Cell transplantation for treatment of acute myocardial infarction: unique capacity for repair by skeletal muscle satellite cells.

An adult heart injured by an ischemic episode has a limited capacity to regenerate. We administered three types of adult guinea pig cells [cardiomyocytes (CMs), cardiac fibroblasts (CFs), and skeletal myoblasts (Mbs)] to compare their suitability for repair of acute myocardial infarction. We used confocal fluorescent microscopy and a variety of specific immunomarkers and echocardiography to provide anatomic evidence for the viability of such cells and their possible functional beneficial effects. All cells were transfected with adenovirus-containing beta-galactosidase gene so that migration from the injection sites could be traced. Both freshly isolated CMs as well as CFs were found concentrated in the infarcted zone; these cells survived for at least 2 wk posttransplantation. Transplanted CMs were regularly striated and grew long projections that could form gap junctions with native CMs, which was evidenced by connexin43 labeling. In addition, CM transplantation resulted in increased angiogenesis in the infarcted areas. In contrast, transplanted CFs did not appear to make any gap junctional contacts with native CMs nor did they enhance local angiogenesis. Mbs cultured for 7 days and transfected Mbs were identified 7 days posttransplantation in the infarcted area. During that time and thereafter, Mbs proliferated and differentiated into myotubes that formed new, regularly striated myofibers that occupied most (50-70%) of the infarcted area by 2-3 wk. These newly formed myofibers maintained their Mb skeletal muscle origin as evidenced by their capacity to express myogenin and fast skeletal myosin. This skeletal phenotype appeared to downregulate with time, and Mbs partially transdifferentiated into a cardiac phenotype as indicated by labeling for cardiac-specific troponin T and cardiac myosin heavy chain. By the third week posttransplantation, new myofibers formed apparent contacts with the native CMs via putative gap junctions that expressed connexin43. Myocardial performance of animals that were successfully transplanted with Mbs was improved.

Maintenance of Ca(i)(2+)transients during prolonged cardiac arrest aids rapid contractile recovery.

We explored the effects of contractile arrest maintained for 24-72 h in the presence of 2,3-butanedione monoxime or a Ca(2+) channel blocker (nifedipine or verapamil) on contractile activity, Ca(i)(2+) transients, and myofibrillar protein content and ultrastructure in long-term cultures of spontaneously beating adult guinea-pig cardiomyocytes. The contractions were not affected by 5 mM 2, 3-butanedione monoxime, but they were strongly or fully suppressed by 10 and 18 mM 2,3-butanedione monoxime, respectively, while the Ca(i)(2+)transients triggered by the maintained spontaneous electrical activity were either not changed at all (5 and 10 mM 2, 3-butanedione monoxime) or decreased only slightly (18 mM 2, 3-butanedione monoxime). The uncoupling of excitation from contraction by 10-18 mM for 24-72 h did not affect the content of the myofibrillar proteins. Confocal laser microscopy showed that these exposures affected the assembly of myofilaments, giving an overall deranged appearance to the myofibrils. In spite of this effect, the cells' contractile activity was readily regained within 15-60 min upon the washout of 2,3-butanedione monoxime. The 24-72-h exposures to 5 microM nifedipine or 10 microM verapamil, which blocked fully both the Ca(i)(2+) transients and contractility, did not affect the myofibrillar protein content nor their assembly. However, the recovery of contractile activity after exposure to a Ca(2+)-channel blocker was significantly slower (several days) than after 2,3-butanedione monoxime exposure. Furthermore, cultures exposed to Ca(2+)-channel blockers also had significantly decreased sensitivity to beta-adrenergic stimulation. Altogether, these data indicate the importance of regular Ca(2+) influx for the maintenance of the functional integrity of adult cardiomyocytes during prolonged periods of contractile arrest.

The antioxidant effects of a novel iron chelator salicylaldehyde isonicotinoyl hydrazone in the prevention of H(2)O(2) injury in adult cardiomyocytes.

OBJECTIVE: This study was designed to investigate the cardioprotective effect of the novel lipophilic iron chelator salicylaldehyde isonicotinoyl hydrazone (SIH) against the oxidative stress exerted by H(2)O(2) through the production of OH radical via the Fenton reaction and to compare them with those of the hydrophilic iron chelator deferoxamine (DFO) and the Na(+)/H(+) exchange inhibitor methylisobutyl amiloride (MIA). METHODS: We used long-term cultures of spontaneously beating adult guinea-pig ventricular cardiomyocytes developed and characterized previously in our laboratory. We assessed their contractile activity by video-recording as well as the underlying Ca(i)(2+) transients by Fura 2 fluorescence. In some experiments we also recorded these functional parameters, plus the electrical activity (action potentials) in response to electrical stimulation via suction pipettes, in individual freshly isolated myocytes. RESULTS: Exposure of the regularly and synchronously beating cultured cardiomyocytes to 100 microM H(2)O(2) initially caused a substantial prolongation of Ca(i)(2+) transients accompanied by an irregular contractile activity, then in contractile arrest and a severalfold increase in cytosolic [Ca(2+)] that occurred, within 30 min of H(2)O(2) application. Similar effects were also observed using freshly isolated cardiomyocytes. The latter effects were first accompanied by significant prolongation of the action potential duration (APD) with superimposed early afterdepolarizations followed by a second phase with a very fast decrease in APD, contractions, as well as Ca(i)(2+) transients and a third phase of inexcitability, contractile arrest, increased cytoplasmic [Ca(2+)] and a final contracture. All these effects were irreversible in both types of preparations but they could be fully prevented by a 15-min preincubation with 200 microM SIH. Similar protective effects were observed with DFO, but in this case a much higher concentration had to be used (1 mM) and much longer (2 h) preincubation was needed. By contrast, 5 microM MIA failed to fully protect the cardiomyocytes, although a significant delay (10 min) of the effects of H(2)O(2) was observed. CONCLUSIONS: The data indicate that SIH provides a very powerful and very fast protection against the oxidative stress exerted by H(2)O(2) presumably via the iron-mediated Fenton reaction producing hydroxyl radical (OH), whereas the protective effect of DFO is hindred by its very slow and rather limited intracellular entry, and the protection that MIA exerts via the inhibition of Na(+)/H(+) exchange against H(2)O(2) much less effective.

Altered transarcolemmal Ca transport modifies the myofibrillar ultrastructure and protein metabolism in cultured adult ventricular cardiomyocytes.

The present study was designed to investigate how prolonged (24-72 h) exposure to modifiers of Ca transarcolemmal transport affects the myofibrillar structure, protein turnover and content of myofibrillar proteins in adult guinea pig cardiomyocytes maintained beating synchronously in long-term cultures. First we established the functional responses (the contractile activity and [Ca]i transients) of the cultured myocytes to acute exposures to several drugs used in this study. The ultrastructural characteristics of these cultures under the various treatments were determined using immunohistochemistry and confocal scanning laser microscopy, and their biochemical properties were evaluated using analysis of total cellular protein content, myofibrillar protein content and SDS-PAGE electrophoretic examination. We compared the effects of 24, 36 and 72 h-long exposures to the various specific Ca-flux modifiers. Increased Ca influx via CaL-channel agonist (Bay K 8644) or via the reversed-mode of the Na/Ca exchanger (veratrine) did not alter the myofibrillar structure or the specific protein profiles or proteosynthesis. However, when cytosolic Ca was increased by three different types of inhibitors of Ca extrusion from the cells via Na/Ca exchange, (Na-free solution, 5 mM NiCl2 and 10(-6) M ouabain), very significant changes in all investigated parameters occurred almost immediately. Twenty-four h-long exposure to Na-free did not affect significantly the total cellular protein (TCP), but the protein synthesis was decreased by 87% and the total myofibrillar protein (TMP) content was decreased by 38%. The myofibrils were heavily fragmented. Similarly, 24 h-long exposure to 5 mM NiCl2 did not affect the TCP, but it reduced protein synthesis by about 90% and decreased the total myofibrillar protein content by 30%. These effects were even more pronounced at 72 h of exposure and they were accompanied with a complete disassembly of myofilaments. Exposure to 10(-6) M ouabain over 72 h resulted in > 80% inhibition of protein synthesis, a 45% decrease in TCP content and a 53% in TMP content. In contrast, 10(-7) M ouabain did not produce any such changes. The changes produced by the Na/Ca-exchange inhibitors were accompanied by only minor changes in DNA content, indicating that the myocytes remained viable. Moreover, these effects were not due to the associated contractile arrest, since exposure to CaL-channel antagonists (5-20 microM nifedipine or 10 microM verapamil) produced only very minor changes in the myofibrillar structure and in protein profiles. Our data demonstrate that short-term (up to 72 h) increased Ca influx or contractile arrest of well-interconnected, spontaneously beating adult cardiomyocytes does not affect their ultrastructural characteristics or their myofibrillar protein turnover greatly, while any situations leading to Ca accumulation (via inhibition of Na/Ca exchange) affect cardiomyocyte function and ultrastructure almost immediately. These data are in sharp contrast to those previously reported from immature, neonatal myocytes.

Postnatal development of the rat intrinsic cardiac nervous system: a confocal laser scanning microscopy study in whole-mount atria.

We used confocal laser scanning microscopy and fluorescent immunohistochemistry to study the developmental pattern and distribution of specific neuronal phenotypes within the intrinsic cardiac nervous system in whole-mount atrial preparations from newborn to 5 week old rats. Individual ganglia and neuronal cell bodies were localized by means of two general neuronal markers: protein gene product 9.5 (PGP) and microtubule-associated protein two (MAP). In rats < or =2 weeks old there were two main subpopulations of intrinsic neurons located in the intraatrial septum and around the origin of the superior vena cava. The more abundant was a population of strongly tyrosine hydroxylase (TH) immunoreactive (IR) neurons (10-40 microm in diameter) most of which were also PGP-IR. The second, less numerous (approximately 60-70% than the TH-IR group) type of neurons exhibited ChAT-IR which colocalized with MAP-IR. Towards the end of the second postnatal week and during the third, the ganglia containing these neurons became more numerous and their localization also included tissues around the origins of the inferior vena cava and the pulmonary veins, as well as both atrial walls close to the AV junction. During the second and third postnatal weeks, when the extrinsic innervation of the adrenergic and cholinergic phenotypes largely increases, the intrinsic innervation also changed greatly, and around the 21st postnatal day it appeared to acquire mature characteristics. The TH-IR neurons changed their characteristics and formed two types of ganglia. The larger ganglia containing large cells (20-40 microm in diameter) expressed TH-IR mostly close to their inner body surface (approximately 80-90% of identified neurons). Most of these neurons also expressed neuropeptide Y (NPY)-IR, specifically around their nuclei. The second type of small strongly TH-IR neurons (approximately 10% of all identified neurons) were contained in smaller groups (20-50 cells) which were usually embedded into much larger ganglia (100-400 cells), containing large (20-50 microm) neurons. Unlike all other intrinsic neurons, these small TH-IR cells did not exhibit any PGP-IR or MAP-IR. The number of ChAT-IR neurons increased at this stage, reaching approximately 90% of the neurons identified by the general neuronal markers. These neurons were surrounded by a rich network of cholinergic varicose nerve fibers, some of which were likely of an extrinsic origin. We have also identified relatively small ganglia expressing immunoreactivity to vasoactive intestinal polypeptide (VIP), and to substance P (SP). The presented data indicate that the phenotypes of intrinsic neurons in the rat heart change greatly during the first month of postnatal development. This may be at least partially related to the development and maturation of functional extrinsic nervous control of the heart.

Distribution of intrinsic cardiac neurons in whole-mount guinea pig atria identified by multiple neurochemical coding. A confocal microscope study.

Functional data indicate that neurons in distinct regions of the heart exert preferential regional cardiac control. To date the regional distribution of specific types of neurons within the intrinsic cardiac nervous system remains unknown, as does their associations with distinct neurotransmitter and/or neuromodulatory profiles. This study was designed to ascertain: (1) the distribution of different classes of neurons within the intrinsic cardiac nervous system as determined by microscopic analysis; (2) the neurochemical profiles of neurons in differing atrial loci; (3) which neurochemicals are co-localized within specific populations of intrinsic cardiac neurons; and (4) the distribution of specific sub-populations of neurons expressing specific immunoreactivities. Taking advantage of confocal laser scanning microscopy and distinct immunoreactive fluorescent markers in various double-label combinations, several sub-populations of intrinsic cardiac neurons were identified. Of all identified neurons, 85-90% were located in ganglia (ganglionic neurons), the rest being isolated (individual neurons). The two general neuronal markers protein gene product 9.5 (PGP 9.5) and microtubule-associated protein (MAP-2) were associated with neurons clustered primarily in the interatrial septum and around the origins of the two vena cavae. Ganglia (group 1) contained three sub-populations of neurons: approx. 80% of ganglionic neurons were large (15-40 microm diameters; group 1a) and approx. 20% had smaller diameters (less than 15 microm; group 1b). All of these neurons were PGP-immunoreactive, exhibiting choline acetyltransferase (ChAT) immunoreactivity (IR), tyrosine hydroxylase (TH) IR, neuropeptide Y (NPY) IR, vasoactive peptide (VIP) IR and substance P (SP) IR. The remaining 5% of ganglionic neurons were small (group 1c; less than 20 microm). These displayed TH immunoreactivity but not MAP, PGP, CHAT, NPY or SP immunoreactivity. Ten to fifteen percent of all neurons loosely distributed outside of ganglia were small (10-25 microm) and located primarily around the origin of the superior vena cava. They displayed immunoreactivity to TH, ChAT, VIP, NPY and SP, but not to MAP-2 or PGP 9.5. These data provide anatomical and immunohistochemical evidence for specific localization of differing populations of intrinsic cardiac neurons with respect to their size, ganglionic distributions and capacity to express multiple neurotransmitters. Although the functional importance of such a regional distribution of differing populations of intrinsic cardiac neurons remains unknown, these anatomical data support the thesis that unique clustering of specific populations of neurons within this nervous system represents the anatomical substrate for complex local cardiac regulatory phenomena occurring at the level of the target organ.

Differences in the structural characteristics of adult guinea pig and rat cardiomyocytes during their adaptation and maintenance in long-term cultures: confocal microscopy study.

In this study, we used laser confocal scanning microscopy and immunofluorescent markers to describe the establishment of long-term (1-5 week) cultures of adult guinea pig cardiomyocytes and adult rat cardiomyocytes. Providing that the preparation of freshly isolated guinea pig cardiomyocytes consists mostly (> 80%) of rod-shaped, Ca(2+)-tolerant, and quiescent cells and these are plated under optimal conditions and density (10(5) cells/cm2), these myocytes have the following characteristics: (1) they remain elongated with regular ultrastructural characteristics and quiescent for 1 week; (2) within 10-14 days, they reestablish intercellular contacts and resume contractile activity, which becomes synchronous all through the confluent layer; (3) their myofibrillar striations remain regular all through the adaptation to culture conditions without any sign of dedifferentiation or redifferentiation; (4) they form adherence junctions (as indicated by their immunoreactivity to an antibody against N-cadherin) over the entire cellular surface; (5) they appear to retain their ability to express atrial natriuretic peptide (ANP), as indicated by immunoreactivity to anti-ANP antibody; (6) this activity seems to be directly related to the surface area of the myocytes in contact with the substrate. In contrast, rat cardiomyocytes cultured under very similar and optimal conditions exhibit very different characteristics during their adaptation in long-term cultures: (1) although 85-90% of freshly isolated cells are also rod-shaped and Ca2+ tolerant they exhibit slow spontaneous contractions; (2) they round up during the first few days, and during the first week they dedifferentiate, losing their regular striated appearance; (3) they spread, becoming irregularly shaped, and, unlike the guinea pig cardiomyocytes, they do not form confluent layers, no matter what the plating density is; (4) they atrophy at a very early stage in the cultures, so that by the fourth week, they have lost most of their myofibrils; (5) the initial rounding up is largely eliminated by exposure for 24 h to 1 microM ryanodine or 20 mM butanedione monoxime, compounds that suppress the spontaneous contractions. In conclusion, our studies demonstrate that adult guinea pig cardiomyocytes adapt and survive in long-term (1-5 week) cultures much better than do adult rat cardiomyocytes, indicating that the long-term cultures of adult guinea pig ventricular myocytes provide a valuable experimental model which opens new possibilities for studying the cellular and molecular regulation of myocardial function under the acute or chronic influence of various intrinsic and/or extrinsic factors.

Further studies of the mechanism(s) of polyunsaturated-fatty-acid-mediated increases in intracellular cAMP formation in N1E-115 neuroblastoma cells.

Following earlier observations that increasing the polyunsaturated fatty-acid (PUFA) content of N1E-115 neuroblastoma cells elevated basal and adenosine (Ado)-stimulated intracellular cyclic AMP (cAMP) formation, we carried out studies to determine the mechanism(s) by which PUFA exerted their modulatory effects. Basal increases in cAMP in the PUFA-enriched (PUFA+) cells were evident with short (60 sec) exposure to a phosphodiesterase inhibitor (Ro 20-1724), and increased to a maximum at 20 min; they were not observed in the absence of Ro 20-1724. Forskolin-stimulated cAMP formation in the presence of the Ro compound was 2- to 3-fold higher in the PUFA+ cells. Basal elevations in cAMP were reduced by approximately 70% by exposing the PUFA+ cells to Ado deaminase (ADA) or to an Ado antagonist, and were further increased by inhibiting ADA, which suggested that they could be producing endogenous Ado that activated stimulatory Ado receptors. However, this did not appear to involve PUFA-mediated stimulation of 5'-nucleotidase activity or inhibition of [3H]Ado uptake. Overall, the results of this study indicated that multiple mechanisms are involved in PUFA modulation of cAMP formation.

Non-eicosanoid functions of essential fatty acids: regulation of adenosine-related functions in cultured neuroblastoma cells.

Studies have demonstrated that augmenting the omega 6 polyunsaturated-fatty-acid (PUFA) content of N1E-115 neuroblastoma cells by media supplementation with linoleic acid results in greater than or equal to 2-fold increases in basal levels of intracellular cyclic AMP (cAMP). Data suggested some involvement of increased production of adenosine from endogenous metabolites; however, increases in adenosine were not related to increased activity of 5'-nucleotidase or decreased uptake of extracellular adenosine. PUFA-dependent elevations in basal cAMP were evident within 1 min of exposure to a phosphodiesterase inhibitor; this phenomenon did not appear to be due to PUFA-dependent changes in Ca2+ uptake or to increases in sensitivity of adenylate cyclase to Ca2+. Forskolin-stimulated cAMP formation was 3-fold higher in PUFA-enriched cells than in control cells, which suggested a direct effect on the functioning of the catalytic unit. Linoleic acid supplementation resulted in a 2-fold increase in the maximum amounts of cAMP produced in response to the stable adenosine analogue, 5'-N'ethylcarboxy-amidoadenosine (NECA). The altered stimulatory response did not involve eicosanoid formation, but may have been related to an increase in the number of stimulatory adenosine receptors, as judged by binding of [3H]NECA. These studies indicate that membrane PUFA modulate adenosine-related functions in neuroblastoma cells, and suggest that a complex series of mechanisms is involved in this regulation.

Adenosine-dependent regulation of cyclic AMP accumulation in primary cultures of rat astrocytes and neurons.

The regulation of intracellular cyclic AMP (cAMP) formation by adenosine (Ado) and its analogues has been examined in primary cultures of rat-brain astrocytes and neurons. In the presence of the phosphodiesterase inhibitor, Ro 20-1724, basal levels of cAMP ranged from 40-120 pmol/mg protein in both cell types. Levels were not altered by treating the cells with Ado deaminase, which suggested that they did not produce appreciable amounts of endogenous Ado under standard culture conditions. In the astrocytes, microM quantities of agonists increased cAMP up to 30-fold higher than basal values; the relative potencies were typical of an A2 Ado receptor (NECA greater than Ado greater than R-PIA). Neuron-enriched cultures exhibited a maximum fourfold increase in cAMP in response to NECA; this was decreased a further eightfold when the cultures had prolonged exposure to the antimitotic agent, c-Ara, to eliminate greater than 98% of the nonneuronal cells. Low (nM) amounts of the Ado agonists inhibited cAMP formation in both cell types. In the astrocytes, the order of potency of inhibition of isoproterenol-stimulated cAMP formation was typical of an A1 receptor (R-PIA greater than Ado greater than NECA); maximum inhibition was 55-65%. Isoproterenol did not increase cAMP in the neuronal cultures. However, forskolin-stimulated formation was effectively (approximately 50%) inhibited by A1 Ado agonists; inhibition was not affected by prolonged treatment with c-Ara. From this study we tentatively concluded that rat astrocytes and neurons both contain inhibitory A1 Ado receptors, but that the stimulatory "A2" subtype is localized mainly on astrocytes.

Effects of membrane polyunsaturated fatty acids on adenosine receptor function in intact N1E-115 neuroblastoma cells.

We have examined the effects of increasing membrane polyunsaturated fatty acids (PUFAs) on adenosine receptor function in intact N1E-115 neuroblastoma cells. Addition of linoleic acid to the culture medium for 48 h resulted in an approximate threefold increase in the amount of omega 6 fatty acids esterified to membrane phospholipids. Basal cAMP accumulation was significantly higher in the PUFA-enriched cells than in controls, although the differences could be diminished by approximately 75% by treatment of the cells with adenosine deaminase or 8-phenyltheophylline. Exposure of the cultures to the stable adenosine analogue 5'-N-ethylcarboxyamide adenosine (NECA) resulted in concentration-dependent increases in cAMP accumulation. Data from saturation experiments indicated that the maximum amount of cAMP that could be formed in response to NECA in the PUFA-enriched cells was twice that in control cells. Also, the amount of agonist required to elicit half maximal stimulation in the supplemented cells was significantly less than in the control cells (mean values for EC50, 0.85 and 1.43 microM, respectively). The results of this study demonstrate that membrane PUFA have the ability to modify interactions between adenosine receptors and adenylate cyclase in neural cells, a fact that is of potential importance in considering the central role that adenosine plays as a neuromodulator in the nervous system.

Effects of adenosine analogues on basal, prostaglandin E1- and forskolin-stimulated cyclic AMP formation in intact neuroblastoma cells.

We have examined the effects of R-phenylisopropyladenosine (R-PIA) and other adenosine analogues on basal, prostaglandin E1 (PGE1)- and forskolin-stimulated cyclic AMP (cAMP) formation in intact N1E-115 neuroblastoma cells, to determine whether the cells contain A1 adenosine receptors that are negatively coupled with adenylate cyclase. Basal levels of cAMP (68 +/- 7 pmol/mg protein; mean +/- SE, N = 15) were not altered by low concentrations of R-PIA. The apparent lack of inhibition was not due to increases in cAMP due to activation of a stimulatory A2 receptor by endogenously-synthesized adenosine. By comparison, low levels of R-PIA did reduce significantly (P less than 0.05) PGE1-dependent increases in cAMP formation (maximum response to PGE1, 972 +/- 77 pmol cAMP/mg protein; EC50 for PGE1, 0.2 microM). Inhibition was dose dependent, and resulted in a 30-50% maximum reduction in production stimulated by PGE1. Nanomolar concentrations of R-PIA elicited half-maximal inhibition; the inhibitory response was blocked by 8-phenyltheophylline (8-PT). The order of potencies of several adenosine analogues in eliciting this response suggested that inhibition was mediated by an A1 adenosine receptor. Examination of the effects of R-PIA on forskolin-stimulated cAMP formation yielded several interesting findings. First, stimulation by the diterpene by itself was blocked by both adenosine deaminase (ADA) and 8-PT (40 and 25% inhibition respectively). Low concentrations of R-PIA (less than 10(-6) M) had no effect on forskolin-stimulated cAMP production. At higher levels (greater than or equal to 10(-6) M) the analogues acted synergistically with the diterpene, to yield cAMP levels that were up to 3-fold higher than the additive effect of the two agents. Potentiation was stereospecific, Ca2+ dependent, and was blocked by 8-PT. The results of this study suggest that, in N1E-115 neuroblastoma cells, inhibitory A1 receptors are not stimulated in response to non-specific elevations in cAMP, but are associated with specific stimulatory receptors such as those activated by PGE1.