The apolipoprotein(a) promoter contains a retinoid response element.

Retinoids were previously shown to lower apolipoprotein(a) [apo(a)] mRNA levels, suggesting that the apo(a) promoter contains a retinoid response element (RRE). Scanning the apo(a) promoter for sequences related to the consensus RRE half-site (5'-PuG(G/T)TCA-3') uncovered four sites that could potentially function as RREs at -2915, -1875, -1036, and -407. The activity of these sites was assessed by their ability to compete with a very strong consensus DR5 RRE for binding to retinoic acid receptor (RARalpha) and retinoid X receptor (RXRalpha) heterodimers using electrophoretic mobility-shift assays. Only the -1036 site (5'-TGACCTTGTGATCC-'3) was an effective competitor of the DR5 RRE; therefore, it was designated as apo(a) RRE. Apo(a) RRE competed with DR5 RRE for RARalpha/RXRalpha binding with 1/10 the affinity of DR5 RRE, while a scrambled apo(a) RRE was inactive. These results suggested that apo(a) RRE is a potential candidate for mediating the effect retinoids have on apo(a) mRNA expression.

Preferential addition of newly synthesized membrane protein at axonal growth cones.

The addition of plasma membrane proteins to a growing axon could occur by preferential insertion at the tip (the growth cone), by uniform insertion along the axon, or by insertion at the cell body and bulk flow along the axon. To differentiate between these possibilities we used a defective herpesvirus vector to express an exogenous protein, the lymphocyte transmembrane protein CD8 alpha, in cultured rat hippocampal neurons. The newly synthesized protein first appeared on the axonal surface almost exclusively at the growth cone. Preferential addition at the growth cone was also observed in minor processes (immature dendrites), but not in mature dendrites. Over several hours, CD8 alpha reached a uniform distribution over the entire neuronal surface, presumably by diffusion within the membrane and possibly endocytic recycling. As well as providing materials for axonal growth, the selective addition of membrane vesicles at the growth cone may contribute to the polarized distribution of axonal surface molecules.

Expression of human amyloid precursor proteins in cultured neuronal cells through the use of HSV-1 defective vectors.

Postmortem investigations of Alzheimer's patients reveal senile plaques that contain, among other molecules, deposits of beta-amyloid protein. The role of the beta-amyloid deposits remains unclear but identification of mutations in the amyloid precursor protein (APP) gene within the beta-amyloid portion in hereditary forms of the disease provide evidence that these deposits are involved in the pathological state. To more fully investigate this hypothesis attempts have been made to create transgenic mice to overexpress the beta-amyloid protein but these models have not been successful in modeling the disease. We have chosen to utilize the HSV-1 defective vector system which allows the expression of experimental genes in neuronal cells to overexpress APPC100. We have cloned the rat tyrosine hydroxylase (TH) promoter into a defective HSV plasmid. Cloning the firefly luciferase gene under the control of the TH promoter (demonstrates that the promoter is active after infection of human SY5Y cells or rat PC12 cells. A synthetic APP cDNA which represents the last 100 amino acids of the carboxy terminus of APP including the beta-amyloid protein was synthesized and inserted under the control of the TH promoter. Infection and subsequent nuclease protection assays demonstrate expression of the synthetic gene in the infected cells. Current research focuses on detection of the expressed protein within the infected cells and determination of the time period for continued expression.

Developmentally regulated expression of an exon containing a stop codon in the gene for glutamic acid decarboxylase.

In the adult rat brain, the gene for glutamic acid decarboxylase (GAD; L-glutamate 1-carboxy-lyase, EC 4.1.1.15) is expressed predominantly as a 3.7-kilobase transcript. Earlier data showed that embryonic brain expresses an RNA transcript distinct from the adult form; however, the exact structure of this form was not elucidated. Here, transcripts expressed in the embryonic but not the adult brain were cloned and analyzed. These transcripts include an exon not expressed in the adult inserted into coding sequence. The embryonic exon contains a stop codon that is in-frame with the coding sequence. The exon is found in genomic DNA within the GAD gene where it is flanked by introns with conventional splice sites. On the basis of these structural data, we propose the hypothesis that, early in brain development, transcripts encoding a truncated form of GAD are expressed. The deduced protein cannot function as a decarboxylase because the stop codon in the embryonic exon occurs upstream of the binding site for pyridoxal phosphate, an essential cofactor. Thus, alternative splicing plays a crucial role in the pathway leading to the development of functional GABAergic neurons. The central nervous system-derived cell lines B65 and C6 express a mixture of the adult and embryonic forms of GAD mRNA. They therefore are useful clonal models of central nervous system cells in the early phases of differentiation.

Characterization of a cDNA coding for rat glutamic acid decarboxylase.

cDNA clones have been isolated for rat glutamic acid decarboxylase (glutamate decarboxylase; EC 4.1.1.15) (GAD) and 3216 bp of the sequence have been determined. This sequence extends the previously reported feline GAD cDNA sequence both in the 5' (67 bp) and 3' (887 bp) directions and contains the polyadenylation signal and tail. The cDNA codes for a 67 kDa mol. wt. protein beginning from the putative initiator methionine found in the feline sequence. Extensive homology to feline GAD was identified at the amino acid level (97% identity) within the coding region. This interspecies homology is high compared to other neurotransmitter synthesizing enzymes and suggests selective pressure to maintain the primary sequence throughout the full length of the protein. Homology is found 5' to the putative initiator methionine. Extensive stretches of homology are also found in the 3' non-coding region. These conserved non-coding regions may play a role in GAD mRNA regulation. The rat cDNA sequence will facilitate investigations into the structure and regulation of the GAD gene.

Guanine nucleotide regulation of the interconversion of the two-state hepatic glucagon receptor system of rat.

To investigate whether guanine nucleotides regulate interconversion of the two-state hepatic glucagon receptor we have utilized kinetic assays of glucagon binding to partially purified rat liver plasma membranes. Dissociation of glucagon at 30 degrees C exhibited biexponential character in either the absence or presence of GTP, indicating that the system previously seen in intact hepatocytes is independent of intracellular modulators. In each case the receptors underwent a time-dependent conversion from a low affinity to a high affinity state. However, GTP decreased the fraction of receptors in the high affinity state. The rank order for stabilizing the low affinity state was Gpp(NH)p greater than GTP greater than GDP much greater than GMP = no nucleotides. Data from competition binding assays with increasing concentrations of GTP allow calculation of equilibrium constants which are 3.32 nM for glucagon and receptor in the absence of GTP, 18.6 nM for glucagon and receptor in the presence of GTP, 1.55 microM for the association of receptor and GTP presumably linked to an N protein, and 8.86 microM for the association of the glucagon-receptor complex and GTP again presumably linked to an N protein, Glucagon binding to receptor is noncooperative in both the absence and presence of GTP, distinguishing this system from the beta-adrenergic system. With GTP, binding to the low affinity state is favored because of the relative affinities reported. Therefore, GTP regulates the activation by slowing the conversion of the receptor from a low affinity to high affinity form.

Partial agonism in the glucagon receptor system is a consequence of the two-state rat hepatic receptor.

We have previously demonstrated that the glucagon receptor binds hormone to form a low affinity complex which, by a time- and temperature-dependent mechanism, is converted to a high affinity complex (Horwitz, E.M., Jenkins, W.T., Hoosein, N.M., and Gurd, R.S. (1985) J. Biol. Chem. 260, 9307-9315). In this report we have investigated the effects of agonist concentration, potency, and intrinsic activity on the characteristics of the two, interconvertible states of the glucagon receptor. As the glucagon concentration is increased from 0.02 to 0.50 nM, the initial velocity of binding increases. The conversion of a low affinity to a high affinity complex is the rate-limiting step in the overall binding reaction and approaches its maximal velocity as the hormone concentration exceeds 0.20 nM. At equilibrium, 87-90% of the hormone-receptor complexes are in the high affinity state at all hormone concentrations examined. [S-methyl-Met27]glucagon, a full agonist with reduced potency, binds to the two-state system in a manner analogous to that of native glucagon. The binding of N alpha-biotinyl-N epsilon-acetimidoglucagon, a partial agonist with reduced potency, effects a two-state system where the high affinity state accounts for only 35% of the total hormone-receptor complexes at equilibrium. We conclude that the formation of the high affinity complex is the rate-limiting step involved in glucagon binding; reduction in binding potency with full agonism is due to a reduction in the affinity of the ligand for the unoccupied receptor and not to an alteration of the interconversion of the two states, and decreased intrinsic activity is due to a quantitative decrease in conversion of the low to high affinity state.