Faculty Bibliography

beta 1 Integrin-like immunoreactivity was localized in rat brain using a polyclonal antibody raised in rabbit against rat liver beta 1 integrin. One-dimensional immunoblotting of whole rat brain membranes indicated that this antiserum recognized a single molecular species at 116,000 M(r), indicative of rat beta 1 integrins. Specific staining of beta 1 integrin-like immunoreactivity was found in the vascular structures of the brain, including microvessels, the ventricular ependymal cells, and pia mater. The pineal gland stained densely, and diffuse staining was present throughout the gray matter of the brain. This diffuse staining had a patterned appearance in certain structures, such as the apical dendritic field of CA1 in hippocampus, and occasional labeling of astrocytes, verified by labeling with GFAP on adjacent sections, was noted.

In this study, the role that active tension development plays in the formation and maintenance of cardiac myocyte myofibrillar structure and cellular shape was investigated. By use of the calcium channel blocker verapamil, spontaneous contractile activity of neonatal rat heart myocytes was inhibited for 24 to 96 hours. Confocal microscopy of rhodamine phalloidin-stained cells revealed that, within 24 hours of contractile arrest, actin filaments of myofibrils were no longer aligned with one another at their I bands and Z lines. Cellular shape was also affected, with the cells developing a less stellate appearance while remaining attached to the substrate as well as to one another. By 48 hours, actin fibrils were largely absent from these cells. The disappearance of actin was confirmed by measurements of actin synthesis and accumulation rates and by pulse-chase biosynthetic labeling experiments. It was revealed that, although actin synthesis was significantly reduced in arrested myocytes, the rapid disappearance of total cellular actin was largely due to increased rates of actin degradation. Contractile arrest produced by L-type calcium channel blockade with verapamil (or other calcium channel blockers) accelerated actin degradation to a greater extent than K+ depolarization. Chloroquine partially suppressed the accelerated rate of actin degradation, indicating that lysosomal proteolysis may be involved in actin degradative processing. Protein kinase C activation also partially inhibited the accelerated rate of actin degradation but did not restore actin filaments in arrested myocytes. The reformation of actin fibrils and their reassembly into striated myofibrils occurred when contractile activity was restored by removal of verapamil from the culture medium. The period of time required for myocytes to reassemble actin filaments and to regain their elongated morphology was proportional to the period of time that the cells were inhibited from contracting. Data are presented to indicate that active tension development by neonatal cardiac myocytes in culture is critical to the maintenance of filamentous actin structure via mechanisms involving actin assembly, disassembly, and degradation.

Significant physiological changes occur in the heart following birth including increased arterial blood pressure and heart rate. Concurrently, biochemical and structural alterations are evident in the neonatal heart in response to these dynamic physiological properties. Prominent among these is the elaborate development of the cardiac extracellular matrix, composed primarily of interstitial collagen. The collagenous fibers, together with other matrix components, form an elastic, stress-tolerant network which functions in the dissipation of force throughout the heart wall. The present studies have used biochemical and molecular techniques to show the temporal and spatial patterns of interstitial collagen accumulation and expression during late fetal and neonatal development of the rat heart. The use of biochemical and particularly molecular methodologies allows the analysis of the expression of matrix components at a resolution previously not attained by structural studies alone. These data show relative increases in interstitial collagen immediately following birth as well as spatial differences in collagen mRNAs within the heart. The data presented provide further evidence for a role of mechanical stimulation in the regulation of collagen gene expression during this period of heart development.

To investigate the origin and development of the cardiac conduction system, the distribution of HNK-1 immunoreactivity in embryonic rat hearts was studied in histological sections and in three-dimensional computer reconstructions. Earliest HNK-1 reactivity was found along the endocardial surface of the fusing tubular heart at 9.5 embryonic days (ED) and subsequently within individual myocytes scattered widely along the looped tubular heart. Immunopositive myocytes appeared along the earliest ventricular trabeculae as they coalesced to form the developing interventricular septum during day 11, spreading to either side to give rise to the right and left bundle branches in the 12.5 ED heart. In the venous pole of the heart, primordia of the sinus node, and of the transient left sinus node, appeared immunopositive from 12.5 ED, coalescing during ED 13 along the anterior wall of the right sinus horn or developing coronary sinus, respectively. In the atria, several distinct tracts of immunoreactive myocytes were defined by 14.5 ED, ramifying from the sinoatrial junction to the atrial appendages or to the atrio-ventricular (AV) junction near the AV node. The timing and distribution of these immunostaining patterns suggest that ventricular conduction tissue develops within the earliest trabecular and septal myocardium, and is distinct from later immunopositive atrial tracts and extracardiac cell populations, such as neural crest, that appear to contribute to formation of the sinus node and autonomic innervation of the heart.

Cell-matrix interactions probably play a cooperative role with cell-hormone interactions to ensure normal differentiation of the adenohypophysis. The extracellular matrix (ECM) surrounding adult adenohypophysial cells contains laminin but its embryonic development has not been described. This study was carried out to test the hypothesis that adenohypophysial cells are associated with components in the ECM prior to cellular differentiation in the adenohypophysis. Fetuses were removed from Golden Syrian hamsters every 12 h from embryonic days 8.5-14. Coronal and sagittal 5-microns-thick sections of paraplast-embedded embryos were stained for fibronectin, laminin, or collagen IV using avidin-biotin immunoperoxidase staining. The adenohypophysial anlage was observed initially as a group of epithelial cells (Rathke's pouch) in the roof of the stomatodeum in contact with a basement membrane. The basement membrane stained positively for laminin, collagen IV and fibronectin. Lighter staining for laminin, fibronectin and collagen IV was observed between the developing adenohypophysial cells in Rathke's pouch. Proliferative activity was apparent in the antero-inferior region of Rathke's pouch and resulted in the formation of the bulk of the adenohypophysis. Mesenchyme infiltrates the region between the base of Rathke's pouch and the oral epithelium, thus separating the two. The basement membrane surrounding the pouch appears to become discontinuous in the regions of high proliferative activity. These results show that ECM components appear early during the development of adenohypophysial cells prior to their cellular differentiation into hormone-containing cells. This association between ECM components and developing adenohypophysial cells provides the anatomical basis for cell-ECM interactions to influence adenohypophysial development and differentiation.

During development, extracellular matrix (ECM) molecules are thought to play a major role in regulating the formation of the heart. The change in the heart from a simple tube to a complex, four-chambered organ requires the modification of both the cellular components as well as the surrounding ECM. Matrix metalloproteinases (MMP), which include collagenases, are enzymes present in the ECM that have the potential to modify the existing ECM during the development of the heart. Using both monoclonal and polyclonal antisera against collagenase, specific temporal and spatial patterns have been documented during critical periods of heart development. The cytokine interleukin 1 alpha (IL-1 alpha), a potent inducer of the MMP expression, was also shown to have a similar staining pattern in the developing heart. The monoclonal anti-rat collagenase (Mab) intensely stained the surfaces of the myocytes in the trabeculae and the ventricular and atrial walls of the 11.5 or 12.5 embryonic day (ED) rat hearts. In contrast, the polyclonal anti-human collagenase (Pab) stained not only the cardiomyocytes but also the hypertrophic endocardial cells. Pab appeared to stain the leading edge of the mesenchymal cells that migrate into the cardiac jelly of the 11.5 or 12.5 ED hearts. Immunohistochemical staining showed IL-1 alpha on the endocardial endothelium and the surface of cardiomyocytes near the cardiac jelly just before or coincident with the appearance of migrating cells. IL-1 alpha was detected on the endocardial endothelium, cardiomyocytes in the trabeculae, and the ventricular and atrial walls, as well as in the myocardial basement membrane of the truncal or atrioventricular region. However, no staining could be detected on the migrating cells in the cardiac cushions. These results indicate the presence of collagenase and IL-1 alpha on the surface of cardiomyocytes and mesenchymal cells at times when the heart is undergoing acute remodeling during septation and trabeculation. These data suggest a role for collagenase/cytokine interaction in tissue remodeling during critical stages of cardiac embryogenesis where modification of the ECM is essential to cardiac morphogenesis.

The temporal and spatial distribution of the basement membrane component laminin was examined in vivo in developing rat hearts at 11.5 and 15 days of embryonic development (ED), and in neonates and adults, by pre-embedding ultrastructural immunocytochemistry. In addition, the patterns observed at 11.5 days ED were compared to the distribution of laminin in embryos maintained in whole-embryo culture. At 11.5 days ED laminin was localized in punctate patches on the surface of the plasma membrane, with large gaps between areas of staining. The development of myocytes and localization of laminin in the whole embryo-cultured embryos was similar to that found in the in vivo embryos. At 15 days ED, laminin localization was limited to distinct patches of developing extracellular matrix material associated with the sarcolemma. Gaps between areas of localization were shorter than in the 11.5-day hearts. In neonates, distribution of laminin localization was more extensive with fewer gaps and was associated with the developing basement membrane. In adult hearts, laminin was localized along the entire length of the basement membrane and was heaviest in areas of morphological specialization, such as Z-bands, where collagen bundles contacted the sarcolemma.

Ubiquitin, a highly conserved 76-residue protein found in all eukaryotic cells, can be covalently bound to a wide variety of proteins in the nucleus, cytosol, cytoskeleton, and plasmalemma. This diversity of target proteins reflects a diversity of functions for ubiquitin conjugation. Previous studies have showed enhanced localization of ubiquitin conjugates to Z-bands of normal skeletal muscle and increased ubiquitination in atrophic muscles. These results have implicated a ubiquitin-mediated pathway in protein turnover and degradation in striated muscle. To investigate whether such a pathway might also exist in cardiac striated muscle, we used an affinity-purified polyclonal antibody (conjugate specific) and indirect immunofluorescence to localize ubiquitin conjugates in neonatal and adult rat cardiac myocytes both in vitro and in vivo. In both cultured myocytes and heart tissue, fluorescent ubiquitin conjugates were found in the nucleus as aggregates, in the cytoplasm in a striated pattern indicative of Z-bands, and in intercellular junctions at the intercalated discs between myocytes. Although the acceptor proteins and the physiological significance of ubiquitination at these locations are unknown, the targeting of ubiquitin to specific sites within the nucleus, myofibrils, and sarcolemma could provide a means for selective processing of individual components within these larger macromolecular assemblies, thus implying a regulatory role for ubiquitin conjugation in turnover or stability of proteins in the heart.

These investigations tested the hypothesis that secretion of prolactin (PRL), LH, and FSH in vitro is influenced by the substratum on which adult or fetal adenohypophyseal cells are cultured. Adenohypophyses were removed from adult male Golden Syrian hamsters and from fetal hamsters on day 16 of gestation. The glands were dissociated and cultured in a 1:1 mixture of Dulbecco's Modified Eagle's Medium (DMEM) and Ham's F-12 medium containing 10% fetal bovine serum (FBS), 25 mM Hepes, and antibiotics. The cells were cultured on three substrata: glass, laminin, and the reconstituted basement membrane of the Engelbreth-Holm-Swarm (EHS) tumor (Matrigel). Medium was collected and replaced every 48 h for 14-22 days. Concentrations of PRL, LH, and FSH in medium were measured by RIA. The substratum influenced hormone secretion. PRL concentrations were elevated in cultures of adult cells on Matrigel in each of four experiments. Adenohypophyseal cells on Matrigel maintained a rounded shape longer than cells on glass or laminin. In studies using fetal adenohypophyseal cells, PRL concentrations were elevated significantly in medium from cultures on Matrigel at and after 2 days as were concentrations of LH and FSH after 6 days. Additional experiments showed that the higher PRL concentrations in medium surrounding adult cells plated on Matrigel were not due to the release of soluble factors from Matrigel, differential cell attachment on Matrigel, the differential presence of adenohypophyseal fibroblasts, nor differential rates of cell proliferation. The results show that Matrigel maintains the secretion of PRL from adult adenohypophyseal cells in vitro more effectively than glass or laminin substrata and support the hypothesis that cell-matrix interactions mediate the observed differences. The results also show that in long-term cultures (14-22 days), fetal adenohypophyseal cells secrete significantly more PRL, LH, and FSH on Matrigel than they secrete when cultured on glass or laminin. Thus, Matrigel influences the function and possibly the maturation of adenohypophyseal cells in vitro. Furthermore, although laminin is the most abundant component in Matrigel, the effects of Matrigel on lactotrophs and gonadotrophs in vitro are probably not attributable solely to its laminin content.