Wheat germ agglutinin (WGA) belongs to a family of dimeric chitin binding lectins specific for N-acetyl-D-glucosamine (GlcNAc) and N-acetyl-D-neuraminic acid (NeuNAc). The polypeptide chain consists of a tandem repeat of four conserved 4.3 kDa domains (A, B, C, D) stabilized by four disulfide bonds. Saccharide binding occurs in the dimer interface where domains of different subunits are in contact (A with D, B with C). An aromatic amino acid rich pocket on one domain (sufficient for saccharide binding) and a polar region on the contacting domain constitute a complete binding site. Saccharide binding affinities may differ among the four unique sites (eight/dimer) due to sequence divergence. Two equivalent sites/monomer were detected in solution. However, conclusive evidence is lacking as to their locations on the dimer. To delineate individual specificities and dimerization requirements, it was desirable to isolate and characterize each site independently. This thesis describes an expression system by which single WGA domains can be efficiently generated as functional proteins. The B-domain was cloned first, because binding was observed at this site in several WGA-oligosaccharide crystal complexes. Two B-domain sequences were cloned varying at residue 28 (Ala→Ser). In a putative domain dimer the polar Ser28 would mimic an H-bond observed to stabilize NeuNAc in the WGA B-site from the contacting C-domain. The domains were expressed as fusion proteins from which they were proteolytically separated and isolated in high yields. The recombinant domains were shown to associate with chitin (poly-GlcNAc). The correct tertiary structure was inferred by saccharide binding ability and antibody recognition. All cysteines were found to be in disulfide linkages. Isothermal titration calorimetry showed that (GlcNAc)3&4, binding to both B-domain mutants is seven-fold weaker than to WGA (Kd3.5x 10-4M versus 0.54x 10-4M). Binding of N-acetylneuraminyl-lactose was undetectable. Gel filtration, Mass spectral and NMR analysis indicated that the recombinant domains exist as monomers in solution. Thus, the complete WGA binding site was not reproduced and the low affinity reflects only the interactions of the saccharide with the aromatic pocket.

It was hypothesized that the C-terminal sequences for mutant p53 would be required for oligomerization. and oligomerization may be critical for gain-of-function. An N-terminal deletion mutant of p53 that deletes amino acids 1-293 was used as a tool to perform hetero-oligomerization studies. This mutant retains the entire oligomerization domain but dispenses off the transactivation domain and a large portion of the sequence- specific DNA-binding domain. Co-transfection experiments show that p53 del. 1-293 forms hetero-oligomeric complexes with p53-D281G. Also. co-expression of p53 del. 1- 293 with p53-D281G inhibited p53-D28lG-mediated transactivation of the EGFR and MDRl promoters suggesting that hetero-oligomerization inactivates transcriptional functions of mutant p53. The interaction of p53 deli 1-293 and p53-D281G reduced transactivation potential of p53-D281G in stably transfected 10(3) murine cells. Therefore, the data presented supports the idea that proper oligomeric forms of mutant p53 are required for its transactivation function. Expression of mutant p53-D2810 also resulted in increased growth rate (H1299 cells), decreased chemosensitivity (H1299 and 21PT cells) and increased plating efficiency (Saos-2 cells). Expression of a transactivation deficient mutant p53 did not induce gain-of-function properties (increased growth rate and decreased chemosensitivity). Unlike the other gain-of-function properties tested, soft agar plating efficiency in Saos-2 cells was not significantly affected by the expression of a transactivation deficient mutant p53, suggesting that transactivation may not be the only factor affecting this gain-of-function property In order to identify the genes responsible for the observed phenotypes, global gene expression analyses were carried out using p53-null H1299 cell stably transfected to express mutant p53 (-Rl75H, -R273H and -D281G). A thorough and stringent analysis revealed 150 genes up-regulated by the expression of mutant p53. Up-regulation of a number of these genes was confirmed by QPCR and transient transcriptional promoter analyses; expression of the transactivation deficient mutant p53-D2810 (L22Q/W23S) did not result in up-regulation of the tested genes further supporting the idea that transactivation of genes is directly related to gain-of-function phenotypes. Using the ASNS gene as a model, this transactivation by mutant p53 was concentration dependent and that the increased transcription did indeed result in increased protein levels.

Neutral cholesterol ester hydrolase (CEH) is a key enzyme in regulating hepatic free cholesterol. Using the CEH specific cDNA sequence in the 5'-untranslated region as a primer, 1.3 kb of sequence upstream of the ATG initiation codon was amplified and cloned. Primer extension analysis with total RNA from rat primary hepatocytes identified a transcription initiation site, 60 bases upstream from the initiation codon. No typical TATA-box sequences were found upstream from the transcription start site. However, a consensus GC-box, which can bind the positive transcription factor SP1, was found 35 bases upstream from the transcription start site. In addition the promoter also contained several hormone responsive half elements, sterol response elements, ubiquitous transcription factor binding sites and liver specific elements. To determine the promoter activity of the rat CEH gene, the 1.3 kb of 5’-flanking region was fused to a luciferase reporter gene. Smaller 5’-deletion constructs were obtained by generation of unidirectional nested deletion breakpoints in the full-length construct with Exonuclease III. Basal promoter activity, as well as transcriptional regulation by hormones, signal transduction pathways and agents perturbing cholesterol metabolism were studied in human hepatoblastoma HepG2 cells and cultured primary rat hepatocytes by transient transfection assays of the promoter activity of the deletion constructs. Functional glucocorticoid response elements, phorbol ester responsive sequences and sterol responsive sequences were mapped with both the culture systems. Results indicate that the first 599 base pairs upstream of the initiation codon and the region between nucleotides -1317 and -1190 regulate the effects of various physiological stimuli. The effects of various stimuli used in this study were similar in the two cell lines. The rat CEH gene appears to be finely regulated by distinct signals converging to consensus promoter regulatory sequences.

Sites of unrepaired DNA damage block DNA synthesis by replicative polymerases and interrupt replication fork progression. This is a problem because stalled replication forks can collapse leading to DNA double-stranded breaks, chromosome instability, and eventually cell death. The translesion DNA synthesis (TLS) pathway rescues stalled replication forks by using polymerases that are specialized for replicating damaged DNA, called non-classical polymerases. The use of TLS is risky however, because it can result in replication errors leading to mutations and eventually carcinogenesis. Genetic studies have shown that the non-classical polymerase Rev1 plays a key role in error-prone TLS, but not much is known about its function. Rev1 is a modular protein that has three functional regions: an N-terminal BRCT (BRCA1 C-terminal) domain, a polymerase domain, and a C-terminal domain (CTD). Rev1 promotes replication fork rescue by other non-classical polymerases through interactions mediated by its CTD; however, the roles of the other two domains in promoting TLS are not well understood. Non-classical polymerases have evolved to be specific for the bypass of a small number of lesions or families of closely related lesions called cognate lesions. Genetic studies using yeast have suggested that the abasic site, one of the most common forms of DNA damage, is a cognate lesion of Rev1. However, steady state kinetic studies suggest Rev1 is very inefficient at incorporating nucleotides opposite an abasic site. To resolve this controversy, I examined the pre-steady state kinetics of nucleotide incorporation by yeast Rev1. I found that Rev1 is capable of rapid nucleotide incorporation, but that only a small fraction of the protein molecules possessed this robust activity. I characterized the nucleotide incorporation activity of the catalytically robust fraction of Rev1 and found that it efficiently incorporated dCTP opposite a template abasic site. My studies strongly suggest that the abasic site is a cognate lesion for Rev1. The rev1-1 allele encodes a mutant form of yeast Rev1 with a G193R substitution in the BRCT domain that reduces error-prone TLS and results in low levels of DNA damage-induced mutagenesis. Unfortunately, little is known about the function of the Rev1 BRCT domain and how this mutation disrupts this pathway. To gain insight into the function of the Rev1 BRCT domain, I solved the X-ray crystal structure of the domain and showed that substitutions in residues constituting its phosphate-binding pocket do not affect error-prone TLS. This suggests that the role of the Rev1 BRCT domain does not involve binding a phosphorylated interaction partner. I also found that the G193 residue is located in a conserved turn of the BRCT domain, and my in vivo and in vitro studies suggest that the G193R substitution disrupts Rev1 function by destabilizing the fold of the BRCT domain.

DNA damage causes replication forks to stall, which can ultimately lead to double strand breaks, chromosomal rearrangements, and cell death. Proliferating cell nuclear antigen is a clamp protein which encircles DNA and acts as a sliding platform for proteins that will act on the DNA, including polymerases. Replication fork stalling causes PCNA to be mono-ubiquitylated, resulting in a polymerase switch from the classical polymerases that perform replication to the Y-family polymerases which conduct translesion synthesis (TLS) to bypass the DNA damage and allow replication to continue. Y-family polymerases include DNA polymerases eta, kappa, iota, and Rev1. The mechanism by which these polymerases are assembled into multi-protein complexes which include PCNA has previously not been well-understood. The studies outlined below demonstrate that ternary complexes involving PCNA, Rev1, and pol η can adopt multiple conformations which can rapidly interconvert. Additionally, I have demonstrated that ubiquitin stimulates the catalytic activity of DNA polymerase eta through a novel interaction with the catalytic core. These studies represent significant progress towards understand the regulation of TLS polymerase activity through both recruitment to and activation by ubiquitylated PCNA.

No description available.

No description available.

No description available.

No description available.

No description available.