The bacterium Escherichia coli uses DNA-segregating machinery known as the ParMRC system to ensure the stable inheritance of low copy-number plasmids by daughter cells during cell division. In this system, ParM, an actin-like protein, forms a filamentous spindle between ParR/parC complexes that are assembled on sister plasmids; segregation of the plasmids is achieved by growth of the ParM filaments and the resulting push of plasmids to opposite cell poles. Studies combining in vitro biochemical assays live cell fluorescence microscopy and cryo-electron microscopy has led to deep mechanistic insights into the action of the ParMRC system. It is thought that ParM filaments (attached to the ParR/parC complexes) elongate through a mechanism of “insertional polymerization,” where the growth of the ParM filament occurs preferentially at the end associated with the ParR/parC complex. Owing to good structural information available in the literature, ParMRC serves as an interesting model system for molecular dynamics simulations. These simulations may provide a better insight into the insertional polymerization mechanism of ParMRC system. My current study focuses on the development of a computational model, a long-term goal, to perform molecular simulations of the ParMRC system. My work here covers the progress of my study by including two main components of the ParMRC system: 1) modeling and molecular simulations to study the elongation and pairing of ParM single and double filaments respectively; and 2) Homology modeling of ParR. Encouragingly, molecular simulations of ParM single and double filaments recapitulated some of the experimental elements of ParM and provided us with some interesting observations. We anticipate that the work presented here may serve as a good start to perform molecular simulations of the ParMRC system.

Single subunit T7 RNA polymerase is relatively small and shares mechanistic features with all RNA polymerases, making it an ideal model to study transcription. The T7 RNA polymerase recognizes and binds specifically to a unique sequence in the DNA, melts the DNA at the start site and initiates de novo dinucleotide synthesis. In vitro selection for full function of the T7 RNA polymerase promoter has yielded the consensus sequence, indicating that the latter is optimized for the overall process. However, given potentially competing individual requirements between distinct stages during transcription, the consensus sequence might not be the most tightly binding sequence. In particular, some of the binding energy may be sacrificed to drive unfavorable DNA melting and function may pose additional sequence constraints. I have used SELEX to identify the DNA sequences most tightly binding to T7 RNA polymerase and the results reveal that they are AT-rich in the initial bubble region from positions -4 to +3. Introduction of mismatches in this region lowers the barrier to initial bubble melting, leading to higher binding affinities. That the consensus promoter is G-rich downstream of and including +1 likely reflects the functional need to stabilize the open conformation. The conformational rearrangement of polymerase facilitates passage from an initially unstable stage to a stable, processive elongation phase now free of the promoter. Indeed, if halted within a homopolymeric template sequence, slippage of the RNA to allow subsequent extension competes favorably with dissociation. This RNA extension is strongly affected by the complex stability, in that removal of a driving force for forward movement of the complex along the DNA promotes slippage, while facilitation of forward movement reduces slippage. The halted elongation complex continues extending RNA by slippage until the substrates are consumed entirely. After depletion of the slippage substrate, the homopolymeric RNA can slide along the template towards its 3' end, forming an arrested elongation complex. Most importantly, this work demonstrates that complex stability arises less from the thermodynamic stability of the hybrid duplex and more from the kinetic stability afforded by the topological locking of the RNA around the template strand DNA.

The reversible methylation of certain residues in the transmembrane receptors mediates adaptation to changes in attractant concentration. In Tsr, four glutamic acid residues (E297, E304, E311, and E493) have been identified as major methylation sites. However, no studies have been conducted on these sites to examine any differences in their rates of methylation. In this study, the methylation rates of these sites were measured by using engineered receptors, in which each has only one major site available for modification. Site four was found to be the most rapid site to get methylated followed by site one, two, and three. This distribution was compared to the one determined for the aspartate receptor and site four was found to be the major difference between these two receptors. Serine and the adaptor protein, CheW, produced similar site-dependent increases in the rate of methylation of all sites. At saturating concentration of serine or CheW, the rate increases were largest at site one (a slow site in the absence of ligand), and least at site four (the fastest site in the absence of ligand). The effects of serine and CheW were also synergistic, except for site four. In the presence of both CheW and serine, the individually. Altogether, the results of this study imply that the major sites of methylation in Tsr (except site four) are involved in the ligand-specific adaptation process. Moreover, CheW appears to play an important role in determining the kinetics of adaptation in the bacterial cell.

The forward translocation model (paused or stalled polymerase moves forward in the absence of transcription, dissociation of the complex happens when hybrid is too short to stabilize the complex) is tested by measuring the stabilities of stalled elongation complexes on DNA constructs that either favor or inhibit forward translocation. Results are consistent with the model; conditions that favor forward translocation lead to decreased stability, while conditions that disfavor forward translocation lead to increased stability. Results also show that the rewinding of the upstream edge of the transcription bubble is a major driving force for forward translocation. In separate biochemical assays, it is demonstrated that the stability of stalled complexes can also be dominated by "bumping" from a trailing polymerase transcribing from the same direction. Trailing polymerases can efficiently displace a stalled polymerase. As predicted by current models for promoter clearance, the instability caused by "bumping" is position dependent. When the first polymerase is stalled 12 bp or less from the promoter, a second polymerase is blocked from binding the promoter. When the first polymerase is stalled between 12 to 20 bp, the second polymerase can bind and make abortive transcripts, but it cannot displace the first complex since it is in an unstable initiation conformation. Only when the leading polymerase is stalled beyond 20 bp, can the second polymerase bind, initiate and displace the leading complex. The function of the promoter region that is melted in the initiation bubble (-4 to +4, relative to the transcription start site) of the T7 RNA polymerase is probed by SELEX (in vitro selection). Results demonstrate that there is no convergence among the selected tight binding sequences. So, those bases are not direct contributors for binding. But the selected sequences show a bias toward AT rich after many rounds of SELEX. This is consistent with the idea that bending or melting at this region facilitates promoter-polymerase binding. Finally, SELEX is also used to select for DNA sequences that will allow stably stalled elongation complex. Preliminary results are consistent with the forward translocation model.

Lyme disease is the most prevalent vector-borne disease in the United States. Lyme disease is mediated by the spirochete Borrelia burgdorferi and transmitted by Ixodes scapularis ticks. Recently, critical protein-protein interactions responsible for the unique host:vector:pathogen symbiosis have been identified. A key protein involved in both transmission and persistence of the spirochete is the antigenic tick salivary protein Salp15. Salp15 has been shown to be important in host immunosuppression by interaction with the CD4 glycoprotein as well as spirochete protection by interaction with a bacterial outer surface protein, OspC. Although these critical protein-protein interactions have been identified, a more in-depth biophysical analysis is lacking. To better understand the mechanism of action of these proteins, we have studied the interaction of Salp15 with CD4 and OspC using a variety of biochemical techniques. We have isolated two forms of Salp15, monomeric and dimeric Salp15. The binding interaction of Salp15 with CD4 is only observed with D-Salp15. D-Salp15 is a disulfide mediated multimer of Salp15. Additionally, we do not observe a direct interaction between Salp15 and OspC in our experimental procedures, perhaps due to the oligomeric state of Salp15. The protein-protein interaction experiments have significantly contributed to the understanding of the molecular pathogenesis of Lyme disease and warrant further structural investigation. The isolation and characterization of the active form of Salp15 is the first step in creation of potential factors to modulate immune responses. Further identification of the molecular interactions between proteins in these complexes will be vital for understanding the mechanism of immunosuppression as well as understanding the complex interactions in Lyme disease.

Polyelectrolyte (PE)-protein interaction and its applications were investigated on protein aggregation and selective protein binding. More specifically, we studied the effect of protein charge anisotropy on the inhibition of protein aggregation and selective protein binding in the system with pure electrostatic interactions. First of all, the aggregation mechanisms of three proteins (Zn/Zn-free insulin, bovine serum albumin (BSA), and β-lactoglobulin (BLG)) were studied by kinetic analysis of turbidimetric and dynamic light scattering (DLS) results. For insulin, zinc effects on multimerization and aggregation were also examined by size exclusion chromatography (SEC) and DLS. These proteins were found to undergo different aggregation mechanisms which results in different inhibition effects by bio-polyelectrolyte heparin under conditions of pH and ionic strength known to favor the formation of stable heparin-protein complexes through electrostatic interactions. The turbidimetric titration results in the presence of heparin reveal the heparin effects are dependent upon the heparin-protein binding affinity which is controlled by the charge anisotropy of proteins explored by DelPhi electrostatic modeling. Secondly, the selective binding of protein isoforms (BLGA and B) were evaluated on different substrates to challenge the assumption that long range electrostatics lack of good selectivity as short range interactions. More specifically, we studied BLG A and B binding on different substrates with various geometries: cationic polymer, nanoparticle, and nanoparticle-coated surface by using a wide range of techniques including isothermal titration calorimetry (ITC), surface plasmon resonance (SPR), turbidimetry, size exclusion chromatography (SEC), and ion exchange chromatography (IEC) etc. The selective protein binding by cationic PE was further demonstrated to be able to successfully enrich BLGA by a factor of 2 from normal BLG A/B mixture by PE coacervation, The relationship between protein charge anisotropy (patch) and binding affinity and between binding affinity and protein (BLGA and B) separation selectivity was elucidated by different separation techniques (SEC, IEC), ultrafiltration, and isothermal titration calorimetry (ITC).

Catecholamines (CA)--dopamine (DA), norepinephrine (NE) and epinephrine (E)--are taken up and stored by organelles in the chromaffin cell of the adrenal medulla. The energy for accumulating the high concentration of CA in the granule is derived from the protonmotive force established by proton pumping ATPase in the granule membrane. Active transport of CA is accomplished by electrogenic exchange of two protons for each CA('+) via a reserpine sensitive CA transporter in the granule membrane. In experiments to further probe the role of ATP in CA uptake and release we have found that ATP can stimulate net uptake, not just exchange, of CA in freshly isolated granules and in lysed resealed vesicles (ghosts). We have also found that ATP limits release of CA from granules or ghosts by a process other than reuptake. This process depends primarily on hydrolysis of ATP and activity of the proton pump, but is reserpine insensitive and therefore does not depend on CA transporter function. It is proposed that release of CA occurs by diffusion across the granule membrane and that the rate of diffusion is controlled by the effective concentration of CA('o). CA('o) concentration is decreased at lowered pH, and in the presence of ATP as a consequence of the formation of weak electrostatic CA:ATP complexes. Experiments to test the ability of several CA receptor ligands to bind to the CA transporter have revealed that (alpha)(,1) adrenergic receptor ligands, prazosin, and phenoxybenzamine are the most effective inhibitors of CA transport, followed closely by (beta) adrenergic and dopamine receptor ligands, propranolol, alprenolol, isoproterenol, NPA and ADTN. The (alpha)(,2) agonist clonidine, and the serotonin receptor antagonist methysergide were an order of magnitude less effective. Antagonists were generally more effective in their inhibition of NE uptake than agonists. These results suggest that the binding site of the transporter shares similarities with the binding site of CA receptors. Covalent labeling of granule membrane proteins, and separation on SDS PAGE gels revealed 3 bands labeled in common by (('3)H) phenoxybenzamine and (('3)H)NE. These were at molecular weights about 20K, 32K and 45K.

Abstract not available

(1) N-Acetyl{('3)H}lysyl-tRNA('Lys) and N-acetyl{('3)H}glutamyl-tRNA(,2)('Glu) were cross-linked to the P site of Escherichia coli 70S tight-couple ribosomes by irradiation with light of 300 to 400 nm. Covalent attachment was dependent on the presence of polynucleotide message and nearly 85% of the ('3)H-labeled amino acids could be released from the cross-linked complexes by puromycin. Cross-linking was stimulated in the presence of acetone and reversed by subsequent exposure to light of 254 nm. Centrifugation of covalent complexes through sucrose gradients revealed that tRNA was attached only to 30S subunits, and further analysis demonstrated that linkage was to 16S RNA, but not to ribosomal proteins. Partial RNase T(,1) hydrolysis of N-acetylaminoacyl-tRNA--16S RNA complexes allowed localization of the site of cross-linking to the 8S RNA fragment derived from the 3' 40% of the 16S RNA. When N-acetyl-aminoacyl-tRNA--8S RNA complexes were electrophoresed in polyacrylamide gels under denaturing conditions, a major fraction of the tRNA was found to be cross-linked to one or more rRNA subfragments of 100-125 nucleotides. (2) N-Acetylvalyl-tRNA(,1)('Val) was bound to the P site of uniformly ('32)P-labeled 70S tight-couple ribosomes and cross-linked to 16S RNA in the 30S subunit by irradiation with near ultraviolet light. Following partial RNase T(,1) digestion of N-acetylvalyl-tRNA(,1)('Val)--16S{('32)P}RNA complexes, the rRNA fragments to which the tRNA cross-links were isolated by two-dimensional gel electrophoresis according to the diagonal method. After the first dimension, the tRNA-rRNA cross-link was cleaved by photolysis at 254 nm. Upon electrophoresis in the second dimension, RNA segments previously attached to tRNA migrated to positions beneath the diagonal. The appearance of oligonucleotides below the diagonal was dependent on the presence of tRNA in the initial reaction mixture and required photoreversal of the cross-link. Sequence analysis revealed that the smallest of three nested rRNA fragments containing the site of tRNA attachment encompassed residues 1362 through 1497 of the 16S RNA. Following complete RNase T(,1) hydrolysis of N-acetylvalyl-tRNA(,1)('Val)--16S (('32)P)RNA complexes, the cross-linked tRNA-rRNA oligonucleotide was isolated by polyacrylamide gel electrophoresis. The cross-linked adduct contained a ('32)P-labeled nonanucleotide from the 16S RNA and an unlabeled pentadecanucleotide from the tRNA. The sequence of the nonanucleotide was determined to be U-A-C-A-C-A-C-C-G, which occupies positions 1393-1401 of the 16S RNA and is located within an evolutionarily conserved region. Secondary digestion analysis of the covalent tRNA-rRNA oligonucleotide revealed that the cross-linked residue in the 16S RNA was C(,1400). The site of cross-linking in the tRNA was ascertained using N-acetylvalyl-tRNA(,1)('Val)--16S RNA complexes prepared from nonradioactive ribosomes. After complete RNase T(,1) digestion of these complexes, the RNA was labeled at the 3' end with {5'-('32)P}pCp. The covalent tRNA-rRNA oligonucleotide isolated from the mixture released a single end-labeled component upon photoreversal of the cross-link. Chemical sequence analysis demonstrated that this product was the anticodon-containing pentadecanucleotide of tRNA(,1)('Val), C-A-C-C-U-C-C-C-U-cmo('5)U-A-C-m('6)A-A-G(,39), and that the site of covalent attachment to 16S RNA was the 5' anticodon nucleotide, cmo('5)U(,34). The structure of the cross-linked adduct is proposed to be a cyclobutane dimer between cmo('5)U(,34) of tRNA(,1)('Val) and C(,1400) of the 16S RNA.

Two endogenous plant compounds, (DELTA)-1-tetrahydrocannabinol (THC) and gibberellic acid (GA), have been reported to have endocrine effects in mammals. THC, the psychotomimetic constituent of marijuana, has been reported to influence the male reproductive system in several animal species. Its effects were manifested in a loss of testosterone biosynthesis in Leydig cells and in a decrease in pituitary output of luteinizing hormone and follicle stimulating hormone. Similar studies in the female have not shown a consistent pattern, especially the current controversy of estrogen-like activity of THC upon the reproductive tract and accessory sex organs. THC has been reported to have estrogenic activity in the female rat by several investigators. Several other groups have reported no such activity in the rat uterus, mouse mammary tissue, and uterine tissue from the rhesus monkey. To date, no definitive study has been undertaken in one animal species, monitoring several endpoints of estrogen activity, to resolve this conflict. GA has also been implicated in possessing hormonal activity, namely, glucocorticoid-like activity under conditions of stress, restoration of the reproductive tract histology in ovariectomized rats, and weak estrogenic activity and estrogen synergism in mice. There have been no reports of the biochemical mechanism of GA action. In order to ascertain if either of these compounds had estrogenic or antiestrogenic activity and if so, the molecular mechanism of action, three indicators of estrogenic activity were investigated in the immature female rat. Temporally, these indicators were the competition for the cytosolic estrogen receptor protein, the in vitro synthesis of an estrogen-enhanced protein (IP) and the in vivo uterine wet and dry weight gain. These indicators encompassed estrogenic responses from the initial physicochemical event to the ultimate expression of uterine mitotic activity. In the immature rat system, the cytosolic estrogen receptor showed saturation kinetics with a K(,d) of 3 x 10('-10)M and a maximal number of binding sites of 35 pmol/g uterine tissue. This binding was competitive with unlabeled estradiol. The estrogen-enhanced IP synthesis was monitored by polyacrylamide gel electrophoresis. Maximal induction appeared as an increase in the protein ('3)H/('14)C leucine-incorporation ratio with a relative mobility (R(,m)) of 0.65. Subcutaneous administration of estradiol resulted in a significant increase in both wet and dry uterine weight following a three-day dosage regimen. Competition for the cytosolic estrogen receptor was not observed with either THC or GA over a concentration range of 10('-10) to 10('-6)M. Neither compound was able to elicit IP synthesis at a concentration of 10('-5)M as reflected in a ('3)H/('14)C baseline curve at R(,m) 0.65. Neither compound was able to elicit any significant increase in uterine wet and dry weight when administered at 2 and 10mg/kg body weight. When co-administered with estradiol, no antagonism was seen. There did appear to be a slight synergism with estradiol but further investigation showed this to be insignificant. Neither THC nor GA appeared estrogenic in tests for estrogen responsiveness in rat uterine tissue under the experimental conditions imposed. Since the animal population investigated was immature, it is possible that these compounds may express estrogen-like activity at later stages of development. Perhaps the appearance of a mature drug metabolizing system with consequent biotransformation could explain some of the conflicts reported in the literature.