The process of transcription is quite complicated and involves many more steps than the classic textbook view of: Bind, melt, transcribe, release. We have studied the complex nature of transcription from the initiation of transcription through the transition from an unstable initiation complex to a stable, processive, elongation complex. We show that promoter release is not required to make the transition to a stable elongation complex. We show conclusively that a proposed model of transient excursion is not adequate for explaining the abortive cycling process that occurs in all known polymerases. We have proposed and tested a model that unites the processes of promoter release, collapse of the initially melted bubble, and proper RNA displacement. Additionally, we have challenged the existing model for the mechanism of inhibition by T7 lysozyme on T7 RNA polymerase.

Architecturally well-defined polymeric materials with precisely controlled chain length, sequence, stereochemistry and interchain interactions can be produced using the fidelity of biological protein synthesis. A set of periodic protein polymers of repeating unit sequence (AlaGly)$\sb3$-X-Gly, where X is Asn, Phe, Ser, Val, or Tyr, has been produced to examine the relation between amino acid residue volume and crystalline unit cell dimensions. The proteins were overexpressed in Escherichia coli and purified on the basis of acid/ethanol precipitation or insolubility in aqueous sodium dodecyl sulfate. The monodisperse populations of purified polypeptides were processed in the form of oriented crystalline mats by precipitation from formic acid under mechanical shear. Analysis by infrared spectroscopy and x-ray diffraction showed that the artificial proteins adopt a chain-folded lamellar structure comprised of anti-parallel $\beta$-sheets with polar orientation and three-residue folds at the lamellar surface; as seen for ((AlaGly)$\rm\sb3GluGly\rbrack\sb{36}$ (Krejchi, 1997). The x-ray diffraction signals for each of the (AlaGly)$\sb3$-X-Gly polymers were indexed on an orthorhombic unit cell with invariant a (hydrogen bond direction) and c (chain direction) axes. However, the b-axis (sheet stacking direction) spacing increased linearly with the volume of the substituted amino acid, indicating a linear relationship between the average intersheet stacking dimension and the volume of the residue at position X. Analysis of the Phe variant utilizing proton spin diffusion in solid state NMR spectroscopy, provided direct evidence for the confinement of Phe residues at the lamellar surface. The chain-folded lamellar architecture adopted by this family of periodic polypeptides accommodates a wide range of residues differing in charge, steric bulk, and hydrophobicity. These results provide a new approach to the controlled engineering of intermolecular interactions in polymeric solids.

Clostridial neurotoxins (BoNTs) are among the most toxic substances presently known. They are a unique group of metalloproteases which catalyze single site cleavage of specific proteins involved in the docking and fusion of synaptic vesicles with plasma membrane for neurotransmitter release. BoNTs are classified into seven distinct serotypes, designated as A to G. Botulinum neurotoxins type A (BoNT/A) is produced by Clostridium botulinum type A as a complex with a group of neurotoxin associated proteins (NAPs). Our results suggest that the bacterial production of NAPs along with BoNT is designed for the NAPs to play an accessory role in the neurotoxin function, in contrast to their previously known limited role in protecting the neurotoxin in the GI tract and in the external environment. We have examined the structure of BoNT/A in aqueous solution, and found the structure in aqueous solution differs dramatically from that resolved by x-ray crystallography, both at secondary and quaternary levels. In terms of secondary structure, BoNT/A in aqueous solution has about 47% β-sheet structure as revealed by infrared spectroscopy, while x-ray crystallography revealed only 17% β-sheet structure. In terms of quaternary structure, BoNT/A exists as a dimer in aqueous solution, which contrasts with the reported monomeric structure of BoNT/A based on the x-ray crystallography. The dimeric form of BoNT/A can self-dissociate into monomeric form at a concentration lower than 50 nM. This concentration-dependent structural change has a significant impact on the endopeptidase activity of BoNT/A: the catalytic efficiency of the monomeric BoNT/A is about 4-fold higher than that of its dimeric form. This difference implies a sterically restricted catalytic site of BoNT/A in the dimeric form of BoNT/A. Reduction of disulfide bond between heavy chain and light chain of BoNTs is required for their enzymatic activity. We are investigating structural differences of BoNT/A under reducing and non-reducing conditions. Reduction of disulfide bond between heavy and light chains of BoNT/A induced a molten globule conformation at physiological temperature (37°C). This molten globule conformation of BoNT/A has been proven to be the enzymatically active structure.

The Cu$\rm\sb{A}$ center, located in subunit II of cytochrome c oxidase, is the primary site for electron entry from cytochrome c. The recent crystal structure confirms this site to be a binuclear copper center with a dithiolate, dihistidyl coordination. A methionine and the carbonyl oxygen of Glu$\sp{198}$ are also ligands. To further characterize the structure and function of the Cu$\sb{A}$ center, site directed mutagenesis has been implemented in subunit II of S. cerevisiae. Substitutions of direct ligands to the Cu$\rm\sb{A}$ site result in loss of cellular respiration. Cytochrome a from cytochrome c oxidase is not detected in the visible absorption spectrum. Furthermore, subunit II does not accumulate to wild-type levels as observed by immunodetection. These results suggest that the enzyme is not assembling, most likely due to the disruption of copper binding, and that subunit II is being degraded. In contrast, substitutions to Glu$\sp{198},$ whose carbonyl oxygen is a ligand to one of the coppers, merely results in cells with reduced respiratory function. In an attempt to perturb the Cu$\rm\sb{A}$ site without disrupting protein stability, nearby amino acids were substituted. Asp$\sp{158}$ is known to hydrogen bond to the Cu$\rm\sb{A}$ ligand His$\sp{161}$, and Ser$\sp{197}$ is adjacent to the Cu$\rm\sb{A}$ ligand Cys$\sp{196}$. Substitutions at these positions result in reduced cellular respiration and may perturb the Cu$\rm\sb{A}$ site. Substitutions to conserved residues Gly$\sp{194}$ and Gly$\sp{201}$ result in a loss of cellular respiration and the underaccumulation of subunit II, while substitutions at other conserved positions have no observable effect on respiration. These results are discussed in relation to the homologous quinol oxidases and nitrous oxide reductases. To begin to characterize residues which may be involved in binding cytochrome c, substitutions were made to various conserved carboxylates which had been implicated in binding cytochrome c. The recent crystal structure demonstrates that Glu$\sp{198}$ and Asp$\sp{158}$ are not available for direct interaction with cytochrome c. However, the crystal structure does point to other conserved carboxylates. Asp$\sp{112}$ may be at the periphery of a cytochrome c binding site since replacement of this residue by Arg results in only a small decrease in cellular respiration.

Initiation of transcription is a central step in cellular regulation. In order to understand better the fundamental machanisms in this complex process, kinetic and structural studies have been carried out on a variety of modified promoter constructs using simple model RNA polymerase from bacteriophage T7. Investigation of the template strand elements in the initiation domain by introduction of non-nucleosidic spacers demonstrates that most of the apparent protein-DNA contacts in the crystal structure are not essential for initiation. In this respect, the part of the template strand from −4 to −2 serves as a “tether”, holding the templating bases near the active site. In cases where templating bases are not connected directly, the non-template strand duplexed downstream of the active site can serve a similar role. Some of the specificity in the positioning of the templating bases appears to be derived from the protein contact with the template strand base at position −1. Precise positioning of the templating bases near the active site appears to involve strong non-specific interactions of phosphodiester groups between positions −1 and +1, and +2 and +3. T7 RNA polymerase demonstrates strong preference for the initiating substrate nucleotide GTP. Experiments with the nucleotide analog 7–deaza–GTP demonstrate that this preference is due to a protein interaction with the non-Watson Crick face of the base involving the nitrogen-7 and oxygen-6 of guanine. In the presence of GTP as the sole substrate, on the template encoding GGGXX…,T7 RNA polymerase can engage in slippage transcription, making a ladder of RNA products ranging from 2 to about 14 nt in length. Termination of the ladder at 13–14 nt appears to involve a cooperative formation of structure in the nascent RNA, leading to the disruption of the transcribing complex. Slippage transcription from promoters encoding long runs of G provides evidence that the RNA:DNA heteroduplex initially can reach a length of nine nucleotides. The kinetics of synthesis of dinucleotide is characterized in a model system using numerical integration of mechanism-based rate equations. The analyses show that product rebinding cannot be ignored and can come to compete with the regular RNA production within a single enzyme turnover. Finally, the product release rate shows substrate dependence, suggesting that binding of an incoming nucleoside triphosphate can facilitate release of the bound RNA, most likely via direct competition at the active site.

It has been shown that the epidermal growth factor receptor (EGFR) can directly interact with F-actin via an actin-binding domain (ABD) located at amino acid residues 984–996. In this study, we investigate the interaction of the EGFR with actin using an actin cosedimentation assay in a reconstituted cell free system. We find that soluble components of the cytosol from A431 cells dramatically enhance the EGFR/actin interaction. This result is found for EGFR purified by EGFR1 immunoprecipitation, over an EGFR1 affinity column or over a wheat germ agglutinin column. This EGFR/actin interaction is shown to be both cytosol protein concentration and actin concentration dependent. Several lines of evidence suggest that this interaction is not mediated by the previously defined ABD of the EGFR but rather requires both the kinase domain and a portion of the C-terminal regulatory domain of the EGFR. Our results suggest the binding of EGFR to F-actin is under the regulation of one or multiple soluble cytosolic proteins through a mechanism that involves the complete tertiary structure of the cytoplasmic domain of the EGFR, but not the previously defined ABD. Altering the extent of the phosphorylation of the EGFR does not change the ability of the receptor to precipitate with F-actin, suggesting that the interaction between the EGFR and actin is independent of the phosphorylation of the EGFR. Intriguingly, the actin co-precipitated EGFR shows a decreased level of phosphorylation compared with the receptor in the supernatant in an actin cosedimentation assay. In addition, depletion of some F-actin-binding proteins from the cytosol enhances the EGFR involved protein tyrosine phosphorylation in the same assay. Cell imaging experiments show that disruption of the actin cytoskeleton in the intact cell using latrunculin B inhibits EGF stimulated internalization of the EGFR. These results suggest that the actin cytoskeleton system plays an important role in the regulation of the EGFR function. A hypothesis for the regulation of the signaling of the EGFR by the actin cytoskeleton is thus proposed for the future study.

Hsp70 molecular chaperones have key functions in the cell for folding, repair and degradation of proteins. Hsp70s are a conserved family of proteins that consist of an N-terminal ATPase domain (NBD) and a C-terminal substrate-binding domain. ATP binding leads to reduced substrate-binding affinity, and reciprocally, substrate binding activates their ATPase rate. This interdomain communication is essential for Hsp70 chaperone function. Previous work suggests a key role for a stretch of conserved hydrophobic residues, 389VLLL 392, in the interdomain linker of E. coli Hsp70, DnaK, in allosteric coupling. Strikingly, we observed that the presence of these residues on the ATPase domain of DnaK (NBD392) led to a markedly enhanced rate of hydrolysis relative to either the intact DnaK protein or a construct lacking this segment (NBD388). Thermal denaturation experiments showed differences in the stability of the two ATPase domain constructs and indicated altered rigid-body movements of subdomains upon linker interactions with the ATPase domain at pH 7.0. We have further illustrated by charge state distributions with ESI-MS (electrospray ionization mass spectrometry) that the linker docking causes conformational rearrangements to a more closed conformation at this pH. We also found that this conformational switch can be reversed by varying the pH. H/D exchange mass spectroscopy of the two ATPase domain constructs revealed different exchange kinetics for the constructs in the nucleotide free-state at pD 7.6. NBD388 displayed a single slowly-exchanging population, whereas NBD392 showed two populations: one which exchanged at a rate similar to NBD388, and one which exchanged rapidly, suggestive of a global unfolding event. The differences in exchange kinetics were reversed by the addition of ADP. These data confirmed our previous findings that NBD392 exists in dynamic equilibrium between an "open", more globally dynamic state and a "closed", less dynamic state, whereas NBD388 samples only the more "open" state. Taking all of our results together, we hypothesize a possible mechanism of allostery: Ligand-induced shift in an open/closed state equilibrium along with dynamic changes in the ATPase domain may be communicated to the substrate-binding domain via the linker residues, implicating the conserved linker as an allosteric switch.

Nucleosomes are the most fundamental unit of chromatin structure, which is composed of ~∼145 bp of DNA wrapped around an octamer of core histones H2A, H2B, H3 and H4. Chromatin structure plays a role in many cellular processes and the eukaryotic cell has devised many different ways to alter the structure of the nucleosome as part of the cellular mechanisms that regulate important processes that occur in the cell. Some of these processes include modification events such as DNA methylation, and histone protein modifications, and in some cases includes the incorporation of two copies of a specialized variant histone into each nucleosome to create a specialized nucleosome at the centromeres. CenH3s represent a family of variant histone proteins that function to replace histone H3 in specialized nucleosomes found only at the centromere. So far CenH3 proteins have been identified at every eukaryotic centromere tested and these proteins are essential. Due to the relative simplicity of its centromere and associated kinetochore proteins, the S. cerevisiae CenH3 protein, Cse4p, was first characterized genetically. Cse4p contains a histone fold domain that is 60% identical to the HFD of histone H3, as well as an N-terminus that is unique. It was recently discovered that the localization of Cse4p to the centromere is dependant only on the HFD of the protein and not on the function of the N-terminus. All current evidence supports the idea that Cse4p interacts with histone H4 and the other core histone proteins at the centromere locus, in the absence of histone H3. It is still not known how very low levels of Cse4 proteins can effectively compete with the much more abundant histone H3 proteins to bind to histone H4 and localize to the centromere. I expressed the recombinant Cse4p, histone H3, histone H4 and a cse4 mutant protein (cse4-286p) and purified these proteins in order to compare the biophysical characteristics of Cse4p and the Cse4p/H4 complex with those of histone H3 and the H3/H4 complex. Results of these studies indicate that Cse4p follows an assembly pathway that is distinct from the histone H3 pathway. Mutations that interfere with the Cse4p assembly pathway cause cell death. To determine if pathways similar to the proposed Cse4p-specific assembly pathway occur that involve the CenH3 proteins from other species, I searched several public genomic databases and compiled 160 distinct CenH3 HFD sequences. The newly identified CenH3 sequences were used to generate a multiple sequence alignment (MSA) to look for a functional connection between the CenH3 HFD primary sequences and the homodimerization of Cse4p, which is mediated through the 4-helix bundle. Statistical coupling analysis (SCA) of the MSA reveals that amino acids that are involved in making DNA contacts exhibit covariance with amino acids that probably play a role in homodimerization. These results suggest a connection between homodimerization of Cse4p and DNA-protein interactions. Another cluster of covarying Cse4p residues were identified, which most likely play a role in Cse4p interactions with histone H4. One amino acid residue in this cluster was previously identified in budding yeast cells to cause a temperature sensitive phenotype when mutated, which was rescued by over-expression of the histone H4 protein. This proposed new centromeric nucleosome assembly pathway provides an elegant, testable model to determine how CenH3 proteins compete with histone H3 to interact with histone H4.

The C-protein alpha and beta antigens of group B Streptococcus (GBS) are protective, surface-associated proteins. These potential virulence factors are candidates for use in a GBS conjugate vaccine. The genes encoding the alpha (bca) and beta (bcb) antigens had previously been cloned and expressed in Escherichia coli. Both the native and the cloned bca gene products are expressed as laddering peptides. This dissertation describes the characterization of bca and bcb gene clones (pJMS23 and pJMS1), the localization of two protective epitopes within the alpha antigen, the elucidation of the alpha antigen laddering mechanism, and the identification and characterization of spontaneous alpha antigen mutants in GBS. The bca gene was localized in a subclone of pJMS23 and the bcb gene was localized in pJMS1, to facilitate molecular characterization of these genes. Previous studies of the recombinant bcb gene product from pJMS1, indicated that the entire gene was not present. To obtain the full-length bcb gene, another bcb gene clone (pDEK4) was isolated from a library of GBS DNA. The gene product from this clone was characterized biochemically and immunologically. Two regions of the alpha antigen were characterized that contain protective epitopes, the N-terminus and the repeat region. The epitope bound by alpha antigen monoclonal antibody, 4G8, was localized to the repeat region. Antibodies raised to the recombinant alpha antigen N-terminus are opsonic and protect mice against alpha antigen-bearing strains of GBS. The basis of alpha antigen laddering was studied in both GBS and E. coli. Although, laddering appears to be the result of proteolytic processing in GBS, it could not be determined if the recombinant alpha antigen is processed in E. coli. In GBS, a serine proteolytic activity that formed laddering peptides was localized to the intracellular fraction of GBS. Spontaneous mutants of the alpha antigen were identified and characterized at the phenotypic and genotypic levels. These mutants contain deletions localized to the tandem repeat region. However, these deletions did not affect resistance to opsonophagocytosis in the absence of alpha antigen-specific antibodies. Thus, the size of the alpha antigen does not affect susceptibility to opsonophagocytosis in the absence of antibodies.

The ER quality control machinery maintains the fidelity of the protein maturation process by sorting aberrant proteins for ER-associated protein degradation (ERAD), a process requiring retranslocation from the ER lumen to the cytosol and degradation by the proteasome. To understand the role of N-linked glycans in ERAD, degradation of wild-type (Tyr) and mutant (Tyr(C85S)) tyrosinase was examined. Here, we demonstrated that both wild-type and mutant tyrosinase were substrates of the 26S proteasome. Tyr(C85S), however, was less stable, and the cell line harboring the C85S mutation exhibited an up-regulated unfolded protein response as measured by XBP-1 mRNA splicing. Inhibiting mannose trimming or accumulating Tyr(C85S) in a monoglucosylated form led to stabilization, supporting a role for lectin chaperones in ER retention and mannose trimming in proteasomal degradation. In contrast, preventing glucose trimming caused rapid disappearance of protein. Upon closer examination employing procedures which monitored the appearance of degradation product (small peptides) rather than the disappearance of full-length protein, ablating lectin chaperone binding induced the formation of aggregates. Colocalization of tyrosinase with BiP and PDI, but not calnexin, implicated the latter two in aggregate dissolution. The fact that aggregates were disassembled and cleared from the ER at a rate similar- to non-aggregated species and degraded by the proteasome suggests a model of glycoprotein degradation in which non-lectin molecular chaperones function in the quality control of glycoproteins, at least in part, in the absence of lectin chaperones.