JBC

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Originally published In Press as doi:10.1074/jbc.M703513200 on October 9, 2007

J. Biol. Chem., Vol. 282, Issue 48, 34984-34993, November 30, 2007
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
282/48/34984    most recent
M703513200v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Li, J.
Right arrow Articles by Boheler, K. R.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Li, J.
Right arrow Articles by Boheler, K. R.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

The Pro-angiogenic Cytokine Pleiotrophin Potentiates Cardiomyocyte Apoptosis through Inhibition of Endogenous AKT/PKB Activity*

Jinliang Li{ddagger}12, Hong Wei{ddagger}1, Alan Chesley§1, Chanil Moon{ddagger}, Melissa Krawczyk{ddagger}, Maria Volkova{ddagger}, Bruce Ziman{ddagger}, Kenneth B. Margulies, Mark Talan{ddagger}, Michael T. Crow§, and Kenneth R. Boheler{ddagger}2

From the {ddagger}Laboratory of Cardiovascular Science, National Institute on Aging, Baltimore, Maryland 21224, the §Department of Medicine, The Johns Hopkins University, Baltimore, Maryland 21224, and the University of Pennsylvania School of Medicine, Philadelphia, Pennsylvania 19104

Received for publication, April 26, 2007 , and in revised form, September 5, 2007.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 
Pleiotrophin is a development-regulated cytokine and growth factor that can promote angiogenesis, cell proliferation, or differentiation, and it has been reported to have neovasculogenic effects in damaged heart. Developmentally, it is prominently expressed in fetal and neonatal hearts, but it is minimally expressed in normal adult heart. Conversely, we show in a rat model of myocardial infarction and in human dilated cardiomyopathy that pleiotrophin is markedly up-regulated. To elucidate the effects of pleiotrophin on cardiac contractile cells, we employed primary cultures of rat neonatal and adult cardiomyocytes. We show that pleiotrophin is released from cardiomyocytes in vitro in response to hypoxia and that the addition of recombinant pleiotrophin promotes caspase-mediated genomic DNA fragmentation in a dose- and time-dependent manner. Functionally, it potentiates the apoptotic response of neonatal cardiomyocytes to hypoxic stress and to ultraviolet irradiation and of adult cardiomyocytes to hypoxia-reoxygenation. Moreover, UV-induced apoptosis in neonatal cardiomyocytes can be partially inhibited by small interfering RNA-mediated knockdown of endogenous pleiotrophin. Mechanistically, pleiotrophin antagonizes IGF-1 associated Ser-473 phosphorylation of AKT/PKB, and it concomitantly decreases both BAD and GSK3β phosphorylation. Adenoviral expression of constitutively active AKT and lithium chloride-mediated inhibition of GSK3β reduce the potentiated programmed cell death elicited by pleiotrophin. These latter data indicate that pleiotrophin potentiates cardiomyocyte cell death, at least partially, through inhibition of AKT signaling. In conclusion, we have uncovered a novel function for pleiotrophin on heart cells following injury. It fosters cardiomyocyte programmed cell death in response to pro-apoptotic stress, which may be critical to myocardial injury repair.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 
Growth factors and cytokines have essential roles throughout embryonic and fetal development. In adults, however, their expression is often limited to a few cell types, but in response to injury, some of these genes become re-activated to meet the demands of repair. Pleiotrophin (PTN),3 also known as osteoblast-specific factor-1, heparin-binding growth-associated molecule, heparin-binding growth factor-8, or neurite growth-promoting factor (1, 2), is one such factor. The ptn gene product, which is identical in rodents and humans, is an 18-kDa secreted heparin-binding cytokine that contains a signal peptide and two lysine-rich clusters (3–5). It was originally described as a development-regulated growth factor expressed in neuroectodermal and mesodermal cell lineages, but it is also expressed at sites of early vascular development and in healing wounds (6, 7). Postnatally, PTN is down-regulated, and in most adult rodent and human tissues, it is present at very low levels. Its reactivation in adults strongly promotes angiogenesis and vascular growth in tumors (4, 8–10), and it has served as a marker of cancer erosion, angiogenesis, and metastasis (11, 12). In ectodermal derivatives, it promotes neurite outgrowth (13) and glial cell differentiation (5). Following acute ischemic injury to brain, it is up-regulated (14), and it enhances neuron survival (15). In fibroblasts and epithelial and endothelial cells, PTN is mitogenic (3, 4), but it prevents apoptosis in serum-starved NIH3T3 cells (16).

The pleiotropic functions of PTN are thought to be regulated primarily through autocrine/paracrine effects mediated through receptor-activated signaling (17–20). Binding to one of its five known receptors (receptor protein-tyrosine phosphatase β/{zeta} (RPTP β/{zeta}), anaplastic lymphoma kinase, syndecan-1, syndecan-3, and syndecan-4) has been shown to activate either the Ras-mitogen-activated protein kinase (MAPK) or the phosphatidylinositol 3-kinase (PI3K)-AKT signaling axes (16, 21, 22). Activation of the MAPK pathway by PTN inhibits programmed cell death in NIH3T3 cells, whereas activation of the PI3K pathway is generally believed to be protective. Both pathways may be required for or involved in the proliferation of NIH3T3 cells (16).

In heart, the functions of PTN have not yet been elucidated. We previously showed that ptn is up-regulated early during the in vitro formation of cardiomyocytes (CMs) from pluripotent stem cells and that it is expressed in developing heart (23). In atrial natriuretic factor-null mice subjected to pressure overloaded, Wang et al. (24) reported that Ptn transcripts are significantly increased. Similarly, we showed that Ptn is increased and remains elevated in human hearts with left ventricle assist device despite a number of phenotypic shifts associated with left ventricle assist device support (25), suggesting that the accumulation of Ptn mRNAs is not directly regulated by hemodynamic overload. PTN does, however, seem to promote cardiac vascularization following ischemia (26) and increase bromodeoxyuridine incorporation into mouse heart cells in vivo (27). These latter results led to the suggestion that injection of recombinant PTN might represent a potential therapeutic agent that could initiate new vessel formation in damaged myocardium, even though its effects on CMs have not been rigorously investigated.

In this study, we have examined the role of PTN on contractile cells isolated from heart. Because PTN is believed to be a secreted growth factor with anti-apoptotic and proliferative activities, we tested the hypothesis that addition of PTN would promote CM growth (hypertrophy) and survival. We demonstrate, however, that PTN is released from CMs in response to cell stress and can potentiate CM apoptosis through inhibition of AKT signaling.


   EXPERIMENTAL PROCEDURES
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
RESULTS
 DISCUSSION
 REFERENCES
 
Rodent and Human Tissue Samples—Hearts were obtained from rats at the ages indicated in the text according to an approved animal study protocol (019-KRB-RaMi) at the National Institute on Aging. Human ventricular myocardial samples were obtained from 18 male patients with severe heart failure secondary to idiopathic dilated cardiomyopathy or ischemic heart disease and from 18 unused male donor hearts (25, 28). The average ages of the two groups were 50.2 ± 16.0 and 53.8 ± 19.1 years. Ten additional human samples, with an average age of 53.5 ± 11.2 years, were also obtained for ELISA from both male and female patients with heart failure.

Rat Myocardial Infarction Model—Forty six male Sprague-Dawley rats (Charles River Laboratories, Inc.), 8 weeks of age (body weight 257 ± 14.2 g), were housed and studied in accordance to the National Institutes of Health Guide for the Care and Use of Laboratory Animals. Rats (Charles River Laboratories, Inc.) were divided into 2 groups, sham and myocardial infarction (MI). Surgical procedures were as described, and the average infarct size was consistent with our previous report (29). Rat hearts were harvested 1 day, 1 week, and 4 weeks post-surgery (7–9 hearts/group/time point).

Primary Cultures of Cardiomyocytes (CMs)—Neonatal ventricular cardiac myocytes (NNCMs) were cultivated as described (30, 31) for 24 h, and switched to serum-free media for 18 h. Adult rat ventricular cardiac myocytes (ACMs) were plated (90–95% purity) and maintained in ACCT medium (32). Neonatal rat heart fibroblasts were isolated after pre-plating and passaged 2 or 3 times prior to experimentation. Cells were exposed to human recombinant PTN (catalog number 450-15; PeproTech, Rocky Hill, NJ) as indicated. To induce apoptosis, NNCMs and ACMs were exposed to hypoxia for 24 and 2 h, respectively. ACMs were then re-exposed to ambient oxygen for 4.5 h. Cells were exposed to hypoxia in a modular incubator chamber (Billups-Rothenberg, Del Mar, CA) by introduction of a mixed gas consisting of 95% N2 and 5% CO2. For irradiation, NNCMs were exposed to UV light after removal of the incubation medium. Cells were placed at a fixed distance under the UV lamps of a UV Stratalinker or similar device, and the exposure times were kept constant for each dose of 50–250 J/m2, depending on the experimental requirements. Media were then reapplied to the cells and incubated under standard cultivation conditions. For experiments involving PTN, hypoxia or irradiation was induced only after a 1-h pre-incubation with recombinant PTN or vehicle.

Immunocytochemistry and Apoptosis—Myocyte and fibroblast cultures were examined by immunostaining (31). Primary antibodies against sarcomeric {alpha}-actinin (Sigma), cardiac myosin heavy chain (MHC) (Abcam, Cambridge, MA), PTN (R & D Systems, Minneapolis, MN), and secondary antibodies conjugated with fluoro-350, -488, and -568 (Invitrogen) were employed. MetaMorph software (Universal Imaging Corp.) was used to determine cell surface area by two-dimensional planimetry (n = 4). Areas were normalized to the number of nuclei in sarcomeric {alpha}-actinin-positive cells. As a control for cardiac hypertrophy, cultured NNCMs were stimulated with 10 µM phenylephrine.

Transverse sections of paraffin-embedded left ventricles from MI- and sham-operated animals were examined by immunostaining (see above). Terminal deoxynucleotidyltransferase-mediated dUTP nick end labeling (TUNEL) was employed using an in situ cell death detection kit with TMR Red (Roche Applied Science). TUNEL-positive nuclei and total nuclei were counted from peri-infarct and noninfarct regions. Data are expressed as a percentage of TUNEL-positive nuclei to the total number of nuclei evaluated per section.

Nuclear fragmentation was detected in live cells after addition of Hoechst 33342, whereas necrotic cells were excluded by co-staining with propidium iodide. For DNA laddering, genomic DNA was isolated, and laddering was detected following size fractionation on agarose gels. All measurements, in triplicate, were normalized to total cellular protein (31).

Real Time PCR, Western Blotting, and ELISA—Total RNA was reverse-transcribed, and PCRs were run in the presence of SYBR green (28). The threshold cycle (Ct) of each sample was calculated, and normalized data are presented as a {Delta}Ct (Ctreference gene - Ct) value, where Gapd was used as a reference. Primer sequences for real time quantitative PCR are shown in Table 1. Western blotting was performed with antibodies to PTN, phospho-AKT (Ser-473), phospho-GSK3β (Ser-9), and phospho-BAD (Ser-136) obtained from R & D Systems (PTN, Minneapolis, MN), Cell Signaling Technology (AKT and BAD, Beverly, MA), and Santa Cruz Biotechnology (actin and GSK3). Protein signals were normalized to total protein loading following staining with Amido Black (Invitrogen), and phosphorylated proteins were also normalized to sarcomeric {alpha}-actin. The presence of PTN in the concentrated cell culture media (supernatants) was detected by Western blotting, following sample concentration using an Amicon Ultracel-5k centrifugal device (Millipore, Billerica, MA). For ELISA, tissue samples were incubated with capture antibody (catalog number AF-252-PB, R& D Systems., Minneapolis, MN). PTN was detected with a biotinylated pleiotrophin antibody (R & D Systems) and streptavidin-conjugated peroxidase (Invitrogen), followed by reaction with tetramethylbenzidine-H2O2 (Pierce). Absorbance was measured at 450 nm.


View this table:
[in this window]
[in a new window]

  		TABLE 1
Primers employed for real time quantitative PCR

 
Adenovirus and Adenoviral Infection—A recombinant replication-defective adenovirus encoding a hemagglutinin (HA) epitope-tagged constitutively activated AKT-1 and a control virus expressing no gene product (Ad/CMV/DEST-null) were constructed by recombination cloning from a plasmid encoding myr-AKT1-HA (gift of T. Francke, Columbia University) into pAd/CMV/V5-DEST vector (Invitrogen) to generate pAd/CMV/DEST-myrAkt1-HA. The viruses were commercially amplified, purified, and titered (Vector Laboratories). CMs were infected with virus in serum-free media, and infectivity of pAd.CMV/DEST-myrAkt1-HA was determined by immunostaining with an antibody against the HA tag (Santa Cruz Biotechnology).

RNA Interference of Ptn Expression—Post-transcriptional gene silencing was performed using chemically synthesized duplex RNA oligonucleotides against rat Ptn (catalog number LQ-092194-01, Dharmacon, Lafayette, CO). Myocytes were placed in Opti-MEM (Invitrogen) to which 100 nM siRNA was added along with the GeneSilencer reagent (GenLantis Corp., San Diego). Four hours later, cultivation media were added. Four siRNAs targeting Ptn transcripts were tested, but only two (siRNA2 and -4) were effective in pilot experiments.

Statistical Analysis—Data are expressed as means ± S.D. A Student's t test, Mann-Whitney test, or analysis of variance with repeated independent measurement was employed to determine statistical significance. p values of less than 0.05 were considered to be significant. Correlation coefficients were determined using functions in Microsoft Excel.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
DISCUSSION
 REFERENCES
 
Pleiotrophin in Human and Rat Hearts—In heart, pleiotrophin gene (ptn) expression is dynamically and inversely regulated during development and in injury. During rodent development, pleiotrophin transcripts (Ptn) decrease by >36-fold (p < 0.05) between fetal (16–18 dpc) and adult stages (2–3 months of age) of development (Fig. 1A), and significant decreases in PTN proteins were observed in neonatal versus adult heart (ELISA, 0.89 ± 0.07 µg/g neonatal versus 0.33 ± 0.07 µg/g adult heart tissue; Fig. 1B). In nonfailing adult human myocardium, the low levels of Ptn decrease as a function of age; however, Ptn transcripts are significantly (p < 0.05) elevated by up to 70-fold in failing human myocardium relative to nonfailing controls. By ELISA, we furthermore determined that PTN proteins are present at 2.44 ± 0.70 µg/g in failing human hearts (Fig. 1B). This latter quantity is similar to that reported in human brain and gliomas, which have been reported to have some of the highest levels of PTN among adult human tissues (33).

In a rodent injury model of myocardial infarction (MI) induced by coronary ligation, ptn gene products were significantly elevated relative to controls (Fig. 1C). In these experiments, hearts had an average infarct size of 15–25%, but no significant change in left ventricle to body weight ratio could be demonstrated during the time frame examined (ratios: sham 2.56 ± 0.49 x 10-3 versus ligation 2.87 ± 0.31 x 10-3 for 1 week, n = 7; sham 2.25 ± 0.15 x 10-3 versus ligation 2.41 ± 0.16 x 10-3 for 4 weeks, n = 9). In sham-operated animals, PTN immunostaining was detectable in myocardium post-surgery, but following MI, robust PTN staining was observed in both infarcted and peri-infarct myocardium (Fig. 1D). Consistent with its known properties as a secreted cytokine, PTN immunostaining was observed primarily in pericellular regions; however, cellular staining was also detected. In infarcted myocardium, PTN immunostaining was generally limited to non-MHC-positive cells. In the peri-infarct myocardium, PTN signals were observed both in MHC-positive cells and in close proximity to TUNEL-positive cells. Neither TUNEL nor PTN immunostaining was readily observed in the adjacent viable myocardium (Fig. 1D). In the peri-infarct myocardium at 1 day post-MI, 35.6 ± 1.2% of the nuclei were TUNEL-positive, but at 1 and 4 weeks post-ligation, only 12.8 ± 4.9% (sham, 0.93 ± 0.76%) and 9.6 ± 3.6% (sham, 0.97 ± 0.86%) of the nuclei, respectively, were TUNEL-positive. Notably, only about 40% of the TUNEL-positive cells in this region were positive for PTN (Fig. 1D), and only a fraction (<30%) of these cells were cardiac MHC-positive. The majority of PTN-positive cells thus appear to be non-CMs. Although we cannot determine the primary source of PTN, which could be either resident cells or infiltrates of inflammatory cells present in damaged myocardium, its distribution demonstrates that PTN is elevated in the peri-infarct myocardium and is in close proximity to contractile cells.


Figure 1
View larger version (63K):
[in this window]
[in a new window]

  		FIGURE 1.
Pleiotrophin is dynamically regulated during development and in models of heart failure. A, Ptn transcript abundance was measured by quantitative reverse transcription-PCR in heart tissues obtained from fetal (16–18 dpc), neonatal, and adult rats. B, PTN protein levels were determined by ELISA from neonatal and adult rats, and data are expressed relative to total tissue protein. Also by ELISA, PTN protein levels proved to be highly abundant in failing human myocardium. (*, p < 0.05 versus neonatal heart). C, PTN proteins were significantly up-regulated following myocardial infarction, relative to sham-operated animals. An example of a Western blot and averaged data are shown. *, p < 0.05 versus sham). D, cardiac MHC staining (green) was employed as a marker of contractile cells in sham-operated and infarcted myocardium. In these heart sections, PTN (blue) immunostaining was poorly detectable in both sham-operated and nonoperated (inset) hearts, but it was markedly increased in peri-infarct regions of infarcted myocardium. Similarly, TUNEL-positive (red) cells were absent in controls but readily observed in both infarcted and peri-infarct myocardium. When overlaid, PTN staining was coincident with ~40% of TUNEL-positive signals at 1-week post-ligation (purple staining), but <30% of the TUNEL-positive cells were positive for both PTN and MHC (brownish color). Although many CMs in the peri-infarct region were PTN-positive, the data clearly show the presence of other PTN-positive cells, perhaps associated with inflammatory cells or macrophages which have been shown to contain high levels of PTN (14). The infarct region is shown by the presence of an asterisk and an absence of MHC-positive cells. The peri-infarct region is to the right of this region and the viable myocardium (MHC-positive) is to the left.

 
Recombinant PTN and Its Pro-apoptotic Effects on Cardiac Myocytes in Vitro—Similar to the results observed during development, isolated NNCMs contained significantly higher (>10-fold) levels of ptn gene products (RNA and protein) than ACMs (Fig. 2, A and B). In contrast, cardiac-derived fibroblasts (FBs) only contained <1 and ~3% of the amount of PTN detected in NNCMs and ACMs, respectively. PTN immunostaining was also much stronger in freshly isolated NNCMs than in either ACMs or neonatal FBs (Fig. 2C). To ensure that the observed signals were specific for contractile cells, NNCM cultures were fixed and co-stained with both PTN and {alpha}-actinin, a sarcomeric protein specific for muscle cells (Fig. 2C, inset). In these experiments, contaminating FBs only comprised 4.8 ± 2% of the total number of cells in the primary NNCM cultures. Interestingly, PTN immunostaining was observed in NNCMs both in the cytoplasmic and nuclear regions of the cells (Fig. 2C), similar to reports showing co-localization of PTN with nuclear markers in neurons (34). Transcripts to all five known PTN receptors (RPTP β/{zeta}, anaplastic lymphoma kinase, and syndecans -1, -3, and -4) were also present in hearts and in isolated CMs (Table 2), suggesting that PTN might be able to activate a signaling cascade in contractile cells.


View this table:
[in this window]
[in a new window]

  		TABLE 2
Presence of known PTN receptor transcripts in heart The transcript abundance of pleiotrophin and known PTN receptors were determined by real time quantitative reverse transcription-PCR (n = 3). {Delta}Ct values are shown. Ct represents the amplification cycles where the growing fluorescence signal would reach the threshold setting. {Delta}Ct is a normalized Ct value with respect to Gapd. Negative values in this table indicate that the transcripts were more prevalent (lower Ct value) than transcripts encoding Gapd. On average, a difference of 3–4 {Delta}Ct units indicated a 10-fold difference in expression. Transcripts encoding RPTP were just detectable in heart, but those to syndecans 3 and 4 were relatively abundant in both whole heart and in NNCMs.

 
To determine whether exogenously added PTN could act as a growth factor to promote cellular hypertrophy or survival, recombinant PTN from 1 to 5 µg/ml was added to NNCM cultures, and cells were observed for 1 week. During this time frame, no change in cell morphology or spontaneous beating rate could be demonstrated at any dose, and no significant change in Anf transcript abundance was measured (-{Delta}Ct, 6.41 ± 0.25 PTN-treated versus 6.61 ± 0.22 control). The cell surface area (%) of PTN-treated cells did not differ from that of vehicle-treated cells (98.9 ± 17.1% versus control, n = 4, p > 0.05), whereas phenylephrine led to a pronounced increase in cell size (Fig. 3A). However, we did observe an increase in the number of floating cells within 24 h of PTN addition, and some NNCMs began to display traits associated with cell death. In fact, the addition of PTN significantly increased the number of {alpha}-actinin-positive cells that contained condensed or fragmented nuclei (Fig. 3A). Apoptotic nuclei were observed in 8.1 ± 0.9 and 19.9 ± 2.0% of NNCMs at days 1 and 5, respectively, and in control experiments, a significantly smaller fraction of condensed or fragmented nuclei were observed (day 1, 2.94 ± 0.1%; day 5, 9.56 ± 2.1%, n = 4). DNA laddering with a pattern typical of caspase-mediated DNA fragmentation also increased as a function of time and PTN concentration (Fig. 3B). The minimum effective dose was 1 µg/ml at day 1, but even 50 ng/ml of PTN increased the amount of DNA laddering by day 5. This latter concentration is substantially less than that measured in vivo in both humans and rodent samples, and the effective dose was similar to the amount measured by ELISA in neonatal heart (~0.94 µg/ml tissue volume). Importantly, the observed DNA laddering could be completely blocked by pretreatment with 40 µmol/liter benzyloxycarbonyl-VAD-fluoromethyl ketone (a pan-caspase inhibitor) (Fig. 3C).

Next we determined whether PTN could potentiate the pro-apoptotic effects of cell injury on contractile cells. Because this cytokine was elevated in hearts following ischemic injury, we postulated that PTN would be up-regulated in response to injury and that this up-regulation might be implicated in injury repair or cardiac remodeling. To test this hypothesis in vitro, hypoxia was induced by exposing NNCMs to a gas mixture composed of 95% N2 and 5% CO2. After 24 h, the number of apoptotic nuclei doubled (hypoxia versus PTN + hypoxia, 11.6 ± 1.8 versus 23.6 ± 2.9% (Fig. 3D, n = 4)), and the incidence of DNA laddering increased significantly (p < 0.05) following the addition of PTN. Moreover, NNCMs exposed to UV irradiation (100 J/m2) had a >2-fold increase in the number of apoptotic nuclei relative to controls (UV versus PTN + UV, 20.2 ± 4.2 versus 42.1 ± 4.5%) (Fig. 3D) and a significant increase in DNA laddering (p < 0.05, n = 4).


Figure 2
View larger version (50K):
[in this window]
[in a new window]

  		FIGURE 2.
Pleiotrophin gene products are expressed in isolated NNCMs and ACMs. A, quantitative reverse transcription-PCR analysis of Ptn mRNA transcripts in cardiac cells. Ptn transcripts were more abundant in NNCMs than in ACMs (*, p < 0.05 versus NNCM, n = 4); however, the levels are only ~5.0% that found in whole heart (see Table 2). Data are expressed as a percentage relative to NNCMs (100%). B, similarly, PTN proteins detected on Western blots are much greater in NNCMs than in ACMs; however, no appreciable levels of PTN could be detected in either neonatal or adult FBs (*, p < 0.05 versus NNCM). C, PTN immunostaining was readily detected in NNCMs, but only weakly so in ACMs and NN FBs. To distinguish between NNCMs and potentially contaminating FBs, sarcomeric {alpha}-actinin-positive staining was used as a marker of muscle cells, and its absence was indicative of non-CMs or FBs. The rod-shaped morphology was sufficient to distinguish cellular specificity of ACMs. Hoechst 33342 was employed as a nuclear marker. In the overlays, PTN immunostaining was observed both in cytoplasmic and nuclear regions of NNCMs, whereas in ACMs, immunostaining was observed primarily in cytoplasmic and perinuclear regions of the cells.

 
The increase in cell death observed 48 h post-hypoxia occurred concomitant with a 2-fold (2.0 ± 0.2) increase in cellular PTN protein (Fig. 3E, cells), demonstrating that PTN is also up-regulated in response to this common cardiac stress. Because these cultures contained both CMs and a small percentage of contaminating FBs, the source of PTN was determined by comparing the results from NNCMs with neonatal FB cultures, which were essentially free of any contaminating CMs. In these experiments, PTN was only detectable in NNCMs, and we could not detect any endogenous PTN in FB cultures. Even after hypoxia or UV irradiation, PTN could only be detected in NNCMs, thus indicating that the primary source of this cytokine in this experimental model system is from CMs and not from FBs. Importantly, PTN could also be detected in the media of cultured NNCMs, but not in cultured FBs (Fig. 3E, n = 3). Although no significant difference could be demonstrated between control and UV-irradiated NNCMs at 24 h, a small increase was detected with hypoxia at this time. The amount of PTN present in the media was, however, markedly increased between 24 and 48 h post-stress (see Fig. 3E (hypoxia) and next section (UV irradiation)) and was particularly noteworthy following hypoxia, which demonstrated a 13.2 ± 0.4-fold increase in the amount of PTN.

We then went on to examine adult cardiomyocytes (ACMs) to determine whether PTN could also potentiate cell death in mature contractile cells (Fig. 3, F and G). Because ACMs proved more sensitive and died more readily than NNCMs, a number of pilot experiments were performed. Cells were exposed to hypoxia from 1 to 24 h, and the number of rounded (non-rod-shaped) cells was determined. We observed that a 1-h exposure to hypoxia only increased cell death marginally, but exposures of 3 h or more greatly increased the number of rounded, nonviable cells. Cells treated with PTN were therefore exposed to hypoxia for only 2 h followed by re-exposure to ambient air (i.e. reoxygenation) for 4.5 h. Under these conditions, PTN significantly increased the number of apoptotic nuclei (p < 0.05, n = 4) from 11.6 ± 3.9% (hypoxia and reoxygenation) to 20.5 ± 4.6% (PTN + hypoxia and reoxygenation) (Fig. 3, F and G), demonstrating that PTN generally enhances the incidence of CM programmed cell death in response to pathologically relevant apoptotic stimuli.

Endogenous PTN Contributes to Cellular Apoptosis—Because PTN was released from CMs in response to apoptotic stimuli (see Fig. 3E), siRNAs were employed to determine whether knockdown of endogenous Ptn could inhibit the apoptotic response of CMs to cellular stress. These experiments were performed both with a nontargeting siRNA and GFP vector as a negative control (GFP) and with siGlo as a fluorescent indicator. The latter showed that >90% of the NNCMs could be successfully transfected (Fig. 4A), and no significant difference in Ptn transcripts could be demonstrated between siGlo and GFP transfection controls versus nontransfected cells. However, Ptn transcripts were significantly reduced in NNCMs within 24 h of transfection and prior to exposure to UV irradiation by both siRNA2 and siRNA4 relative to siGlo or GFP (Fig. 4B). Importantly, the percentage of apoptotic cells did not differ among any of the experimental groups (p > 0.05). Similarly, no significant reduction in PTN protein could be demonstrated at this time with either siRNA2 or siRNA4 relative to controls (not shown).


Figure 3
View larger version (45K):
[in this window]
[in a new window]

  		FIGURE 3.
PTN promotes apoptosis in CMs. A, NNCMs treated with 1.5 µg/ml PTN for 1 day show a significant increase in the number of fragmented and condensed nuclei (arrows) but not in cell surface area. The middle panels show magnified views of cells containing nuclei scored positive for apoptosis (inset). Phenylephrine (Phe) rapidly increases the surface area of NNCMs. B, increasing concentrations of PTN augment genomic DNA fragmentation. C, PTN-induced apoptosis is inhibited by the pan-caspase inhibitor benzyloxycarbonyl-VAD-fluoromethyl ketone (Z-VAD-FMK). D, PTN significantly increases the number of apoptotic nuclei in NNCMs in a time-dependent manner and following hypoxia and UV exposure (*, p < 0.05 versus control, n = 4 for each). E, PTN proteins were detectable in the NNCM-conditioned cultivation media but not in the neonatal FB-conditioned media. In response to hypoxia (hyp), a small increase in the amount of PTN was observed both in cells and in the media of NNCM cultures following hypoxia, but at 48 h post-hypoxia, a ≥2-fold increase (p < 0.05) in cellular PTN and a 10–13-fold (p < 0.05, n = 3) increase in the amount of PTN present in the media was detected (left panel). Similarly, no significant increase in cellular or secreted PTN was observed 24 h post-UV irradiation relative to controls (ctl)(right panel), but with longer cultivation times (see Fig. 4C), PTN levels increased. F, ACMs treated with PTN (5 µg/ml) and exposed to hypoxia and reoxygenation showed an increase in the number of condensed and fragmented nuclei (arrows). G, PTN (5 µg/ml) significantly enhances the apoptotic response of ACMs following hypoxia-reoxygenation (*, p < 0.05, n = 4).

 
In response to UV irradiation, however, endogenous levels of PTN increased by up to 50% relative to nonirradiated NNCMs transfection controls, but following knockdown of Ptn by either siRNA2 or siRNA4, PTN protein levels were significantly reduced (p < 0.05) relative to GFP + UV controls (Fig. 4C). In fact, PTN was almost undetectable in UV-irradiated cells transfected with these siRNAs, and the amounts were below those observed in the non-UV, siRNA-treated cells. The knockdown of PTN was also specific for Ptn because actin proteins were not affected by the presence of these siRNAs. Most importantly, the silencing RNAs significantly reduced the number of apoptotic cells by 40–45% relative to UV-irradiated controls (Fig. 4D), thus demonstrating that endogenous PTN, perhaps through an autocrine mediated pathway, directly contributes to programmed cell death in NNCMs in response to cell stress.


Figure 4
View larger version (41K):
[in this window]
[in a new window]

  		FIGURE 4.
Knockdown of Ptn by siRNA inhibits UV-induced apoptosis. A, phase (left) and fluorescent (right) images of NNCMs transfected with siGlo RNA. B, quantitative reverse transcription-PCR analysis of Ptn mRNAs in NNCMs after siRNA transfection and before UV irradiation. All other cells were transfected with the indicated siRNA or control. (*, p < 0.05 versus GFP control, n = 3). C, Western blot showing the abundance of PTN in NNCMs after transfection with the indicated vectors and following UV irradiation. Control refers to nontransfected cells, and a siGlo transfection in non-UV treated cells is also shown. Sarcomeric actin was used to monitor loading. D, incidence of apoptosis following transfection with siRNAs targeting Ptn was significantly (*, p < 0.05, n = 3) reduced relative to the control transfections in UV-irradiated cells. The results from non-UV irradiated transfected cells are shown for comparisons.

 
Mechanisms of PTN-induced Apoptosis—Because PTN has been shown to function as an autocrine/paracrine factor to modulate the MAPK and the PI3K-AKT signaling in NIH3T3 cells (16), we tested its effects on NNCMs in the presence of the MAPK/ERK kinase (MEK) inhibitor PD98059 (15–25 µmol/liter) and the PI3K inhibitor LY294002 (20 µmol/liter). In these experiments, we could not demonstrate any inhibition by PD98059 on PTN-induced apoptosis, either by DNA laddering or by scoring for apoptotic nuclei (p > 0.05), but the combination of PTN and LY294002 promoted apoptosis. When added alone, LY294002 also activated apoptosis, similarly to what we had previously reported in NNCMs (31), but this apoptotic response was less than the two together (Fig. 5A), suggesting that the combined effects of PTN and LY294002 must involve PI3K inactivation (i.e. signal cross-talk).

To test this possibility, we added IGF-1 to the cells. This led to enhanced phosphorylation of AKT at serine 473, but in combination with PTN (1.5 µg/ml), AKT phosphorylation was markedly decreased (Fig. 5B). Analysis of serine 473 phosphorylation indicated that recombinant PTN significantly reduced basal AKT phosphorylation in a dose- and time-dependent manner without altering total cellular AKT protein content (Fig. 5, C and D). The decrease in phosphorylation, which is generally indicative of reduced AKT activity (35), was observed within 5–15 min and maintained for up to 12–24 h after cytokine addition (Fig. 5E). The effects of PTN on the AKT signaling axis in CMs were thus persistent, and perhaps cumulative. When AKT phosphorylation was plotted as a function of apoptosis (DNA laddering), a correlation coefficient of 0.997 was calculated from the quadratic polynomial regression analysis, further indicating a close relationship between PTN-mediated dephosphorylation of AKT and apoptosis. In contrast, Erk1/Erk2 showed a transient increase in phosphorylation at positions Thr-183 and Thr-185, excluding the possibility that the changes in phosphorylation were simply because of altered phosphatase activity. C2C12 myoblasts also showed an increase in AKT phosphorylation after treatment with PTN (Fig. 5C), showing that the decreased phosphorylation of AKT and associated apoptosis were specific to CMs.

The inhibitory effects of PTN on AKT signaling were examined further by analyzing two direct downstream targets, glycogen synthase kinase 3-β (GSK3β) and BCL2 antagonist of cell death (BAD) (36, 37). GSK3β is a downstream transducer of PI3K-AKT, and when phosphorylated at Ser-9, this protein shows reduced kinase activity. After the addition of PTN, phosphorylation of serine 9 in GSK3β was significantly reduced in CMs (Fig. 6A), and this reduction, relative to controls, was sustained for at least 6 h (Fig. 6B). The increase in GSK3β activity and the pro-apoptotic response (DNA fragmentation) elicited by PTN could be blocked by pretreatment with LiCl, at a dose (3 mmol/liter) that specifically inhibits GSK3β activity (Fig. 6C). Similarly, BAD, another downstream target of AKT, showed altered phosphorylation. BAD phosphorylation by the PI3K/AKT pathway promotes its association with other members of the BCL2 family to limit apoptosis; however, addition of PTN reduced BAD phosphorylation at serine 136 in NNCMs within 5–15 min (Fig. 6D), followed by an increase in DNA laddering. Thus PTN not only inhibits AKT phosphorylation, it can specifically alter the phosphorylation of multiple downstream targets of AKT.


Figure 5
View larger version (35K):
[in this window]
[in a new window]

  		FIGURE 5.
PTN inhibits the phosphorylation of AKT in NNCMs. A, LY294002 (20 µmol/liter) alone promotes apoptosis. PTN (1.5 µg/ml) does not significantly increase the degree of apoptosis elicited by LY294002; however, at higher doses of PTN (5 µg/ml), the increase was significant (p < 0.05). B, IGF-I (0.5 nmol/ml) strongly induces the phosphorylation of AKT at serine 473, but PTN (1.5 µg/ml) partially inhibits this phosphorylation event. C, AKT phosphorylation at serine 473 is decreased in a concentration-dependent manner by the addition of PTN, but total AKT protein is unchanged. In contrast and after treatment with PTN, C2C12 skeletal myoblasts (positive control) show an increase in serine 473 phosphorylation, and ERK1/2 phosphorylation is transiently increased. D, decrease in AKT phosphorylation at position 473 (i.e. active AKT) elicited by PTN (1.5 µg/ml) in NNCMs is time-dependent. E, summary data showing that the decrease in AKT phosphorylation at position 473 occurs within 5–15 min of PTN addition, and the reduction in active AKT can last for up to 12 h. (* indicates p < 0.05, n = 3.)

 
If PTN does, in fact, modulate CM apoptosis through inhibition of AKT, then the expression of a constitutively active form of AKT should be sufficient to inhibit this effect. We therefore employed an adenoviral construct that expressed a hemagglutinin-tagged and constitutively active form of AKT (pAd.CMV/DEST-myrAkt1-HA). A virus expressing no gene product (pAd.CMV/DEST-null) was used as the control. In these experiments, we determined that >90% of the infected NNCMs expressed HA-tagged myr-AKT1 (Fig. 7A). Adenoviral infection and expression of myr-AKT1 reduced, but did not significantly alter, the number of nonirradiated NNCMs scored positive for apoptosis. Expression of myr-AKT1 did, however, significantly decrease both the number of apoptotic nuclei and the degree of DNA laddering observed after addition of recombinant PTN alone (Fig. 7B). Following UV irradiation, NNCMs infected with pAd.CMV/DEST-myrAkt1-HA showed a 63.8% decrease in the number of apoptotic cells when compared with pAd.CMV/DEST-null infected cells. Expression of myr-AKT1 also reduced the number of apoptotic cells and the amount of DNA laddering observed in UV irradiation + PTN-treated cells (Fig. 7B). In fact, the relative reduction in apoptosis in response to myr-AKT1 overexpression was greater in the UV + PTN group than in the UV group alone. Similar results were also observed in response to hypoxia. The overexpression of myr-AKT1 reduced the degree of DNA laddering relative to hypoxia alone (Fig. 7B). These latter data confirm that the mechanisms elucidated in this study are general phenomena and that the pro-apoptotic potential of PTN can be blocked by expression of constitutively active AKT1.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
REFERENCES
 
In this study, we show that the secreted growth factor PTN is dynamically regulated in hearts of rodents and men. As predicted for a development-regulated cytokine, PTN is reduced during normal heart development, but in rat hearts subjected to coronary ligation, this cytokine is significantly up-regulated. In nonfailing adult human heart, Ptn is poorly expressed, but in failing hearts, ptn gene products are strongly up-regulated. In fact, PTN proteins in failing myocardium are almost 100-fold higher than that reported from most adult human tissues (33). Although it is difficult to predict in adults how activated cytokine genes will affect cellular events in response to injury, we originally postulated that contractile cells would respond to this growth factor through the induction of hypertrophy. Instead, we find that recombinant PTN in a dose- and time-dependent manner triggers a slow increase in the number of apoptotic CMs, but when combined with pro-apoptotic stimuli, it significantly potentiates CM programmed cell death.

This finding raises the question of whether pleiotrophin up-regulation is beneficial in the context of heart failure. In adult myocardium, PTN is minimally expressed in or around CMs, and even though recombinant PTN can promote apoptosis in vitro, such a response would not necessarily be expected in vivo, at least not at low concentrations. Intact myocardium, unlike cultured cells, contains an extracellular matrix composed largely of collagen, basement membranes, and proteoglycans. Because chrondroitin-sulfate and heparin-sulfate proteoglycans bind PTN to regulate its function in brain, it is likely that PTN would also be bound by similar moieties in vivo, which may limit any pro-apoptotic effects (38, 39). This cytokine is, however, readily detectable in heart cells located adjacent to blood vessels,4 and it is widely distributed in peri-infarct and infarcted myocardium. Because it has already been shown to promote neovasculogenesis in heart following injury, its primary function must be to improve blood flow, which would clearly be beneficial to ischemic hearts (26). We furthermore speculate that relatively high quantities of PTN in damaged heart will promote the death and removal of damaged contractile cells, which in vivo may be important for scar formation and remodeling. In fact, PTN has been associated with tissue repair and remodeling. In cerebral ischemia, for example, PTN is strongly up-regulated in developing microvasculature, endothelial cells, and macrophages (14), and its secretion stimulates breast cancer progression through remodeling of the tumor microenvironment (40). Because heart failure is also associated with significant remodeling, our data suggest that the effects of PTN are not limited to revascularization.


Figure 6
View larger version (23K):
[in this window]
[in a new window]

  		FIGURE 6.
PTN modifies the phosphorylation of downstream targets of AKT. A, treatment of NNCMs with recombinant PTN (1.5 µg/ml) results in a persistent decrease in GSK-3β phosphorylation at serine 9. B, summary data of GSK-3β dephosphorylation as a function of time following the addition of PTN. (* indicates p < 0.05, n = 3.) C, LiCl, at a concentration (3 mmol/liter) that specifically inhibits GSK-3β activity, decreases both GSK-3β dephosphorylation and DNA fragmentation elicited by 1.5 µg/ml PTN in NNCMs. D, within 30 min, PTN (1.5 µg/ml) significantly reduces BAD phosphorylation at serine 136.

 


Figure 7
View larger version (85K):
[in this window]
[in a new window]

  		FIGURE 7.
Expression of constitutively active AKT1 inhibits PTN-potentiated apoptosis of NNCMs. A, greater than 90% of cells were infected with pAd.CMV/DEST-myrAkt1-HA as determined by immunostaining with an antibody against the HA tag. B, in both UV-treated and hypoxic NNCMs, overexpression of myr-AKT1, relative to cells infected with pAd.CMV/DEST-null virus, reduced the observed intensity of DNA laddering in PTN-treated cells.

 
Although PTN inhibition of AKT activity and the increase in apoptosis is quantifiable and highly significant, this response is unique to CMs. This distinction may be attributed to a variety of possibilities. To date, most of the growth and prosurvival effects of PTN have been described in dividing cells, and although NNCMs, and to a lesser extent ACMs, retain some capacity to divide, the frequency at which this occurs is relatively low and almost nonexistent in vitro. PTN has also been reported to activate MAPK signaling in NIH3T3 cells to inhibit apoptosis (16), but the inability of PD98059 to alter the apoptotic response in CMs indicates that this PTN-associated mechanism is either absent or different in these cells. These apparent mechanistic differences may also be due, at least partially, to comparisons between established and primary cell lines. Alternatively, structural domains within PTN, which are known to dock with multiple PTN receptors, may elicit unique responses among cells (41).

Importantly, the primary source of PTN in damaged heart is currently unknown. We demonstrate that it is produced and released by contractile cells in vitro, suggesting that it may be acting as an autocrine factor; however, our immunostaining data of infarcted myocardium, where <30% of the PTN-positive cells were also {alpha}-MHC-positive, also suggested that its secretion is not limited to CMs. In brain, both the endothelial cells associated with neovascularization and macrophages are known to express high levels of PTN within 3 days of ischemic injury. Because PTN is also up-regulated in heart following an MI, it is likely that these cell types contribute to the increased presence of PTN in damaged myocardium.

Currently, the effects of PTN are known to involve at least five receptors and both the PI3K-AKT/PKB and MAPK signaling cascades. In this study, we have only provided direct evidence in CMs for inhibition of the PI3K-AKT/PKB signaling axis. More specifically, PTN-mediated inhibition of AKT in vitro occurs concomitantly with dephosphorylation of downstream targets BAD and GSK3β, and the activation of caspases. The pro-apoptotic and potentiating responses elicited by PTN occur at concentrations similar to that seen in vivo, and we have demonstrated that these effects can be inhibited by either adenovirus-mediated expression of constitutively active AKT or by siRNA-mediated knockdown of PTN. Because CMs are the primary source of PTN, at least in vitro, these data lead us to conclude that PTN can act as a signaling molecule to inhibit AKT activity and promote apoptosis. The mechanism underlying the inhibition of AKT signaling is less clear, however. We have shown that transcripts to all five known PTN receptors are present in heart and in NNCMs, but of these, only syndecan-3 and syndecan-4 are prominently expressed. Of particular interest is syndecan-4, which has been shown previously to be elevated in CMs in response to ischemia (42). Zhang et al. (43) have furthermore shown that endothelial expression of syndecan-4 in heart selectively enhances microvessel responsiveness to FGF2 through increased generation of nitrous oxide (NO), and although NO release involves AKT-1-dependent activation of endothelial nitric-oxide synthase, no direct link between AKT and syndecan-4 has thus far been demonstrated in CMs. It thus remains unknown which receptors mediate the pro-apoptotic versus pro-angiogenic responses in heart or even whether the pro-angiogenic functions of PTN are because of AKT/PKB activation. Future studies will therefore be needed to test these possibilities in vivo, but if PTN-mediated angiogenesis functions independently of AKT signaling, then the selective activation of AKT in myocytes after ischemia should foster myocyte survival without inhibiting the pro-angiogenic responses of PTN.

In conclusion, we have identified a new function to the expanding list of actions that pleiotrophin has in cells and tissues. In cardiac myocytes, this cytokine is reactivated in response to hypoxia, and in a dose- and time-dependent manner, it potentiates programmed cell death of both NNCMs and ACMs through inhibition of AKT/PKB signaling. Its up-regulation in human heart failure and in myocardial ischemic may therefore contribute to the death of damaged CMs, as well as foster cardiac remodeling and revascularization of the injured heart.


   FOOTNOTES
 
* This work was supported by the Intramural Research Program of the National Institutes of Health, National Institute on Aging (to K. R. B. and M. T.), and by National Institutes of Health Grants RO1-AG017022 (to K. B. M.) and HL-073935 (to M. T. C.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Back

1 These authors should be considered as first authors. Back

2 To whom correspondence may be addressed: National Institutes of Health/NIA/GRC/LCS, 5600 Nathan Shock Dr., Baltimore, MD 21224. Fax: 410-558-8150; E-mail: bohelerk{at}grc.nia.nih.gov.

3 The abbreviations used are: PTN, pleiotrophin protein; ptn, the pleiotrophin gene; Ptn, pleiotrophin mRNA/transcript; CM, cardiomyocytes; ACMs, adult cardiomyocytes; NNCM, neonatal cardiomyocyte; MI, myocardial infarction; FBs, fibroblasts; BAD, BCL2 antagonist of cell death; GSK3β, glycogen synthase kinase 3-beta; HA, hemagglutinin; TUNEL, terminal deoxynucleotidyltransferase-mediated dUTP nick end labeling; RPTP β/{zeta}, receptor protein tyrosine phosphatase β/{zeta}; MAPK, Ras-mitogen activated protein kinase; PI3K, phosphatidylinositol 3-kinase; Gapd, glyceraldehyde phosphate dehydrogenase; ELISA, enzyme-linked immunosorbent assay; IGF, insulin-like growth factor; siRNA, small interfering RNA; ERK, extracellular signal-regulated kinase; GFP, green fluorescent protein; MHC, myosin heavy chain. Back

4 J. Li and K. R. Boheler, unpublished data. Back



   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Rauvala, H. (1989) EMBO J. 8, 2933-2941[Medline] [Order article via Infotrieve]
  2. Sato, M., Miyado, K., Okazaki, M., and Kimura, M. (1997) Biochem. Biophys. Res. Commun. 238, 831-837[CrossRef][Medline] [Order article via Infotrieve]
  3. Tezuka, K., Takeshita, S., Hakeda, Y., Kumegawa, M., Kikuno, R., and Hashimoto-Gotoh, T. (1990) Biochem. Biophys. Res. Commun. 173, 246-251[CrossRef][Medline] [Order article via Infotrieve]
  4. Fang, W., Hartmann, N., Chow, D. T., Riegel, A. T., and Wellstein, A. (1992) J. Biol. Chem. 267, 25889-25897[Abstract/Free Full Text]
  5. Deuel, T. F., Zhang, N., Yeh, H. J., Silos-Santiago, I., and Wang, Z. Y. (2002) Arch. Biochem. Biophys. 397, 162-171[CrossRef][Medline] [Order article via Infotrieve]
  6. Souttou, B., Raulais, D., and Vigny, M. (2001) J. Cell Physiol. 187, 59-64[CrossRef][Medline] [Order article via Infotrieve]
  7. Christman, K. L., Fang, Q., Kim, A. J., Sievers, R. E., Fok, H. H., Candia, A. F., Colley, K. J., Herradon, G., Ezquerra, L., Deuel, T. F., and Lee, R. J. (2005) Biochem. Biophys. Res. Commun. 332, 1146-1152[CrossRef][Medline] [Order article via Infotrieve]
  8. Choudhuri, R., Zhang, H. T., Donnini, S., Ziche, M., and Bicknell, R. (1997) Cancer Res. 57, 1814-1819[Abstract/Free Full Text]
  9. Bastias, J., Wei, M. X., Huynh, R., Chaubet, F., Jozefonvicz, J., and Crepin, M. (2002) Anticancer Res. 22, 1603-1613[Medline] [Order article via Infotrieve]
  10. Czubayko, F., Schulte, A. M., Berchem, G. J., and Wellstein, A. (1996) Proc. Natl. Acad. Sci. U. S. A. 93, 14753-14758[Abstract/Free Full Text]
  11. Riegel, A. T., and Wellstein, A. (1994) Breast Cancer Res. Treat. 31, 309-314[CrossRef][Medline] [Order article via Infotrieve]
  12. Wu, H., Barusevicius, A., Babb, J., Klein-Szanto, A., Godwin, A., Elenitsas, R., Gelfand, J. M., Lessin, S., and Seykora, J. T. (2005) J. Cutan. Pathol. 32, 125-130[CrossRef][Medline] [Order article via Infotrieve]
  13. Hampton, B. S., Marshak, D. R., and Burgess, W. H. (1992) Mol. Biol. Cell 3, 85-93[Abstract]
  14. Yeh, H. J., He, Y. Y., Xu, J., Hsu, C. Y., and Deuel, T. F. (1998) J. Neurosci. 18, 3699-3707[Abstract/Free Full Text]
  15. Hida, H., Jung, C. G., Wu, C. Z., Kim, H. J., Kodama, Y., Masuda, T., and Nishino, H. (2003) Eur. J. Neurosci. 17, 2127-2134[CrossRef][Medline] [Order article via Infotrieve]
  16. Bowden, E. T., Stoica, G. E., and Wellstein, A. (2002) J. Biol. Chem. 277, 35862-35868[Abstract/Free Full Text]
  17. Maeda, N., Nishiwaki, T., Shintani, T., Hamanaka, H., and Noda, M. (1996) J. Biol. Chem. 271, 21446-21452[Abstract/Free Full Text]
  18. Milev, P., Chiba, A., Haring, M., Rauvala, H., Schachner, M., Ranscht, B., Margolis, R. K., and Margolis, R. U. (1998) J. Biol. Chem. 273, 6998-7005[Abstract/Free Full Text]
  19. Stoica, G. E., Kuo, A., Aigner, A., Sunitha, I., Souttou, B., Malerczyk, C., Caughey, D. J., Wen, D., Karavanov, A., Riegel, A. T., and Wellstein, A. (2001) J. Biol. Chem. 276, 16772-16779[Abstract/Free Full Text]
  20. Mitsiadis, T. A., Salmivirta, M., Muramatsu, T., Muramatsu, H., Rauvala, H., Lehtonen, E., Jalkanen, M., and Thesleff, I. (1995) Development (Camb.) 121, 37-51[Abstract]
  21. Souttou, B., Ahmad, S., Riegel, A. T., and Wellstein, A. (1997) J. Biol. Chem. 272, 19588-19593[Abstract/Free Full Text]
  22. Powers, C., Aigner, A., Stoica, G. E., McDonnell, K., and Wellstein, A. (2002) J. Biol. Chem. 277, 14153-14158[Abstract/Free Full Text]
  23. Anisimov, S. V., Tarasov, K. V., Riordon, D., Wobus, A. M., and Boheler, K. R. (2002) Mech. Dev. 202, 25-74
  24. Wang, D., Oparil, S., Feng, J. A., Li, P., Perry, G., Chen, L. B., Dai, M., John, S. W., and Chen, Y. F. (2003) Hypertension 42, 88-95[Abstract/Free Full Text]
  25. Margulies, K. B., Matiwala, S., Cornejo, C., Olsen, H., Craven, W. A., and Bednarik, D. (2005) Circ. Res. 96, 592-599[Abstract/Free Full Text]
  26. Christman, K. L., Fang, Q., Yee, M. S., Johnson, K. R., Sievers, R. E., and Lee, R. J. (2005) Biomaterials 26, 1139-1144[CrossRef][Medline] [Order article via Infotrieve]
  27. Chen, H. W., Yu, S. L., Chen, W. J., Yang, P. C., Chien, C. T., Chou, H. Y., Li, H. N., Peck, K., Huang, C. H., Lin, F. Y., Chen, J. J., and Lee, Y. T. (2004) Heart 90, 927-934[Abstract/Free Full Text]
  28. Boheler, K. R., Volkova, M., Morrell, C., Garg, R., Zhu, Y., Margulies, K., Seymour, A. M., and Lakatta, E. G. (2003) Proc. Natl. Acad. Sci. U. S. A. 100, 2754-2759[Abstract/Free Full Text]
  29. Moon, C., Krawczyk, M., Ahn, D., Ahmet, I., Paik, D., Lakatta, E. G., and Talan, M. I. (2003) Proc. Natl. Acad. Sci. U. S. A. 100, 11612-11617[Abstract/Free Full Text]
  30. Iwaki, K., Sukhatme, V. P., Shubeita, H. E., and Chien, K. R. (1990) J. Biol. Chem. 265, 13809-13817[Abstract/Free Full Text]
  31. Chesley, A., Lundberg, M. S., Asai, T., Xiao, R. P., Ohtani, S., Lakatta, E. G., and Crow, M. T. (2000) Circ. Res. 87, 1172-1179[Abstract/Free Full Text]
  32. Communal, C., Singh, K., Pimentel, D. R., and Colucci, W. S. (1998) Circulation 98, 1329-1334[Abstract/Free Full Text]
  33. Pufe, T., Bartscher, M., Petersen, W., Tillmann, B., and Mentlein, R. (2003) Arthritis Rheum. 48, 660-667[CrossRef][Medline] [Order article via Infotrieve]
  34. Taravini, I. R., Ferrario, J. E., Delbe, J., Ginestet, L., Debeir, T., Courty, J., Murer, M. G., Gershanik, O. S., and Raisman-Vozari, R. (2005) Brain Res. 1066, 196-200[CrossRef][Medline] [Order article via Infotrieve]
  35. Alessi, D. R., Andjelkovic, M., Caudwell, B., Cron, P., Morrice, N., Cohen, P., and Hemmings, B. A. (1996) EMBO J. 15, 6541-6551[Medline] [Order article via Infotrieve]
  36. Datta, S. R., Dudek, H., Tao, X., Masters, S., Fu, H., Gotoh, Y., and Greenberg, M. E. (1997) Cell 91, 231-241[CrossRef][Medline] [Order article via Infotrieve]
  37. Yin, H., Chao, L., and Chao, J. (2005) J. Biol. Chem. 280, 8022-8030[Abstract/Free Full Text]
  38. Bao, X., Mikami, T., Yamada, S., Faissner, A., Muramatsu, T., and Sugahara, K. (2005) J. Biol. Chem. 280, 9180-9191[Abstract/Free Full Text]
  39. Spinale, F. G., Zellner, J. L., Johnson, W. S., Eble, D. M., and Munyer, P. D. (1996) J. Mol. Cell. Cardiol. 28, 1591-1608[CrossRef][Medline] [Order article via Infotrieve]
  40. Chang, Y., Zuka, M., Perez-Pinera, P., Astudillo, A., Mortimer, J., Berenson, J. R., and Deuel, T. F. (2007) Proc. Natl. Acad. Sci. U. S. A. 104, 10888-10893[Abstract/Free Full Text]
  41. Bernard-Pierrot, I., Delbe, J., Rouet, V., Vigny, M., Kerros, M. E., Caruelle, D., Raulais, D., Barritault, D., Courty, J., and Milhiet, P. E. (2002) J. Biol. Chem. 277, 32071-32077[Abstract/Free Full Text]
  42. Li, J., Brown, L. F., Laham, R. J., Volk, R., and Simons, M. (1997) Circ. Res. 81, 785-796[Abstract/Free Full Text]
  43. Zhang, Y., Li, J., Partovian, C., Sellke, F. W., and Simons, M. (2003) Am. J. Physiol. 284, H2078-H2082

Add to CiteULike CiteULike   Add to Complore Complore   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us   Add to Digg Digg   Add to Reddit Reddit   Add to Technorati Technorati    What's this?



This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
282/48/34984    most recent
M703513200v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted