affecting its pRb kinase activity in other cell types.⁴⁷ We did not detect any kinase activity even with the mildest detergent conditions, such as 0.1% Tween 20, which has been reported to preserve the catalytic activity of cdk4.⁴⁷ Using the same lysis buffer for cyclin D₁, we easily detected pRb kinase activity in the immune complex of cyclins D₃, E, and A and cdk4. Therefore, it is unlikely that failure of detection of pRb kinase activity in the cyclin D₁ immune complex is due to technical problems. It is rather likely that the cyclin D₁–cdk4 complex is catalytically inactive in cardiac myocytes. cdk4 requires modification to be fully activated even after complex formation with cyclins.^{48 49} Because phosphorylation of threonine 172 by cdk-activating kinase and possibly dephosphorylation of tyrosine 17 by cdc25A positively regulate cdk4,^{48 49 50} such positive modification may not occur in myocytes. Alternatively, activity of the cyclin D₁–cdk4 complex might be inhibited by cdk inhibitors.¹⁹ A similar finding has been reported in skeletal myotubes.⁵¹ Although high levels of cyclin D₃ expression and its association with cdks were detected in differentiated L6 cells, cyclin D₃–cdk complexes were reported to have no kinase activity.⁵¹ Therefore, it is intriguing to hypothesize the existence of strong cyclin D–cdk inhibitors in terminally differentiated cells.

Activity of cdks can be inhibited by cdk inhibitors, such as p21^CIP1/WAF1, p27^KIP1, p57^KIP2, and the INK4 family.¹⁹ Because protein expression of INK4 and p27^KIP1 is stimulated by TGF-ß1,¹⁹ it is possible that Ang II upregulates the expression of cdk inhibitor(s) through TGF-ß1, and the cdk inhibitors may in turn suppress cdk2 and cdc2 activation or inactivate cdk4 after its transient activation by Ang II. Alternatively, because Ang II downregulates protein expression of cyclins D₁, D₃, and A and cdc2, failure of activation or downregulation of cdks by Ang II could occur independent of cdk inhibitors. At present, we do not know which cdk inhibitors exist in cardiac myocytes. Further studies are necessary to elucidate the role of cdk inhibitors in terminal differentiation of cardiac myocytes and cardiac hypertrophic responses.

	Selected Abbreviations and Acronyms

 

Ang II = angiotensin II BrdU = bromodeoxyuridine cdk = cyclin-dependent protein kinase ECL = enhanced chemiluminescence GST = glutathione S-transferase HRP = horseradish peroxidase IE gene = immediate-early gene MAP kinase = mitogen-activated protein kinase PKC = protein kinase C pRb = retinoblastoma gene product TGF = transforming growth factor

	Acknowledgments

 
This work was supported in part by an AHA Michigan Grant-in-Aid (J.S.), AHA Michigan Postdoctoral Fellowship (H.A.), and by an NIH grant (S.I.). We thank Dr T.J. Kulik for critical reading of the manuscript.

Received July 26, 1996; accepted October 23, 1996.

	References

Top Abstract Introduction Materials and Methods Results Discussion References

 

1. Morgan HE, Baker KM. Cardiac hypertrophy. Circulation. 1991;83:13-25.[Free Full Text]
2. Zak R. Development and proliferative capacity of cardiac muscle cells. Circ Res. 1974;2(suppl):17-26.
3. Ueno H, Perryman B, Roberts R, Schneider MD. Differentiation of cardiac myocytes after mitogen withdrawal exhibits three sequential states of the ventricular growth response. J Cell Biol.. 1988;107:1911-1918.[Abstract/Free Full Text]
4. Claycomb WC. Control of cardiac muscle cell division. Trends Cardiovasc Med. 1992;2:231-236.
5. Kirshenbaum LA, Schneider MD. The cardiac cell cycle, pocket proteins, and p300. Trends Cardiovasc Med.. 1995;5:230-235.
6. Sadoshima J, Xu Y, Slayter HS, Izumo S. Autocrine release of angiotensin II mediates stretch-induced hypertrophy of cardiac myocytes in vitro. Cell. 1993;75:977-984.[Medline] [Order article via Infotrieve]
7. Shubeita HE, McDonough PM, Harris AN, Knowlton KU, Glembotski CC, Brown JH, Chien KR. Endothelin induction of inositol phospholipid hydrolysis, sarcomere assembly, and cardiac gene expression in ventricular hypertrophy. J Biol Chem.. 1990;265:20555-20562.[Abstract/Free Full Text]
8. Ito H, Hirata Y, Adachi S, Tanaka M, Tsujino M, Koike A, Nogami A, Murumo F, Hiroe M. Endothelin-1 is an autocrine/paracrine factor in the mechanism of angiotensin II-induced hypertrophy in cultured rat cardiomyocytes. J Clin Invest. 1993;92:398-403.
9. Sadoshima J, Izumo S. Mechanical stretch rapidly activates multiple signal transduction pathways in cardiac myocytes: potential involvement of an autocrine/paracrine mechanism. EMBO J.. 1993;12:1681-1692.[Medline] [Order article via Infotrieve]
10. Sadoshima J, Izumo S. Signal transduction pathways of angiotensin II–induced c-fos gene expression in cardiac myocytes in vitro: roles of phospholipid-derived second messengers. Circ Res.. 1993;73:424-438.[Abstract/Free Full Text]
11. Sadoshima J, Qiu Z, Morgan JP, Izumo S. Angiotensin II and other hypertrophic stimuli mediated by G protein–coupled receptors activate tyrosine kinase, mitogen-activated protein kinase, and 90-kD S6 kinase in cardiac myocytes: the critical role of Ca²⁺-dependent signaling. Circ Res.. 1995;76:1-15.[Abstract/Free Full Text]
12. Sadoshima J, Izumo S. The heterotrimeric Gq protein-coupled angiotensin II receptor activates p21^ras via the tyrosine kinase-Shc-Grb2-Sos pathway in cardiac myocytes. EMBO J. 1996;15:775-787.[Medline] [Order article via Infotrieve]
13. Sadoshima J, Izumo S. Rapamycin selectively inhibits angiotensin II–induced increase in protein synthesis in cardiac myocytes in vitro: potential role of 70-kD S6 kinase in angiotensin II–induced cardiac hypertrophy. Circ Res.. 1995;77:1040-1052.[Abstract/Free Full Text]
14. Yamazaki T, Komuro I, Kudoh S, Zou Y, Shiojima I, Mizuno T, Takano H, Hiroi Y, Ueki K, Tobe K, Kadowaki T, Nagai R, Yazaki Y. Mechanical stress activates protein kinase cascade of phosphorylation in neonatal rat cardiac myocytes. J Clin Invest.. 1995;96:438-446.
15. Bogoyevitch MA, Marshall CJ, Sugden PH. Hypertrophic agonists stimulate the activities of the protein kinases c-Raf and A-Raf in cultured ventricular myocytes. J Biol Chem.. 1995;270:26303-26310.[Abstract/Free Full Text]
16. Simpson P. Proto-oncogene and cardiac hypertrophy. Annu Rev Physiol.. 1988;51:189-202.[Medline] [Order article via Infotrieve]
17. Blenis J. Signal transduction via the MAP kinases: proceed at your own RSK. Proc Natl Acad Sci U S A.. 1993;90:5889-5892.[Abstract/Free Full Text]
18. Nigg EA. Cyclin-dependent protein kinases: key regulators of the eukaryotic cell cycle. Bioessays. 1995;17:471-480.[Medline] [Order article via Infotrieve]
19. Sherr CJ, Roberts JM. Inhibitors of mammalian G1 cyclin-dependent kinases. Genes Dev.. 1995;9:1149-1163.[Free Full Text]
20. Kato J, Matsushime H, Hiebert SW, Ewen ME, Sherr CJ. Direct binding of cyclin D to the retinoblastoma gene product (pRb) and pRb phosphorylation by the cyclin D-dependent kinase CDK4. Genes Dev. 1993;7:331-342.[Free Full Text]
21. Nevins JR. E2F: a link between the Rb tumor suppressor protein and viral oncoproteins. Science. 1992;258:424-429.[Abstract/Free Full Text]
22. Sadoshima J, Izumo S. Molecular characterization of angiotensin II–induced hypertrophy of cardiac myocytes and hyperplasia of cardiac fibroblasts: a critical role of the AT₁ receptor subtype. Circ Res.. 1993;73:413-423.[Abstract/Free Full Text]
23. Sadoshima J, Jahn L, Takahashi T, Kulik TJ, Izumo S. Molecular characterization of the stretch-induced adaptation of cultured cardiac cells: an in vitro model of load-induced cardiac hypertrophy. J Biol Chem.. 1992;267:10551-10560.[Abstract/Free Full Text]
24. Jahn L, Sadoshima J, Izumo S. Cyclins and cyclin-dependent kinases are differentially regulated during terminal differentiation of C2C12 muscle cells. Exp Cell Res.. 1994;212:297-307.[Medline]