This item is only available as the following downloads:
1 LACTOSE/GALACTOSE METABOLISM IN STREPTOCOCCUS GORDONII By NICOLE MARTINO A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2011
2 2011 Nicole Martino
3 To my mom, dad, sister, b rother and a unt
4 ACKNOWLEDGMENTS I would like to thank my advisor, Dr. Robert Burne, for his immeasurable guidance, support and kindness throughout this entire process. I am truly grateful for the opportunity to work in his lab. I would also like to thank my supervisory committee, Dr. Paul Gulig and Dr. Graciela Lorca, for investing the ti me and effort necessary to provide their honest and open opinions as well as providing me with so much support I am extremely grateful to Dr. Lin Zeng for providing me with his valuable insight and advice as well as answering all of my questions. He invested a lot of time into teaching me everything I needed to know to succeed. I also greatly appreciate the help I received from other members of the lab including Kinda, Matt, Dr. Bryan Korithoski, Chris, Dr. Liu and Dr. Ahn. I especially appreciate my fam ily for supporting me throughout the past two years and giving me unconditional love and support through the good and difficult times. I dont know what I would have done without my mother and aunt! They were both instrumental to my success in this program and I truly appreciate everything they have done for me.
5 TABLE OF CONTENTS page ACKNOWLEDGMENTS .................................................................................................. 4 LIST OF TABLES ............................................................................................................ 8 LIST OF FIGURE S .......................................................................................................... 9 ABSTRACT ................................................................................................................... 11 CHAPTER 1 INTRODUCTORY REMARKS ................................................................................ 13 General Overview Of The Oral Microbiome In Health And Disease ....................... 13 Genus Streptococcus and Streptococcus gordonii ........................................... 13 Composition of the Normal Oral Flora .............................................................. 14 Colonization of the Oral Cavity and Biofilm Formation ..................................... 14 Ecological Determinants in the H uman Oral Cavity .......................................... 16 Interspecies Interactions in Dental Plaque ....................................................... 19 Importance of Sugar Metabolism ............................................................................ 21 In Prokaryotes .................................................................................................. 21 In the Oral Ca vity .............................................................................................. 21 Sugar Transport and Utilization .............................................................................. 23 Lactose and Galactose Structure and Function ................................................ 23 Sugar Transport Systems ................................................................................. 24 PEP Group Translocation Phosphotransferase System (PTS) ...................... 25 Tagatose6 Phosphate Pathway ...................................................................... 26 Leloir Pathway .................................................................................................. 26 Genetic Organization of the Lactose and Galactose Gene Clusters ....................... 27 Streptococcus mutans UA159 .......................................................................... 27 Streptococcus gordonii DL 1 Challis ................................................................ 28 Regulation of PTS Activity and the Tagatose6 Phosphate Pathway ..................... 29 Carbon Catabolite Repression ......................................................................... 29 Regulation of the lac Genes ............................................................................. 32 Specific Aims .......................................................................................................... 32 2 MATERIALS AND METHODS ................................................................................ 38 Bacterial Strains and Growth Conditions ................................................................ 38 DNA Manipulations ................................................................................................. 39 Growth Rate Assays ............................................................................................... 41 Analysis o f Promoter Gene Fusion Strains ............................................................. 42 RNA Isolation and Gene Expression via quantitative Real Time RT PCR .............. 43 Expression, Purification and Dialysis of Recombinant N terminal 6X His tagged LacR protein ........................................................................................................ 45
6 Electrophoretic Mobility Shift Assays ...................................................................... 48 Mixed Species Liquid Culture Competition Assay .................................................. 48 3 CHARACTERIZATION OF CONTIGUOUS LACTOSE/GALACTOSE OPERONS IN S. GORDONII DL 1 CHALLIS ......................................................... 53 Introduction ............................................................................................................. 53 Results .................................................................................................................... 54 Bioinformatics Review of the Lactose and Galactose Gene Clusters ............... 54 Growth Phenotype of the S. gor donii Lactose and Galactose Gene Clusters .. 57 TV 0.5% glucose ..................................................................................... 58 TV 0.5% galactose .................................................................................. 58 TV 0.5% lactose ...................................................................................... 59 Operonspecific Gene Expression via Quantitative Real Time RT PCR .......... 60 Discussion .............................................................................................................. 61 4 GENE REGULATION IN RESPONSE TO LACTOSE/GALACTOSE TRANSPORT AND ENZYMATIC METABOLISM ................................................... 81 Introduction ............................................................................................................. 81 Results .................................................................................................................... 82 Analysis of lacA1 and lacA2 Promoter Activity Using cat Gene Fusions .......... 82 Gene Expression of LacA1, LacA2, LacR and LacT ........................................ 85 In Vitro Binding Analysis of S. gordonii LacR Recombinant Protein to the lacA1 and lacA2 Promoters ........................................................................... 86 Discussion .............................................................................................................. 87 5 IMPACT OF GALACTOSE UTILIZATION ON THE INTER SPECIFIC COMPETITION BETWEEN CARIOGENIC S. MUTANS UA159 AND COMMENSAL S. GORDONII DL 1 ORAL STREPTOCOCCI .............................. 104 Introduction ........................................................................................................... 104 Results .................................................................................................................. 105 Growth Comparison of S. gordonii DL 1 and S. mutans UA159 ..................... 105 Mixed Species Liquid Culture Competition Assay .......................................... 106 TV + 0.5% galactose without phosphate buffer ........................................ 107 TV + 0.5% galactose + 50 mM phosphate buffer ..................................... 107 TV + 0.5% glucose without phosphate buffer ........................................... 108 TV + 0.5% glucose + 50 mM phosphate buffer ........................................ 109 Discussion ............................................................................................................ 109 6 SUMMARY AND FUTURE DIRECTIONS ............................................................ 122 Summary .............................................................................................................. 122 Future Directions .................................................................................................. 124 LIST OF REFERENCES ............................................................................................. 129
7 BIOGRAPH ICAL SKETCH .......................................................................................... 138
8 LIST OF TABLES Table page 2 1 Bacterial strains used in this study. .................................................................... 50 2 2 Primers used in this study. ................................................................................. 51 2 2 Continued ........................................................................................................... 52 3 1 Potential promoter sites involved in the lac operon of S. gordonii based on software prediction (Softberry BPROM). ......................................................... 68 3 2 A list of the genes involved in the lactose and galactose PTS and tagatose6 phosphate pathway in related species. ............................................................... 69 3 3 Calculated doubling times o f the wild type strain and mutated lac gene strains of S. gordonii based on growth curve data. ............................................. 70 4 1 CAT activity of the lacA1 promoter of S. gordonii DL1 in the background of the wild type, lacT (M1stop), and strains. ....................................... 95 4 2 CAT activity of the lacA2 promoter of S. gordonii in the background of the wild type, lacT (M1stop), lacT (M1stop) and strains. ........... 96 5 1 Calculated doubling times of the S. gordonii DL1 wild type and S. mutans UA159 wild type strains based on growth curve data. ...................................... 114 5 2 Average measured pH values from the mixed species competition assay testing viability. ................................................................................................. 119 5 3 Average measured pH values from the mixed species competition assay testing persistence. ........................................................................................... 121
9 LIST OF FIGURES Figure page 1 1 A general overview of the phosphoenolpyruvatedependent sugar:phosphotransferase system (PEP dependent PTS) involved in carbohydrate transport across the cell membrane. ............................................. 34 1 2 Schematic showing predicted pathways for catabolism of lactose and galactose by S. gordonii following transport via a sugar specific PTS and shunted through the tagatose6 phosphate pathway. ......................................... 35 1 3 Genetic organization of the lac operon in S. mutans UA1 59. ............................. 36 1 4 Genetic organization of the lac operon in S. gordonii DL1. ................................. 37 3 1 Growth of S. gordonii DL1 wild type, Lac, and lacT (M1stop) strains in TV 0.5% Glucose with an oil overlay. ......................... 71 3 2 Growth of S. gordonii DL1 wild type, Gal, and strains in TV 0.5% Glucose with an oil overlay. .............................................................. 72 3 3 Growth of S. gordonii DL1 wild type, and A2B2 strains in TV 0.5% Glucose with an oil overlay. ..................................... 73 3 4 Gro wth of S. gordonii DL1 wild type, Lac, and lacT (M1stop) strains in TV 0.5% Galactose with an oil overlay. ...................... 74 3 5 Growth of S. gordonii DL1 wild type, Gal, and strains in TV 0.5% Galactose with an oil overlay. ........................................................... 75 3 6 Growth of S. gordonii DL1 wild type, and A2B2 strains in TV 0.5% Galactose with an oil overlay. .................................. 76 3 7 Growth of S. gordonii DL1 wild type, Lac, and lacT (M1stop) strains in TV 0.5% Lactose with an oil overlay. .......................... 77 3 8 Growth of S. gordonii DL1 wild type, Gal, and strains in TV 0.5% Lactose with an oil overlay. ............................................................... 78 3 9 Growth of S. gordonii DL1 wild type, and A2B2 strains in TV 0.5% Lactose with an oil overlay.. ..................................... 79 3 10 Expression levels of the lacG and EIICGal transcripts via quantitative Real Time RT PCR when grown in various carbohydrate sources.. ........................... 80 4 1 Critical elements compr ising the putative promoter regions of lacA1 and lacA2 based on bioinformatics analysis. ............................................................. 94
10 4 2 CAT activity of the 385 bp region directly upstream of the ATG star t site of lacA1 ................................................................................................................. 97 4 3 CAT activity of the 308 bp region directly upstream of the ATG start site of lacA2 .. ................................................................................................................ 98 4 4 CAT activity of the 308 bp region directly upstream of the ATG start site of lacA2 ................................................................................................................. 99 4 5 Expression levels of the lacR, lacT lacA1 and lacA2 transcripts via quantitative Real Time RT PCR when grown in various carbohydrate sources. 5). ....................................................................................................... 100 4 6 Expression and purification of the S. gordonii LacR recombinant protein. ........ 101 4 7 EMSA of the S. gordonii recombinant LacR protein and the lacA1 promoter. .. 102 4 8 EMSA of the S. gordonii recombinant LacR protein and the lacA2 promoter.. 103 5 1 Growth of S. gordonii DL1 and S. mutans UA159 in TV 0.5% Glucose with an oil overlay.. .................................................................................................. 115 5 2 Growth of S. gordonii DL1 and S. mutans UA159 in TV 0.5% Galactose with an oil overlay. ............................................................................................ 116 5 3 Growth of S. gordonii DL1 and S. mutans UA159 in TV 2% Galactose with an oil overlay. ................................................................................................... 117 5 4 Mixed species Competition Assay testing viability of the S. gordonii DL1 and S. mutans UA159 strains when grown in TV supplemented with 0.5% glucose or galactose, with or without phosphate buffer. ................................... 118 5 5 Mixed species Competition Assay testing persistence of the S. g ordonii DL1 and S. mutans UA159 strains when grown in TV supplemented with 0.5% glucose or galactose, with or without phosphate buffer. ................................... 120 6 1 Model depicting Lac R and carbohydrate catabolite repression of the lacA1 and lacA2 promoters in S. gordonii when glucose is present. .......................... 126 6 2 Model depicting the regulation of the lac A1 and lacA2 promoters in S. gordonii when galactose is present. ................................................................. 127 6 3 Model depicting the regulation of the lacA1 and lacA2 promoters in S. gordonii when lactose is present. ..................................................................... 1 28
11 Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science LACTOSE/GALACTOSE METABOLISM IN STREPTOCOCCUS GORDONII By Nicole Martino August 2011 Chair: Robert A. Burne Major: Medical Sciences Streptococcus gordonii is a Gram positive, facultative anaerobic commensal bacterium commonly found in dental plaque. Its genome contains two gene clusters encoding of tagatose pathway genes ( lacABCD) and sugar phosphotransferase system (PTS) enzyme II components likely responsible for the transport of lactose (EIILac) and galactose (EIIGal). Also encoded with the EIIGalcontaining operon is a DeoR like transcription regulator (Lac R) and with the EIILaccontaining operon a putative PRD (PTS regulated domain) containing transcription antiterminator. In contrast, the Streptococcus mutans genome contains only one tagatose operon and a homologous LacR protein that are required for galac tose metabolism. Our goal was to determine the contributions of these genes to lactose/galactose metabolism and to test whether S. gordonii has a selective advantage over S. mutans attributable to highaffinity galactose transport. Genetic analyses of the lactose and galactose gene clusters in S. gordonii were carried out using various mutants and Real Time qRT PCR and promoter reporter fusions. Growth analysis showed that EIILac transports lactose and galactose. Analyses of Real Time qRT PCR results indica te that the expression of EIIGal is upregulated in galactose and lactose. Results from promoter cat (chloramphenicol acetyltransferase)
12 fusions implicate LacR as a negative regulator of the lacA1 and lacA2 promoters and show that CcpA is a negative regulat or of the lacA2 promoter. In vitro binding assays confirmed that LacR could bind to both the lacA1 and lacA2 promoter regions. A mixedspecies liquid culture competition assay was performed with S. gordonii and S. mutans grown in glucose and galactose medi a with and without phosphate buffer. Results showed that S. gordonii maintained an advantage over S. mutans due to its galactose PTS. However, S. mutans was better able to persist due to its acidogenic and aciduric properties. Collectively these results support that the galactose and lactose systems are differentially regulated in S. gordonii and that a highaffinity galactose PTS is advantageous when S. gordonii is competing against the caries pathogen S. mutans
13 CHAPTER 1 INTRODUCT ORY REMARKS General Overview Of The Oral Microbiome In Health And Disease Genus Streptococcus and Streptococcus gordonii The bacterial genus Streptococcus is comprised of Gram positive, nonmotile, facultative, anaerobic cocci. Belonging to the phylum Firmicutes, these bacteria divide along a single axis and often form chains of cells. Most streptococci are catalaseand oxidasenegative. Both pathogenic and commensal organisms are encompassed in this genus. Streptococci are broadly class ified based on a hemolysis pattern, with the viridans group of streptococci displaying alpha hemolysis due to its ability to oxidize the iron in hemoglobin. When grown on a blood agar plate, a green halo forms around the streptococcal colonies and is termed partial hemolysis. The viridans streptococci are further classified into groups based on gene sequence homology and a few defining characteristics, including the ability to ferment mannitol and sorbitol, produce glucans from sucrose and sustain growth at low pH ( 22) The re are five viridans groups including the anginosus, mitis, salivarius, bovis and mutans streptococci ( 46 ) Streptococcus gordonii belongs to the mitis group, which also includes Streptococcus mitis Streptococcus pneumoniae, Streptococcus oralis Streptococcus sanguis and Streptococcus parasanguis ( 46) Previously, S. gordonii was considered to be a part of the S. sanguis species until Kilian and coworkers defined it as a new species based on DNA sequence information ( 48 ) However, both species are still considered t o be commensal organisms associated with oral health. Streptococcus mutans belongs to the mutans group and is considered the primary etiological agent of dental caries. Earlier studies have indicated an inverse relationship in plaque samples
14 between the pr esence of S. mutans and strains classified as S. sanguis (including S. gordonii ) ( 31, 55 ) Composition of the Normal Oral Flora Recognized as one of the most densely populated regions of the body, the oral cavity harbors upwards of approximately 700 species of indigenous bacteria with about half of those organisms uncultivated in vitro thus far ( 47, 74) Streptococcus spp, Lactobacillus spp and other Gram positive filamentous rods (e.g. Actinom yces spp) constitute the vast majority of the resident flora in the mouth; however, Gram negative cocci (e.g. Veillonella spp, Neisseria spp) and Gram negative rods (e.g. Bacteroides spp, Fusobacterium spp, Vibrio spp, Spirillum spp, Prevotella spp, Porphy romonas gingivalis ) are also abundant. Spirochetes ( Treponema, Borrelia ) are commonly found in the gingival crevice and Candida albicans is the yeast most commonly isolated throughout the mouth ( 1 39 ) Oral streptococci are abundant in dental plaque found on the surface of the teeth. According to a study conducted in Pisa, Italy, S. mutans prevalence was measured at 14% in caries free patients, 27% in caries inactive patients and 67% in caries active pat ients ( 13) Mitis group streptococci, including S. gordonii and S. sanguis are associated with good oral health and higher proportions are present in caries free and caries inactive patients ( 14, 51 71) Colonization of the Oral Cavity and Biofilm Formation There are a number of distinct microcosms in the oral cavity such as saliva, the tongue dorsum, buccal mucosa, gingival crevice and supragingival space. Each ecol ogical niche provides a unique microenvironment with characteristics that render it capable of sustaining colonization and growth of certain microbial communities. For
15 example, the normal flora of the gingival crevice is dissimilar from that of the flora on the tooth surface. Colonization of these microcosms is based on ecological determinants that include preferential adherence to a particular surface, nutritional need, tolerance of environmental stresses and inhibitory substances and competition from neig hboring organisms ( 39 70 ) S. gordonii is most commonly isolated from the supragingival environment, which is located on the surface of the tooth enamel. Supragingival biofilms, often referred to as plaque, consists of a community of microorganisms with the capability to adhere to each other and to surfaces. Successful biofilm formation is an essential step in progression to a carious lesion. Biofilm formation begins when the acquired pellicle, a thin film over the tooth enamel, accumulates after cleaning of the surface. This coating is comprised of salivary glyocproteins and other proteinaceous and nonproteinaceous material ( 10) Pioneer (early) colonizers, including S. gordonii and Actinomyces spp, possess adhesins that allows for binding via receptors to the host pellicle. In a timedependent manner, late colonizers begin to coaggregate with previously attached early colonizers on the tooth structure, resulting in a multi tiered cluster of cells with a surrounding matrix of bacterial and host components called the glycocalyx ( 60) Glucosyltransferase enzymes produced by most oral streptococcal species hydrolyze sucrose and polymerize the glucose monomers into sticky extracellular glucans, which contribute to glycocalyx synthesis and adherence to the enamel ( 96) Mature biofilms have a high degree of structural complexity, bacterial heterogeneity and differ metabolicall y, physiologically and biochemically from planktonic cells ( 27 28 ) The levels of gene expression in an established biofilm differ from planktonic cells. In addition, a mature biofilm develops an
16 increased resistance to antibiotics, host defenses and environmental stresses ( 27 28, 44) Because S. mutans is both acidogenic (creating abundant amounts of acid) and aciduric (able to sustain growth in especially acidic conditions), biofilms with high proportions of this species (considered the disease state) are at an increased risk for producing copious amounts of acid. Prolonged acid exposure is responsible for demineralization of the tooth enamel and the primary cause for developing a carious lesion ( 49 84 85 99) Ecological Determinants in the Human Or al Cavity Inside the human mouth is an ever changing environment. There are a multitude of different bacterial species suspended in saliva due to washoff from the many different surfaces ( 50 ) However, most bacteria cannot remain in the saliva permanently due to its constant flow and movement. It is imperative for an organism to adapt by adhering to a surface. If not properly adherent, the microorganism can easily be washed away with the curr ent of saliva and passed into the intestinal tract via swallowing. In order to circumvent physical removal, the correct niche must be located and adherence achieved via receptors on the acquired host pellicle or neighboring microbes ( 66) Attachment ensures that the microorganism is able to compete in the surrounding environment; however, it does not guarantee survival. Fermentation of carbohydrates is the main mechanism in which viridans streptococci gai n a source of energy. The availability of nutritional resources depends on the host diet, food debris, metabolic products of other bacteria, and constituents in saliva ( 42) Of those mentioned, saliva is the main biological fluid in the oral cavity and the most constant source available to most microbes in vivo Therefore, salivary
17 glycoconjugates (espe cially O and N linked oligosaccharides of glycoproteins and mucins) can serve as an important nutrient resource for those bacteria harboring the genetic machinery necessary to exploit it. Secreted and/or surfaceassociated glycosidases are enzymes that modify and cleave glycoconjugates found in the saliva. Both S. pneumoniae and S. oralis have been shown to possess glycosidases capable of acid glycoprotein (AGP), first with neuraminidase to cleave off the terminal sialic acid galactosidase to cleave off the second galactose moiety ( 18, 21, 86) The study on S. oralis also reported that almost the entire oligosaccharide was deglycosylated from the glycoprotein ( 21 ) The release of these sugar moieties served as a nutritional source and supported growth even when no free carbohydrates were present. S. gordonii galactosidase activity, bgaA and bgaC. Thus, if S. gordonii w ere capable of catabolizing glycoconjugates in saliva, this mechanism would serve as an additional nutrition source. It is important to keep in mind that the release of sugar moieties from glycoproteins would produce only low concentrations of free sugar i n the environment as opposed to large spikes in sugar concentrations such as after a meal. Further conjecture would hint that a highaffinity transport system directed at the uptake of sugars embedded in glucoconjugates has the potential to confer a select ive advantage over species that do not have the same genetic components. Therefore, it may be advantageous for S. gordonii to possess a highaffinity galactosespecific PTS since galactose is often the second moiety cleaved from glycoproteins after sialic acid ( 41)
18 Adaptation to environmental stresses such as pH, oxygen tension and inhibitory substances within the mouth is vital to the survival of both planktonic and biofilm associated microbes. The ability to withstand these pressures will determine which species flourish. Most microorganisms are sensitive to the effects of acid. Any pH not in the tolerable range is likely to kill the microbe or lessen its colonization and/or virulence ( 65) Exceptions include the mutans streptococci and lactobacilli, which are acidogenic and aciduric (reference). These species are responsible for much of the acid production by the degradation of carbohydrates via the EmbdenMeyerhof Parnas (EMP) pathway, which creates acid as an end product ( 16) When acid accumulates above the acceptable limits for acidsensitive species, the balance between commensal and cariogenic bacteria favors disease. Oxygen concentration in the supragingival space varies greatly depending on the age, activity and type of biofilm formed. In a newly forming biofilm, oxygen flows relatively freely throughout the structure due to its thin layer of cells ( 65) In contrast, it is much more difficult for oxygen to diffuse through a mature biofilm because of its thickness; therefore, an oxygen gradient forms ( 19) Due to the heterogeneous nature of dental plaque, aerobic bacteria become positioned closer to the edge of the biofilm w hile anaerobic bacteria remain deeper in the biofilm ( 65) This is because oxygen levels tend to decrease as biofilm depth increases ( 65 ) Oral bacteria have a high oxygen metabolism and, as a result, toxic products are produced by metabolic activity that are susceptible to the li mitations of diffusion ( 19) This can increase the exposure time to stress and is especially harmful for anaerobic bacteria.
19 Inhibitory substances secreted by the host and other bacteria can threaten a microbial population as well. Alt hough mature biofilms have a lower risk of being affected by these substances ( 28) both planktonic and biofilm associated cells must contend with this threat. Bacteriocins and hydrogen peroxide (H2O2) production by various microbial species, secretion of proteases, generation of alkaline products, and antibiotics are all examples of the various types of substances that can interfere with the growth of bacteria ( 54 ) In some cases, two species will produce opposing bacteriocins or inhibitory substances in an attempt to outcompete each other for a particular space ( 42, 54 ) The microbe with the better adaptive mechanism often wins. All of the factors mentioned in this section generate constant competition among the microbial community and the outcome determines whether the oral cavity will be one of health or disease. Interspecies Interactions in Dental Plaque Because dental plaque contains a multitude of bacterial species, it is logical to anticipate extensive bacterial interactions among the community. Furthermore, these interactions not only extend to the process of biofilm formation, but also to growth and survival. Interactions between cariogenic bacteria, such as S. mutans and noncariogenic bacteria, such as S. gordonii can modulate the pathogenic potential of a biofilm ( 49 ) Multiple researchers have reported evidence suggesting both cooperative and antagonistic interactions among species in the oral microbiome ( 42, 52, 54, 59 73 ) Microorganisms established in a biofilm have evolved a host of metabolic pathways designed to more efficiently utilize the varying carbohydrate sources encountered in the respective niches. For example, although many streptococcal species possess genes for glucosyltransferase enzymes, only mutans streptococci
20 possess extracellular glucanbinding proteins ( 11) These extracellular glucanbinding proteins may increase the cariogenic potential of S. mutans by mediating the binding of glucan to the bacterial cell and promoting adhesion to the biofilm matrix ( 11) In addition, the ability to metabolize sucrose into lactic acid more rapidly than other bacteria provides another competitive advantage for S. mutans ( 54) In contrast, synergistic interactions in a bi ofilm have been described as well. For example, Actinomyces naeslundii and S. oralis have been shown to poorly colonize saliva coated surfaces when incubated separately, but form considerable biofilms on the same surfaces when incubated together ( 73) The combination of metabolic activities from each spec ies was sufficient to support a mutualistic relationship and sustain growth of a biofilm. The metabolites produced from each microbe may likely be affecting other organisms in the biofilm. For example, plaque with large proportions of S. mutans often presents low quantities of S. sanguinis ( 59 ) Production of high amounts of lactic acid by the former is detrimental to nonaciduric species such as S. sanguinis thereby favoring S. mutans In response, S. sanguinis and S. gordonii produce H2O2, which cannot be metabolized by S. mutans and acts as a nonspecific antimicrobial agent ( 53, 54) In addition, the metabolic products of one mi crobe may support the growth of another. For example, the lactic acid generated by S. mutans is metabolized into H2O2 and pyruvate via the enzyme lactate oxidase by Streptococcus oligofermentans which is associated with caries free patients ( 90, 91 ) Both commensal and pathogenic bacteria produce inhibitory substances and/or enzymes in order to effectively compete against neighboring species. Kuramitsu
21 demonstrated that S. gordonii is able to degrade the competence stimulating peptide (CSP) made by S. mutans via the challisin protease, leading to attenuation of bacteriocin production and inhibition of sucrosedependent biofilm formation ( 55 ) Bacteriocins are proteinaceous toxins usually affecting closely related organisms ( 25) S. mutans is able to produce at least five bacteriocins (mutacins I V). Mutacins I, II and III are lanthioninecontaining lantibiotics with a broad killing spectrum whereas mutacins IV and V are unmodified lantibiotics with mutacin IV specifically targeting the mitis group of streptococci ( 54) Importance of Sugar Metabolism In Prokaryotes The most critical objective for a prokaryote is to persist and grow by utilizing biosynthetic pathways to produce the necessary constituents of a living cell. These pathways require energy in the for m of ATP. Therefore it is of interest for the cell to obtain energy in the most efficient manner. Heterotrophic bacteria often catabolize carbohydrates to acquire energy. These carbohydrates can be taken into the cell in the form of monosaccharides or disaccharides, depending on the enzymes available to the microbe. Each species possesses genes encoding a variety of enzymes and this determines which sugars are available for uptake and catabolism. It also determines the niche occupied by the species since uptake of the correct substrates in the microenvironment is required for energy and survival. In the Oral Cavity The primary function of the oral cavity is to provide the host with nutrition, often in the form of carbohydrates. It therefore follows that bact eria colonizing the mouth should regard carbohydrate utilization as an important aspect of survival and growth since
22 there are so many different substrates to choose from. Because of the extensive variety of nutritional resources available, there is a cons tant struggle among different species to utilize the most efficient transport mechanisms in hopes of outcompeting neighbors and keeping energy expenditures as low as possible. This is evidenced by the fact that many species in the mouth possess multiple substratespecific carbohydrate transport mechanisms ( 7 93) The importance of carbohydrate metabolism first became evident in 1860 when Pasteur reported that bacteria produce lactic acid via sugar fermentation ( 75 ) In vitro studies then demonstrated that exposure to the acidic by products of sugar metabolism can induce dental caries ( 64) Finally, Miller published results indicating that bacteria from human saliva can produce lactic acid from dietary constituents and that the quantities made were sufficient enough to cause demineralization of the tooth enamel ( 68) A few decades l ater, Stephan provided the first in vivo evidence linking bacterial carbohydrate metabolism with acid production ( 84) He attributed the findings to a causal relationship between the sharp decrease in plaque pH measured after a sugar rinse and the presence of a mixed population of bacteria ( 84 85) Sugar metabolism by way of the EMP pathway (glycolysis + homolactic acid fermentation) creates lactic acid as an end product, which is responsible for the decline in pH ( 81) However, Stephan also noted a gradual increase in pH with an eventual plateau as acid production diminished. It was shown years later that the gradual rise in pH was due to the generation of alkaline products by acidsensitive microbes present nearby ( 98)
23 With the information gained from the previously mentioned studies, it appears that there is a fluctuating balance between acidtolerant and acidsensitive bacteria within a dental biofilm. This balance stems from the fact that acidtolerant species tend to rapidly metabolize sugar and lower the pH while acid sensitive species tend to counteract that by producing basic products to increase the pH. Only when the composition of the microbial community begins to deviate from that of the normal healthy flora does the cariogenic potential of the biofilm increase. The normal healthy flora includes bacterial species that either benefit the host or do not benefit the host but also do not cause disease. Microbes associated with disease such as S. mutans and lactobacilli (dental caries) or Porphyromonas gingivalis (periodontal disease) are not found in high numbers, if any, of dental plaque isolated from healthy teeth ( 1 ) Therefore, a deviation from the normal healthy flora would include increased numbers of pathogenic species or the conversion of opportunistic pathogens into full fledged pathogens. Sugar Transport and Utilization Lactose and Galactose Structure and Function Lactose is a disaccharide composed of glucose and galactose fused together via a galactose. Commonly found in milk and other dairy products, lactose acquired its name from lactis the Latin word for milk. Its mol ecular formula is C12H22O11lactose and lactose. The difference between these isomers lies in the orientation of the hydrogen and hydroxyl groups of carbon number 1 in the glucose moiety. In reality, lactose actually ex ists in equilibrium as a combination of both of these forms.
24 Galactose is a monosaccharide composed of 6 carbon atoms and is a C 4 epimer of glucose. It is often found in dairy products, pectins, gums and mucilages as well as sugar beets and various fruits and vegetables. Its molecular formula is C6H12O6. Aside from transportation across the bacterial membrane for use as energy, galactose is also often incorporated as the second moiety in glycoproteins on the bacterial cell membrane surface ( 41 ) galactosidase enzymes made by the cell to release the galactose for energy utilization ( 21 86 ) The free galactose is then available for uptake by the cell. Sugar Transport Systems Lactose and galactose are both transported through complex systems embedded in the bacterial membrane. In oral streptococci there are three transport systems known to exist for these sugars: (1) primary active transport, (2) secondary active transport and (3) group translocation (see this reference for review ( 33) ). Primary and secondary active transport systems pump in the sugar without modifying it chemically. Primary transport systems are common to all bacteria and require energy from ATP for the internalization of the sugar substrate. This sugar specific system is termed an ATP binding cassette (ABC) transporter and usually includes four domains that are often fused together, two inserted into the membrane and two located in the cytoplasmic sp ace. The two integral membrane proteins serve as the channel for the sugar substrate to pass through while the two cytoplasmic proteins act as ATPases, which harvest energy from ATP to power the movement across the membrane. Secondary transport systems do not use the energy from ATP to power the transfer across the membrane, but rather use a concentration gradient to force the sugar molecule into the cell. Two different types of secondary transport systems are
25 often employed for lactose and/or galactose upt ake: (1) a protoncoupled antiporter, or (2) lactosegalactose antiporter. The last transport system, group translocation, is the most efficient from a bioenergetics standpoint. Also known as the phosphoenolpyruvatedependent phosphotransferase system (PEP PTS), this multi component system transfers a phosphoryl group to inbound sugar substrates via a highenergy PEP molecule (Figure 1 1). The chemical composition of the carbohydrate is therefore altered. A signal cascade comprised of two general cytoplasmi c components and one transport permease aids in the transfer of the phosphate group from PEP to the sugar. This mechanism is found in both Gram negative and Gram positive bacteria with minor differences. While the cytoplasmic constituents are not substratespecific, the transport permease is. Usually each PTS only allows for one or two specific sugars to be transported, although a particular sugar can be transported by more than one PTS. For example, S. mutans has been shown to transport mannose, glucose and galactose through the mannose PTS (EIIMan). PEP Group Translocation Phosphotransferase System (PTS) The process starts with phosphoenolpyruvate, a glycolytic intermediate (Figure 11) (see this reference for review ( 93) ). The high energy bond created from detachment of the phosphoryl group is enough to power a signal cascade and allow for importation of the sugar. The phosphoryl group is transferred to a histidine residue located on the first cyt oplasmic protein, Enzyme I (EI). EI subsequently transfers it to histidine residue 15 on the second cytoplasmic protein, histidinephosphocarrier protein (HPr). HPr continues the cascade by donating the phosphate to the carbohydratespecific Enzyme II (EII) complex. This complex consists of a hydrophobic integral membrane domain
26 (subunit C, EIIC) and two hydrophilic cytosolic domains (subunit A and B, EIIAB). EIIA catalyzes the phosphate transfer from HPr and passes it on to either the histidine or cysteine of EIIB. The sugar enters via the translocating EIIC domain and docks at a sugar binding site. EIIB then provides the final phosphorylation onto the sugar at the position of carbon 6. The sugar is now free to move into the cytoplasm where it can be metabolized through the proper pathway. Tagatose 6 Phosphate Pathway Once lactose or galactose has been phosphorylated, the cell will direct it towards the correct pathway for catabolism (Figure 12). In the case of lactose6 phosphate (Lac 6 P), a cytosolic phospho galactosidase ( lacG ) will catalyze its breakdown into glucose and galactose 6 phosphate (Gal 6 P). Glucose enters the glycolytic pathway whereas Gal 6 P enters the tagatose pathway. Because galactose is already transported into the cell as Gal 6 P, there are no necessary steps to facilitate its entry into the tagatose pathway. The first enzymes in the pathway, both galactose6 phosphate isomerases ( lacA, lacB ), convert Gal 6 P into tagatose6 phosphate via an aldose ketose reaction. Then an additional phosphate is attached at the carbon 1 position of tagatose6 phosphate to create tagatose1,6 bisphosphate after tagatose6 phosphate kinase ( lacC) harvests energy from a molecule of ATP. The final enzyme in the pathway tagatose1,6 diphosphate aldolase ( lacD), cleaves tagatose1,6 bisphosphate into glyceraldehyde3 phosphate (G 3 P) and dihydroxyacetone phosphate (DHAP). Leloir Pathway Galactose can also be brought into the cell through a second mechanism of transport. A nonPTS permease can uptake the hexose and lead it into the alternative
27 Leloir pathway (see this reference for review ( 36) ). The permease does not alter the galactose substrate by phosphorylation. Instead, the sugar is transported in its free form. Once galactose has entered the Leloir pathway, it is phosphorylated at the carbon 1 position by ATP dependent galactokinase (GalK) to produce galactose1 phosphate (Gal 1 P). The next enzyme in the pathway, galactose1 phosphate uridylyltransferase (GalT), requires the co fact or UDP glucose. Gal 1 P is then transformed into the galactose derivative UDP galactose + glucose1 phosphate (Glc 1 P) via GalT. Finally, UDP glucose 4 epimerase (GalE) recycles UDP galactose by converting it back into UDP glucose while phosphoglucomutase converts Glc1 P into glucose6 phosphate. Galactose transport can occur through both the PTS and nonPTS permeases; however, usually one mechanism is more dominant than the other. There have been conflicting reports on which mechanism plays a dominant r ole in the metabolism of galactose in S. mutans Ajdic and Pham suggest that the Leloir pathway plays a dominant role due to abolished growth in galactose when galK is inactivated ( 7 ) In contrast, Zeng and coworkers argue that the lactose PTS plays a dominant role. Their findings show that a lacG def icient strain is unable to grow on galactose and over expression of the Leloir pathway in that same strain did not rectify growth on galactose ( 103) Therefore, it appears that in S. mutans the dominant transport system for galactose uptake is the lactose PTS. Genetic Organization of the Lactose and Galactose Gene Clusters Streptococcus mutans UA159 The S. mutans UA159 genome contains two main types of sugar transporters: the PEPdependent PTS or the ABC transporter. In a microarray analysis of the S. mutans genome after growth in 13 different sugars, Ajdic and Pham found that
28 glucosides, and sugar alcohols were all transported through the PTS while the ABC transporters were mostly specialized for oligosaccharide transport ( 7 ) Lactose enters S. mutans through the lactose PTS and continues to the tagatose pathway. Galactose may enter through the lactose PTS or through an unidentified nonPTS permease that leads the sugar into the Leloir pathway. It is likely that galactose enters the cell through both mechanisms. The genes for the lactose PTS, lacRABCDFEG are situated adjacently i n the genome (Figure 13). The lactosespecific transporter is annotated by lacF (EIIA) and lacE (EIIBC). A phospho galactosidase ( lacG ) catabolizes Lac 6 P before it enters the tagatose6 phosphate pathway. The enzymes in this pathway include the two su bunits of the heteromeric galactose6 phosphate isomerase ( lacAB ), a tagatose6 phosphate kinase ( lacC) and a tagatose 1,6 bisphosphate aldolase ( lacD). A transcriptional regulator ( lacR) sits just upstream of the entire operon ( 79 ) All of these genes are based on annotation and not experimentation. Streptococcus gordonii DL 1 Challis S. gordonii DL1 harbors a similar genetic layout to S. mutans UA159 in terms of the lactosespecific PTS and tagatose pathway genes (Figure 14). However, in contrast to S. mutans the S. gordonii genome contains a unique set of genes that appear to be part of a dedicated galactosespecific PTS, including SGO_1522 ( EIIA), SGO_1521 (EIIB) and SGO_1520 (EIIC). In addition, a second set of tagatose pathway genes are situated just upstream of the EIIGal genes. Another difference between the genetic organizations of the lac genes in S. mutans UA159 and S. gordonii DL1 lie s in a regulatory transcriptional antiterminator SG0_1515 ( lacT ), which is situated between the first set of tagatose genes and the lactosespecific PTS. Although the lacT gene is not
29 found in the S. mutans UA159 genome, it is present in the genome of anot her oral microbe, Lactobacillus casei ( 33) In addition to the lactose PTS, S. gordonii also contains genes for the Leloir pathway. Therefore, galactose is likely transported by both a PTS and nonPTS permease. However, because it appears from the S. gordonii lac genomic structure that a dedicated galactosespecific PTS exists, the main route of transport into the cell could likely be through the PTS. For the purposes of this thesis, the tag atose pathway genes just up stream of the lactose specific PTS have been termed the lactose cluster while the second set of tagatose pathway genes just upstream of the galactosespecific PTS have been designated as the galactose cluster. Regulation of PTS Activity and the Tagatose 6 Phosphate Pathway Carbon Catabolite Repression Under normal circumstances in the mouth, bacteria are often bathed in multiple sugar sources due to the variety of carbohydrates consumed in the diet and sugar moieties contained in glycoconjugates of saliva. During these conditions, the bacteria must decide the best course of action in regards to satisfying their nutritional requirements from the environment in the most efficient manner. Therefore, many bacterial species have developed a mechanism that preferentially utilizes a particular sugar source while repressing expression of the genetic components associated with an alternative carbon source ( 83) This mechanism is known as carbon catabolite repression (CCR) and is executed by both Gram negative and Gram positive microbes. In many cases, glucose is a preferred sugar source over sugars such as lactose and galactose. The main components involved in CCR of Gram positives include HPr, HPr
30 kinase/phosphatase (HPrK/P), the global transcriptional regulator catabolite control protein A (CcpA) and various carbohydratespecific EII of the PTS ( 37 83 ) Global control and regulation by these proteins is based on the physiological state of the cell and the available carbon sources ( 76) HPr is the primary protein involved in CCR in that it relays infor mation about the nutritional needs of the cell based on its phosphorylation state. When HPr is phosphorylated by EI at histidine residue 15 (P His HPr) it is responsible for signaling the transport of carbohydrates into the cell via PTS permeases. In addit ion, HPr is capable of being phosphorylated at serine residue 46 (P Ser HPr), through the enzymatic activity of HPrK/P. The fluctuating levels of P His HPr and P Ser HPr constantly change depending on signals in the cytosolic environment ( 76) CcpA, a member of the LacI GalR family of bact erial transcriptional regulators, is implicated in controlling genes related to energy, transport, metabolism and other functions ( 4 ) Although CcpA often acts as a repressor, this protein is capable of functioning as an activator of transcription as well depending on the location of the catabolite response element ( cre ) binding sequence in relation to the gene promoter ( 61) If the cre site is located downstream of the promoter then CcpA is most often a repressor whereas if the cre site is located upstream of the promoter then CcpA is most often an activator ( 37) Binding of CcpA to a cre sequence located proximal to promoter regions is stimulated by P Ser HPr and together they form a complex that corepresses genetic transcription of a gene by binding ( 38) CCR in Gram positive bacteria can be achieved via two different mechanisms: CcpA dependent or CcpA independent. In CcpA dependent CCR, phosphorylation by
31 EI produces P His HPr, which is necessary for the uptake of carbohydrates through the PTS. As the carbohydrates are metabolized, glycolytic intermediates such as fructose1,6 bisphosphate (FBP) and glucose6 phosphate (G 6 P) are released When these intermediates accumulate to a threshold concentration in the cell, HPrK/P is activated and an AT P dependent phosphorylation of HPr occurs at the serine residue (P Ser HPr) ( 80) With CcpA dependent CCR activated, P Ser HPr and CcpA bind to the cre sequence located in the promot er of the gene of interest and corepress transcription of the operon ( 83) When the concentration of FBP in the cell drops too low and the concentration of inorganic phosphate (Pi) builds up, HPrK/P works in the reverse direction to dephosphorylate P Ser HPr and produce pyrophosphate (PPi) + free HPr, which then becomes available to again bind phosphate at the histidine residue for carbohydrate transport ( 76) In CcpA independent CCR, components of the PTS are involved in regulating genetic transcription. In S. mutans the carbohydratespecific mannose permease (ManL), fructose permeases (FruI/FruCD), and the fructose/mannose permease (EIILev) have been identi fied as contributors in the CcpA independent repression of fruA and levDEFG ( 2 97 101 ) The mechanism for this type of CCR begins similarly to that of CcpA dependent CCR. The uptake of preferred sugar sources by the respective PTS creates FBP, which in turn activates ATP dependent phosphorylation by HPrK/P to create P Ser HPr ( 83) At this point, P Ser HPr has been postulated to bind to LevR and inhibit its activation of the fruA and levDEFG promoters via an allosteric mechanism, thus repressing transcription of the genes ( 102) Furthermore underphosphorylation of the EIIA and EIIB subunits of the ManL (and possibly FruI/FruCD, EIILev) transporter also represses transcription of the genes ( 102 )
32 Regulation of the lac Genes There is not much informati on available in the literature in regards to the regulation of the lactose and/or galactose PTS for S. gordonii ; however, research studies completed on species with similar lac genetic organization may provide insight into how S. gordonii regulates these gene clusters. Secondary regulators often exist such as repressors and antiterminators, which work in conjunction with CCR mediated controls ( 37) The repressor associated with the lactose PTS is LacR. The lacR gene, along with the rest of the lactose PTS, is present in S. mitis S. sanguinis S. pyogenes S. aureus S. mutans S. pneumoniae and L. lactis ( 33, 62, 87, 94 100 ) Principally, lacR is annotated as a transcriptional repressor, which is supported by evidence showing that LacR binds to its own promoter as well as the lacA promoter in L. lactis ( 32, 35, 95) Comparable roles for LacR have been hypothesized for S. aureus and S. mutans ; however, L. casei does not possess a lacR gene ( 33) A transcriptional antiterminator (LacT) similar to that of the BglG/SacY family is putatively present in the genome of al l species mentioned except S. mutans S. aureus and L. lactis The activity of LacT in L. casei is controlled by EIILac and the common PTS elements and is modulated by phosphorylation of two PTS regulation domains (PRD) ( 40) There have been no other studies published on the regulatory mechanism of LacT, indicating that this area of research is very novel. Specific Aims Aim I: Characterization of contiguous lactose/galactose operons in S. gordonii Aim II: Gene regulation in response to lactose/galactose utilization and enzymatic metabolism.
33 Aim III: Impact o f lactose/galactose utilization on the inter specific competition between cariogenic S. mutans and commensal S. gordonii oral streptococci.
34 Figure 1 1. A general overview of the phosphoenolpyruvatedependent sugar:phosphotransferase system (PEP dependent PTS) involved in carbohydrate transport across the cell membrane. A phosphate molecule is transferred from phosphoenolpyruvate to Enzyme I (EI). EI transfers the phosphate to residue histidine 15 of the histidine phosphocarrier protein (HPr). Subunit A of the Enzyme II (EII) complex passes the phosphate from P His HPr to subunit B of the EII complex The incoming sugar is transported and concomitantl y phosphorylated by the transmembrane protein subunit C of the EII complex. PEP Pyruvate E1 HPr A B C Enzyme II Complex (Subunits A, B, C) S S -P Cytoplasm Extracellular Space Cell Membrane
35 Figure 12. Schematic showing predicted pathways for catabolism of lactose and galactose by S. gordonii following transport via a sugar specific PTS and shunted through the tagatose6 phosphate pathway. Once the sugar enters the cell, lacG will catabolize Lac 6 P into glucose and Gal 6 P. Gal 6 P enters the tagatose pathway where the galatose6 phosphate isomerases ( lacA, lacB ) convert it into tagatose6 phosphate. Tagatose6 phosphate kinase adds a phosphate to the carbon one position to produce tagatose1,6 bisphosphate. Lastly, tagatose1,6 disphosphate aldose ( lacD) converts the sugar intermediate into glyceraldehyde3 phosphate and dihydroxyacetone phosphate. Lactose PTS Galactose PTS Lactose 6 phosphate Galactose 6 phosphate Tagatose 6 phosphate Tagatose 1,6 bisphosphate Glyceraldehyde 3 phosphate Dihydroxyacetone phosphate lacG lacA, lacB lacC lacD ATP ADP, 2H + H 2 0 Glucose (P galactosidase) (gal 6 P isomerase) (tagatose 6 P kinase) (tagatose 1,6 diphosphate aldolase)
36 Figure 13. Genetic organization of the lac operon in S. mutans UA159. The lac operon contains the genes for a repressor ( lacR), the tagatose 6 phosphate pathway ( lacABCD), the lactose transporter ( lacF, lacE ) and phosphogalactosidase ( lacG ) lacR lacA lacB lacC lacD lacF lacE lacG Lactose PTS Tagatose genes
37 Figure 14. Genetic organization of the lac operon in S. gordonii DL1. The lactose cluster contains the genes for the first set of tagatose6 phosphate pathway enzymes ( lacABCD1 ), a transcriptional antiterminator ( lacT ), lactose transporter ( EIIALac, EIIBCLac) and pho spho galactosidase ( lacG ). The galactose cluster contains the genes for a repressor ( lacR), a second set of tagatose6 phosphate pathway enzymes ( lacABCD2 ) and the galactose transporter ( EIIAGal, EIIBGal, EIICGal). lacR lacB 2 lacD 2 EIIB Gal lacA 1 lacC 1 lacT EIIBC Lac Lactose PTS Galactose PTS lacA 2 lacC 2 EIIA Gal EIIC Gal lacB 1 lacD 1 EIIA Lac lacG Tagatose genes Set 2 Tagatose genes Set 1
38 CHAPTER 2 MATERIALS AND METH ODS Bacterial Strains and Growth Conditions The bacterial strains used in this study are listed in Table 21. All strains wer e stored in 50% glycerol at 80C. All S. gordonii and S. mutans strains were maintained on brain heart infusion (BHI) agar plates (Difco Laboratories, Detroit, MI) and broth cultures were routinely grown in either BHI or TryptoneVitamin (TV) (Difco Laboratories, Detroit, MI ) base medium. All cultures were incubated at 37C in a 5% CO2 chamber. Selection and maintenance of antibiotic resistant strains on agar was achieved by supplementing media with erythromycin (Em) (10 g/ mL ), kanamycin (Km) (1 mg/ mL ), or spectinomycin (Sp) (1 mg/ mL ) as needed. BHI broth cultures supplemented with antibiotics employed half those concentrations. E. coli strains were maintained on LuriaBertani (LB) agar plates under aerobic conditions at 37C. All liquid cultures were grown in LB broth in an aerobic chamber with vigorous agitation of the cells at either 30C or 37C depending on experimental conditio ns. Vigorous agitation is defined as shaking at 250 rpm in a vessel where the culture occupied less than 20% of the total vessel volume. Antibiotics utilized in the selection of relevant strains include Km (25 g/ mL ) and ampicillin (Ap) (100 g/ mL ) for bot h agar and broth conditions. Induction of the lac D thiogalactoside (IPTG) (SigmaAldrich, St. Louis, MO) yielded expression of recombinant protein in an E. coli M15 expression system At an OD600 of 0.5 0.6, IPTG was added to a fin al concentration of 0.005 mM and the cells were incubated another three hours before harvesting. Uninduced cultures were treated exactly the same with the exception that no IPTG was added.
39 DNA Manipulations Coding sequences of the genes in the lac and gal clusters of S. gordonii were deleted in whole or in part and replaced with a nonpolar Em, Km, or Sp resistance cassette via allelic exchange, as described previously ( 56) Briefly, two DNA sequences flanking the gene of interest were amplified via PCR using gene spec ific primers with integ rated restriction cut sites (Table 22) These fragment s along with the appropriate a ntibiotic r esistance vector, were cut using restriction enzymes (New England BioLabs, Ipswich, MA; Invitrogen, Carlsbad, CA) and ligated together using T4 DNA ligase overnight at 16C. This m utated, linear DNA was then transformed into naturally competent S. gordonii DL 1 cells using 10% horse serum ( Sigma Aldrich, St. Louis, MO ) ( 63, 67) Resulting strains were subjected to PCR verification and DNA sequencing before use in experiments. Specific genes targeted for mutation include lacR lacA2 B 2 EIIGal, lacA1 B 1 EIILac ( lacEF ) and lacG (Table 2 1 ). Double deletion mutants were made by transformation of one strain with the chromosomal DNA of the second strain. Plasmid PlacA2 cat was constructed by fusing a 308bp sequence from the promoter region of lacA2 with the promoterless chloramphenicol acety ltransferase ( cat ) gene ( 43) located on plasmid pJL84 ( 105 ) The promoter cat fusion was excised from the plasmid using restriction enzymes SacI and BamHI and subcloned into the integration vector pMJB8. This vector integrates a single copy of the promoter fusion into the S. gordonii genome at a remote site ( gtfG locus). This plasmid encodes a Km marker for selection purposes. Plasmid PlacA1 cat w as constructed in the same manner by fusing a 385bp sequence from the promoter of lacA1 with the cat gene from plasmid pYQ4, a plasmid derived from pMJB8. pYQ4 can be used directly to integrate into the S. gordonii
40 chromosome at the gtfG locus. Restrictio n enzymes used include BamHI and KpnI This plasmid contains an Em marker for selection purposes. A 4 bp mutation was created in the cre of the lacA2 promoter by site directed mutagenesis. Briefly, two pairs of primers were used to generate two separate DNA fragments via PCR: the original pair of primers used to make the PlacA2cat DNA and another pair designed to incorporate the 4bp mutation into the DNA fragments (Table 2 2). Therefore, the two separate DNA fragments each contained the mutation, one fr agment at the 5 end and the other fragment at the 3 end. The two fragments were then combined in recombinant PCR along with the original pair of primers, producing a single product containing both fragments and the desired 4bp mutation. This mutated lac A2 promoter region was then fused to the cat gene in plasmid pYQ4 to create the cre PlacA2 cat plasmid ( 104) The plasmid encodes Em resistance and integrates into the S. gordonii genome at the gtfG locus. All three of these plasmids were then transformed into various mutated genetic backgrounds and the impact assessed using the chloramphenicol acetyltransferase (CAT) reporter assay of Shaw ( 82) A site directed point mutation was generated in the lacT gene whereby the translatio nal start codon (Met8) was replaced with a stop codon. This strain was designated lacT (M1stop). Briefly, a 1 2 kb DNA fragment containing the lacT point mutation was generated using recombinant PCR. The mutated lacT DNA fragment was transformed into wildtype S. gordonii DL 1 along with an indicator plasmid carrying the PlacA2 cat reporter fusion and a Km resistance cassette ( 101 ) A lower concentration of the plasmid was used (at least 1000fold) than of the mutated DNA fragment, with integration occurring at the gtfG locus. Transformants were then plated on BHI + Km
41 and several colonies were screened for the desired mutation by using an allelespecific mismatch amplification mutation analysis (MAMA) PCR ( 23) and verified via sequencing. This type of mutation allows for transcription to occur while halting translation. MAMA PCR was used to verify integration of the mutated lacT DNA fragment into the S. gordonii genome. The reaction included three primers: (1) a 5 primer (SGO15155), (2) a 3 MAMA primer (SGO1515M8stop 3) and (3) a 3 control primer (SGO15153) that anneals at a distal site (>0.5 kb) on the same strand (Table 22). The MAMA primer is designed to detect sitedirected mutations and usually contains one mismatch with the wild type allele and two mismatches with the mutant allele within the last thr ee nucleotides of the primer sequence. Therefore, when wildtype DNA is tested via MAMA PCR, the MAMA primer binds preferentially to generate the shorter DNA fragment. This indicates that no point mutations exist in the tested DNA region. When cells contai ning mutated lacT DNA are tested, amplification of the larger control product is observed. This is a result of the MAMA primers inability to anneal to the mutated lacT DNA region and indicates that the sitedirected mutation is present in the template strand. S. gordonii colonies were screened in a 50 mL reaction that contained 0.6 mM primer 1, 0.4 mM primer 2 (MAMA primer) and 0.2 mM primer 3, treated at 95C for 5 minutes, followed by 30 cycles as follows: 95C for 25 seconds, 55C for 25 s and 72C for 2 min ( 102) Growth Rate Assays Phenotypic growth analysis was documented for all oral streptococci strains with a Bioscreen C reader (Oy Growth Curves Ab, Ltd., Helsinki, Finland). Individual colonies were inoculated in triplicate into BHI and grown overnight. Strains were then
42 sub cultured the next morning into fresh BHI media and grown to an OD600 of 0.5 (mid exponential phase) Samples were then subcultured into fresh TV medium containing glucose, galactose, or lactose at a concentration of 0.5%. Samples were loaded onto a 100 well plate at 300 L each with a 50 L mineral oil overlay to mimic anaerobic conditions. The plate w as then incubated in the Bioscreen C at 37 C and the optical density measured at 600 nm (OD600) every 30 min with 10 s of shaking before each reading. Analysis of Promoter Gene Fusion Strains S. gordonii cat fusion strains for the lacA1 and lacA2 promoter s were assayed in various mutated backgrounds. Strains were grown overnight in TV medium containing glucose, galactose, lactose, glucose + galactose, or glucose + lactose at a concentration of 0.5%. In the morning, the cells were subcultured into 30 mL of the same medium and grown to early exponential phase (OD600 of 0.4 0.5). Cells were harvested and washed with 10 mM Tris HCl pH 7.8 followed by resuspension in 750 L of the same buffer on ice. The concentrated cell suspension was mixed with glass beads ( 0.1 mm diameter) equal to a volume of 500 L. Homogenization and mechanical breakdown of the cells was achieved by subjecting the cells to a Bead Beater ( Biospec Products, Inc., Bartlesville, OK) for 30 s pul ses twice at 4C with a 2 min incubation period on ice between pulses. Centrifugation produced a pellet of the cellular debris while leaving proteins suspended in the supernatant fluid. The cellfree extract was used to measure the CAT specific activity via the spectrophotometric method outlined by Shaw ( 82)
43 This kinetic assay determines the rate at which acetyl CoA acetylates the antibiotic chloramphenicol (Cm) via the chloramphenicol acetyltransferase ( cat ) enzyme, generating a free CoA sulfhydryl group end product in the process. Addition of 5,5 Dithiobis 2 nitrobenzoic acid (DTNB) activates a second reaction in which the free CoA sulfhydr yl group and DTNB react to form a mixed disulfide of CoA and thionitrobenzoic acid + a molar equivalent of free 5thio 2 nitrobenzoate (TNB). The second reaction is used to quantitatively measure cat enzymatic rates with a spectrophotometer since the react ion of DTNB with the sulfhydryl group on CoA results in an increase in absorbance at 412 nm (due to the TNB anion) ( 82) The reaction mixture contains the protein to be measured, 0.1 mM acetyl CoA and 0.4 mg/ mL of DTNB in 100 mM Tris HCl, pH 7.8. The reaction mixture was adjusted to 37C in a spectrophotometer equipped with a temperaturecontrolled cuvette chamber. Subsequently, Cm was added to initiate the reaction. The assay was run with a control containing no Cm and triplicates of each sample containing Cm at a final concentration of 0.1 mM. The rate of increase in absorption at 412 nm was meas ured and used to calculate the CAT specific activity. The BCA protein assay was used to measure total protein content of the cell lysate samples with BSA employed as the standard. RNA Isolation and Gene Expression via quantitative Real Time RT PCR RNA was isolated from cultures of wild type and genetically modified S. gordonii strains to evaluate genetic expression via quantitative Real Time RT PCR ( 6 ) Cells were grown in 10 mL of fresh medium and harvested for RNA isolation during the midexponential phase. Cells were frozen in 1 mL Bacterial RNA Protectant in 80C storage until time for extraction. Media conditions tested included TV with 0.5% of the appropriate sugar (either glucose, galactose, or lactose). Total RNA was then isolated
44 using protocols described elsewhere ( 26) Briefly, cells were pelleted, resuspended in TE buffer containing SDS and homogeniz ed in the presence of glass beads and acidic phenol. The cell lysat e was centrifuged for 15 min at top speed and cellular RNA was extracted from the supernatant fluid via the RNeasy Mini Kit (Qiagen, Germantown, MD). Total RNA was estimated in triplicate using a spectrophotometer. The RNA was also analyzed by electrophoresis on a 0.8% TAE gel to determine quality of the isolated nucleic acid. High quality RNA was then converted to cDNA using Reverse Transcriptase PCR (RT PCR). Each RT reaction combined 50 ng random hexamer primers, 10 mM dNTP mix, 1 g total extracted RNA and DEPC treated water up to 10 L. This reaction was incubated at 65C for 5 min and then placed on ice for 1 min The cDNA synthesis master mix combined 10X RT buffer, 25 mM MgCl2, 0.1 M DTT, RNaseOUT and SuperScript III RT. Addition of 10 L of cDNA synthesis master mix to each sample reaction completed the set up of the RT PCR reaction. The mixture was incubated at 25C for 10 min followed by 50C for 50 min. The reaction was then ter minated at 85C for 5 min and chilled on ice. Any residual RNA was removed by adding 1 L of RNase H to each sample and incubating for 20 min at 37C. Each cDNA sample was diluted 10 times for lat er use in the quantitative Real Time PCR reaction. Sense and anti sense primers (Table 22) were designed for a specific gene and used to create a standard curve template via standard PCR. This amplified PCR product was purified, measured and diluted into a series ranging from 0.5x101 to 0.5x108 copies/L. A 96 wel l plate was loaded with the 8point standard curve series (doubles), each sample condition (triplicates) and a negative control (doubles). A master
45 mix was prepared for each sample with 12 L of SYBR Green super mix, 100 M sense primer, 100 M anti sense primer and 8.6 L water. Each well contained 4 L of nucleic acid plus 21 L of master mix for a total of 25 L. Once the plate was set up, the quantitative Real Time PCR was run using the Biorad iCycler program (Hercules, CA). All data were normalized to 1 g of total RNA. Expression, Purification and Dialysis of Recombinant N terminal 6X His tagged LacR protein The QIAexpressionist system was used to create a recombinant LacR protein. This system uses the low copy pQE 30 vector, which is based on the T5 promoter transcriptiontranslation system. The plasmid includes an optimized promoter operator element consisting of phage T5 promoter and two lac operator sequences which increase lac repressor binding to ensure efficient repression of the T5 promoter. It also contains a synthetic ribosomal binding site, a 6xHis tag coding sequence located at the 5 end of the cloning region, multiple cloning site and translational stop codons in all reading frames, two lactamase gene ( bla ) conferring ampicillin resistance, cat gene (no promoter not normally expressed) and a ColE1 origin of replication. The lacR region was amplified via PCR using primers lacR 5 BamHI (GCTAGAGAGG AGG ATCCATGGGTAA) and lacR3 SalI (GCCCTTTAAAGAGCTGTCGACTAACT) and subsequently PCR purified. Both the lacR PCR product and the pQE 30 plasmid were s equentially digested for 3 hours at 37C with restriction enzymes BamHI and SalI Each digestion reaction was v erified on a 0.8% agarose gel and gel purified. Both the lacR DNA and digested plasmid DNA were vacuum dried together and set up in a ligation reaction overnight at 16C
46 containing 7 L water, 2 L buffer with ATP and 1 L T4 DNA ligase. The ligation product was then transformed by electroporation into competent E. coli M15[pREP4] cells, cultured in 1 mL LB broth for 2 hours in an aerobic shaker and plated on LB agar containing Km and Ap for selection. The resulting colonies were screened via PCR to verify the correct size of the plasmid. Primers used for this PCR were pQE30forward (5 CCGA AAAGTGCCACCTGACGTCTAAG 3) and pQE30reverse (5 AGGTCATTACTGGATCTA TCAACAGG 3). Four colonies were chosen for sequencing and storage in 80C. Sequencing of three colonies verified the correct lacR sequence. To determine if the recombinant LacR protein was soluble, 10 mL of LB broth containing Km and Ap was inoculated overnight with the transformed lacR expressing E. coli M15 cells. This culture was diluted (1:20) into 100 mL of fresh medium and grown to OD600 of 0.6. An uninduced control was collected, pelleted and placed in the 80C freezer. Adding 1 mM IPTG induced expression of the LacR protein. The culture was incubated for another 3 hours. A 1 mL induced control s ample was collected, pelleted and placed in the 80C freezer. The rest of the culture was centrif uged at 4,000 x g for 20 min and frozen at 80C until further use. An SDS PAGE gel was prepared to evaluate the induction of LacR in the pellet versus the supernatant samples. Three samples were prepared. The uninduced control and induced pellet were both combined with 2X SDS buffer and boiled for 10 min The third sample included 50 mL of induced culture combined with 750 L lysis buffer. The cells were then homogenized w ith 500 L glass beads for 30 s and spun down at top speed for 10 min in 4C. The supernatant was extracted from the sample and combined with 2X SDS buffer. All three
47 samples were run on the SDS PAGE gel, stained with coomassie blue and destained for visualization of protein bands. To purify a large scale (1 liter) E. coli expression culture, cells were grown overnight with Km and Ap. In the morning, cells were diluted into 1 L fresh media and grown aerobically in 30C to OD600 of 0.6. Addi ng IPTG to a final concentration of 0.005 mM induced expression of LacR. The cells were harvested after 3 hours, centrif uged at 4,000 x g for 30 min and stored in 80C. To prepare the protein lysate under native conditions, the pellet was thawed for 15 mi n and suspended in 4 mL lysis buffer containing 10 mM imidazoleHCl. The mixture w as then incubated for 30 min on ice along with lysozyme (1 mg/ mL ) and RNase A (10 g/ mL ). The cells were added to 500 L glass beads and homogenized for 30 s twice with a 2 m in cooling period in between. The sample was centrifuged at top speed for 15 min and the cleared lysate collected. To the lysate, 200 L of 50% Ni NTA slurry was added and mixed gently on a rotary shaker at 200 rpm for 90 min in 4 C. After this incubation period, the mixture was spun down via centrifugation and the supernatant fluid containing the recombinant protein was collected. The nickel slurry was then washed 5 times with increasing concentrations of imidazole buffer. All washes were saved for later evaluation. Wash 1 = 10 m M Wash 2 = 30 mM Wash 3 = 30 mM Wash 4 = 50 mM Wash 5 = 50 mM The protein was then eluted from the nickel slurry eight times using 100 L of an el ution buffer containing 250 mM i midazole. All eluates were saved for later evaluation.
48 A protein gel was run with samples from each step of the purification process. This was to allow for proper tracki ng of the protein at all times. The eluate with the purest protein sample was then dialyzed against PBS overnight. Electrophoretic Mobility Shift Assays Electrophoretic mobility shift assay (EMSA) was used to evaluate the ability of the recombinant LacR protein to bind to the lacA1 and lacA2 promoters of S. gordonii Biotinylated and unbiotinylated DNA probes were created using standard PCR for both promoter sequences. Each binding reaction was comprised of 50 mM MgCl2, 2 g poly dI dC and 5X binding buffer (50 mM Hepes, pH 7.9, 250 mM KCl, 5 mM EDTA, 25 mM DTT, 50% glycerol) as well as varying concentrations of LacR purified protein, biotinyla ted/unbiotinylated DNA probe (for either the lacA1 or lacA2 promoter sequence) and water up to 10 L. The binding reactions were subsequently incubated on ice for 30 min after which time each was loaded onto a 4% nondenaturing PAGE gel (5X TBE, 30% acrylamide, 2% bisacrylamide, 30% APS and TEMED) and run for 1 hour at 35 mA. Following electrophoresis, the DNA was transferred to a positively charged hybridization membran e via Genescreen Plus (Boston, MA) and UV crosslinked. The biotinylated DNA was then detected with a Chemiluminescence EMSA kit ( Pierce Biotechnology, Rockford, IL) by probing with StreptavidinHorseradish Peroxidase Conjugate and the Chemiluminescent Substrate. An autoradiograph captured the resulting image after exposure for 30 s to 2 min. Mixed Species Liquid Culture Competition Assay S. gordonii DL 1 and S. mutans UA159 genomes were both modified by replacing the fruA gene in each with either Em (DL1) or Km (UA159) resistance determinants, thus allowing for enumeration of these strains individually on selective media. Each
49 strain was cultured in triplicate overnight in 10 mL BHI supplemented with antibiotics. The next morning, each sample was subcultu red into 10 mL fresh BHI in a 1:50 fold dilution. The cells were grown for 3 hours, the OD600 was determined and S. gordonii and S. mutans samples were mixed together in a 1:1 cell ratio (OD600 = 0.5) into fresh TV supplemented with either 0.5% glucose or galactose, with or without 50 mM phosphate buffer (final concentration). This was designated as t = 0 hrs. The samples were allowed to incubate for 6 hours (t = 6 hrs) at 37C, 5% CO2. At this time, the absorbance and pH were measured. The samples were ser ially diluted from 101 108 and plated on BHI + Em and BHI + Km. Samples were then diluted 1:50 into fresh media. Both the rediluted (viability) and original (persistence) cultures were then incubated overnight (t = 22 hrs). At this time, both the viability and persistence cultures were measured for OD600 and pH. Again the samples were serially diluted and plated on BHI containing Em or Km. In the last part of the assay, the persistence culture was kept as is while the viability culture was rediluted once more in a 1:50 fold dilution. These samples were incubated for 8 hours until t = 30 hrs (endpoint). In this last step, the samples were again serially diluted and plated on BHI containing Em or Km. All plates were incubated for 2 days in the anaerobic chamber at 37C, 5% CO2 and then CFUs were counted for dilution plates that contained between 30300 colonies.
50 Table 21. Bacterial strains used in this study. Strain Phenotype or description Reference or source E. coli strains: DH10B General cloning strain Invitrogen M15[pREP4] General purpose protein expression host Qiagen LacR M15[pREP4]/pQE 30 expressing LacR This study S. gordonii strains: DL1 Wild type host ATCC 49818 Lac EII Lac ::Em This study Em lacG ::Em This study Km lacG ::Km This study lacA1B1 ::Km This study lacT (M1stop) lacT ::Km point mutation at amino acid Met1 This study Gal EII Gal ::Em This study Em lacR ::Em This study Km lacR ::Km This study lacA2B1 ::Em This study A2B2 lacA1B1 A2B2 ::Km Em This study WT::P lacA1 cat S. gordonii P lacA1 integrated into gtfG locus of DL1 This study lacT (M1stop)::P lacA1 cat S. gordonii P lacA1 integrated into gtfG locus of lacT (M1stop) This study ::P lacA1 cat S. gordonii P lacA1 integrated into gtfG locus of This study ::P lacA1 cat S. gordonii P lacA1 integrated into gtfG locus of This study WT::P lacA2 cat S. gordonii P lacA2 integrated into gtfG locus of DL1 This study lacT (M1stop)::P lacA2 cat S. gordonii P lacA2 integrated into gtfG locus of lacT (M1stop) This study ::P lacA2 cat S. gordonii P lacA2 integrated into gtfG locus of This study ::P lacA2 cat S. gordonii P lacA2 integrated into gtfG locus of This study ::P lacA2 cat S. gordonii P lacA2 integrated into gtfG locus of DL1 This study WT:: cre P lacA2 cat S. gordonii cre P lacA2 integrated into gtfG locus of DL1 This study :: cre P lacA2 cat S. gordonii cre P lacA2 integrated into gtfG locus of This study :: cre P lacA2 cat S. gordonii cre P lacA2 integrated into gtfG locus of This study fruA ::Em Tong et al. (2011) S. mutans strains: UA159 Wild type host ATCC 700610 fruA ::Km This study
51 Table 22. Primers used in this study. Primer Sequence Application EIIGal 1 5 GGAACAAGGTGCAGATGCTGTTAAGTT 3 Inactivation of EII Gal EIIGal 2HIII 5 CTTGATTATAAGCTTGTAGACAGAAGACCAA 3 Inactivation of EII Gal EIIGal 3 5 CCAAGGAGACACCAGCTGTTAAACCAA 3 Inactivation of EII Gal lacE 1 5 GAGAAGATTTCTCGGAATATTTATCTCC 3 Inactivation of lacE lacE 2 5 GCTGCATAGTCCACAAAGTGCTCAATA 3 Inactivation of lacE lacG 1 5 GACTTTGCGGCTCTAGAAGCAGCTA 3 Inactivation of lacG lacG 2 5 GGGTGCGCGCCTATGAAATAAGG 3 Inactivation of lacG lacG 3BamHI 5 GAAACGCAGGATCCCTATCCAAAGAAA 3 Inactivation of lacG lacA1 1 5 GGTAACGCAACAACCGGTGATA 3 Inactivation of lacA1 lacA1 2EcoRI 5 GCTCCAATAAGAATTCCCATGATCATTCTC 3 Inactivation of lacA1 lacA1 3EcoRI 5 CGTGATATGAATTCAGCTCTCTATGCT 3 Inactivation of lacA 1 lacA1 4 5 CTTCCCGAGATACTAACTACCTCA 3 Inactivation of lacA1 lacA2 1 5 GGTCCTGGAACATCGGTCGAA 3 Inactivation of lacA2 lacA2 2XbaI 5 TCCAGCAGCATCTAGACCTATAATAATTG 3 Inactivation of lacA2 lacA2 3 5 CATCAAGCAGATGCAAACTTCTTTAC 3 Inactivation of lacA2 lacA2 4 5 GGAATATCTACCTTGTAGAAAGTATCAC 3 Inactivation of lacA2 lacR 1 5 CCGCCAAGGATCCTCTCATAACGA 3 Inactivation of lacR lacR 2BamHI 5 GTTCCAGCTGGATCCACAAATTTAAGAA 3 Inactivation of lacR lacR 3SphI 5 GCTTATACTCAGAAGCATGCCTTAGAAC 3 Inactivation of lacR lacR 3BamHI 5 GCTTATACTCAGAAGGATCCCTTAGA 3 Inactivation of lacR lacR 4 5 CGGACCAGCCCCATAAGCAT 3 Inactivation of lacR SGO1515 5 5 GCACCAAGCGCCTGCCAGACTGTCT 3 lacT Point Mutation SGO1515 M8stop5 5 CGAATCATACATCCTTAGAACAACAATGTTGCTC 3 lacT Point Mutation SGO1515 M8stop 3 5 GAGCAACATTGTTGTTCTAAGGATGTATGATTCG lacT Point Mutation SGO1515 M8stop3MAMA 5 GTTTGGCTAGAGCAACATTGTTGTACA 3 lacT Point Mutation SGO1515 3 5 TCGCCTGCCCAAGTGATTTTGC 3 lacT Point Mutation PlacA1 5KpnI 5 ATTGTCCCGGTACCATTTCTGTTTATCAGA 3 P lacA1 Amplification PlacA1 3BamHI 5 CCAATAATAATTGCCATGGATCCTCTCCTTTGTT 3 P lacA1 Amplification; lacA1 and lacA2 Probes PlacA2 5SacI 5 GGCAAGGAGGAGCTCACGGCTATCT 3 P lacA2 Amplification PlacA2 3BamHI 5 CCTATAATAATTGCCATGGATCCTCTCCTTTGTT 3 P lacA2 and cre P lacA2 Amplification PlacA2 5KpnI 5 GGCAAGGAGGGTACCACGGCTATCT 3 cre P lacA2 Amplification PlacA2 MCRE 3 5 CAGAAAGAATTATAGTAAGATACGTTCTTTTAATATATTAA 3 cre P lacA2 Amplification PlacA2 MCRE 5 5 TTAATATATTAAAAGAACGTATCTTACTATAATTCTTTCTG 3 cre P lacA2 Amplification
52 Table 22. Continued Primer Sequence Application lacR 5BamHI 5 GCTAGAGAGGAGGATCCATGGGTAAGAATCA 3 LacR Expression lacR 3SalI 5 GCCCTTTAAAGAGCTGTCGACTAACTGTTTAT 3 LacR Expression PlacA1 5Biotin 5 /5Biosg/ATTGTCCCCCTCCCATTTCTGTTTATCAGA 3 lacA1 Probe (Hot) PlacA1 5 5 ATTGTCCCCCTCCCATTTCTGTTTATCAGA 3 lacA1 Probe (Cold) PlacA2 5Biotin 5 /5Biosg/GGCAAGGAGGATTTCACGGCTATCT 3 lacA2 Probe (Hot) PlacA2 5 5 GGCAAGGAGGATTTCACGGCTATCT 3 lacA2 Probe (Cold) lacG S 5 GACAGGCTATGGAGAGGTCAATC 3 RT PCR lacG AS 5 TGGTGTATCAAAGTGGTGAAGGG 3 RT PCR SGO1520 S 5 GATAACAACGGAGTAAGCCAAGG 3 RT PCR SGO1520 AS 5 TTGGAGCATTTAGGAGGTCGTC 3 RT PCR lacR S 5 TGGAACTGTAACGGTTGCTGAG 3 RT PCR lacR AS 5 CTCTTCGCCCCACCAAATACTC 3 RT PCR SGO1515 S 5 TCTGGACCGAAGCCGTGATG 3 RT PCR SGO1515 AS 5 CCGTCTTCTGGGCAATGATGTG 3 RT PCR lacA1 S 5 GGTCAGGATTTTGTTGATGTGACCC 3 RT PCR lacA1 AS GGACCAGCCCCATAAGCATCGAT 3 RT PCR lacA2 S 5 GGTGCAGATGCTGCTGGAAAT 3 RT PCR lacA2 AS CACCTCAGCTGCAACTGCCAAT 3 RT PCR
53 CHAPTER 3 CHARACTERIZATION OF CONTIGUOUS LACTOSE/GALACTOSE OPERONS IN S. GORDONII DL 1 CHALLIS Introduction Unpublished data from Zeng and coworkers in the Burne laboratory indicate d that S. gordonii is able to grow well on galactosecontaining medium at a 10fold lower concentration than S. mutans Upon examining the S. gordonii genome, two tandem lac gene clusters were found, both containing a set of tagatose genes ( lacABCD) as well as a PTS specific for either galactose or lactose. The redundant nature of these two gene clusters is unlike the organization of the lac genes in th e S. mutans genome, which only contains one set of tagatose genes and a lactosespecific PTS. Of most importance, S. mutans lacks the transcriptional antiterminator and galactosespecific transporters found in S. gordonii Thus, S. gordonii appears to have a selective advantage over the caries producing pathogen. Based on the growth data and the major differences in genomic structure between the two species, it is conceivable that this advantage may be due to the specific genes that S. gordonii harbors dedi cated to a highaffinity galactose transport system. On the other hand, S. mutans seems to have either failed to acquire this transport mechanism or lost it through evolution. Therefore, since S. gordonii is more sensitive to galactose consumption ( 103) this species could potentially compete successfully against the pathogen in an environment where galactose is present in small quantities. This aim was dedicated to characterizing the two gene clusters in S. gordonii as a comparison against other species and for further use in later aims. In order to understand more about how S. gordonii utilizes galactose and lactose, a bioinformatics analysis was necessary to understand and compare homologous genes from other species. Growth assessments were important for appointing a specific
54 phenotype t o each gene. Finally, further evidence supporting the growth data was obtained through Real Time RT PCR, helping to decipher which gene cluster was inducible by which sugar. Results Bioinformatics Review of the Lactose and Galactose Gene Clusters An NCBI n ucleotide BLAST search comparing the genome of S. gordonii strain DL 1 Challis to S. mutans strain UA159 in the online database Oralgen ( http://oralgen.lanl.gov ) resulted in the finding that S. gordonii contains two sets of genes dedicated to t he tagatose6 phosphate pathway whereas the S. mutans genome only contains one set. Although both species have a lactosespecific transporter contained in the lac operon, S. gordonii also possess a second transport system desig nated as a putative galactosespecific PTS component. The first set of tagatose genes found in S. gordonii appears to be homologous (between 6677% identity according to an NCBI BLAST search) to those in S. mutans ; however, a second set is situated adjacent to the first set, just upstream of the galactose transporter genes (SGO1520 SGO1522; EIIABCGal). Interestingly, the galactose transporter genes are completely absent in the S. mutans UA159 genome. Based on an Oralgen database search, S. pyogenes also con tains genes annotated for a galactosespecific PTS along with a paralog ous set of the tagatose genes, excluding lacC. LacD.1 of S. pyogenes which is contained in the second set of tagatose genes, plays an important role as a global regulator of virulence factor expression rather than the traditional role in the tagatose6 phosphate pathway ( 62) In addition, L. casei contains a galactosespecific PTS ( 24) and an NCBI Gene search suggests S. pneumoniae also has a potential gal actose specific PTS.
55 A transcriptional antiterminator designated by our lab as lacT is present in the S. gordonii DL1 genome, but is noticeably absent from the S. mutans UA159 genome This transcriptional antiterminator belong s to the BglG/SacY family of p roteins associated with regulation of sugar metabolis m ( 34) It contains a Co AntiTerminator (CAT) RNA binding domain located at the amino terminus that facilitates binding to the ribonucleic antiterminator (RAT) sequence of nascent mRNA transcripts. This binding stabilizes the RAT structure and effectively prevents formation of a termination structure, allowing RNA Polymerase to continue transcription of the genes downstream ( 34) Two PTS regulat ory domains (PRD) of approximately 100aa in length constitute most of the remaining part of the gene. Each domain contains a highly conserved histidine residue at position 7 that serves as the phosphorylation site for the PRDs ( 92) The site is phosphorylated when there is no sugar entering the cell ( 9 45 ) This phosphorylation inactivates lacT and allows termination of transcription to occur since the genetic machinery needed to metabolize the sugar is not required. Upon transport of sugar across the cell membrane, most of the phosphate in the cytosol is used by the general components of the PTS (EI and HPr) to phosphorylate the incoming substrate ( 29 ) This leaves relatively little phosphate left over for bind ing to histidine 7 in the PRDs of lacT causing dephosphorylation. Thus, lacT is activated and the antiterminator protein attaches to mRNA to prevent termination of transcription. Software prediction (Softberry BPROM) evaluating the intergenic region located just upstream of lacT suggested a potential pr omoter site (Table 31). A cre sequence has also been found upstream of lacT suggesting that lacT is regulated by another protein. L. casei has a lacT similar to S. gordonii It has been reported that in L. casei lacT is transcribed along with lacEFG
56 as one single mRNA of 4.4 kb ( 8 ) Be cause the S. gordonii lac TEFG genes are similar in size to L. casei and lacT has a potential promoter with a cre sequence, lacT may have a comparable function. A transcriptional regulator LacR located upstream of the second set of tagatose genes is predicted by Oralgen to have a helix turn helix domain of about 5060 amino acids. This domain is located in the amino terminal region of the protein and binds DNA for effective repression of the lactose operon. The carboxy terminal end may contain either an effector binding domain or an oligomerization domain. Effector molecules for this type of regulator are often phosphorylated and act as intermediates in the appropriate metabolic pathway. This specific gene has homology to the d eoR gene sequence of Esch erichia coli, which functions as a transcriptional repressor of the deo operon ( 72, 79) The LacR protein is present in multi ple genera, including Streptococcus Staphylococcus and Bacillus and often participates in the regulation of sugar catabolism. A protein alignment of LacR located in the S. mutans UA159 genome revealed 59% identi ty to LacR of S. gordonii A BLAST search indicated that S. sanguinis strain SK49 has a protein that most closely relates to S. gordonii LacR with 86% identity. The exact mechanism of regulation by LacR has not been elucidated for G ram positive bacteria as of yet. Software prediction also revealed a potential promoter site in the region just upstream of lacR (Table 31). Comparison of the two isomerases, LacA1 and LacA2, in an NCBI protein BLAST search disclosed that the two proteins are 97% identical. The species with the closest homology to LacA1 was S. sanguinis strain SK36 with an identity of 100%. Softberry
57 BPROM also predicted potential promoter regions in front of both lacA1 and lacA2 (Table 31). A few other streptococci, lactococci and l actobacilli retain various combinations of these genes as well ( 17, 24 69 78, 79) As mentioned earlier, the S. pyogenes genome most closely resembles the one found in S. gordonii with two sets of the tagatose genes (excluding lacC.1 ), a lactose PTS, a separate galactose PTS and a BglG family transcriptional antiterminator ( 78 ) In contrast the S. mutans genome contains only one operon involving the lactose PTS, the tagatose genes and regulatory component LacR ( 79) Also f ound through searches in the Oralgen and NCBI databases, the following species have similar transport systems as well: L actococcus lactis L actobacillus casei S. sanguinis S. mitis S. pneumoniae and S taphylococcus aureus (Table 32). S. gordonii is a good model organism for studying lactose and galactose utilization because of the similarity between its lac genes and those of other important human pathogens s uch as S. pyogenes and S. pne u moniae and its ease of genetic manipulation. Growth Phenotype of the S. gordonii Lactose and Galactose Gene Clusters Various mutant strains were created by deleting genes of interest from the genome and replacing each with an antibiotic resistance cassette via allelic exchange. All strains were grown in TV supplemented with 0.5% of the appropriate sugar and a mineral oil overlay to reduce oxygen tension. Incubation in a BioscreenC reader allowed for measurement of the optical density (OD600) every 30 min which was used to calculate doubling times (Table 33).
58 TV 0.5% glucose When grown in TV supplement ed with 0.5% glucose (Figure 31 ), the wild type (WT) S. gordonii strain grew to a final OD600 of 0.72 with a doubling time of 92 6.2 min EIILaclacA1B1 and lacT (M1stop) displayed a similar final optical density and growth rate as WT in lacG mutant exhibited a much longer lag phase of about 12 hours, a slightly slower growth rate with a doubling time of 120 15 min and a lower final yield of 0.53. It is normally characteristic of streptococcal growth curves to depict a slight dip in optical density just after reaching peak growth. This phenomenon is due to cell lysis, as described in experiments performed on S. mutans ( 5 ) lacG mutant experienced cell lysis more extensively than all of the other strains. Results for deletion mutants from the galactose gene cluster (Figure 32 ) indicate EIIGal and lacA2B2 both displayed simi lar growth rates and final yields to WT. A lacR with a doubling time of 120 5.2 min lacA1B1A2B2 grew i dentically to the WT (Figure 33 ). TV 0.5% galactose When grown in TV supplemented with 0.5% galactose (Figure 34 ), the WT strain grew to a final OD600 of 0.91 with a doubling time of 190 2.2 min and a lag phase of about 5 hours. Most of the deletion mutants from the lactose cluster showed signs of defective growth on galactose. The lactoseEIILac, grew to a similar yield as WT with an OD600 of 0.89, however, the lag phase lasted about 16 hours (vs. a lag of 5 hours in WT) and the doubling time was slower at 260 lacG LacA1B1 strain also had a lag phase of about 15 hours, a slightly decreased growth rate with a doubling
59 time of 250 22 min and a yield of 0.84. The lacT (M1stop) strain grew slower than WT with a doubling time of 220 16 min and had a comparable final yield. Growth results for the galactose cluster deletion mutants were not as straightforward as expected (Figure 3EIIGal strain grown in galactose was inconsistent. Final yield was always similar to WT, but the growth rate and lag phase sometimes varied between replicates With a calculated doubling time of 170 4.2 min, this mutant appears to grow faster than WT; however, in certain instances the strain actually grew slower than WT and had a lag phase of about 25 hours. It is recommended that additional experiments be completed before fully assessing the impact of this mutation on the growth of S. gordonii in a galactosecontaining medium. The strain displayed a comparable growth rate to WT with a doubling time of 200 18 min. When lacA2B2 was deleted, the growth rate increased with a doubling time of 180 19 min ; however, final yield was the same as WT. When lacA1B1 A2B2 was grown on galactose, no growth was obs erved (Figure 36 ). TV 0.5% lactose When grown in TV supplement ed with 0.5% lactose (Figure 37 ), the WT strain grew to a final OD600 of 0.69 with a doubling time of 140 32 min and a lag phase of about 4 hours. As expected, all of the mutant strains within the lactose gene cluster were severely defective when grown in lactose. When the lactosespecific transporter EIILac), almost no growth was observed with a doubl ing time of 2600 720 min lacG strain displayed very little growth with a final optical density of 0.18. With a lag phase of 15 hours and a doubling time of 360 26 min lacA1B1 strain reached a final yield of 0.59. The mutant lacking the
60 transcriptional antiterminator, lacT (M1stop) had the highest final yield (OD600 of 0.62) and a doubling time of 240 36 min ; however, its lag phase extended to 20 hours. As for the galactose cluster, deletion mutants did not appear to compromise bacteri al growth on lactose (Figure 38 ). In fact, all three strains ( Gal, LacR and lacA2B2) grew more rapidly than WT in lactose. The final optical density of the Gal, LacR and lacA2B2 strains were equal to WT. As seen in Table 33, the doubling time of the galactosespecific transporter mutant was 93 9.8 min LacR strain was 120 3.3 min and lacA2B2 strain was 100 2.8 min All of these strains experienced a lag phase of 4 hours similar to the WT strain. Conversely, when both sets of isomerase lacA1B1A2B2 ), no growth was observed with a final yield of 0.11 (Figure 3 9 ). Operonspecific Gene E xpression via Quantitative Real Time RT PCR Real time RT PCR was performed to evaluat e the expression of a representative gene from what we believed to be two distinct operons to allow for clarification as to which cluster was induced by which carbohydrate source and hint at possible operon arrangements The lacG and EIICGal genes were chos en as the representative genes based on the fact that these genes are situated at the end of each gene cluster. Therefore, if the clusters were inducible by a particular sugar, RNA Polymerase would have to pass through all of the other genes before transcr ibing either lacG or EIICGal. RNA was extracted from S. gordonii WT cells that were grown in TV supplemented with 0.5% glucose, galactose, or lactose and converted to cDNA via RT PCR. Using Real Time PCR, this cDNA was used to measure the expression level of lacG and EIICGal. Standard curves were prepared to measure copy numbers of cDNA. The data were normalized to copies/ ( g of total RNA )
61 Figure 310 shows the expression of lacG in a WT background. When compared to growth in glucose, lacG expression was induced 24fold when grown in galactose (p=0.04) Even higher expression was noted when the cells w ere grown in lactose, with a 390fold induction compared to glucose (p=0.2) Figure 310 indicates the expression of EIICGal in a WT background. Comparing growth in glucose and galactose, EIICGal expression was induced 28fold (p=7.6x105) In a comparison of glucoseand lactosegrown cells, a 2.5fold increase in induction occurred (p=0.04) All comparisons were calculated as significant based on pair wise Students t tests. Discussion EIILac strain is not affected by growth in glucose, this strain is moderately affected by growth in galactose and severely affected by growth in lactose. T hese conclusions suggest that the lactose PTS does not transport glucose. The extensive lag phase and severely diminished growth rate displayed when grown in lactose is a strong indication that EIILac is the primary transporter of lactose through the cell membrane. However, since some growth in lactose is achieved, the cell is must be using another mechanism to bring lactose inside, such as through another transporter or it is producing secreted enzymes that break down lactose outside the cell Interestingly, growth in galactose was also slower in the EIILac strain along with a lag phase of about 20 hours, indicating that EIILac likely transports galactose into the cell, albeit in lower quantities than it brings in lactose. No growth was detectable for the lacG mutant in either lactose or galactose. While this is understandable in the case of the former sugar, it is perplexing in the case of the latter. LacG is required for the catabolism of Lac 6 P into glucose and Gal 6 P; therefore, a strain would be unable to fulfill that role when grown in the presence
62 of lactose substrate and the cells would not be able to grow. It seems though that lacG should not affect grow th in galactose since phosphogalactosidase is not required for the metabolism of Gal 6 P. The lacG mutant also has a n extended lag phase when grown in glucose. Again, the lacG enzyme is not necessary for the metabolism of glucose. The findings regarding growth on glucose and galactose imply that lacG is more important than pr eviously thought and it is likely responsible for providing a second, equally important and possibly regulatory, function. Similar to the strain, the lacA1B1 mutant displayed impaired growth on galactose and lactose with a long lag phase. The lacA1B1 enzymes are responsible for converting Gal 6 P into tagatose6 P in the first step of the tagatose pathway. Since both galactose and lactose are converted to Gal 6 P before entering the tagatose pathway, a deletion of lacA1B1 would prevent Gal 6 P from be ing metabolized by the tagatose pathway. When this occurs, Gal 6 P begins to accumulate inside the cell leading to inhibition of growth. Zeng and coworkers have proposed that this occurs in S. mutans where Gal 6 P builds up to such high levels that it actually becomes inhibitory to the cell ( 103 ) Growth is not however totally abolished since there is a second set of tagatose pathway genes capable of metabolizing the Gal 6 P apparently at a lower rate than LacA1B1 Since the strain containing the lacT mutation has a r educed growth rate and long lag phase when grown in lactose, it appears that lacT may enhance utilization of this sugar W e propose that w hen lactose is being transported that LacT would be dephosphorylated and capable of binding to nascent mRNA to prevent termination of transcription. In a lacT mutant the cell is missing this crucial step and the tagatose
63 genes are mostly not transcribed because termination structures in the mRNA are able to form without lacT there to interfere. In this scenario, Lac 6 P continues to enter the cell and is catabolized into glucose and Gal 6 P, but the Gal 6 P cannot be metabolized because the L acABCD gene products have not been synthesized. Therefore, Gal 6 P a gain builds up to toxic levels. As mentioned previously, it has been shown in L. casei that lacT is transcribed as part of a polycistronic mRNA with lacEFG This may also be the case in S. gordonii since software prediction indicates a putative promoter site in the region directly upstream of lacT If this were the case, then when lacT is deleted, lacEFG would not be transcribed, thereby removing production of the transporters and phosphogalactosidase genes Therefore, it is proposed that lacT can act to regulate transcr iption of the tagatose genes and autoregulate so as to control expression of the genes for the lactose transporter and phosphogalactosidase. This explanation is supported by studies by Fujita and coworked which showed that each antiterminator controls the expression of a cognate EII of the PTS and that EII in turn negatively regulates its antiterminator ( 37) The fact that the lacT (M1stop) strain grew only slightly slower than WT in galactose suggests that it does not play an important role in galactose metabolism. Deletion of the putative galactose transport and metabolism genes provided unexpected results. Although the Gal strain grew similarly to wild type when in glucose medium the same strain grew much faster than wildtype in lactose. It is therefore possi ble that EIIGal plays a role in the negative regulation of the tagatose pathway, which is still necessary for the metabolism of lactose; thus explaining the
64 faster growth rate in lactose by the Gal strain. Absence of EIIGal would then derepress the tagatose genes, leading to increased expression. Therefore, the cell would have an enhanced capacity for metabolizing lactose galactosidases may be cleaving lactose into glucose and galactose before it enters the cell. This rele ased galactose can then be taken up by the lactose(and possibly mannose) PTS and metabolized via the tagatose6 phosphate pathway thus metabolizing more galactose than a wildtype cell. Growth results were often inconsistent when the Gal strain was grown in galactose. More often than not the growth rate of the EIIGal mutant was slightly faster than WT. However, at times the EIIGal strain displayed a slower growth rate with a lag phase of about 25 hours compared to WT. The extreme difference in growt h often occurred within replicates of the same sample. If the galactose transporter were specialized for highaffinity transport, then the cell would express the EIIGal gene the most under conditions where there are only small concentrations of galactose in the environment such as may occur in saliva. Therefore, if the cell is bombarded with high quantities of galactose, the lactose transporters are utilized instead. Si nce 0.5% galactose is considered a much higher concentration than the physiological levels found inside the oral cavity, this may account for the inconsistency in growth rates and lag phase. In regards to the slightly faster grow rate displayed by the Gal strain in galactose, it is possible that the cell sensed removal of the galactose transporter. In effect, the cell attempted to over compensate for the loss of the galactose PTS by transcribing more copies of the EIILac (and possibly EIIMan) genes so t hat galactose could continue to be brought in through other transporters. The absence of EIIGal may not have affected growth rate significantly because the galactose
65 transporter is a highaffinity system. In effect, the cell may not express the galactose P TS if the concentration of galactose in the environment is considered too high. In this scenario, both the lactoseand mannosePTS would likely transport the galactose inside the cell. The lacR mutant grew slower in glucose compared to WT. The stra in also presented a clumping phenotype and tended to aggregate when grown in BHI, which is a medium that contains glucose. Further investigation revealed that the cells lacking LacR grow in much longer chains than WT (about 1520 cells per chain compared t o 25 in WT). Under stress inducing conditions, S. mutans forms long chains and aggregates in culture, resulting from impairment of cell septation and/or autolysis ( 58) In addition, a lacR deficient strain of S. mutans grown in glucose was observed to have lower glucose PTS activity compared to the parental strain ( 3 ) Therefore, LacR may be involved in other processes in the cell such as in the regulation of the glucose PTS. When grown in galactose, the strain and WT exhibited similar growth rates. Since the role of LacR is to repress the genes in the tagatose pathway, removal of thi s protein would result in constitutive expression of lacABCD. Therefore, when the strain is grown in galactose, one would expect an increased growth rate since the tagatose pathway genes are constantly being expressed and available to metabolize more Gal 6 P than a WT cell. Because the growth rate of the lacR mutant is not significantly affected in galactose, it seems likely that there is another layer of regulation not yet accounted for. Based on the strain growth data in galactose, the tagatos e genes seem to be expressed at similar levels to the WT. Therefore, even with LacR removed, lacABCD expression could still be under the control of another regulatory protein.
66 The strain displayed a faster growth rate in glucose compared to WT. As with the lacR deficient strain, lacA2B2 may also serve a function relating to the regulation of the glucose PTS. The LacA2B2 mutant also displayed a faster growth rate in galactose and lactose. If lacA2B2 were absent, one would expect growth rate to be slower since one of the tagatose pathways has been shut off. However, similar to the EIIGal strain the loss of lacA2B2 may be sensed by the cell. To compensate for the loss of one of the tagatose pathways, the cell relays a message to increase expression of the genes in the other tagatose pathway. This way, the incoming sugar does not build up inside the cell as Gal 6 P and inhibit growth. The double mutant lacA1B1 A2B2 strain did not grow at all in either galactose or lactose likely because both sets o f these genes are important for metabolizing Gal 6 P. It is also clear from these data that the tagatose pathways are the only way in which either galactose or lactose is metabolized since no growth was supported. Data generated by Real Time PCR on lacG ge ne expression confirmed that the phosphogalactosidase enzyme is induced in lactose (390fold induction), a result that was expected since the cell requires lactose to be broken down into glucose and Gal 6 P before it can be metabolized. Without the LacG enzyme, the incoming lactose never gets processed. At first glance, it was difficult to understand why there was a significant induction in lacG expression under galactose conditions (24fold induction) since this enzyme is not required for galactose meta bolism. However, once it was combined with the growth data, it seemed likely that lacG might have another role in activation of gene expression in addition to its enzymatic activity. In S. mutans it has been shown that lacG must be catalytically active fo r optimal lac gene expression. It has been proposed
67 that lacG converts Gal 6 P into another compound that functions as the inducer for the expression of the lac genes ( 103 ) It is likely that lacG serves a similar purpose in S. gordonii by converting Gal 6 P into some other intermediate that acts as an inducer for LacR, thereby derepressing transcription of the tagatose pathway genes. Transcription of the galactose transporter gene is significantly higher (28 fold induction) in cells grow ing in galactose as compa r ed t o glucose, which supports a role for EIIGal as the galactose transporter The 2.5 fold induction of EIIGal gene expression when grown in lactose was also statistically significant compared to glucose One suggestion for why EIIGal is transcribed when grown in lactose may be that the galactosidases mentioned earlier cleave lactose into glucose and galactose outside of the cell providing an inducing signal.
68 Table 31. Potential promoter sites involved in the lac operon of S. gordonii based on software prediction (Softberry BPROM). Sequence ( 10 region) Sequence ( 35 region) lacT TGCTATAAT TTTACA lacR GGATAAAAT TTGAA lacA1 GATTAAAAT TTTTCA lacA2 TAGTATATT TTGACA
69 Table 32. A list of the genes involved in the lactose and galactose PTS and tagatose6 phosphate pathway in related species. Lactose PTS Galactose PTS lacABCD lacG lacR lacT Streptococcus mutans Streptococcus sanguinis Streptococcus mitis (putative lacA only) Streptococcus pyogenes (contains 2 sets) (2) (2) Staphylococcus aureus Lactococcus lactis Lactobacillus casei (plasmid) (chromosome) ? (putatively on chromosome) (plasmid) ? (plasmid) Streptococcus pneumoniae (2) is defined as present in the genome. is defined as not present in the genome.
70 Table 33. Calculated doubling times of the wildtype strain and mutated lac gene strains of S. gordonii based on growth curve analysis Strain Glucose T d Glucose OD600 Galactose T d Galactose OD600 Lactose T d Lactose OD600 Wild type 92 6.2 0.72 190 2.2 0.91 140 32 0.69 Lac 110 2.9 0.72 260 7.4 (16 h lag) 0.89 2600 720 0.37 120 15 (12 h lag) 0.53 0.08 0.18 96 2.3 0.76 250 22 (15 h lag) 0.84 360 26 (15 h lag) 0.59 lacT (M1stop) 88 8.3 0.70 220 16 0.93 240 36 (20 h lag) 0.62 Gal 88 3.3 0.79 170 4.2 0.90 93 9.8 0.71 120 5.2 0.71 200 18 (12 h lag) 0.91 120 3.3 0.69 78 3.2 0.78 180 19 0.93 100 2.8 0.73 A2B2 91 2.4 0.78 0.10 0.11 Td is defined as the doubling time. OD600 is defined as the optical density measured at a wavelength of 600 nm in a spectrophotometer. is denoted when a strain displayed minimal growth and the doubling time was too large for calculation.
71 Figure 31. Growth of S. gordonii DL1 wild type Lac, and lacT (M1stop) strains in TV 0.5% Glucose with an oil overlay. Optical density at 600 nm was measured every 30 min using a Bioscreen C Each point represents the mean of three separate cultures in triplicate. Doubling times an d standard deviations are located in Table 33.
72 Figure 32. Growth of S. gordonii DL1 wild type Gal, and strains in TV 0.5% Glucose with an oil overlay. Optical density at 600 nm was measured every 30 min using a Bioscreen C Eac h point represents the mean of three separate cultures in triplicate. Doubling times and standard deviations are located in Table 33.
73 Figure 33. Growth of S. gordonii DL1 wild type , and A2B2 strains in TV 0.5% Glucose wit h an oil overlay. Optical density at 600 nm was measured every 30 min using a Bioscreen C Each point represents the mean of three separate cultures in triplicate. Doubling times and standard deviations are located in Table 33.
74 Figure 34. Growth of S gordonii DL1 wild type Lac, and lacT (M1stop) strains in TV 0.5% Galactose with an oil overlay. Optical density at 600 nm was measured every 30 min using a Bioscreen C Each point represents the mean of three separate cultures in triplicate. Doubling times and standard deviations are located in Table 33.
75 Figure 35. Growth of S. gordonii DL1 wild type Gal, and strains in TV 0.5% Galactose with an oil overlay. Optical density at 600 nm was measured every 30 min using a Bioscreen C Each point represents the mean of three separate cultures in triplicate. Doubling times and standard deviations are located in Table 33.
76 Figure 36. Growth of S. gordonii DL1 wild type , and A2B2 st rains in TV 0.5% Galactose with an oil overlay. Optical density at 600 nm was measured every 30 min using a Bioscreen C Each point represents the mean of three separate cultures in triplicate. Doubling times and standard deviations are located in Table 3 3.
77 Figure 37. Growth of S. gordonii DL1 wild type Lac, and lacT (M1stop) strains in TV 0.5% Lactose with an oil overlay. Optical density at 600 nm was measured every 30 min using a Bioscreen C Each point represents the mean of three separate cultures in triplicate. Doubling times and standard deviations are located in Table 33.
78 Figure 38. Growth of S. gordonii DL1 wild type Gal, and strains in TV 0.5% Lactose with an oil overlay. Optical density at 600 nm was measured every 30 min using a Bioscreen C Each point represents the mean of three separate cultures in triplicate. Doubling times and standard deviations are located in Table 33.
79 Figure 39. Growth of S. gordonii DL1 wild type , lacA2B2 and A2B2 strains in TV 0.5% Lactose with an oil overlay. Optical density at 600 nm was measured every 30 min using a Bioscreen C Each point represents the mean of three separate cultures in triplicate. Doubling times and standard deviations are located in Table 33.
80 Figure 310. Expression levels of the lacG and EIICGal transcripts via quantitative Real Time RT PCR when grown in various carbohydrate sources. Cells were grown to midexponential phase in TV 0.5% glucose, galactose, or lactose at 37C with 5% CO2. Results shown are the mean and standard deviations of three separate cultures assayed in triplicate for each strain. Pair wise Student t tests were used to determine significant differences ( p < 0.05). 1.00E+03 1.00E+04 1.00E+05 1.00E+06 1.00E+07 Glucose Galactose LactoseCopies/ug total RNA Sugar Type EIICGal lacG * *
81 CHAPTER 4 GENE REGULATION IN RESPONSE TO LACTOSE/GALACTOSE TRANSPORT AND ENZYMATIC METABOLISM Introduction As described in Chapter 3, there are several genes identified in the Oralgen database as putative regulatory components for the lac operons in S. gordonii Although a lactosespecific PTS is present in many Gram positive bacteria, the presence of two seemi ngly separate gene clusters dedicated to both galactose and lactose utilization is unique to only S. pyogenes S. pneumoniae, L. casei and S. gordonii ( 15, 103 ) To further complicate matters, those four species are not unified in the arrangement of their lac genes in the genome. Although all four appear t o contain a double set of tagatose pathway genes, including the lactose PTS, galactose PTS, lacR, lacT and lacG these genes are not always adjacent to each other in the genome. In the case of L. casei some of the lac genes are located on a plasmid while the rest appear to be chromosomal ( 24 33 ) Galactose is one of the major carbohydrate components found in host glycoproteins ( 41) including many abundant salivary proteins As a result, galactose is always available in low concentrations in the oral environment In turn, species such as S. gordonii galactosidases capable of cleaving galactose from glycoproteins may utilize a highaffinity galactose PTS to obtain carbon sources for energy ( 18 21) With the advantage of having constant access to a supply of carbohydrates, even during fasting periods by the host S. gordonii would have a competitive advantage over S. mutans and other oral bacteria that appear to lack a mechanism for highaffinity internalization of galactose.
82 It is equally important to understand the regulatory mechanisms behind the S. gordonii lac operons. If it is accurate that commensal S. gordonii gains a selective advantage over pathogenic S. mutans with enhanced growth in galactose, then understanding the differences in regulation between the two species could help in the development of ways to exploit this process to improve oral health. Thus, the goals of the studies described in C hapt er 4 were (1) to characterize lacA1 and lacA2 promoter activity, (2) to use lacA1 and lacA2 promoter gene fusions to assess the regulatory effects of three lac genes in a variety of sugars, (3) to measure mRNA transcript levels of key putative lac regulators in a variety of sugars, and (4) to determine the ability of the putative repressor LacR to bind to both the lacA1 and lacA2 promoters. Results Analysis of lacA1 and lacA2 Promoter Activity Using cat Gene Fusions Bioinformatics analysis revealed the existence of a promoter situated directly upstream of both lacA1 and lacA2 (Figure 4 1). To determine the validity of this discovery, reporter gene fusions were constructed for use in Chloramphenicol Acetyltransferase (CAT) assay s. The 385 bp region directly upstream of the ATG start site of lacA1 containing the putative promoter was cloned in front of a promoterless chloramphenicol acetyltransferase ( cat) gene on plasmid pYQ4 and integrated into the S. gordonii genome at the gtfG locus. A similar fusi on was made by cloning the 308bp region upstream of the ATG start site of lacA2 in front of the cat gene on plasmid pMJB8, with integration into S. gordonii also occurring at the gtfG locus. Once the promoter fusion was cloned into S. gordonii DL1 and it was evident that there was promoter activity, the promoter gene fusion was subcloned into various mutant strains
83 to assess the impact of the loss of each lac gene on the regulation of the promoters. For a quick reference to all CAT activity numbers in table form, see Table 41 and Table 42. Focusing on the lacA1 promoter, the WT::PlacA1cat strain verified that growth in galactose or lactose resulted in acti vation of the promoter with 2300 and 1900 units of CAT activity, respectively (p=0.3, p=0.5) (Figu re 4 2). When grown in the presence of the repressing sugar glucose and galactose, promoter activity decreased to 80 units of CAT activity (p=0.03) In contrast when the cells were grown in a combination of glucose and lactose, promoter activity was reduced to approximately half of the activity level (1000 units) seen in only lactose. Mutation of lacT did not affect promoter activity in galactose or a glucose pl us galactose mixture. However, promoter activity was abolished when the cells were grown in a mixture of glucose plus lactose as compared to wild type Because growth is severely limited in lactose, the lacT mutant was not assayed for promoter activity under those conditions. Removal of lacR produced drastic effects, with promoter activity increased in glucose, galactose and lactose (Pglc=0.02) The lacG mutant is not able to grow in galactose or lactose; therefore, this strain was assayed in the presence o f each sugar in combination with the repressing sugar glucose. Loss of lacG resulted in decreased activity by about half in glucose plus galactose as compared to the wildtype strain, while there was no activity in glucose plus lactose conditions. Again, t he lacG mutant was not assayed in galactose or lactose due to poor growth on that substrate. Evaluation of the lacA2 promoter proved to be more complicated than originally anticipated (Figure 43). Comparison between WT and mutant strains only showed
84 small changes in promoter activity. In the WT::PlacA2cat strain, activity was much lower than in WT::PlacA1cat The lacA2 promoter was inducible by galactose with 32 units, but not by lactose with only 4 units of CAT activity. In a lacT mutant background, act ivity was only slightly reduced to 23 units of CAT activity when the cells were grown in galactose. Removal of lacR caused modest derepression in glucose (8 units) and galactose (37 units). A double deletion of lacRT resulted in slight increases in activi ty compared to the lacR background with 13 units in glucose and 54 units in galactose. Compared to the wildtype strain, the lacG deficient strain did not affect lacA2 promoter activity in glucose plus lactose, but it did increase significantly in glucose and glucose plus galactose (Figure 44). Because none of the potential gene regulators associated with this system appeared to have any major effects on promoter activity of lacA2 it was theorized that regulation of this promoter is CcpA dependent. A ::PlacA2 cat strain resulted in significant derepression of the promoter in all carbohydrate conditions (Figure 43). Of note, promoter activity in lactose increased to a similar level as in galactose with 120 and 130 units of CAT activity reported, respectively. To investigate whether CcpA was masking the regulatory actions of LacR, a ccpA/lacR ::PlacA2 cat strain was also assayed (Figure 43). Surprisingly, promoter activity was decreased greatly in both galactose and lactose conditions. In order to eliminate any effects on the rest of the genome caused by a loss of CcpA, a sitespecific four base pair mutation starting 49bp upstream from the ATG start site ( AGC GTT CGT ATC ) was created in the predicted CcpA binding site (catabolite response element cre ) of the lacA2 promoter using recombinant PCR techniques (Figure 41). It was not until this strain, WT::cre PlacA2 -
85 cat was assayed that the true promoter activity was revealed (Figure 43). In all three carbohydrate conditions, promoter activity increased significantly. Without an active binding site for CcpA, the lacA2 promoter yielded 44 units of CAT activity in glucose and 230 units of CAT activity in both galactose and lactose. Oftentimes, a promote r is under the regulation of multiple transcription factors. Although the lacR mutation did not dramatically affect lacA2 promoter activity previously, mutation of the CcpA binding site combined with a lacR deletion caused substantial derepression of the promoter activity in glucose, galactose and lactose conditions. When the lacG mutant was introduced into the strain with the cat fusion lacking the cre, promoter activity was similar to that of the WT:: cre PlacA2 cat strain in glucose and glucose plus galactose, but was lower in the mutant growing in glucose plus lactose (Figure 43). Gene Expression of LacA1, LacA2, LacR and LacT To complement the data acquired via the CAT assays, quantitative Real Time PCR was used to measure the expression of various la c genes in the background of WT, and lacT (M1stop) strains when grown in glucose, galactose, or lactose (Figure 4 5). E xpressi on of the lacR gene was significantly up regulated in galactose in a lacT (M1stop) mutant compared to wild type (p=0.03) In the lacT mutant strain, copy numbers of the lacR transcript in galactose were significantly higher in magnitude than in glucose (p=0.03) Transcript levels of lacT were not significantly different in the or lacT (M1stop) strains. Increased expres sion level in galactose compared to glucose of the wild type strain was the only noteworthy distinction (p=0.04) Transcription of lacA1 in the wild type strain was markedly upregulated in both galactose and lactose (p=0.0.3
86 and p=0.03, respectively) This same trend appeared in the lacT mutant strain, with lacA1 expression being upregulated in galactose (p=2.8x103) Comparison of the wild type and lacT (M1stop) strains also indicated a significant difference in LacA1 expression in galactose (p=0.02) Lastly, transcript levels of lacA2 in the wild type strain were significantly different when the cells were grown in glucose, galactose and lactose with the highest induction measured in galactose (pglc gal=4.0x103, pglc lac=7.3x103, pgal lac=9.1x103) The last significant variance in gene expression was between the wild type and strains growing in galactose, where the copy number of lacA2 was significantly higher than the wildtype strain (p=0.04). In Vitro Binding Analysis of S. gordonii LacR R ecombinant Protein to the lacA1 and lacA2 Promoters Being that LacR is the putative repressor protein for the lac system in S. gordonii attention was focused on delving deeper into the regulatory role of this gene. As mentioned previously, LacR contains a helix turn helix region at its amino terminal end that is capable of binding to DNA for the purpose of repressing gene transcription. An online database called RegPrecise ( http://regprecise.lbl.gov/RegPrecise/ ) was employed to find several consensus LacR binding sequences for the S. gordonii genome (Table 41). Using these sequences, one putative LacR binding site was found within the lacA1 promoter as well as six putative LacR binding sites within the lacA2 promoter (Figure 41). T o determine if LacR could bind directly to either promoter, a 6X His tagged recombinant LacR protein sequence was generated using pQE 30 in E. coli M15[pREP4] cells. Expression was induced with IPTG and the protein was subseq uently purified via nickel affinity chromatography (Figure 4 6). The 28 KDa
87 protein was then dialyzed before addition into the binding reaction. After a 30min incubation period on ice, the reactions were run on a nondenaturing polyacrylamide gel T he DNA was transferred to a hybridization membrane and UV crosslinked. The biotinylated DNA was then detected with chemiluminescence by probing with StreptavidinHorseradish Peroxidase Conjugate and Chemiluminescent Substrate. All of the EMSA results presented h ere are preliminary findings and further work must be completed with the proper controls to accurately confirm these results. Addition of 2.08, 4.16 and 6.24 pmol of the recombinant LacR protein to a biotinylated probe of the lacA1 promoter resulted in inc reasing shift of the probe (Figure 4 7). When both biotinylated and unbiotinylated lacA1 probe were mixed in the same reaction tube, competition from the unbiotinylated probe reduced the binding of LacR incrementally (based on a 1:1 and 1:10 ratio of bioti nylated:unbiotinylated probe). Similar results were obtained when the biotinylated probe of the lacA2 promoter was tested in the same conditions (Figure 48). Densitometry analysis confirmed that there was in fact shift for both promoters. Discussion Based on the results presented here, the lacA1 and lacA2 promoters are differentially regulated. Some of the key differences in the structure of the promoters include a CcpA binding site and six putative LacR binding sequences in the lacA2 promoter, but only one predicted LacR binding site and no cre sequence in the lacA1 promoter. CAT assay data revealed that the lacA1 promoter is activated by both galactose and lactose, indicating a role of the first set of tagatose pathway genes ( lacABCD1 ) in metabolism of both sugars. In addition, the lacA1 promoter is likely not regulated by CcpA at least in a direct manner Aside from the fact that there are no cre
88 sequences in the lacA1 promoter, promoter activity of the wildtype strain in galactose alone was very high, while promoter activity in glucose plus galactose was abolished. Had CcpA been exerting an effect on the lacA1 promoter, some activity would still have been noted in glucose plus galactose. Also noted, cells grown in glucose plus lactose resulted in promoter activity of about half that measured in lactose alone. The most reasonable explanation for this is that lactose is preferred over galactose and therefore better transported as a sugar substrate under repressing conditions and a bett er inducer of the lacA1 promoter The transcriptional antiterminator LacT does not appear to regulate the lacA1 promoter since activity in the lacT mutant strain in galactose was comparable to that of WT. Once more, regulation of the lacA1 promoter seems to be CcpA independent due to the fact that promoter activity in a lacT mutant is again abolished in glucose plus galactose. Although a lactose culture was not assayed due to poor growth, a glucose plus lactose culture also measured no promoter activity. T hat result aligns with the previous line of reasoning concerning LacTs autoregulation. Because the lactose transporters are located directly downstream of LacT, autoregulation by the transcriptional antiterminator also affects the transcription of the dow nstream lactose transporter as well. As it follows, if LacT is removed from the cell, there will be no transcription of the lactose transporters. So normally when the wildtype strain is grown in glucose plus lactose, a small amount of lactose is able to enter the cell via the lactose transporter, resulting in some lacA1 promoter activity. However, in the lacT (M1stop) strain no promoter activity is measured due to the fact that the lactose transporters are not synthesized without LacT present.
89 Evaluating t he strain is slightly more complicated. Promoter activity of lacA1 is higher in the lacR deficient strain compared to wild type for glucose, galactose and lactose. This was expected in glucose because, since the LacR protein is not present, the lacA1 operon cannot be repressed and will be expressed constitutively. Conversely, when comparing WT to the strain in terms of galactose and lactose growth conditions, increased promoter activity was observed as well. If LacR is not present, the promoter activity should not change considerably from wildtype levels since these two sugars already induce the lacA1 promoter. The most suitable explanation for why lacA1 promoter activity is considerably higher in the strain stems from the hypothesis that LacG converts Gal 6 P into an inducer for LacR. In the WT strain, Gal 6 P is not operating as an inducer at its most efficient level since much of the carbohydrate is actually entering the tagatose pathway. Therefore, some LacR is still binding to the lacA1 promoter and exerting a repressive effect. It is only when LacR is completely removed from the cell that the lacA1 promoter activity increases dramatically. Therefore, LacR acts as a repressor for the lacA1 promoter A lacG deletion decreased lacA1 promoter activity by at least half in every sugar tested. This lends support to the theory that LacG is helping to activate the lacA1 promoter by converting Gal 6 P into an inducer for LacR. Cells grown in glucose plus la ctose actually displayed almost no lacA1 promoter activity. Similar to the reasoning behind this in the lacT mutant, LacG is required for the catabolism of Lac 6 P into glucose and Gal 6 P. Without this enzyme, there is no Gal 6 P available to metabolize a nd the lacABCD genes in this pathway are not necessary.
90 The lacA2 promoter is under tighter regulatory control via at least two different mechanisms uncovered here. The first mechanism involves CcpA and is based on a more global control of carbohydrate ut ilization and catabolite repression. The second mechanism is specific for the lac operon and involves LacR, LacT and LacG. In the WT strain, the lacA2 promoter is activated in galactose but less so in lactose, which is most likely due to CcpA dependent eff ects as well as better induction by Gal 6 P. Even then, the activity is extremely low compared to the levels measured from the lacA1 promoter. This same trend is evident in the strain, albeit a slight derepression. Absence of lacT produced a small r epressive effect in both glucose and galactose. The double mutant lacR and lacT (M1stop) strain caused just slightly more derepression than the strain alone. All of this data led to the conclusion that LacR acts as a repressor and lacT as an activator of the lacA2 promoter, but that the effect of each is not very influential in regulating the lacA2 promoter. This was puzzling since LacR and LacT are the two transcriptional factors associated with the second set of tagatose genes. It appeared that in g lucose and glucose plus galactose, the absence of lacG had a negative effect on the lacA2 promoter, but no significant effect in glucose plus lactose. Interestingly, this data agreed with the growth data in that LacG plays some kind of role in glucose metabolism. Although it appears that LacG is repressing the lacA2 promoter in glucose plus galactose, this effect could also be due to the glucose present only and have nothing to do with the galactose. Most surprising though was the fact that a cell lacking l acG did not interfere with lacA2 promoter activity in glucose plus lactose, where one would assume that lacA2 promoter activity would decrease due to no enzymatic
91 cleavage of lactose for use in the tagatose pathway. This could be explained by our theory that EIIGal specializes in the transport of low concentrations of galactose. With LacG absent in lactose conditions, Lac 6 P cannot be catabolized into Gal 6 P. galactosidases can still release galactose from lactose Once EIIGal transports the sugar inside, the lacA2 promoter is stimulated. The other possibility for regulation of the lacA2 promoter may be related to global control of transcription by CcpA, as many other carbohydrate operons are regulated in this manner. Dr. Zeng, from the Burne lab, performed a CAT assay on a CcpA deficient strain showing significantly increased lacA2 promoter activity in glucose, galactose and lactose (Zeng and Burne, unpublished). This derepression unmasked the role of CcpA as a direct repressor of the lacA2 promoter. After this revelation, we turned to LacR to investigate if an effect of LacR on lacA2 expression could be observed in a CcpA deficient genetic background. Despite expectations that promoter activity would increase dramatically, the ::PlacA2 cat strain produced the opposite effect with activity dropping to baseline in galactose and lactose. We deduced from the lacA1 promoter data and ::PlacA2 cat results that LacR was acting as an activator under induc ing conditions and as a repressor under repressing sugar conditions. Upon further inquiry, this premise proved to be incorrect. Since CcpA affects hundreds of genes, its deletion could produce unintended effects across the entire genome. A point mutation in the cre sit e ensuring only sitespecific effects to the lacA2 promoter produced activity levels of the same magnitude in galactose and lactose, thus confirming that CcpA has a direct effect in negatively regulating the lacA2 promoter.
92 Testing the cre PlacA2 cat gene fusion in a lacR mutant revealed significantly increased levels of lacA2 promoter activity in glucose, galactose and lactose. Following the same logic as in the lacA1 promoter, when LacR is removed the operon will be constitutively express ed in glucose while activity will be enhanced in galactose and lactose. Since removal of LacR causes derepression, it can be concluded that LacR is repressing the lacA2 promoter. Lastly, when the :: cre PlacA2 cat strain was assayed in glucose and glucose plus galactose, lacA2 promoter activity was not affected. However, a significant repression occurred in glucose plus lactose, supporting the role of LacG as an activator for the lacA2 promoter. Much of the Real T ime PCR data is in agreement with the findings from the promoter gene fusions. The lacA1 gene was expressed significantly higher in galactose and lactose of WT, indicating that the first set of tagatose genes is important for the metabolism of both carbohydrates. It also confirms that LacR represses the lacA1 promoter since, in glucose, lacA1 is expressed constitutively when LacR is taken out of the equation. Expression data also verifies that LacT does not regulate the lacA1 promoter as evidenced by no difference in expression of the lacA1 gen e between wild type and the lacT mutant. RNA transcript copy numbers for lacA2 in the WT strain align well with the promoter activities obtained with CAT assays. It is noticeable, however, that lacA1 is more highly expressed than lacA2 in the WT strain. At the time of this experiment, it was not known that the lacA2 promoter was under the control of CcpA. Now that CcpA has been implicated in the negative repression of the lacA2 promoter, the decreased lacA2 expression (compared to lacA1 expression) in wildtype seems likely due to the fact that
93 CcpA was functional and probably repressing the lacA2 promoter. In a strain with a mutation in the cre of the lacA2 promoter, it is expected that lacA2 expression would rise significantly in both galactose and lactose. Real T ime PCR data also suggests a negative interaction between LacR and LacT since a lacT mutation caused increased expression of lacR in galactose. From this data, it is plausible that LacT somehow negatively regulates lacR, either directly or indirec tly but there is currently no evidence to support this idea Conversely, lacT gene expression is not affected by the deletion of LacR, implying that the repressor does not regulate lacT transcription. It is more likely that lacT is auto regulated. The reason why there is lacT expression in the lacT mutant is because of the manner in which this strain was made. Instead of deleting the sequence from the genome, a point mutation was made at the ATG start site. In effect, transcription is not affected, but translation is blocked. Promoter fusion assays and/or Real Time PCR data suggest a repressive role for LacR in both the lacA1 and lacA2 promoters. In vitro binding assays using a recombinant LacR protein demonstrated a band shift when either lacA1 or lacA2 DNA probes were supplied. In a cold probe competition, binding of LacR to both promoters was reduced incrementally as the ratio of unbiotinylated to biotinylated probe was increased. This data provides support for the role of LacR as a repressor of both the lacA1 and lacA2 promoters. Although these results are preliminary, when data from the CAT assays, qu antitative Real Time PCR and in vitro binding assays are combined, the conclusion can be made that LacR negatively regulates the lacA1 and lacA2 promoters by binding to putative lacR consensus sequences.
94 Figure 41. Critical elements comprising the putative promoter regions of lacA1 and lacA2 based on bioinformatics analysis. The lacA1 promoter contains one LacR binding sequence 114bp upstream of the ATG start site. The lacA2 promoter contains six LacR binding sequences starting 116bp upstream of the ATG start site and a cre sequence starting 54bp upstream of the ATG start site An o nline database named RegPrecise ( http://regprecise.lbl.gov/RegPrecise/ ) was use d to find the consensus LacR binding sequences for the S. gordonii genome. The promoter prediction program Softberry BPROM was used to find the promoter regions. ATG 10 35 TIS AAACAAAAA 1 Putative lacR binding site lacA 1 DNA sequence upstream of the ATG 6 Putative lacR binding sites (circled in red) lacA 2 DNA sequence upstream of the ATG ATG N 14 10 35 TIS ATATATTAA GTATATT AT GATGAATT T TGACA AAAAA ACAAATAAT AAAAAAACA N 19 Mutated CRE binding site: AGC GTT CGT ATC AAGAA AGCGTT TT
95 Table 41. CAT activity of the lacA1 promoter of S. gordonii DL1 in the background of the wild type lacT (M1stop), and strains Glc Glc SD Gal Gal SD Lac Lac SD Glc/ Gal Glc/ Gal SD Glc/Lac Glc/ Lac SD Wild type 84 15 2300 170 1900 180 80 12 1000 480 lacT (M1stop) 37 3.0 2200 150 NA NA 37 4.3 48 6.2 10000 3500 15000 7400 11000 5700 NA NA NA NA 45 3.7 1200 39 NA NA 43 12 50 23 SD is defined as the standard deviation among triplicate samples.
96 Table 42. CAT activity of the lacA2 promoter of S. gordonii in the background of the wild type lacT (M1stop) , lacT (M1stop) and strains Glc Glc SD Gal Gal SD Lac Lac SD Glc/Gal Glc/Gal SD Glc/Lac Glc/Lac SD Wild type 0.50 0.07 32 0.91 3.8 0.93 0.33 0.13 1.2 1.0 lacT (M1stop) 0.07 0.06 23 2.6 NA NA NA NA NA NA 8.3 0.53 37 0.89 2.4 1.3 NA NA NA NA lacT (M1stop) 13 0.43 54 3.0 NA NA NA NA NA NA 1.3 0.34 NA NA NA NA 1.5 0.51 1.6 0.25 WT:: cre PlacA2 cat 44 9.3 200 6.0 230 7.5 41 1.0 120 22 ::cre PlacA2 cat 430 22 300 35 140 29 350 17 320 60 ::cre PlacA2 cat 32 17 NA NA NA NA 32 23 30 9.3 SD is defined as the standard deviation among triplicate samples.
97 Figure 42. CAT activity of the 385bp region directly upstream of the ATG start site of lacA1 Cells were grown to midexponential phase in TV 0.5% carbohydrate at 37C with 5% CO 2 collected by centrifugation and then measured for CAT activity. Results shown are the mean and standard deviations of three separate cultures for each strain. Pair wi se Student t tests were used to determine significant differences (p < 0.05). ND = Not Determined. 0 5000 10000 15000 20000 25000 WT lacT lacR lacG CAT Specific Activity Glucose Galactose Lactose Glucose/Galactose Glucose/LactoseND ND ND ND
98 Figure 43. CAT activity of the 308bp region directly upstream of the ATG start site of lacA2 Cells were grown to midexponential phase in TV 0.5% carbohydrate at 37C with 5% CO2, collected by centrifugation and then measured for CAT activity. Results shown are the mean and standard deviations of three separate cultures for each strain. Pair wise Student t tests were used to determine signific ant differences ( p < 0.05).
99 Figure 44. CAT activity of the 308bp region directly upstream of the ATG start site of lacA2 Cells were grown to midexponential phase in TV 0.5% carbohydrate at 37C with 5% CO2, collected by centrifugation and then measured for CAT activity. Results shown are the mean and standard deviations of three separate cultures for each strain. Pair wise Student t tests were used to determine significant differences ( p < 0.05). 0 0.5 1 1.5 2 2.5 Wild-type lacG CAT Specific Activity Glucose Glucose/Galactose Glucose/Lactose
100 Figure 45. Expression levels of the lacR, lacT lacA1 and lacA2 transcripts via quantitative Real Time RT PCR when grown in various carbohydrate sources. Cells were grown to midexponential phase in TV 0.5% glucose, galactose, or lactose at 37C with 5% CO2. Results shown are the mean and standard deviations of three separate cultures assayed in triplicate for each strain. Pair wise Student t tests were used to determine any significant differences ( p < 0.05).
101 Figure 46. Expression and purification of the S. gordonii LacR recombinant protein. The 6X His LacR protein vector was transformed into competent E. coli M15[pREP4] cells. Those cells were grown aerobically in 30C and expression of LacR was induced with 0.005 mM of IPTG. Cells were collected by centrifugation and purified usin g nickel affinity chromatography The samples were run on an SDS PAGE gel fo r verification of the 28 k Da LacR protein. Flow W1 W3 W5 E1 E2 E3 E4 E5 E6 E7 E8 Thru 75 kD 50 37 25 kD
102 Figure 47. EMSA of the S. gordonii recombinant LacR protein and the lacA1 promoter. The lacA1 promoter contains one putative LacR binding sequence. All columns contain 10 fmole of biotinylated lacA1 p robe. Column 1 contains lacA1 probe alone. Columns 2, 3 and 4 contain increasing concentrations of purified 6X His LacR protein of S. gordonii at 2.08, 4.16 and 6.24 pmole, respectively. Columns 5 and 6 contain lacA1 probe and 6X His LacR in the same concentrations as column 4 plus 10 and 100 fmole of unbiotinylated lacA1 probe respectively Quantitative results shown below image represent shift as a percent ratio of bound to unbound protein.
103 Figure 48. EMSA of the S. gordonii recombinant LacR protein and the lacA2 promoter. The lacA2 promoter contains one cre sequence and six putative LacR binding sequences. All columns contain 10 fmole of biotinylated lacA2 probe. Column 1 contains lacA2 probe alone. Columns 2, 3 and 4 con tain increasing concentrations of purified 6X His LacR protein of S. gordonii at 2.08, 4.16 and 6.24 pmole, respectively. Columns 5 and 6 contain lacA2 probe and 6X His LacR in the same concentrations as column 4 plus 10 and 100 fmole of unbiotinylated lac A2 probe, respectively Quantitative results shown below image represent shift as a percent ratio of bound to unbound protein.
104 CHAPTER 5 IMPACT OF GALACTOSE UTILIZATION ON THE INTER SPECIFIC COMPETITION BETWEEN CARIOGENIC S. MUTANS UA159 AND COMM ENSAL S. GORDONII DL 1 ORAL STREPTOCOCCI Introduction After initial growth experiments performed in the Burne laboratory suggested that S. gordonii is able to sustain growth on galactose at a concentration that is 10 times lower than what is necessary for S. mutans ( 103) we became interested in understanding the mechanism behind the results. Evaluation of the S. gordonii genome revealed two separate PTS, one each for lactose and galactose, along with a redundant set of tagatose genes. This genetic setup differs from S. mutans which only has a lactose PTS and one set of tagatose genes. Further research into the S. gordonii galactosidases. Results from Chapters 3 and 4 of this document support the hypothesis that S. gordonii has a galactosespecific PTS capable of hig h affinity transport. Competition for available carbon resources is a critical ecological determinant in dental biofilm s, especially during periods of famine when nutrient levels are the lowest. Saliva provides a continuous source of nutrients; therefore, it is advantageous for oral microbes to acquire the genetic machinery necessary to metabolize oligosaccharides of glycoconjugates. Galactose is one of the major constituents in the oligosaccharides of many glycoconjugates ( 41) Therefore, a microbe with the ability to compete for galactose from salivary constituents may hold a distinct advantage over neighboring microbes that cannot utilize this carbohydrate or that cannot efficiently transport galactose at lower concentrations
105 Since we are suggesting that S. gordonii possesses a highaffinity galactosespecific transport system, the goal of Chapter 5 was to design an in vitro mixed species liquid culture competition assay assessing the viability and persistence of both S. gordonii and S. mutans when grown together in galactose or glucose media. In order to ensure that the results were pH independent, the assay was conducted with and without phosphate buffer for both sugar conditions. Results Growth Comparison of S. gordonii DL 1 and S. mutans UA159 Both strains were grown in TV supplemented with either 0.5% or 2% of the desired sugar and a mineral oil overlay to reduce oxygen tension, since oxygen antagonizes the growth of S. mutans as does H2O2 production from oxygen by S. gordonii Incubation in a BioscreenC reader allowed for measurement of the optical density every 30 min which was used to calculate doubling times (Table 51). When grown in TV supplemented with 0.5% glucose (Figure 51), both strains grew to the same f inal yield, though S. gordonii grew at a faster rate. S. gordonii grew to a final OD600 of 0.61 with a doubling time of 88 2.3 min and a lag phase of 7 hours. S. mutans grew to a final OD600 of 0.56 with a doubling time of 180 17 min and a lag phase of 12 hours. When grown in TV supplemented with 0.5% galactose (Figure 52), S. gordonii produced a higher yield with a final OD600 of 0.84 and a much more rapid doubling time of 170 30 min S. mutans grew to a final OD600 of 0.52 with a doubling time of 4 10 61 min Both strains left lag phase at 15 hours. Although the final OD600 for S. gordonii was 0.84, this strain actually reached a peak OD600 of 1.0 apparently due to the fact that S. gordonii is more prone to lysis during stationary phase than S. mutans
106 When grown in TV supplemented with 2% galactose (Figure 53), the results were similar to that seen in TV supplemented with 0.5% galactose. S. gordonii again outcompeted S. mutans by growing at a faster rate (240 3.1 min ) and generating a higher final optical density (OD600 = 0.87). S. mutans grew to a final OD600 of 0.64 with a doubling time of 340 22 min As mentioned in the paragraph above, S. gordonii actually reached a peak OD600 of 1.0 before cells began to lyse in the stationary phase. M ixed Species Liquid Culture Competition Assay Based on the growth phenotypes displayed above when medium was supplemented with either glucose or galactose, a mixedspecies liquid culture competition assay was created to assess the bacterial interaction be tween planktonic S. gordonii DL1 and S. mutans UA159 cells. Cells were grown to midexponential phase at an OD600 of 0.5 and mixed together in one liquid cultur e tube in a 1:1 ratio (t = 0 h). Serial dilutions were plated to ensure that the correct ratio w as achieved. At several time points (t = 6, 22, 30 h), the OD600 and pH were measured, serial dilutions were plated for the purpose of CFU counting and the sample was subcultured in a 1:100 dilution of fresh media of the same kind. These samples measured the viability of the cells. At the same time, the original culture with a measured ratio of 1:1 was also incubated and subjected to all of the same measurements at the same time points. However, the media were never changed, nor was any buffer added. This sample measured the ability of the cells to persist over time. The fruA gene, which does not interfere with metabolism of glucose or galactose ( 20 105 ) was used as a selective marker. In S. gordonii fruA was replaced with an Em resistance cassette while in S. mutans fruA was replaced with a Km resistance cassette. All conditions were tested in triplicate.
107 TV + 0.5% galactose without phosphate buffer In TV supplemented with 0.5% galactose and no phosphate buffer, S. gordonii outcompeted S. mutans in both viability and persistence (Figure 54, 5 5). As far as viab ility is concerned, at t = 6 h both S. gordonii and S. mutans had grown in number, although the CFU count fa vored S. gordonii After this point, S. gordonii growth appeared to stall as the CFU count remained the same throughout the rest of the viability assay. On the other hand, S. mutans consistently declined in numbers by one log at t = 22 and 30 h The media became fairly acidic overnight at pH 4.8 but stabilized around pH 6.0 at the end of the assay (Table 52). Results from the persistence assay, showed that at t = 22 h S. gordonii outnumbered S. mutans by only a small margin. At a pH of 4.7 the media became very acidic over night (Table 5 3). By t = 30 h S. mutans outnumbered S. gordonii by about one log difference. TV + 0.5% galactose + 50 mM phosphate buffer In TV supplemented with 0.5% galactose and 50 mM phosphate buffer, S. gordonii outcompeted S. mutan s in both viability and persistence (Figure 54, 5 5). In terms of viability at t = 6 h S. gordonii appeared to hold an advantage with a CFU count one log higher than S. mutans The pH of the media was fairly neutral at 6.7 (Table 52). From this point on, S. gordonii continued to hold the advantage with the CFU count between the two strains widening further at each time interval. At t = 22 h, S. gordonii was more than two logs hi gher than S. mutans Although the media had been buffered with fresh 50 mM phosphate buf fer at t = 6 h the culture was acidic in the morning, reaching a pH m easurement of 5.4. At t = 30 h, there was again a two log difference between the two species as S. gordonii remained viable. T he pH recorded w as a neutral 6.9.
108 In the persistence portion of the assay, S. gordonii again gained the advantage and continued to grow in numbers until the final time point. However, the margin between the strains appeared to be slightly different from that of the viability por tion of the assay. At t = 22 h, the S. gordonii CFU count was s lightly higher than at t = 0 h. However, the S. mutans cells were severely depleted by two logs Even though the pH measured 5.4 after overni ght incubation, it was not enough to assist S. mutans in regaining strength. At t = 30 h S. gordonii continued to persist while S. mutans decreased further by another two logs. The pH remained around 5.3 since no phosphate buffer had been added since t = 0 h (Table 53). TV + 0.5% glucose without phosphate buffer In TV supplemented with 0.5% glucose and no phosphate buffer, S. gordonii was still able to outcompete S. mutans at all time points in the viability test (Figure 5 4, 5 5). At t = 6 h both strains had grown past their initial CFU counts Despite the acidic conditions of pH 4.9, S. gordonii was able to sustain growth and maintain an advantage over S. mutans (Table 52). At t = 22 h both strains had decreased similarly in number with a pH measurement of 4.6. The last time point, t = 30 h, sh owed no signs of change from t =22 h while the pH from this culture was less acidic at 6.6. The persistence assay is the only assay that S. mutans was able to outcompete S. gordonii At t = 22 h S. mutans had grown slightly while S. gordonii cells already started to die. The pH at this point was 4.4 (Table 53). At t = 30 h S. mutans cell counts remained similar to the previous time point while more S. gordonii cells continued to die. The pH did not chang e and was measured at 4.5.
109 TV + 0.5% glucose + 50 mM phosphate buffer In TV supplemented with 0.5% glucose and 50 mM phosphate buffer, S. gordonii once more outcompeted S. mutans both in viability and persistence (Figure 54, 5 5). During the viability portion of the assay, S. gordonii began to show a slight advantage at t = 6 h The pH of the culture at this time measured 5.8 (Table 52). At t = 22 h a two log difference in cell count was recorded. Because the culture sat overnight, the pH dropped to 5.8. However, the acidic environment did not benefit S. mutans which dropped two logs down in CFU count. Finally at t = 30 h, S. gordonii cell counts remained the same whereas S. mu tans cell counts decreased further. A final pH of 6.2 indicated that the phosphate buffer was indeed neutralizing most of the acid in the culture tube. Although the persistence results for this condition still indicate that S. gordonii outcompeted S. mutan s, the cell count ratio is altered slightly from that seen in the persistence assay for the galactose plus phosphate condition. At t = 22 h the S. gordonii CFU count climbed 4 logs higher than t he original culture at t = 0 h Despite this large increase, the S. mutans CFU count also rose 1.5 logs. So instead of dying or lysing, S. mutans was still able to persist, albeit less so than S. gordonii The culture was acidic with a pH o f 5.8 (Table 53). At t = 30 h it appears that both strains were beginning t o die or lyse, although S. gordonii still reigned. However, both strains still maintained CFU counts that were higher than the original culture count The pH dropped only slightly to 5.7. Discussion In TV 0.5% galactose, S. gordonii grew about 2.5 times faster than S. mutans In addition, final yield was much higher in S. gordonii As previously mentioned, our lab
110 discovered that S. gordonii is able to grow well in galactose at 10fold lower concentrations than S. mutans UA159. Both of these pieces of data suggest that S. gordonii may be better equipped than S. mutans UA159 to utilize galactose, as evidenced by both a faster growth rate, higher final yield and seemingly higher affinity and specificity for this hexose. In comparison, S. gordonii grew about 1.5 times as fast as S. mutans when grown in TV 2% galactose. S. mutans produced a slightly higher final yield and faster growth rate compared to its growth in 0.5% galactose. In contrast, S. gordonii grew slower in 2% galactose than it did in 0.5% galactose. This substantial change in growth rate of S. gordonii in varying galactose concentrations could be attributed to the change in relative participation between the galactose PTS and ATP dependent galactose permease. If the galactose PTS is indeed a highaffinity transporter, then the cells might actually grow better in lower concentrations of galactose as opposed to higher concentrations if the cells switch to a lower affinity, lower capacity transporter at high galac tose concentrations. Specifically in high concentrations of galactose, activation of the presumably lower affinity galactose permease (and possibly lactoseand mannosePTS) would permit growth, albeit at a slower rate. In the case of S. mutans transport of galactose occurs via the lactose and mannose PTS which appear to have about 100fold lower affinity for galactose than for their cognate PTS substrates ( 103 ) It has been shown in Saccharomyces cerevisiae that there is a reciprocal relationship between highand low affinity transport systems wherein cells grown i n high percentages of glucose almost exclusively display activity from low affinity transporters while cells grown in low percentages of glucose displayed activity mainly
111 from high affinity transporters ( 77) Furthermore, multiple studies have reported that both Streptococcus cremoris and Streptococcus lactis possess a galactose PTS that is low affinity in nature due to cell requirements for high concentrations of substrate for maximal growth rates ( 57, 88, 89) S. cremoris was also found to predominantly transport galactose through the galactose PTS as opposed to the galactose permease ( 88) In contrast, it has been suggested by Thomas and coworkers ( 88) that S. lactis strain ML3 uses a highaffinity permease to transport galactose based their results indicating less dependency on galactose concentration by this strain and Thompson and coworkers ( 89) findings that the affinity for the galactose permease was 10fold higher than for the galactose PTS. Therefore, the slower growth rate of S. gordonii in 2% galactose (compared to 0.5% galactose) could be indicative of activation of the lower affinity galactose permease and lactose and mannosePTS whereas the faster growth rate of S. gordonii in 0.5% galactose (compared to 2% galactose) is indicative of activation of the highaffinity galactose PTS. Results for the competition assay predominantly favored S. gordonii in both the viability and persistence portions of the experiment. In evaluating the viability of the cells, S. gordonii showed a strong ability to maintain survival and recover from stress no matter which sugar it was grown in. S. mutans grew faster in gl ucose conditions, indicating its preference for glucose over galactose substrates; however, S. gordonii appeared to grow equally well in both sugars, indicating its enhanced utilization of galactose. At t = 6 hrs, all testing conditions appeared to show si milar growth patterns for both S. mutans and S. gordonii This indicates that in the first 6 hours of cohabitation neither sugar type nor acid production affected competition between S.
112 mutans and S. gordonii However, as time progressed, it became evident that S. gordonii was able to outcompete S. mutans with regards to sugar type. Although S. gordonii grew better in both glucose and galactose, it is clear that the commensal displayed enhanced growth in galactose that likely contributed to its competitive edge. In contrast, S. mutans grew considerably less in galactose, especially when the media was buffered. In regards to acid production, S. gordonii was able to tolerate the stress despite acidic conditions in unbuffered media. This suggests that S. gordo nii has some mechanism to combat acid production in its surroundings. In fact, i t has been shown in S. gordonii that H2O2 production is insensitive to changes in pH between 5.0 and 7.5 ( 12) Therefore, the release of hydrogen peroxide by S. g ordonii may likely have contributed to the decrease in S. mutans cells via oxidative stress. Based on the pH measurements from this assay, production of acid is decreased when the cells are grown in galactose compared to glucose. There are two plausible ex planations for this: (1) because S. gordonii is better able to compete against S. mutans in galactose, the increase in the number of S. gordonii cells (and subsequent decrease in number of S. mutans cells) means that less acid overall is produced, or (2) S. gordonii displays enhanced H2O2 production when grown in galactose. In contrast, it appears that bacteriocin production by S. mutans if any, did not affect the ability of S. gordonii to stay viable. When evaluating the ability to persist, S. gordonii w as only able to effectively compete with S. mutans when phosphate buffer was added to the media. In regards to carbohydratetype, S. gordonii still maintained a competitive advantage over pathogenic S. mutans ; however, this advantage was only limited to bu ffered medium, although
113 severity of the final outcome was decreased in unbuffered medium This suggests that while S. gordonii is still able to outcompete S. mutans when galactose is present, the commensal cannot persist without continual removal of acid from the environment. This aligns with the acidsensitive phenotype of S. gordonii Therefore, despite the advantage that enhanced galactose metabolism confers to S. gordonii this mechanism alone is not enough to overcome the effects of too much acid. It should be noted, however, that saliva flow inside the mouth contributes greatly to the diffusion and level of acid and can also be considered a constant source of carbohydrate, especially galactose. With that in mind, the viability portion of this competit ion assay likely emulates an in vivo setting more closely than the persistence portion. This data also suggests that the decreased acid production when the cells are grown in galactose may be more due to the fact that S. gordonii outnumbers S. mutans rather than enhanced H2O2 production by S. gordonii when grown in galactose. Once S. mutans gained a competitive advantage over S. gordonii bacteriocin production likely may have also contributed to the persistence of the pathogen.
114 Table 51. Calculated doubling times of the S. gordonii DL1 wild type and S. mutans UA159 wild type strains based on growth curve analysis 0.5% Glucose Td 0.5% Glucose OD600 0.5% Galactose Td 0.5% Galactose OD600 2% Galactose Td 2% Ga lactose OD600 DL1 88 2.3 0.61 170 30 0.84 240 3.1 0.87 UA159 180 17 0.56 410 61 0.52 340 22 0.64 Td is defined as the doubling time. OD600 is defined as the optical density measured at a wavelength of 600 nm in a spectrophotometer.
115 Figure 5 1. Growth of S. gordonii DL1 and S. mutans UA159 in TV 0.5% Glucose with an oil overlay. Optical density at 600 nm was measured every 30 min using a Bioscreen C Each point represents the mean of three separate cultures in triplicate. Doubling times and standard deviations are located in Table 51.
116 Figure 52. Growth of S. gordonii DL1 and S. mutans UA159 in TV 0.5% Galactose with an oil overlay. Optical density at 600 nm was measured every 30 min using a Bioscreen C Each point represents the mean of three separate cultures in triplicate. Doubling times and standard deviations are located in Table 51. 0.05 0.5 0 10 20 30 40 50 60OD600 Time (Hours) DL1 UA159
117 Figure 53. Growth of S. gordonii DL1 and S. mutans UA159 in TV 2% Galactose with an oil overlay. Optical density at 600 nm was meas ured every 30 min using a Bioscreen C Each point represents the mean of three separate cultures in triplicate. Doubling times and standard deviations are located in Table 51. 0.05 0.5 0 10 20 30 40 50 60OD600 Time (Hours) DL1 UA159
118 Figure 54. Mixed species Competition Assay testing viability of the S. gordonii DL1 and S. mutans UA159 strains when grown in TV supplemented with 0.5% glucose or galactose, with or without phosphate buffer. S. gordonii and S. mutans cells were grown separately to mid exponential phase in triplicate, mixed together in a 1:1 ratio and incubated at 37C, 5% CO2. At time points t = 6, 22, 30 h the cultures were diluted into fresh media. In addition, at t = 0, 6, 22, 30 h the optical density and pH of each mixedspeci es culture was measured and serial dilutions were plated for enumeration of the strains individually on selective media. 1.00E+00 1.00E+02 1.00E+04 1.00E+06 1.00E+08 0 6 22 30Log CFU Time (hours) Galactose Phosphate buffer S. gordonii S. mutans 1.00E+00 1.00E+02 1.00E+04 1.00E+06 1.00E+08 0 6 22 30Log CFU Time (hours) Galactose + Phosphate buffer S. gordonii S. mutans 1.00E+00 1.00E+02 1.00E+04 1.00E+06 1.00E+08 0 6 22 30Log CFU Time (Hours) Glucose Phosphate buffer S. gordonii S. mutans 1.00E+00 1.00E+02 1.00E+04 1.00E+06 1.00E+08 0 6 22 30Log CFU Time (hours) Glucose + Phosphate buffer S. gordonii S. mutans
119 Table 52. Average measured pH values from the mixedspecies competition assay testing viability. Time Galactose + Phosphate Galactose Phosphate Glucose + Phosphate Glucose Phosphate t = 6 hours 6.7 5.8 5.8 4.9 t = 22 hours 5.4 4.8 5.8 4.6 t = 30 hours 6.9 6.0 6.2 6.6
120 Figure 55. Mixedspecies Competition Assay testing persistence of the S. gordonii DL1 and S. mutans UA159 strains when grown in TV supplemented with 0.5% glucose or galactose, with or without phosphate buffer. S. gordonii and S. mutans cells were grown separately to midexponential phase in triplicate, mixed together in a 1:1 ratio and incubated at 37C, 5% CO2. At time points t = 0, 6, 22, 30 h, the optical density and pH of each mixedspecies culture was measured and serial dilutions were plated for enumeration of the strains individually on selective media. 1.00E+00 1.00E+02 1.00E+04 1.00E+06 1.00E+08 0 22 30Log CFU Time (hours) Galactose Phosphate buffer S. gordonii S. mutans 1.00E+00 1.00E+02 1.00E+04 1.00E+06 1.00E+08 0 22 30Log CFU Time (hours) Galactose + Phosphate buffer S. gordonii S. mutans 1.00E+00 1.00E+02 1.00E+04 1.00E+06 1.00E+08 0 22 30Log CFU Time (Hours) Glucose Phosphate buffer S. gordonii S. mutans 1.00E+00 1.00E+02 1.00E+04 1.00E+06 1.00E+08 0 22 30Log CFU Time (hours) Glucose + Phosphate buffer S. gordonii S. mutans
121 Table 53. Average measured pH values from the mixedspecies competition assay testing persistence. Time Galactose + Phosphate Galactose Phosphate Glucose + Phosphate Glucose Phosphate t = 22 hours 5.4 4.7 5.8 4.4 t = 30 hours 5.3 4.7 5.7 4.5
122 CHAPTER 6 SUMMARY AND FUTURE DIRECTIONS Summary Dental plaque is an organized and complex structure with diverse microbial communities ( 28) Bacterial gene expression is const antly affected by frequent changes in the oral cavity such as nutrient availability, pH, oxygen tension and bacteriocin production ( 28 66 ) Those microbes capable of adapting and persisting in stressful environmental conditions ultimately determine which species will inhabit the biofilm. Commensal organisms are assoc iated with good oral health. In contrast, cariogenic S. mutans and other lactic acid bacteria are often isolated from carious lesions (30) Therefore, it is desirable to shift the equilibrium towards commensal bacteria such as S. gordonii After learning that S. gordonii grows well in galactose at concentrations 10 fold less than S. mutans a review of the genome revealed two tandem gene clusters containing duplicate copies of the tagatose6 phosphate pathway genes, a lactose PTS and a galactose PTS as well as a repressor, transcriptional antiterminator and pho spho galactosidase. One cluster appears to be dedicated to lactose metabolism while the other may be dedicated to a highaffinity galactose transport system. If the S. gordonii genome does encode a galactose PTS capable of highaffinity transport, it could likely give the cell a selective advantage over pathogenic bacteria such as S. mutans and provide a potential mechanism for directing the microbiome towards health. Experiments described in C hapter 3 showed the initial characterization of the genes in the lactose and galactose clusters by observing growth and measuring the transcription of two representative genes (one gene from each cluster). Major findings
123 included that the tr ansport of both lactose and galactose occurs by the lactose PTS, identification of a role for LacT in the expression of the lactose transporter, a function for lacG in regulating induction of the operons, verification that the two sets of tagatose pathway genes are the only way that lactose and galactose can be metabolized and demonstrating that increased expression of the galactose transporter is observed when cells are grown in galactose. Further investigation into the regulation of the lac gene clusters was described in C hapter 4. Two potential promoters were found, one in front of lacA1 and the other in front of lacA2 Promoter gene fusion experiments showed that both lacA1 and lacA2 have promoter activity. Continued analysis revealed that LacR acts as a negative regulator for both promoters. However, the lacA2 promoter is under the additional negative regulation of the global regulator of CCR, CcpA. A role for LacG was described in galactose metabolism whereby LacG is posited to convert Gal 6 P into some other form that would act as an inducer for LacR. Real Time PCR data also suggested that LacT may negatively regulate lacR in some way. An in vitro binding assay showed that LacR is capable of binding to both the lacA1 and lacA2 promoters. Chapter 5 outli ned a mixedspecies liquid culture competition assay between S. gordonii DL1 and S. mutans UA159 grown in galactose and glucose with and without phosphate buffer. Results indicated that S. gordonii is able to compete effectively against S. mutans under conditions testing viability. The ability of S. gordonii to stay viable was dependent on the type of sugar in the environment. Growth in galactose gave a competitive advantage to the commensal versus growth in glucose. When testing the ability to persist, S. gordonii was only able to outcompete S. mutans if the
124 media was buffered, whereas unbuffered media gave S. mutans a competitive advantage. Therefore, persistence appears to be dependent on the ability to survive in an acidic environment and growth in galac tose alone was not effective enough for S. gordonii to outcompete S. mutans at low pH. This assay is limited in scope, as it only reflects the interactions of two oral microbial species and was performed in vitro as opposed to in vivo It is possible that in an in vivo environment the cells would behave differently. For example, S. gordonii may gain an enhanced ability to survive at low pH when grown in galactose. In this case, the commensal could then continue to compete against S. mutans when the pH incre ased. Despite these limitations, the viability portion of the competition assay more accurately reflect s the natural environment of the oral cavity d ue to the nature of saliva flow. In that case, S. gordonii would be a good candidate organism for use in shifting the ecological balance to one of oral health via its high affinity galactose PTS Future D irections The S. gordonii lac genes presented here introduce a novel genetic organization of a system that is widespread in both Gram positive and Gram negative bacteria. What appear to be multiple layers of regulation only constitute the beginning of this story. In the future, this research can be br anched into multiple directions. A more thorough characterization of the galactose transporter is necessary to confirm that EIIGal is indeed a highaffinity transporter. In addition, measuring the gene expression of EIILac and EIIMan as well as the Leloir pathway enzymes in a Gal strain may provide some insight int o the highaffinity nature of the galactose transporter and possible explain the growth data concerning this strain. Northern blot experiments or Reverse Transcriptase PCR will reveal whether the lacTEFG genes are transcribed as
125 an operon or separately. The long chain formation and clumping phenotype expressed by the lacR mutant suggests that LacR is an important protein that probably has far reaching effects across the genome and warrants further examination. It is also important to continue the investigat ion into how LacR interacts with and regulates the lacA1 and lacA2 promoters T he 5RACE method would allow for map ping of the lacA1 and lacA2 promoters Exploration of the deeper layers of regulation that have been suggested here can provide a role for L a cT and L acG as regulators Investigation into the putative lacT and lacR promoters should include promoter gene fusions to evaluate promoter activity in various deletion strains and Real Time PCR to measure gene expression of suspected regulators. Since there is a cre sequence in the promoter of lacT CcpA regulation over this promoter should be investigated as well. The competition assay could be expanded further by performing the same assay between the S. gordonii wild type strain and the EIIGal strain using varying concentrations of galactose. With this kind of setup it would be possible to evaluate high vs low affinity transport by adjusting the galactose concentration. In order to better imitate an in vivo model, the competition assay can also be repeated in a biofilm culture rather than a planktonic culture.
126 Figure 61. Model depicting LacR and carbohydrate catabolite repression of the lacA1 and lacA2 promoters in S. gordonii when glucose is present. Binding of the LacR repressor to th e lacA1 and lacA2 promoters inhibits transcription of both sets of tagatose pathway genes ( lacABCD1 lacABCD2 ). In addition, Glc 6 P is converted to FBP and ATP, both of which activate the enzyme HPrK /P to phosphorylate HPr and produce HPr(Ser P). CcpA bin ds to HPr(SerP) and together these two proteins corepress the lacA2 promoter to further prevent transcription of the second set of tagatose genes.
127 Figure 62. Model depicting the regulation of the lacA1 and lacA2 promoters in S. gordonii whe n galactose is present. Binding of an unknown inducer molecule to LacR causes an inhibition of repression of the lacA1 and lacA2 promoters, resulting in transcription of the tagatose genes. LacG has been proposed to convert or alter the incoming sugar, Gal 6 P, into the unknown inducer molecule. Furthermore, CcpA is inhibited from binding to the lacA2 promoter due to low availability of the corepressor HPr(SerP). With Gal 6 P present in the cell and transcription of the tagatose genes uninhibited, metabol ism of galactose ensues.
128 Figure 63. Model depicting the regulation of the lacA1 and lacA2 promoters in S. gordonii when lactose is present. Two separate mechanisms inhibit repression of the lacA1 and lacA2 promoters and allow transcription of the tagatose genes: (1) after Lac 6 P is catabolized into glucose and Gal 6 P, LacG converts Gal 6 P into an unknown inducer molecule that binds to LacR and inhibits binding to both promoters, (2) CcpA is inhibited fro m binding to the lacA2 promoter due to low availability of the corepressor HPr(SerP). In addition, LacT activates the lacA1 promoter to increase transcription of the first set of tagatose genes. With transcription of both sets of tagatose genes occurring the Gal 6 P present in the cell (originating from lactose uptake) can be metabolized.
129 LIST OF REFERENCES 1. Aas, J. A., B. J. Paster, L. N. Stokes, I. Olsen, and F. E. Dewhirst. 2005. Defining the normal bacterial flora of the oral cavity. J Clin Microbiol 43: 57215732. 2. Abranches, J., M. M. Candella, Z. T. Wen, H. V. Baker, and R. A. Burne. 2006. Different roles of EIIABMan and EIIGlc in regulation of energy metabolism, biofilm development, and competence in Streptococcus m utans J Bacteriol 188: 37483756. 3. Abranches, J., Y. Y. Chen, and R. A. Burne. 2004. Galactose metabolism by Streptococcus mutans Appl Environ Microbiol 70: 60476052. 4. Abranches, J., et al 2008. CcpA regulates central metabolism and virulence gene expression in Streptococcus mutans J Bacteriol 190: 23402349. 5. Ahn, S. J., and R. A. Burne. 2007. Effects of oxygen on biofilm formation and the AtlA autolysin of Streptococcus mutans J Bacte riol 189: 62936302. 6. Ahn, S. J., J. A. Lemos, and R. A. Burne. 2005. Role of HtrA in growth and competence of Streptococcus mutans UA159. J Bacteriol 187: 30283038. 7. Ajdic, D., and V. T. Pham. 2007. Global transcriptional analysis of Streptococcus mutans sugar transporters using microarrays. J Bacteriol 189: 50495059. 8. Alpert, C. A., and U. Siebers. 1997. The lac operon of Lactobacillus casei contains lacT a gene coding for a protein of the BglG family of transcriptional antiterminators. J Bacteriol 179: 1555 1562. 9. Amster Choder, O., and A. Wright. 1997. BglG, the response regulator of the Escherichia coli bgl operon, is phosphorylated on a histidine residue. J Bacteriol 179: 56215624. 10. Armstrong, W. G. 1968. Origin and nature of the acqui red pellicle. Proc R Soc Med 61: 923930. 11. Banas, J. A., and M. M. Vickerman. 2003. Glucanbinding proteins of the oral streptococci. Crit Rev Oral Biol Med 14: 89 99. 12. Barnard, J. P., and M. W. Stinson. 1999. Influence of environmental conditions on hydrogen peroxide formation by Streptococcus gordonii Infect Immun 67: 65586564. 13. Batoni, G., et al 1992. Epidemiological survey of Streptococcus mutans in a group of adult patients living in Pisa (Italy). Eur J Epidemiol 8: 2382 42.
130 14. Becker, M. R ., et al 2002. Molecular analysis of bacterial species associated with childhood caries. J Clin Microbiol 40: 1001 1009. 15. Bettenbrock, K., U. Siebers, P. Ehrenreich, and C. A. Alpert. 1999. Lactobacillus casei 64H contains a phosphoenolpyruvatedependent phosphotransferase system for uptake of galactose, as confirmed by analysis of ptsH and different gal mutants. J Bacteriol 181: 225 230. 16. Bowden, G. H., and Y. H. Li. 1997. Nutritional influences on biofilm development. Adv Dent Res 11: 8199. 17. Breidt, F., Jr., W. Hengstenberg, U. Finkeldei, and G. C. Stewart. 1987. Identification of the genes for the lactosespecific components of the phosphotransferase system in the lac operon of Staphylococcus aureus J Biol Chem 262: 1644416449. 18. Burnaugh, A. M., L. J. Frantz, and S. J. King. 2008. Growth of Streptococcus pneumoniae on human glycoconjugates is dependent upon the sequential activity of bacterial exoglycosidases. J Bacteriol 190: 221230. 19. Burne R. A., R. G. Quivey, Jr., and R. E. Marquis. 1999. Physiologic homeostasis and stress responses in oral biofilms. Methods Enzymol 310: 441460. 20. Burne, R. A., K. Schilling, W. H. Bowen, and R. E. Yasbin. 1987. Expression, purification, and characteriz ation of an exobetaD fructosidase of Streptococcus mutans J Bacteriol 169: 4507 4517. 21. Byers, H. L., E. Tarelli, K. A. Homer, and D. Beighton. 1999. Sequential deglycosylation and utilization of the N linked, complex type glycans of human alpha1acid glycoprotein mediates growth of Streptococcus oralis Glycobiology 9: 469 479. 22. Carlsson, J. 1967. Presence of various types of nonhemolytic streptococci in dental plaque and in other sites of the oral cavity in man. Odontol Rev 18: 55 74. 23. Cha, R. S., H. Zarbl, P. Keohavong, and W. G. Thilly. 1992. Mismatch amplification mutation assay (MAMA): application to the c H ras gene. PCR Methods Appl 2: 14 20. 24. Chassy, B. M., and J. Thompson. 1983. Regulation and characterization of the galactosephosph oenolpyruvatedependent phosphotransferase system in Lactobacillus casei J Bacteriol 154: 12041214. 25. Chatterjee, C., M. Paul, L. Xie, and W. A. van der Donk. 2005. Biosynthesis and mode of action of lantibiotics. Chem Rev 105: 633 684.
131 26. Chen, Y. Y., C. A. Weaver, D. R. Mendelsohn, and R. A. Burne. 1998. Transcriptional regulation of the Streptococcus salivarius 57.I urease operon. J Bacteriol 180: 57695775. 27. Costerton, J. W. 1995. Overview of microbial biofilms. J Ind Microbiol 15: 137 140. 28. C osterton, J. W., Z. Lewandowski, D. E. Caldwell, D. R. Korber, and H. M. LappinScott. 1995. Microbial biofilms. Annu Rev Microbiol 49: 711 745. 29. Crutz, A. M., M. Steinmetz, S. Aymerich, R. Richter, and D. Le Coq. 1990. Induction of levansucrase in Baci llus subtilis : an antitermination mechanism negatively controlled by the phosphotransferase system. J Bacteriol 172: 10431050. 30. de Soet, J. J., and J. de Graaff. 1998. Microbiology of carious lesions. Dent Update 25: 319324. 31. De Stoppelaar, J. D., J. Van Houte, and D. O. Backer. 1970. The effect of carbohydrate restriction on the presence of Streptococcus mutans Streptococcus sanguis and iodophilic polysaccharideproducing bacteria in human dental plaque. Caries Res 4: 114 123. 32. de Vos, W. M. 19 99. Gene expression systems for lactic acid bacteria. Curr Opin Microbiol 2: 289295. 33. de Vos, W. M., and E. E. Vaughan. 1994. Genetics of lactose utilization in lactic acid bacteria. FEMS Microbiol Rev 15: 217237. 34. Declerck, N., F. Vincent, F. Hoh, S. Aymerich, and H. van Tilbeurgh. 1999. RNA recognition by transcriptional antiterminators of the BglG/SacY family: functional and structural comparison of the CAT domain from SacY and LicT. J Mol Biol 294: 389 402. 35. Eaton, T. J., C. A. Shearman, and M. J. Gasson. 1993. The use of bacterial luciferase genes as reporter genes in Lactococcus : regulation of the Lactococcus lactis subsp. lactis lactose genes. J Gen Microbiol 139: 14951501. 36. Frey, P. A. 1996. The Leloir pathway: a mechanistic imperative for three enzymes to change the stereochemical configuration of a single carbon in galactose. FASEB J 10: 461 470. 37. Fujita, Y. 2009. Carbon catabolite control of the metabolic network in Bacillus subtilis Biosci Biotechnol Biochem 73: 245259.
132 38. Fujita, Y., Y. Miwa, A. Galinier, and J. Deutscher. 1995. Specific recognition of the Bacillus subtilis gnt cis acting cataboliteresponsive element by a protein complex formed between CcpA and seryl phosphorylated HPr. Mol Microbiol 17: 953 960. 39. Gibbons, R. J., and J. V. Houte. 1975. Bacterial adherence in oral microbial ecology. Annu Rev Microbiol 29: 19 44. 40. Gosalbes, M. J., C. D. Esteban, and G. Perez Martinez. 2002. In vivo effect of mutations in the antiterminator LacT in Lactobacillus casei M icrobiology 148: 695 702. 41. Hennet, T. 2002. The galactosyltransferase family. Cell Mol Life Sci 59: 10811095. 42. Hojo, K., S. Nagaoka, T. Ohshima, and N. Maeda. 2009. Bacterial interactions in dental biofilm development. J Dent Res 88: 982990. 43. Ho rinouchi, S., and B. Weisblum. 1982. Nucleotide sequence and functional map of pC194, a plasmid that specifies inducible chloramphenicol resistance. J Bacteriol 150: 815 825. 44. Hoyle, B. D., and J. W. Costerton. 1991. Bacterial resistance to antibiotics: the role of biofilms. Prog Drug Res 37: 91105. 45. Idelson, M., and O. Amster Choder. 1998. SacY, a transcriptional antiterminator from Bacillus subtilis is regulated by phosphorylation in vivo J Bacteriol 180: 660 666. 46. Kawamura, Y., X. G. Hou, F. Sultana, H. Miura, and T. Ezaki. 1995. Determination of 16S rRNA sequences of S treptococcus mitis and Streptococcus gordonii and phylogenetic relationships among members of the genus Streptococcus Int J Syst Bacteriol 45: 406 408. 47. Kazor, C. E., et al 2003. Diversity of bacterial populations on the tongue dorsa of patients with halitosis and healthy patients. J Clin Microbiol 41: 558 5 63. 48. Kilian, M., L. Mikkelsen, and J. Henrichsen 1989. Taxonomic Study of Viridans Streptococci: Description of Str eptococcus gordonii sp. nov. and Emended Descriptions of Streptococcus sanguis (White and Niven 1946), Streptococcus oralis (Bridge and Sneath 1982), and Streptococcus mitis (Andrewes and Horder 1906). Int J Syst Bacteriol 39: 471 484.
133 49. Kleinberg, I. 2002. A mixedbacteria ecological approach to understanding the role of the oral bacteria in dental caries causation: an alternative to Streptococcus mutans and the specific plaque hypothesis. Crit Rev Oral Biol Med 13: 108125. 50. Kolenbrander, P. E. 201 1. Multispecies communities: interspecies interactions influence growth on saliva as sole nutritional source. Int J Oral Sci 3: 4954. 51. Kolenbrander, P. E. 2000. Oral microbial communities: biofilms, interactions, and genetic systems. Annu Rev Microbiol 54: 413 437. 52. Kolenbrander, P. E., et al 2002. Communication among oral bacteria. Microbiol Mol Biol Rev 66: 486505. 53. Kreth, J., J. Merritt, W. Shi, and F. Qi. 2005. Competition and coexistence between Streptococcus mutans and Streptococcus sangui nis in the dental biofilm. J Bacteriol 187: 71937203. 54. Kuramitsu, H. K., X. He, R. Lux, M. H. Anderson, and W. Shi. 2007. Interspecies interactions within oral microbial communities. Microbiol Mol Biol Rev 71: 653 670. 55. Kuramitsu, H. K., and B. Y. W ang. 2006. Virulence properties of cariogenic bacteria. BMC Oral Health 6 Suppl 1: S11. 56. Lau, P. C., C. K. Sung, J. H. Lee, D. A. Morrison, and D. G. Cvitkovitch. 2002. PCR ligation mutagenesis in transformable streptococci: application and efficiency. J Microbiol Methods 49: 193 205. 57. LeBlanc, D. J., V. L. Crow, L. N. Lee, and C. F. Garon. 1979. Influence of the lactose plasmid on the metabolism of galactose by Streptococcus lactis J Bacteriol 137: 878 884. 58. Lemos, J. A., and R. A. Burne. 2002. Regulation and Physiological Significance of ClpC and ClpP in Streptococcus mutans J Bacteriol 184: 63576366. 59. Loesche, W. J., J. Rowan, L. H. Straffon, and P. J. Loos. 1975. Association of Streptococcus mutans with human dental decay. Infect I mmun 11: 12521260. 60. Loo, C. Y., D. A. Corliss, and N. Ganeshkumar. 2000. Streptococcus gordonii biofilm formation: identification of genes that code for biofilm phenotypes. J Bacteriol 182: 13741382. 61. Lorca, G. L., et al 2005. Catabolite repression and activation in Bacillus subtilis : dependency on CcpA, HPr, and HprK. J Bacteriol 187: 7826 7839.
134 62. Loughman, J. A., and M. G. Caparon. 2006. A novel adaptation of aldolase regulates virulence in Streptococcus pyogenes EMBO J 25: 54145422. 63. Macr ina, F. L., C. L. Keeler, Jr., K. R. Jones, and P. H. Wood. 1980. Molecular characterization of unique deletion mutants of the streptococcal plasmid, pAM beta 1. Plasmid 4: 8 16. 64. Magitot, E. 1867. Treatise on dental caries: experimental and therapeutic investigations. 65. Marquis, R. E. 1995. Oxygen metabolism, oxidative stress and acidbase physiology of dental plaque biofilms. J Ind Microbiol 15: 198 207. 66. Marsh, P. D. 2004. Dental plaque as a microbial biofilm. Caries Res 38: 204 211. 67. Mercer, D. K., C. M. Melville, K. P. Scott, and H. J. Flint. 1999. Natural genetic transformation in the rumen bacterium Streptococcus bovis JB1. FEMS Microbiol Lett 179: 485490. 68. Miller, W. D. 1890. Microorganisms of the human mouth. Philadelphia, PA: S.S. White Dental Mfg Co. 69. Neves, A. R., et al 2010. Towards enhanced galactose utilization by Lactococcus lactis Appl Environ Microbiol 76: 70487060. 70. Nobbs, A. H., R. J. Lamont, and H F. Jenkinson. 2009. Streptococcus adherence and colonization. Microbiol Mol Biol Rev 73: 407450. 71. Nyvad, B., and M. Kilian. 1990. Comparison of the initial streptococcal microflora on dental enamel in caries active and in caries inactive individuals. Caries Res 24: 267 272. 72. Oskouian, B., and G. C. Stewart. 1990. Repression and catabolite repression of the lactose operon of Staphylococcus aureus J Bacteriol 172: 3804 3812. 73. Palmer, R. J., Jr., K. Kazmerzak, M. C. Hansen, and P. E. Kolenbrander. 2001. Mutualism versus independence: strategies of mixedspecies oral biofilms in vitro using saliva as the sole nutrient source. Infect Immun 69: 57945804. 74. Paster, B. J., I. Olsen, J. A. Aas, and F. E. Dewhirst. 2006. The breadth of bacterial diversity in the human periodontal pocket and other oral sites. Periodontol 2000 42: 80 87. 75. Pasteur, L., Faulkner, F., and Constabl e Robb, D. 1860. Physiological theory of fermentation.
135 76. Poncet, S., et al 2004. HPr kinase/phosphorylase, a Walker motif A containing bifunctional sensor enzyme controlling catabolite repression in Gram positive bacteria. Biochim Biophys Acta 1697: 123 135. 77. Ramos, J., K. Szkutnicka, and V. P. Cirillo. 1988. Relationship between low and highaffinity glucose transport systems of Saccharomyces cerevisiae J Bacteriol 170: 53755377. 78. Rosch, J. W., and E. Tuomanen. 2007. Adapting a diet from sugar to meat: doubledealing genes of Streptococcus pyogenes Mol Microbiol 64: 257 259. 79. Rosey, E. L., and G. C. Stewart. 1992. Nucleotide and deduced amino acid sequences of the lacR lacABCD, and lacFE genes encoding the repressor, tagatose 6phosphate gene cluster, and sugar specific phosphotransferase system components of the lactose operon of Streptococcus mutans J Bacteriol 174: 61596170. 80. Saier, M. H., et al 1996. Catabolite repression and inducer control in Gram positive bacteria. Microbiology 142 ( Pt 2): 217230. 81. Sandham, H. J., and I. Kleinberg. 1970. Contribution of lactic and other acids to the pH of a human salivary sediment system during glucose catabolism. Arch Oral Biol 15: 12631 283. 82. Shaw, W. V. 1975. Chloramphenicol acetyltransferase from chloramphenicol resistant bacteria. Methods Enzymol 43: 737 755. 83. Smith, E. G., and G. A. Spatafora. 2011. Gene Regulation in S. mutans : Control in a Complex Environment. J Dent Res. Pub lished online 8 July 2011. 84. Stephan, R. M. 1940. Two Factors of Possible Importance in Relation to the Etiology and Treatment of Dental Caries and Other Dental Diseases. Science 92: 578 579. 85. Stephan, R. M., and E. S. Hemmens. 1947. Studies of changes in pH produced by pure cultures of oral microorganisms; effects of varying the microbic cell concentration; comparison of different microorganisms and different substrates; some effects of mixing certain microorganisms. J Dent Res 26: 1541. 86. Terra, V. S., K. A. Homer, S. G. Rao, P. W. Andrew, and H. Yesilkaya. 2010. Characterization of novel betagalactosidase activity that contributes to glycoprotein degradation and virulence in Streptococcus pneumoniae. Infect Immun 78: 348 357. 87. Tettelin, H., et al 2001. Complete genome sequence of a virulent isolate of Streptococcus pneumoniae. Science 293: 498506.
136 88. Thomas, T. D., K. W. Turner, and V. L. Crow. 1980. Galactose fermentation by Streptococcus lactis and Streptococc us cremoris : pathways, products, and regulation. J Bacteriol 144: 672 682. 89. Thompson, J. 1980. Galactose transport systems in Streptococcus lactis. J Bacteriol 144: 683 691. 90. Tong, H., et al 2007. Streptococcus oligofermentans inhibits Streptococcus mutans through conversion of lactic acid into inhibitory H2O2: a possible counteroffensive strategy for interspecies competition. Mol Microbiol 63: 872 880. 91. Tong, H., X. Gao, and X. Dong. 2003. Streptococcus oligofermentans sp. nov., a novel oral isol ate from caries free humans. Int J Syst Evol Microbiol 53: 11011104. 92. Tortosa, P., et al 1997. Multiple phosphorylation of SacY, a Bacillus subtilis transcriptional antiterminator negatively controlled by the phosphotransferase system. J Biol Chem 272 : 1723017237. 93. Vadeboncoeur, C., and M. Pelletier. 1997. The phosphoenolpyruvate:sugar phosphotransferase system of oral streptococci and its role in the control of sugar metabolism. FEMS Microbiol Rev 19: 187 207. 94. van Rooijen, R. J., and W. M. de Vos. 1990. Molecular cloning, transcriptional analysis, and nucleotide sequence of lacR, a gene encoding the repressor of the lactose phosphotransferase system of Lactococcus lactis. J Biol Chem 265: 18499 18503. 95. van Rooijen, R. J., M. J. Gasson, and W. M. de Vos. 1992. Characterization of the Lactococcus lactis lactose operon promoter: contribution of flanking sequences and LacR repressor to promoter activity. J Bacteriol 174: 22732280. 96. Vickerman, M. M., M. C. Sulavik, and D. B. Clewell. 1995. Or al streptococci with genetic determinants similar to the glucosyltransferase regulatory gene, rgg Infect Immun 63: 45244527. 97. Wen, Z. T., and R. A. Burne. 2002. Analysis of cis and trans acting factors involved in regulation of the Streptococcus mutans fructanase gene ( fruA ). J Bacteriol 184: 126 133. 98. Wijeyeweera, R. L., and I. Kleinberg. 1989. Arginolytic and ureolytic activities of pure cultures of human oral bacteria and their effects on the pH response of salivary sediment and dental plaque in vitro Arch Oral Biol 34: 43 53.
137 99. Wilkins, J. C., K. A. Homer, and D. Beighton. 2002. Analysis of Streptococcus mutans proteins modulated by culture under acidic conditions. Appl Environ Microbiol 68: 23822390. 100. Xu, P., et al 2007. Genome of the opportunistic pathogen Streptococcus sanguinis J Bacteriol 189: 31663175. 101. Zeng, L., and R. A. Burne. 2008. Multiple sugar: phosphotransferase system permeases participate in catabolite modification of gene expression in Streptococcus mutans Mol Mic robiol 70: 197 208. 102. Zeng, L., and R. A. Burne. 2010. Seryl phosphorylated HPr Regulates CcpA Independent Carbon Catabolite Repression in Conjunction with PTS Permeases in Streptococcus mutans Mol Microbiol 75 : 11451158. 103. Zeng, L., S. Das, and R A. Burne. 2010. Utilization of lactose and galactose by Streptococcus mutans : transport, toxicity, and carbon catabolite repression. J Bacteriol 192: 24342444. 104. Zeng, L., Y. Dong, and R. A. Burne. 2006. Characterization of cis acting sites controlling arginine deiminase gene expression in Streptococcus gordonii J Bacteriol 188: 941 949. 105. Zeng, L., Z. T. Wen, and R. A. Burne. 2006. A novel signal transduction system and feedback loop regulate fructan hydrolase gene expressi on in Streptococcus mutans Mol Microbiol 62: 187 200.
138 BIOGRAPHICAL SKETCH Nicole Christine Martino was born in 1986 in Fort Lauderdale, Florida. The eldest of three children, she spent 18 years in South Florida with her family before graduating fro m Boyd H. Anderson High School in 2004 with a diploma from the International Baccalaureate (IB) Programme. She earned her B.S. in Food Science and Human Nutrition with an emphasis on Human Nutrition from the University of Florida (UF) in 2008. During this time, she contributed to dental related research on the etiology of canker sores with Dr. Lorena Baccaglini. In October 2007 and April 2008, she presented an abstract and poster of her work at the Hispanic Dental Association (HDA) annual meeting and Americ an Association for Dental Research (AADR ) annual meeting, respectively. After taking a year off to work and travel, Nicole enrolled as a graduate student in the UF College of Medicine. Working towards a Master of Science degree, she completed research under the supervision of Dr. Robert A. Burne for two years. In May 2010, she presented her work in an abstract and poster at the American Society for Microbiology (ASM) General Meeting. Upon completion of her M.S. program, Nicole plans to publish a journal art icle related to her recent work in Dr. Burnes laboratory and will be attending the UF College of Dentistry (UFCD) in Gainesville, Florida in order to earn a D.M.D. degree and enter the professional field of dentistry.