Two mutants belonging to the same complementation group in the yeast Saccharomyces cerevisiae (WD and W24) were isolated which exhibit decreased expression of the heme-dependent CYC1 gene encoding iso-1-cytochrome c. Normal levels of CYC1 expression were restored when the mutants were supplemented with a heme analogue, suggesting that the mutations alter intracellular heme levels. Further analysis of WD revealed that it is also respiratory deficient. The mutations in these two strains are not complemented by any of the previously cloned heme biosynthetic genes. A clone was isolated from a yeast genomic library which confers both respiratory competence to WD and restores CYC1 expression in W24 to normal levels. DNA sequence analysis revealed an open reading frame of 1179 base pairs coding for a protein of 393 amino acids which does not display significant homology to any other sequence in a database search. Disruption of the wild type gene, designated RHE1, elicits a respiratory deficient phenotype and does not result in heme auxotrophy. CYC1 expression in the disrupted strains is significantly reduced compared to the levels observed in W24. Analysis of the regulation of RHE1 expression revealed that this gene is transcriptionally repressed by growth in a non-fermentable carbon source and under heme-sufficient conditions.

$ alpha$1,2-Mannosidases are a family of enzymes with similar amino acid sequences that are required for the formation of complex and hybrid asparagine-linked oligosaccharides in mammalian cells. We have determined the genomic structure of one member of this enzyme family, the murine $ alpha$1,2-mannosidase IB gene. Two BALB/c cosmid genomic clones (Cos.31.1 and Cos.25.1) and three overlapping 129/sv P1 genomic clones were isolated. Analysis of Cos.31.1 and the P1 clones showed that the $ alpha$1,2-mannosidase IB gene spans 180kb of the genome, and consists of 13 exons representing the complete open reading frame of the enzyme. Comparison of the intron/exon boundaries with those found in the partially characterized D. melanogaster mas-1 gene indicates that there is some conservation of $ alpha$1,2-mannosidase genomic structure between species. Fluorescence in situ hybridization with $ alpha$1,2-mannosidase UB genomic DNA (Cos.31.1) localized the gene to mouse chromosome 3F2. Sequence analysis of the Cos.25.1 cosmid clone provided evidence for the existence of another related gene or pseudogene. Fluorescence in situ hybridization with Cos.25.1 localized this sequence to mouse chromosome 4A13.

The zinc cluster family of fungal regulatory proteins shares a DNA binding motif that contains six highly conserved cysteines residues. Many of the zinc cluster proteins have been shown to bind as homodimers to target DNA sequences containing a pair of CGG triplets, with each monomer recognizing a triplet. Orientation and spacing between the CGG triplets specify the recognition sequences for a given zinc cluster. For example, Gal4p, involved in galactose metabolism, recognizes CGG triplets, oriented as a palindrome, or inverted repeat, spaced by 11 base pairs (bp) (CGG N11 CCG). The spacing is specific enough so that a change to 10 or 12 by does not enable high affinity binding by Ga14p. Another motif is the direct repeat, as exemplified by Hap1p (CGG N6 CGG). More recently, Leu3p, involved in leucine biosynthesis, was shown to bind to a everted motif spaced by 4 by (CCG N4 CGG). Here we show that another zinc cluster protein, Uga3p, also recognizes an everted motif spaced by 4 bp. However, these two proteins have distinct DNA targets. We show that additional specificity is due to the nucleotides in between the CGG triplets. Altering the 4 nucleotides located between the CGG switches the specificity from a Uga3p site to a Leu3p site in both in vitro and in vivo assays. Therefore, our results identify a new mechanism that expands the repertoire of DNA targets of the zinc cluster family of proteins.

Interactions between the various intracellular loops and nucleotide binding domains (NBDs) of P-glycoprotein (P-gp), and the extent to which they contribute to protein structure and transport mechanism are widely unknown. Analogy to bacterial members of the ABC transporter family suggests that the nucleotide binding domains interact with a specific site on an intracellular loop and with each other, as a cooperative dimer, in order to energize the transport functions. To investigate this hypothesis, we have used a cysteine scanning mutagenesis strategy on four potentially interacting regions of P-gp. We have analyzed the biological activity of the different mouse P-gp cysteine mutants in a yeast heterologous system. We established that the biological activity of the human MDR1 and MDR1-cysteine-less proteins can also be monitored by the yeast system. We also engineered and tested a MDR1 cysteine-less protein containing a factor Xa protease recognition site located in the third extracellular loop as a tool for future studies. Finally, analysis of the biological activity of the substitution mutants reveals key residues in regions that may be involved in drug binding and/or intramolecular domain interactions.

p56Lck, a member of the Src family of tyrosine kinases, is hypophosphorylated at its inhibitory c-terminal tyrosine residue, Y505, in resting cells. Previous studies showed that the amino-terminal "unique region" regulates this dephosphorylation. In this study, we mapped the motif within the "unique region" maintaining Y505 dephosphorylation as well as identified another function for the "unique region". / The 17 c-terminal residues of the unique region constitute the carboxy-terminal phospho-tyrosine regulatory (CPR) motif. Deletion of any part of it raised the state of Y505 phosphorylation by 5--10 fold in resting cells. CPR activity is independent of the serine phosphorylation sites within the "unique region", or the ability of the "unique region" to (i) bind CD4/CD8, or (ii) restrict the access of Lck to detergent insoluble glycosphingolipid enriched microdomains (DIGs). Further, CPR does not regulate global tyrosine dephosphorylation in Lck, as the positive regulatory tyrosine, Y394 is not regulated by it. Finally we show that the "unique region" hampers the ability of the SH4 domain to target Lck to DIGs. This involves serine 59, as its phosphorylation by PMA reduces the amount of Lck in DIGs, while its mutation to alanine, increases it. / This work brings new insight on the role of the "unique region" in both the negative and positive regulation of Lck.

The Prion Protein (PrP) prevents Bax activation in primary human neurons and in MCF-7 breast carcinoma cells. To identify the PrP elements necessary for this anti-Bax function, we performed structure-function analyses on cytosolic PrP (CyPrP). Partial but not complete deletion of the N-terminal BH2-like octapeptide repeats (BORs) abolish CyPrP's anti-Bax function. Deletion of the third and last alpha-helix (PrP23-199) eliminates CyPrP's protection against Bax-mediated cell death, as do further C-terminal deletions. Substitution of helix 3 residues K204, V210, and E219 by prolines inhibits the anti-Bax function of CyPrP while replacement of these three residues by alanines causes a loss of function only for the K204A mutant. Expression of PrP's helix 3 displays anti-Bax activity in both MCF-7 cells and human neurons. Together, these results indicate that although the BOR domain is implicated, helix 3 is necessary and sufficient for the anti-Bax function of CyPrP. Identification of helix 3 as the structural element for the anti-Bax function thus provides a molecular target to modulate PrP's anti-Bax function in cancer and neurodegeneration. / RÉSUMÉ La protéine du prion (PrP) empêche l'activation de Bax dans les neurones humains primaires et dans les cellules de carcinome du sein MCF-7. Afin d'identifier les éléments de PrP requis pour cette fonction anti-Bax, nous avons étudié la relation structure-fonction de la forme cytosolique de PrP (CyPrP). La délétion partielle mais non complète d'une série d'octapeptides du fragment N-terminal semblables au domaine BH2 abolit la fonction anti-Bax de CyPrP. La délétion de la troisième et dernière hélice ainsi que des délétions plus importantes à partir du bout C-terminal éliminent également l'action protectrice de CyPrP contre la mort cellulaire médiée par Bax. Le remplacement des acides aminés de l'hélice 3 K204, V210 et E219 par des prolines inhibe la fonction anti-Bax de CyPrP. Lorsque ces trois acides sont remplacés par des alanines, seul le mutant K204A perd la fonction anti-Bax. L'expression de l'hélice 3 de PrP montre une activité anti-Bax chez les cellules MCF-7 et les neurones humains. Ensemble, ces résultats indiquent que malgré l'implication des octapeptides dans la fonction anti-Bax de PrP, l'hélice 3 est nécessaire et suffisante pour ce rôle protecteur de CyPrP. L'identification de l'hélice 3 come élément-clé de la fonction anti-Bax de PrP donne une cible moléculaire pour moduler la mort cellulaire. Ainsi, cette hélice peut mener vers des avancées dans la lutte contre le cancer ou contre les maladies neurodégénératives.

The establishment and maintenance of the cardiac phenotype requires the activation of cardiac-specific as well as muscle-restricted genes in a tightly regulated spatial manner. This process is presumably governed by the combinatorial action of cell-restricted as well as ubiquitous transcriptional regulators. In order to unravel new cardiac transcriptional regulatory pathways and to characterize how they participate in the establishment of the cardiac transcriptional program, we used the atrial natriuretic factor (ANF) gene as a marker. This strategy led to the identification of a novel cardiac cis-element, the NKE, which is critical for the activity of the ANF promoter and which binds members of the cardiac NK2 family. The NKE is located in proximity to a critical GATA element and the phasing between these two elements is evolutionary conserved. Thus, we hypothesized that GATA-4 would functionally interact with Nkx2-5. Indeed, this work documents, for the first time, a molecular interaction between the essential cardiac proteins GATA-4 and Nkx2-5. Furthermore, this work provides a basis for GATA specificity in vitro and generates novel insights on how the cardiac genetic program is activated in early cardiogenesis. Finally, we report the cloning of a novel cardiac protein likely to have an important role in cardiogenesis, which was cloned in an effort to discover novel binding proteins that bind the critical CARE element required for high promoter activity in embryonic heart and postnatal atrium. This CARE-binding protein, CATF1, is a multifunctional helicase that defines a new family of DNA/RNA helicases. CATF1 is specifically expressed in the heart in neonates and is likely to be an activator of the ANF promoter. Collectively this work significantly furthers our knowledge on the establishment of cardiac- and chamber-specific transcription in the myocardium which is proposed here to occur via a combinatorial set of interactions among cardiac-restricted and ubiqu

Mouse Ly-6 antigens belong to a family of structurally related GPI-anchored cell surface proteins encoded by tightly-linked, but independently regulated genes on chromosome 15. Functional data so far suggest roles in intercellular communication and trafficking of cells during hematopoiesis and immune responses. High sequence divergence between species has hampered homology cloning of human Ly-6 genes. The major part of this thesis reports the characterization of the first of two human Ly-6 homologues which have been discovered recently. Also reported is work on the generation of recombinant forms of Ly-6 proteins for use in Ly-6 ligand studies and in raising antibodies to the novel human Ly-6-related proteins. cDNA clone 9804, provided by collaborators at CytoMed (Cambridge Mass.), encoded a human Ly-6 sequence most like mouse TSA-1. It was used as a probe to isolate genomic fragments containing 9804 exons. Further cloning and characterization of overlapping fragments revealed that the 9804 gene consists of 4 exons, arranged like other Ly-6 genes, and spanning approximately 5 kb. Screening of human-hamster hybrid cell DNA by PCR associated the 9804 gene with chromosome 8. Fluorescence in situ hybridization, performed by collaborator D. Well, refined the mapping to 8q24.3, a region showing synteny with mouse chromosome 15 in the region containing Ly-6 genes. Northern blot analyses demonstrated a broad tissue distribution of 9804 expression, similar to that of TSA-1. While present in many different human tumor cell lines, 9804 mRNA was not detected in the B cell lines Raji and Daudi. 9804 mRNA levels were inducible in vitro by IFN-alpha and ATRA. The response to ATRA was restricted to myeloid leukemia cell lines and was a late event (36--48 hrs), whereas IFN-alpha produced a stronger and more rapid (8--12 hrs) enhancement in normal PBMC and a broad range of cell lines. The induction required protein synthesis, and was apparently at the level of transcription. Solub

Macrophages are important targets for HIV-1 infection, they are generally the first cells infected and are responsible for the extensive viral dissemination seen in HIV-1 infection. Monocytic cells are long lived and become ferocious producers of virus upon pathogen costimulation. This high level of viral replication may result from the activation of NF-kappaB transcription factors. In addition to orchestrating the host inflammatory response, NF-kappaB strongly transactivates the HIV-1 promoter, making NF-kappaB activation a double edged sword; stimuli that induce an NF-kappaB mediated response also lead to HIV-1 transcription. / We have used two promonocytic cell models to investigate the regulation of NF-kappaB signaling and its functional role in HIV-1 infected myeloid cells. Experiments conducted in U937 and PLB-985 cells and their HIV-1 infected counterparts U9-IIIB and PLB-IIIB, revealed that HIV-1 infected cells express constitutive NF-kappaB·DNA binding activity due to altered regulation of upstream kinases. The double stranded RNA dependent protein kinase (PKR) and the IkappaB kinase (IKK) are constitutively activated in these cells and lead to the phosphorylation and subsequent degradation of the inhibitory IkappaB proteins. Degradation of IkappaBalpha and IkappaBbeta releases NF-kappaB which translocates to the nucleus and activates responsive genes, including kappaB dependent NF-kappaB genes p105/p50, p100/p52 and c-Rel. These NF-kappaB subunits are increased in HIV-1 infected U9-IIIB and PLB-IIIB cells and may contribute to constitutive NF-kappaB activation. In addition, IkappaBbeta is able to enter the nucleus and bind DNA bound NF-kappaB forming a ternary complex. IkappaBbeta shields NF-kappaB from dissociation by IkappaBalpha, maintaining transcriptionally active NF-kappaB. Degradation of IkappaBalpha and IkappaBbeta therefore releases active NF-kappaB, and IkappaBbeta additionally protects DNA bound NF-kappaB from inhibition from newly synthesized IkappaBalpha. / We further examined the function of constitutive NF-kappaB activation and found it to be important for maintaining cell viability of HIV-1 infected cells. TNFalpha induced apoptosis inversely correlated with NF-kappaB activation in monocytic cells and inhibition of constitutive NF-kappaB activation resulted in apoptotic death in HIV-1 infected, but not uninfected cells. Therefore, it seems that HIV-1 induces NF-kappaB activation by modulating upstream kinases so that the cell can resist HIV-1 induced apoptosis. These results may have important implications for the treatment of HIV-1 infection.

The GSG (G&barbelow;RP33, S&barbelow;am68, G&barbelow;LD-1) domain is a tripartite protein module of ∼200 amino acids. It consists of an hnRNP K homology (KH) domain and two flanking regions N- and C-terminal to the KH domain Called the NK region and CK region, respectively. The KH domain embedded in the GSG domain has longer loops 1 and 4 compared to other KH domains. The physiological significance of the GSG domain is demonstrated by the fact that genetic mutations in the GSG domain result in developmental defects in various species. The primary goal of this thesis is to characterize the biochemical properties and functions of the GSG domain. We demonstrated that the GSG domain proteins Sam68, Qk1, GRP33, and GLD-1 are RNA-binding proteins that self-associate into multimers. Sam68 complexes bound homopolymeric RNA and the SH3 domains of p59fyn and phospholipase Cgamma1 in vitro, indicating that Sam68 associates with RNA and signaling molecules as a multimer. The formation of Sam68 complex was inhibited by p59fyn, suggesting that Sam68 oligomerization is regulated by tyrosine phosphorylation. Deletion studies in Sam68 and Qk1 demonstrated that the GSG domain is required for both self-association and RNA binding. Sam68 oligomerization requires the extended loops 1 and 4 of the KH domain and the Qk1 dimerization domain is mapped to the NK region. A Qk1 lethal point mutation, altering glutamic acid 48 to a glycine in the NK region, abolishes Qk1 self-association but has no effect on its ability to bind total cellular RNA. The mutant Qk1 protein is more potent than wild-type Qk1 in inducing apoptosis when expressed in NIH 3T3 cells, suggesting that failure to dimerize may be the molecular mechanism for the embryonic lethality. In addition to mediating self-association and RNA binding, the Sam68 GSG domain plays a role in protein localization. We demonstrated by indirect immunofluorescence studies that Sam68 is concentrated in novel nuclear structures, termed Sam