Eva Cukerman

Impaired Heart Contractility in Apelin Gene–Deficient Mice Associated With Aging and Pressure Overload

Apelin constitutes a novel endogenous peptide system suggested to be involved in a broad range of physiological functions, including cardiovascular function, heart development, control of fluid homeostasis, and obesity. Apelin is also a catalytic substrate for angiotensin-converting enzyme 2, the key severe acute respiratory syndrome receptor. The in vivo physiological role of Apelin is still elusive. Here we report the generation of Apelin gene–targeted mice. Apelin mutant mice are viable and fertile, appear healthy, and exhibit normal body weight, water and food intake, heart rates, and heart morphology. Intriguingly, aged Apelin knockout mice developed progressive impairment of cardiac contractility associated with systolic dysfunction in the absence of histological abnormalities. We also report that pressure overload induces upregulation of Apelin expression in the heart. Importantly, in pressure overload–induced heart failure, loss of Apelin did not significantly affect the hypertrophy response, but Apelin mutant mice developed progressive heart failure. Global gene expression arrays and hierarchical clustering of differentially expressed genes in hearts of banded Apelin−/y and Apelin+/y mice showed concerted upregulation of genes involved in extracellular matrix remodeling and muscle contraction. These genetic data show that the endogenous peptide Apelin is crucial to maintain cardiac contractility in pressure overload and aging.

A Genome-Based Resource for Molecular Cardiovascular Medicine Toward a Compendium of Cardiovascular Genes

BACKGROUND Large-scale partial sequencing of cDNA libraries to generate expressed sequence tags (ESTs) is an effective means of discovering novel genes and characterizing transcription patterns in different tissues. To catalogue the identities and expression levels of genes in the cardiovascular system, we initiated large-scale sequencing and analysis of human cardiac cDNA libraries. METHODS AND RESULTS Using automated DNA sequencing, we generated 43,285 ESTs from human heart cDNA libraries. An additional 41,619 ESTs were retrieved from public databases, for a total of 84,904 ESTs representing more than 26 million nucleotides of raw cDNA sequence data from 13 independent cardiovascular system-based cDNA libraries. Of these, 55% matched to known genes in the Genbank/EMBL/DDBJ databases, 33% matched only to other ESTs, and 12% did not match to any known sequences (designated cardiovascular system-based ESTs, or CVbESTs). ESTs that matched to known genes were classified according to function, allowing for detection of differences in general transcription patterns between various tissues and developmental stages of the cardiovascular system. In silico Northern analysis of known gene matches identified widely expressed cardiovascular genes as well as genes putatively exhibiting greater tissue specificity or developmental stage specificity. More detailed analysis identified 48 genes potentially overexpressed in cardiac hypertrophy, at least 10 of which were previously documented as differentially expressed. Computer-based chromosomal localizations of 1048 cardiac ESTs were performed to further assist in the search for disease-related genes. CONCLUSIONS These data represent the most extensive compilation of cardiovascular gene expression information to date. They further demonstrate the untapped potential of genome research for investigating questions related to cardiovascular biology and represent a first-generation genome-based resource for molecular cardiovascular medicine.

Intermittent hyperinsulinaemia and arterial glycosaminoglycans in dogs.

SummaryThe effect of insulin on the glycosaminoglycan content of the arterial ground substance was compared in age-matched alloxan-diabetic dogs, hypophysectomized dogs and normal controls. Hyaluronic acid and the three sulphated components, heparan, dermatan and isomeric chondroitin sulphates were quantitatively determined in the arch, thoracic and abdominal aorta, and in the carotid, coronary, iliac, renal and mesenteric arteries. Diabetic animals were either well or poorly controlled. Six well controlled dogs with a fasting mean plasma glucose level of 8.5 mmol/l were given around 30 units of porcine insulin/day and showed fluctuations in plasma insulin concentrations between 24 and 175 mU/l. Four poorly controlled dogs with a mean fasting plasma glucose level of 18.5 mmol/l received an average of 14 units of insulin/day, and the fluctuations ranged from 8 to 113 mU/l. Eight normal, untreated controls showed mean fluctuations between 15 and 22 mU/l. Within 14 weeks of insulin treatment, well controlled animals displayed glycosaminoglycan alterations in five arterial segments, usually involving more than one glycosaminoglycan constituent. In poorly controlled animals the number of segments that showed glycosaminoglycan alterations was the same, but abnormalities were limited to one of the sulphated components only. The coronary arteries displayed identical glycosaminoglycan alterations in the two groups of diabetic dogs, namely a significant rise in dermatan sulphate content (p < 0.01, mean±SEM) from the normal value of 1.12±0.03 mg/g dry defatted tissue to 1.31±0.03 mg/g in well controlled animals and to 1.37±0.08 mg/g in poorly controlled animals. In order to ascertain that the abnormalities in the glycosaminoglycan chemistry were related to hyperinsulinaemia rather than hyperglycaemia, six hypophysectomized dogs were treated with 1.5 U/day of porcine protamine zinc insulin, for 3 weeks. Average plasma glucose levels were in the order of 4.3 mmol/l and plasma insulin levels fluctuated between 9 and 27 mU/l; the latter represented twice the peak value seen in five untreated hypophysectomized dogs. The resulting chemical changes in the glycosaminoglycan content were in line with those encountered in diabetic animals, including the rise in dermatan sulphate content of the coronary arteries. These results indicate that: (1) hyperinsulinaemia produced by injections of insulin causes alterations of the arterial glycosaminoglycan content; (2) not all segments of the arterial tree are equally responsive to insulin and (3) the coronary arteries have a particularly insulin-sensitive dermatan sulphate metabolism.

Construction of a human heart cDNA library and identification of cardiovascular based genes (CVBest)

The availability of high quality cDNA libraries is often crucial to the successful identification and characterization of genes. The concepts and potential pitfalls of constructing cDNA libraries are presented. Various applications requiring high quality cDNA libraries are outlined, including large-scale single pass sequencing of cDNA clones to generate expressed sequence tags (ESTs) and differential screening of cDNA libraries. The usefulness of combining such approaches for the discovery of novel disease-related and cardiovascular-based ESTs (CVBest) is discussed.

Arterial Glycosaminoglycans in Diabetic Dogs

The glycosaminoglycan (GAG) composition of a number of large and medium-sized arteries was studied in 6 alloxan-diabetic beagles and was compared with 6 normal, age-matched controls. Diabetic animals were maintained on diet and insulin for 100 days. The aortic arch, thoracic and abdominal segments, external iliac, superior mesenteric, renal, common carotid and coronary arteries were analyzed for hyaluronic acid (HA) and for heparan (HS), dermatan (DS), and chondroitin (CS) sulphates. All diabetic dogs displayed significant alterations. The HA content was reduced in iliac arteries, and together with HS, also in the thoracic aorta. HS or CS were increased in carotid, iliac and renal arteries, DS, a GAG constitutent with very high affinity for low density lipoproteins, was significantly increased in coronary arteries alone. 2 additional animals which are excluded from this series did not become diabetic after alloxanization and showed no change in arterial GAG content. Early changes in the chemistry of the arterial ground substance seem to provide a clue to the precocious development of atherosclerotic disease in diabetes.

The relationship of hormones to arterial glycosaminoglycans and atherosclerosis.

Glycosaminoglycan fractions were measured in representative large and medium sized arteries of normal, hypophysectomized and hormone treated young beagles. Hyaluronate, heparan sulphate, dermatan sulphate and the isomeric chondroitin sulphates were determined in the aortic arch, thoracic and abdominal segments, in the external iliac, superior mesenteric, renal, common carotid and coronary arteries. The hormones used for replacement therapy of hypophysectomized animals were growth hormone, thyroxine, cortisone and the sex hormones testosterone, estrogen and progesterone. The sensitivity to an individual hormone was found to differ in various segments of the arterial tree; the thoracic and abdominal aorta were most responsive but renal and superior mesenteric arteries were relatively inert. The hypothesis is advanced that arteries with a GAG metabolism highly sensitive to hormones are more prone to develop atherosclerosis than arteries that have a limited sensitivity to alterations in endocrine balance.

Localization of a novel zinc finger gene to the human chromosome 7p11.2-p12 by fluorescence in situ hybridization.

A novel zinc finger gene (ZNF182) was isolated from a human fetal cardiac cDNA library, using a consensus C2H2 zinc finger oligonucleotide probe. This gene was assigned to human chromosome 7p11.1-p12 by fluorescencein situ hybridization (FISH). Additional FISH signals were identified on both the long and the short arms of chromosome 19, suggesting the presence of homologous genes at these loci.

Identification of a locus of zinc finger genes in human chromosome 19q13.1-q13.3 region by fluorescencein situ hybridization

A group of zinc-binding cDNA clones from a human fetal heart library was isolated using an oligonucleotide probe to the consensus sequence of the linker region of zinc finger proteins. Genes for novel clones were mapped by fluorescencein situ hybridization. In the process, we identified a previously unrecognized locus for two zinc finger-coding genes in human chromosome 19q13.1-q13.3 (ZNF180, ZNF181), where genomic rearrangements were shown to be accompanied by various developmental abnormalities, DNA repair deficiencies, and cellular malignancies.

Somatostatin Mediates the Effect of Growth Hormone Surges on Splanchnic Biogenic Amines

Abstract Experiments were conducted in trained, conscious dogs fitted with an indwelling portal catheter. Radioenzymatic methods were employed for the quantitative measurement of plasma-free serotonin and catecholamines. An injection of ovine growth hormone (GH, 100 μg/kg) or an equimolar amount of somatostatin (somatotropin release inhibitory factor, SRIF, 7.5 μg/kg) into a saphenous vein led, within the first 15 min, to a transient but significant increase in plasma serotonin and a decrease in the concentrations of dopamine, norepinephrine, and epinephrine. The changes were frequently in excess of 40% of baseline values, and were found only in the portal and not in the peripheral circulation. When the animals were pretreated with an antiserum specifically directed against SRIF, GH surges no longer caused alterations in the portal levels of biogenic amines. Thus, the effects of spike concentrations of GH on plasma serotonin and catecholamines are apparently mediated by SRIF, a novel and unexpected function for a hormone that is known as an inhibitor of GH secretion.

Effect of Hormones on the Glycosaminoglycan Content of Canine Renal Arteries

The effect of hypophysectomy and replacement therapy with individual hormones such as somatotropin, cortisone, estradiol, progesterone and testosterone on the glycosaminoglycan (GAG) composition of renal arteries was studied in young beagles. The renals from normal animals showed a uniformly low total GAG content of about two thirds of the value found for aorta and coronary arteries. Neither hypophysectomy nor hormone replacement altered the GAG content, with the exception of estradiol which showed a marked effect by raising the hyaluronic acid content to nearly double the normal value. These results by being at variance with those obtained for aorta and coronary arteries provide further evidence of the differential hormone sensitivity that exists in the various segments of the canine arterial tree with respect to GAG metabolism.