A Bacteroides gingivalis adhesin mediating attachment to red blood cells and buccal epithelial cells was isolated, cloned and characterized. The isolation procedure involved gentle stirring of the cells followed by ammonium sulphate precipitation, ion-exchange and gel chromatography. The native molecule had a Mr in excess of 10⁶ kDa and was made up of subunits with an Mr of 43 kDa. Antisera raised to the adhesin and its subunits reacted with antigens on the surface of B. gingivalis cells. No reaction with fimbriae was seen. The IgG fractions from these antisera inhibited the adherence of B. gingivalis to host tissue. Proteolytic enzymes destroyed binding capability of whole cells and of the purified adhesin but the molecular weight of the haemagglutinin was not altered. A genomic library of B. gingivalis DNA was created in E. coli JM83. 5500 colonies were screened by a colony immunoassay with anti-S. gingivalis serum and by a direct haemagglutinating assay. 337 clones tested positive by the immunoassay and two clones, 1-3,and 1-49 tested positive for haemagglutinating activity. Both haemagglutinating positive recombinants had inserts of 3.2 kb. One clone, 1-49 was chosen for further characterization. E. coli 1-49 expressed a protein of 43 kDa that was not present in E. coli JM 83 control as seen by SDS-PAGE and Western blot analysis. Anti-1-49 serum inhibited the haemagglutinating activity of B. gingivalis and E. coli 1-49. This serum reacted with surface molecules on B. gingivalis and E. coli 1-49 as seen by immunogold electron microscopy and immunofluorescence, and to the purified haemagglutinin by Western blot analysis. Like the haemagglutinin on B. gingivalis, the haemagglutinating activity of E. coli 1-49 was destroyed by heating and proteolytic enzymes but the apparent size of the molecule as determined by SDS-PAGE was not affected. A bacterial coaggregating adhesin from B. gingivalis was isolated and characterized. The isolation procedure involved adsorption of the solubilized adhesin on S. mitis followed by elution with glycine buffer. SDS-PAGE of the boiled adhesin revealed a protein with an Mr of 46 kDa. Proteolytic digestion destroyed all bacterial aggregating activity and hydrolysed the 46 kd protein. Antisera raised to the 46 kDa protein reacted with surface molecules on all strains of B. gingivalis tested. This antiserum inhibited the coaggregation reaction between B. gingivalis and other bacteria. Vesicles produced by B. gingivalis were found to enhance the binding of S. sanguis to serum coated hydroxy apatite (SeHA). Maximum vesicle mediated binding took place at 37°C and was destroyed by heating. The lipopolysaccharide from several black pigmented bacteroides were isolated and characterized physically, chemically and immunologically. All of the LPS were of the smooth type and contained the sugars rhamnose, glucose, galactose, glucosamine and galactosamine; no KDO or heptose were found. / Science, Faculty of / Microbiology and Immunology, Department of / Graduate

Bibliography: pages 147-169. / Butyrivibrio fibrisolvens H17c is a gram-negative obligate anaerobic bacterium found in the rumen of most ruminants. The aim of this thesis was to investigate the enzymes produced by B. fibrisolvens H17c involved in the degradation of cellulose, xylan, and protein. A library of chromosomal DNA fragments from B. fibrisolvens H17c was established in the plasmid pEcoR251, an Escherichia coli positive selection vector. The library was screened for genes expressing cellulase, xylanase, and protease activity. Two genes expressing endo-β-1,4-glucanase and cellodextrinase activity were cloned in E. coli as host. The gene expressing endo-β-1,4-glucanase activity (end1) was cloned on a recombinant plasmid pES400. The end1 gene was located on a 6.8 kb DNA fragment and expressed from its own promoter in the E. coli host. It was shown that 64% of the endoglucanase activity was located in the periplasm of the E. coli host. TnphoA mutagenesis indicated the presence of a functional E. coli-like signal peptide. The nucleotide sequence of end1 was determined and the amino acid sequence (547 amino acids) deduced. The catalytic domain of End1 showed very good similarity to the catalytic domain of the Clostridium thermoceiium EGE endoglucanase. End1 also has a non-catalytic domain similar to the binding domains of the CenA and Cex cellulases from Ceilulomonas fimi The gene expressing cellodextrinase activity (ced1) was cloned on a recombinant plasmid pES500. This gene was located on a 3.55 kb fragment and was also expressed from its own promoter in the E. coli host. The Ced1 enzyme was also exported to the periplasm of the E. coli host, but did not contain a functional E. coli-like signal peptide. The nucleotide sequence was determined and the deduced amino acid sequence (547 residues) showed high similarity to the catalytic domain of the C. thermocellum EGD endoglucanase. The proteins of End1 and Ced1 showed no similarity. The End1 and Ced1 enzymes were characterized using a range of different substrates. The End1 enzyme showed optimal activity at pH 5.6 and 45°C. Optimal activity for the Ced1 enzyme was obtained at pH 6.6 and 50°C. The proteolytic activity of B. fibrisolvens H17c was characterized using gelatin-SD5-PAGE. Ten bands of protease activity with apparent molecular weights ranging between 42 000 and 101 000 were detected at different stages during the growth cycle. The effect of protease inhibitors indicated that all ten protease bands were serine proteases. Optimal activity was observed between pH 6.0 to 7.5 and at a temperature of 50°C. The proteolytic activity of B. fibrisolvens H17c varied depending on the type of carbohydrate substrate in the medium, and was positively correlated with the growth rate.

The value of the adenylate energy charge [(ATP)+1/2(ADP)/(ATP)+(ADP)+(AMP)] in Azotobacter vinelandii cells was monitored during growth and germination in flask cultures. The miximal value of 0.88 was attained during mid-log phase; this declined gradually to 0.50 by late stationary phase. When these cultures were transferred to encystment media, the adenylate energy charge decreased to an average value of 0.40 as the vegetative cells encysted and remained unchanged during the next 20 days. Encystment cultures wre composed of vegetative cells, encysting cells and mature cysts but the proportionate value of the energy charge could be assigned. Viability of the total population remained 95% or higher during the entire period studied. Azotobacter vinelandii cysts cultivated on phosphate-sufficient media. Although cell protein and nucleic acids were unaffected by phosphate deficiency, cell wall structures, oxygen uptake and sncystment were significantly affected. Phosphate-limited cysts contained much larger amounts of poly-beta-hydroxybutyric acid but had a lower adenylate energy charge than did control cysts. The ATP/ADP ratio was much lower in phosophate-deficient cysts than in the control cysts. The data indicate a "substrate saving" choice of three metabolic pathways available to cells of Azotobacter under different growth conditions.

Pattern formation, or the generation of coordinated, emergent behavior, is ubiquitous in nature. Researchers have long sought to understand the mechanisms behind such systems as zebra stripes, repeating flower petals, and fingers on hands, within fields such as physics and developmental biology. Notably, a diverse array of bacteria species naturally self-organize into durable macroscale patterns on solid surfaces via swarming motility—a highly coordinated, rapid movement of bacteria powered by flagella. Meanwhile, researchers in the synthetic biology field, which aims to rationally engineer living organisms for biotechnological applications, have been engineering synthetic pattern formation in microbes over the last several decades. Engineering swarming is an untapped opportunity to increase the scale and robustness of coordinated synthetic microbial systems. In this thesis, we expand the field of engineered pattern formation by applying the tools of synthetic biology and deep learning to engineer and characterize the swarming of Proteus mirabilis, which natively forms a centimeter-scale ring pattern. We engineer P. mirabilis to “write” external inputs into visible spatial records. Specifically, we engineer tunable expression of swarming-related genes that modify pattern features, and we develop quantitative approaches to decoding. Next, we develop a dual-input system that modulates two swarming-related genes simultaneously, and we apply convolutional neural networks (CNNs) to decode the resulting patterns with over 90% top-3 accuracy. We separately show growing colonies can record dynamic environmental changes which can be decoded with a U-Net model. We show the robustness of the engineered strains’ readout to fluctuations in temperature and environmental water samples. Lastly, we engineer strains which sense and respond to heavy metals. Our pCopA-flgM strain records the presence of 0 to 50 mM aqueous copper with decreased colony ring width. We conclude in this chapter that engineering native swarm patterns can thus be applied for building bacterial recorders with a visible macroscale readout. In parallel, to better characterize the swarm patterns of P. mirabilis, we develop a pipeline using deep learning approaches to segment colony images. We develop easy-to-use, semi-automated ground truth annotation and preprocessing methods. We separately segment the (1) colony background from agar and (2) the internal colony ring boundaries. The first task is achieved with a patch-classification approach; in the process, we find that the combination of the trained CNN and the “majority voting” method of label fusion achieves a test DICE score of 93% and correctly segments even faint outer swarm rings. The second task is accomplished with a U-Net which achieves over 83% test DICE. We show that our trained models easily segment a set of colonies generated at two relevant conditions, enabling automated analysis of features such as area and ring width. We apply our pipeline to analyze the more complex patterns of our engineered strains, such as the pCopA-flgM strain. The work in this chapter altogether advances the ability to analyze swarm patterns of P. mirabilis. We also aim to expand the use of our colony-characterization approaches beyond P. mirabilis to other microbes. Therefore, we present our work using deep learning to classify a set of Bacillus species isolated from soil samples. We generate datasets of the species grown under different conditions and apply transfer learning to train well-known CNN architectures such as ResNet and Inception to classify these datasets. This approach allows the models to easily learn these small datasets, and the models generalize to correctly predict a species which forms branching patterns regardless of exact growth condition. We visualize the attributions of the models with the integrated gradients method and find that model predictions are attributable to colony regions. This work sets the stage for classification, segmentation, and characterization of a wider array of microbial species with distinctive macroscale colony morphologies. Finally, we conclude by discussing ongoing efforts to expand upon the work presented in this thesis towards the sensing of dynamic inputs such as light, engineering of species other than P. mirabilis, and further optimization of the system of an engineered swarm pattern as a macroscale biosensor readout. Such work can contribute not only to the fields of synthetic pattern formation and the study of bacterial swarming, but also to the fields of engineered living materials and bio-inspired design.

Rhodopseudomonas palustris is a metabolically versatile phototrophic α-proteobacterium. The organism experiences a wide range of stresses in its environment and during metabolism. The oxidative an pH stresses of four ECF (extracytoplasmic function) σ-factors are investigated. Three of these, σ0550, σ1813, and σ1819 show responses to light-generated singlet oxygen and respiration-generated superoxide reactive oxygen species (ROS). The EcfG homolog, σ4225, shows a high response to superoxide and acid stress. Two proteins, one containing the EcfG regulatory sequence, and an alternative exported catalase, KatE, are presented to be regulated by σ4225. Transcripts of both genes show similar responses to oxidative stress compared to σ4225, indicating it is the EcfG-like σ-factor homolog and controls the global stress response in R. palustris.

Adaptation to any undesirable change in the environment dictates the survivability of many microorganisms. Such changes generate a quick and suitable response, which guides the physiology of bacteria. Stringent response is one of the mechanisms that can be called a survival strategy under nutritional starvation in bacteria and was first observed in E. coli upon amino acid starvation, when bacteria demonstrated an immediate downshift in the rRNA and tRNA levels (Stent and Brenner 1961). Mutations that rendered bacteria insensitive to amino acid levels were mapped to an ‘RC gene locus’, later termed relA because of the relAxed behavior of the bacteria (Alfoldi et al. 1962). Later on, Cashel and Gallant, showed that two “magic spots” (MSI and MSII) were specifically observed in starved cells when a labeled nucleotide extract of these cells was separated by thin layer chromatography (Cashel and Gallant 1969). These molecules were found to be polyphosphate derivatives of guanosine, ppGpp and pppGpp (Cashel and Kalbacher 1970; Sy and Lipmann 1973), and were shown to be involved in regulating the gene expression in the bacterial cell, demonstrating a global response, thus fine-tuning the physiology of the bacterium. Two proteins in E. coli, RelA and SpoT, carry out the synthesis and hydrolysis of these molecules, respectively, and maintain their levels in the cell (Cashel et al. 1996; Chatterji and Ojha 2001). On the other hand, Gram-positive organisms have only one protein Rel carrying out the functions of both RelA and SpoT (Mechold et al. 1996; Martinez-Costa et al. 1998; Avarbock et al. 1999). Although Rel or RelA/SpoT has been studied from several systems in detail pertaining to the physiological adaptation, less information is available on the egulation of the protein activity under different conditions. Our studies show that the RelMsm is composed of several domains (HD, RSD, TGS and ACT) with distinct function. HD and RSD domains, present in the N-terminal half of the protein, harbor catalytic sites for the hydrolysis and the synthesis of (p)ppGpp, respectively. TGS and ACT domains, on the other hand, are present at the C-erminal half of the protein and have regulatory function. It, therefore, appears that a communication exists between these domains, to regulate protein activity. It was shown earlier, while studying Rel from S.equisimilis, that there exists an interaction between the C-terminal and the N- terminal of the protein which determines the kind of activity (synthesis/hydrolysis), the protein should demonstrate (Mechold et al. 2002). Later, the N-terminal half crystal structure of the same protein suggested an inter-domain “cross-talk” between the HD and the RSD domain that controls the synthesis/hydrolysis switch depending on cellular conditions (Hogg et al. 2004). In the present work, studies have been carried out to understand a Gram- positive Rel in greater detail and to find out how the opposing activities of Rel are regulated so that a futile cycle of synthesis and hydrolysis of (p)ppGpp, at the expense of ATP, can be avoided. The work has been divided into several chapters describing studies on various aspects of the protein. Chapter 1 outlines the history of the stringent response and summarizes the information available about the stringent response in various systems including plants. Several roles that (p)ppGpp plays in different bacteria have been examined. A special mention on the crystal structure of RelSeq has been made with respect to the regulation of activity. Also, the information available regarding the effects of (p)ppGpp on RNA polymerase has been documented. Role of ppGpp in plants has been discussed in great detail with special emphasis on abiotic stresses. Since different functional domains have been identified in RelMsm, the protein has been divided into two halves and they have been discussed separately in the form of two chapters. Chapter 2 describes the N-terminal half of the Rel protein of M. smegmatis in greater detail. Out of the several domains identified, the role of the two domains present in the N-terminal half of the protein has been studied. The N-terminal half shows both synthesis and hydrolysis activities. Importantly, we find that the protein is active even in the absence of accessory factors such as ribosome and uncharged tRNA, unlike RelA of E. coli. Moreover, deletion of the C-terminal half of the protein leads to a much higher synthetic activity, clearly indicating that the C-terminus is involved in regulating the activity of the protein. Both TGS and ACT domains (the two domains found in the C-terminal half of the protein) have been found to play a regulatory role. The results also indicate that all the deleted constructs are active both in vitro and in vivo. Chapter 3 discusses the C-terminal half of the protein and its role in the multimerization observed in RelMsm. We show that multimerization of Rel protein is due to the inter-molecular disulfide cross-linking. Furthermore, we find that the monomer is the active species in vivo. One of the fascinating points about the C- terminal half is that it is largely unstructured. Additionally, the C-terminal half cannot complement the N-terminal part of the protein when provided in trans, demonstrating further, the requirement of an intact protein for bringing about regulation of Rel activity. This requirement in cis suggests the presence of an intra-molecular communication between the N- and the C-termini, as a mediator of protein regulation. Further, presence of uncharged tRNA increases pppGpp synthesis and down-regulates its hydrolysis in the wildtype protein. However, the uncharged tRNA-mediated regulation is absent in the deleted construct with only the N-terminus half, indicating that uncharged tRNA binds to the C-terminal half of the protein. Several cysteine mutants have been constructed to understand their role in the regulation of Rel activity. The results suggest that one cysteine, present at the C-terminus, is required for intra-molecular cross-talk and the uncharged tRNA-mediated regulation. A detailed characterization of the communication between the two halves of the protein has been attempted in Chapter 4. Surface plasmon resonance experiments carried out on the different cysteine mutants discussed in Chapter 3, for uncharged tRNA binding indicate that all the mutants bind to uncharged tRNA with near-equal affinities as the wildtype protein. This study suggests that the non-responsiveness for tRNA seen in one of the cysteine mutants is due to the loss of inter-domain interaction, while the binding of protein to accessory factors is unaffected. Fluorescence resonance energy transfer has been carried out to observe domain movement in the presence of accessory factors. Distances between the different domains scattered in this ~90 kDa protein, measured by FRET technique, are suggestive of an inter-domain cross-talk, specifically between C338 and C692, thereby regulating the activity of this enzyme. We show, for the first time, that the product of this protein, (p)ppGpp can bind to the C-terminal half making it unstructured, and can, therefore, regulate the protein activity. Chapter 5 is an effort to characterize the promoter of rel from M. tuberculosis. This study was undertaken in order to develop an expression system in mycobacteria. The +1 transcription and the translation start sites have been identified. The –10 hexamer for the RNA polymerase binding has also been mapped using site-directed mutagenesis and is found to be TATCCT. This promoter is also unusually close to the +1 transcription start site. The promoter is specific for mycobacteria and does not function in E. coli. Additionally, the promoter is found to be constitutive in M. smegmatis; however, the possibility of it being regulated in M. tuberculosis cannot be ruled out. Appendix section discusses, in short, the phylogenetic analysis of the mycobacterial Rel sequences. Diagrams of the plasmids used in this study have been provided. Mass spectra recorded for the in vitro synthesized and purified pppGpp and the trypsin digest of the full-length Rel protein have also been given. O O O O

Indiana University-Purdue University Indianapolis (IUPUI) / Oxalate decarboxylase (EC 4. 1. 1. 2 OxDC) from Bacillus subtilis is a manganese-dependent enzyme that catalyzes the cleavage of the chemically inactive C-C bond in oxalate to yield formate and carbon dioxide. A mechanism involving Mn(III) has been proposed for OxDC, however no clear spectroscopic evidence to support this mechanism has yet been obtained. In addition, a recent study has shown that N-terminal metal binding site loop variants of OxDC were able to catalyze the oxidation of oxalate to yield hydrogen peroxide and carbon dioxide, which makes OxDc function as another oxalate degradation protein in the cupin superfamily, oxalate oxidase (EC 1.2.3.4 OxOx). In this work, wild-type (WT) Bacillus subtilis OxDC and a series of variants with mutations on conserved residues were characterized to investigate the catalytic mechanism of OxDC. The application of membrane inlet mass spectrometry (MIMS), electronic paramagnetic resonance (EPR) spectroscopy and kinetic isotope effects (KIEs) provided information about the mechanism. The Mn(III) was identified and characterized under acidic conditions in the presence of dioxygen and oxalate. Mutations on the second shell residues in the N-terminal metal binding site affected the enzyme activity properties of the metal. In the N-terminal domain, the functional importance of the residues in the active site loop region, especially Glu162, was confirmed, and evidence for the previously proposed mechanism in which OxDC and the OxDC/OxOx chimeric variant share the initial steps has been found. In addition, the mono-dentate coordination of oxalate in the N-terminal metal binding site was confirmed by X-ray crystallography. A proteinase cleavable OxDC was constructed and characterized, revealing the interaction between the N-terminal and C-terminal domains.

Indiana University-Purdue University Indianapolis (IUPUI) / Pseudomonas aeruginosa is a gram negative opportunistic pathogen with the capacity to cause serious disease by forming biofilms, most notably in the lungs of cystic fibrosis (CF) patients. Biofilms are communities of microorganisms that adhere to a solid surface, undergo global regulatory changes, secrete exopolysaccharides, and are innately antibiotic resistant. Virulence modulation is an important tool utilized by P. aeruginosa to propagate infection and biofilm formation in the CF airway. Many different virulence modulatory pathways and proteins have been identified including the protein, MgtE. MgtE has recently been discovered and has been implicated in virulence modulation, as an isogeneic mutation of mgtE leads to increased cytotoxicity. To further elucidate the role of MgtE in P. aerugionsa infections, transcriptional and translational regulation of this protein following antibiotic treatment has been explored. I have demonstrated that mgtE is transcriptionally upregulated following antibiotic treatment of most of the twelve antibiotics tested utilizing RT-PCR and QRT-PCR. A novel model system was employed, which utilizes cystic fibrosis bronchial epithelial (CFBE) cells homozygous for the ΔF508 mutation for these studies. This model system allows P. aeruginosa biofilms to form on CFBE cells modeling the P. aeruginosa in the CF airway. Translational effects of antibiotic treatment on MgtE have been attempted via Western blotting and cytotoxicity assays. Furthermore, to explore the possibility that mgtE is interacting with a known regulatory pathway, a transposon-mutant library was utilized and the regulatory proteins, AlgR and NarX, among others have been identified as possibly interacting with MgtE. Lastly, an MgtE homologue from Staphylococcus aureus was utilized to further demonstrate the virulence modulatory effects of MgtE by demonstrating the expression of the homologue results in decreased cytotoxicity, exactly like expression of the native P. aeruginosa MgtE. This research explores a newly discovered protein that impacts cytotoxicity and biofilm formation and provides valuable information about P. aeruginosa virulence.

Indiana University-Purdue University Indianapolis (IUPUI) / Borrelia burgdorferi, a bacterium which causes Lyme disease, is maintained in nature through a cycle involving two distinct hosts: a tick vector and a mammalian host. To adapt to these two diverse environments, B. burgdorferi undergoes dramatic alterations in its surface lipoprotein. Two essential lipoproteins, outer surface protein A (OspA) and outer surface protein C (OspC), are reciprocally regulated throughout the B. burgdorferi lifecycle. Very little is known about the regulation of OspA. These studies elucidate the regulatory mechanisms controlling the expression of OspA. Various truncations of the ospA promoter were created and then studied in our novel in vitro model of ospA repression or grown within the host-adapted model. A T-Rich region of the ospA promoter was determined to be a cis-element essential for both the full expression and full repression of ospA.

O operon groESL de C. crescentus apresenta dupla regulação. A indução deste operon por choque térmico é dependente do fator sigma de choque térmico σ32. A temperaturas fisiológicas, a expressão de groESL apresenta regulação temporal durante o ciclo celular da bactéria e o controle envolve a proteína repressora HrcA e o elemento CIRCE (controlling inverted repeat of chaperonin expression). Para estudar a atividade da proteína repressora in vitro, produzimos e purificamos de E. coli a HrcA de C. creseentus contendo uma cauda de histidinas e a ligação especifica ao elemento CIRCE foi analisada em ensaios de migração retardada em gel de poliacrilamida (EMRGP). A quantidade de DNA retardada pela ligação a HrcA aumentou significativamente na presença de GroES/GroEL, sugerindo que estas proteínas modulam a atividade de HrcA. Corroboração desta modulação foi obtida analisando fusões de transcrição da região regulatória de groESL com o gene lacZ, em células de C. crescentus produzindo diferentes quantidades de GroES/EL. HrcA contendo as substituições Pro81 AJa e Arg87Ala, aminoácidos que se localizam no domínio putativo de ligação ao DNA da proteína, mostraram ser deficientes na ligação a CIRCE, tanto in vitro como in vivo. Em adição, HrcA Ser56Ala expressa na mesma célula juntamente com a proteína selvagem produziu um fenótipo dominante-negativo, indicando que a HrcA de C. crescentus liga-se a CIRCE como um oligômero, provavelmente um dímero. As tentativas de obtenção de mutantes nulos para os genes groESL ou dnaKJ falharam, indicando que as proteínas GroES/GroEL e DnaK/DnaJ são essenciais em C. crescentus, mesmo a temperaturas normais. Foram então construídas no laboratório as linhagens mutantes condicionais SG300 e SG400 de C. crescentus, onde a expressão de groESL e de dnaKJ, respectivamente, está sob controle de um promotor induzido por xilose (PxyIX). Estas linhagens foram caracterizadas quanto á sua morfologia em condições permissivas ou restritivas, assim como quanto à capacidade de sobrevivência frente a vários tipos de estresse. As células da linhagem SG300, exauridas de GroES/GroEL, são resistentes ao choque térmico a 42°C e são capazes de adquirir alguma termotolerância. Entretanto, estas células são sensíveis aos estresses oxidativo, salino e osmótico. As células da linhagem SG400, exauridas de DnaKlJ, são sensíveis ao choque térmico, à exposição a etanol e ao congelamento, e são incapazes de adquirir termotolerância. Além disso, tanto as células exauridas de GroES/GroEL quanto as exauridas de DnaK/DnaJ apresentam problemas na sua morfologia. As células de SG300 exauridas de GroES/GroEL formam filamentos longos que possuem constrições fundas e irregulares. As células de SG400 exauridas de DnaK/DnaJ são apenas um pouco mais alongadas que as células pré-divisionais selvagens e a maioria das células não possuem septo. Estas observações indicam bloqueio da divisão celular, que deve ocorrer em diferentes estágios em cada linhagem. / In Caulobacter crescentus, the groESL operon presents a dual type of control. Heat shock induction of the operon is dependent on the heat shock sigma factor σ-32. At physiological temperatures, groESL expression is cell cycle regulated and the control involves the repressor protein HrcA and the element CIRCE (controlling inverted repeat of chaperonin ~xpression). To study the activity of HrcA in vitro, we produced and purified from E. coli a histidine-tagged version of the protein, and specific binding to the CIRCE element was analyzed in electrophoretic mobility shift assays (EMSA). The amount of retarded DNA increased significantly in the presence of GroES/GroEL, suggesting that these proteins modulate HrcA activity. Further evidence of this modulation was obtained using lacZ transcription fusions with the groESL regulatory region in C. crescentus cells producing different amounts of GroES/GroEL. The mutants proteins HrcA Pro81Ala and HrcA Arg87Ala, that contain amino acid substitutions in the putative DNA-bindíng domain of the protein, were found to be deficient in binding to CIRCE in vitro and in vivo. Furthermore, HrcA Ser56Ala expressed together with the wild type protein within the same cell, produced a dominant-negative phenotype, indicating that C. crescentus HrcA binds to CIRCE in an oligomeric form, most likely as a dimer. Attempts to obtain null mutants for groESL or dnaKJ were unsuccessful indicating the importance of GroES/GroEL and DnaK/lDnaJ to the survival of C. crescentus cells. Conditional mutants were then constructed in our laboratory in which groESL and dnaKJ expression is under the control ofaxylose inducible promoter (PxyIX) , giving rise to strains SG300 and SG400, respectively. These strains were characterized in regard to their morphology under permissive and restrictive conditions, as well as their viability under different types of environmental stresses. SG300 cells depleted of GroES/GroEL are resistant to heat shock at 42°C and can acquire some thermotolerance, but they are sensitive to oxidative, saline and osmotic stresses. SG400 cells depleted of DnaKlJ are quite sensitive to heat shock, ethanol and freezing, and are unable of acquiring thermotolerance. Cells depleted of either GroES/EL or DnaKlJ also present morphological problems. SG300 cells depleted of GroES/EL form long and pinched filaments. SG400 cells depleted of DnaKlJ are only somewhat more elongated than wild-type predivisional cells and most cells do not present septum. These observations indicate a cell division arrest, which should occur at different stages in each strain.