Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Atypical roles for campylobacter jejuni AA-ABC transporter components PAQP and PAQQ in bacterial stress.. Lin, Ann En-Ju 2008-02-04

You don't seem to have a PDF reader installed, try download the pdf

Item Metadata


24-ubc_2008_fall_lin_ann_en-ju.pdf [ 1.92MB ]
JSON: 24-1.0066912.json
JSON-LD: 24-1.0066912-ld.json
RDF/XML (Pretty): 24-1.0066912-rdf.xml
RDF/JSON: 24-1.0066912-rdf.json
Turtle: 24-1.0066912-turtle.txt
N-Triples: 24-1.0066912-rdf-ntriples.txt
Original Record: 24-1.0066912-source.json
Full Text

Full Text

ATYPICAL ROLES FOR CAMPYLOBACTER JEJUNIAA-ABCTRANSPORTER COMPONENTS PAQP AND PAQQ IN BACTERIAL STRESSTOLERANCE AND PATHOGEN-HOST CELL DYNAMICSbyAnn En-Ju LinB.Sc. Honours, The University of British Columbia, 2006A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THEREQUIREMENTS FOR THE DEGREE OFMASTER OF SCIENCEinThe Faculty of Graduate Studies(Microbiology and Immunology)THE UNIVERSITY OF BRITISH COLUMBIA(VANCOUVER)AUGUST, 2008© Ann En-Ju Lin, 2008ABSTRACTCampylobacterjejuni is a human pathogen that causes severe diarrhea! disease. However,our understanding of C. jejuni virulence mechanisms and survival during disease andtransmission remains limited. Amino acid ATP Binding Cassette (AA-ABC) transporters in C.jejuni have been proposed as being important for bacterial physiology and pathogenesis. Wehave investigated a novel AA-ABC transporter system, encoded by cj0467-9, by generatingtargeted deletions of cj0467 (membrane transport component) and cj0469 (ATPase component)in C. jejuni 81-176. Analyses described herein have led us to designate these genes paqP andpaqQ, respectively [pathogenesis-ssociated glutamine (q) ABC transporter permease()andATPase(Q)].We found that loss of either component resulted in amino acid uptake defects,most notably diminished glutamine uptake. Both zXpaqP and ApaqQ mutants also exhibited asurprising but significant increase in short-term intracellular survival in macrophages andepithelial cells. Levels of resistance to a series of environmental and in vivo stresses wereexamined. Both mutants were hyper-resistant to aerobic and oxidative stress, and while ApaqPwas also hyper-resistant to heat and osmotic shock, ApaqQ was more susceptible than wild-typeto the latter two stresses. Annexin-V staining coupled with fluorescence microscopy revealedthat macrophages infected with the ApaqP and ZipaqQ mutants underwent a lower level ofapoptosis than cells infected with wild-type bacteria. Macrophages infected with the mutantstrains exhibited a transient decrease in ERK activation compared to wild type-infectedmacrophages, potentially explaining the reduced apoptosis phenotype. The ApaqP mutant did notexhibit a defect for short or longer term mouse colonization, consistent with its increased stresssurvival and diminished host cell damage phenotypes. Collectively, these results demonstrate aunique correlation between an AA-ABC transporter with bacterial stress tolerance, intracellularsurvival, host cell damage, and host signal transduction in response to pathogen infection.11TABLE OF CONTENTSTitle Page iAbstract iiTable of Contents iiiList of Tables viList of Figures viiAbbreviations viiiAcknowledgements ixDEDICATION X1.0 INTRODUCTION 11.1 Campylobacterfejuni- History and epidemiology 21.2 Campylobacteriosis 31.2.1 Symptoms 31.2.2 Medical sequelae 31.2.3 Treatment 41.3 Morphology and characteristics 41.4 Metabolism 51.4.1 Amino acid metabolism in C. jejuni 51.4.2 Glutamine metabolism and pathogenesis 61.4.3 Glutamine contributes to host cell survival 71.5 ATP Binding Cassette (ABC) transporters 71.5.1 AA-ABC tranposrter structures 71.5.2 ABC transporter functions 111.5.3 AA-ABC transporters serve as pathogenesis determinants in C. jejuni 121.5.4 AA-ABC transporters serve as pathogenesis determinant in other pathogens ... 121.6 Physiology and stress response 131.6.1 Heatstress 131.6.2 Aerobic stress 141.6.3 Oxidative (reactive oxygen species) stress 141.6.4 Osmotic stress 151.7 Host-pathogen interaction 161.7.1 Host epithelial cell colonization and invasion 161.7.2 Toxin. 181.7.3 Apoptosis 181.8 Host cell signaling 191.8.1 MAPK signaling pathway 191.8.2 MAPK dependent signaling activation in C. jejuni infected cells191.8.3 Inflammatory response 201.9 Objective and hypothesis 212.0 MATERIALS AND METHODS 222.1 Bacterial strains, cell lines, media and growth conditions 232.2 Construction of Campylobacterjejuni 8 1-176 /paqP andApaqQ targeted deletion mutants 232.3 Complementation of ApaqP deletion mutant 242.4 RNA extraction and reverse transcript- PCR (RT-PCR) analysis 252.5 Southern blot hybridization 252.6 Amino acid transport assays 262.7 Cell infection assay for colonization, invasion, intracellularsurvival and gentamicin MIC determination 272.8 In vivo colonization using a mouse model 282.9 Oxidative, aerotolerance, heat stress and osmotic stress survival assays 282.10 Cellular apoptosis and viability detection 292.11 Cell lysate preparation and Western blotting 292.12 Additional phenotypic assays 303.0 RESULTS 343.1 Overview 353.2 cj0467-9 encodes a putative amino acid (AA) ABC transporter 363.3 Construction of targeted, non-polar cj0467 and cj0469 disruption strains 373.4 The ApaqP and ApaqQ C. jejuni mutants are defective for L-glutamine 39uptake and moderately defective for uptake of other amino acidsiv3.5 C. jejuni zSpaqP and ApaqQ increase short-term intracellular survival in 41R.AW264.7 macrophage, 1NT407 epithelial cells and Caco-2 intestinalepithelial cells.3.6 No significant difference between the ApaqP mutant and WT C. jejuni 44colonization are observed in vivo using a mouse infection model.3.7 ApaqP and ApaqQ C. jejuni mutants exhibit increase resistance to limited 46C02,aerobic conditions.3.8 ApaqP and ApaqQ C. jejuni mutants exhibit an increase in resistance to organichydroperoxide t-BOOH, but not inorganic peroxides H20 and paraquat 483.9 Tolerance to heat and osmotic stress is differentially altered with A paqPand ApaqQ 503.10 Macrophages infected with C. jejuni ApaqP and ApaqQ exhibit reduced apoptosisand ERK activation compared to macrophages infected with WT bacteria 523.11 A summary of phenotypes for which iXpaqP and ApaqQ mutants were notsignificantly different from the wild-type strain 573.12 WT C. jejuni and mutants exhibit similar tolerance to heat shock and pH stresses 614.0 DISCUSSION AND CONCLUSIONS 655.0 FUTURE DIRECTIONS 74REFERENCES 78VLIST OF TABLESTable 3.1 Phenotypes that were indistinguishable between ApaqP, z.paqQ mutantsandWT C. jejuni 81-17658Table 3.2 Antimicrobial susceptibility test60viLIST OF FIGURESFigure 1.1 Schematic diagram of a bacterial ABC transporter10Figure 3.1 Generation of non-polar, single insert 4aqP and 4aqQ disruption strains.... 38Figure 3.2 4aqP and 4aqQ mutants exhibit reduced levels of glutamine 40and glutamate uptake relative to WT.Figure 3.3 4aqP and 4aqQ mutants exhibit a statistically significant increase in 42short-term intracellular survival in macrophages and epithelial cellsFigure 3.4 ApaqP is not defective for mouse colonization in vivo 45Figure 3.5 4aqP and paqQ are more resistant to aerobic stresses than WT 47Figure 3.6 zpaqP mutant exhibits a significant increase in resistance to tert-butylhydroperoxide (t-BOOH) but not to hydrogen peroxide (H20)or paraquat .... 49Figure 3.7 The 4aqP mutant is more resistant to heat and osmotic stress while theIXpaqQ mutant is more sensitive than WT 51Figure 3.8 4aqP and 4aqQ infected cells showed decreased levels of apoptosisand ERK activation compared to WT infected cells 54Figure 3.9 4aqP and 4paqQ infected Caco-2 and 1NT407 epithelial cells showedsimilar ERK and INK activities as WT infected cells 56Figure 3.10 4paqP and I.paqQ mutants survive similarly to WT C. jejuni in richor minimal media 59Figure 3.11 4aqP and 4aqQ mutants exhibit similar survival as the WT strainunder anaerobic conditions over time 60Figure 3.12 4aqP and 4paqQ mutants do not exhibit a motility defect in 0.4% agar 62Figure 3.13 Rate of GGT activity in I.paqP and 4paqQ mutants does not significantlydeviate from WT C:jejuni 62Figure 3.14 zSpaqP and i.paqQ mutants exhibit similar protein secretion profile asWT C. jejuni 63Figure 3.15 4aqP and i.aqQ mutants and WT C. jejuni survive similarly in minimalessential media supplemented with L-glutamine 64viiABBREVIATIONSApaqP C. jejuni pathogenesisassociated glutamine trans5orter permease mutantApaqPc C. jejuni pathogenesisassociated glutamine transporter permeasecomplementApaqQ C.jejuni pathogenesisassociated glutamine transporter ATPasemutantz\spoT C. jejuni spoT mutantAA amino acidABC ATP Binding CassetteAmp ampicillinATP adenosine 5’-triphosphatebp base pairBSA Bovine serum albuminCAT chioramphenicol acetyltransferaseCDT Cytolethal Distending ToxinC. jejuni CampylobacterjejuniCFUs colony forming unitsCm chioramphenicolCV crystal violetDAPI 4’ ,6-Diamidino-2-phenylindoleDMEM Dulbecco modified Eagle MediaDNA deoxyribonucleic aciddH2O distilled waterECL enhanced chemiluminescenceE. coli Escherichia coliEDTA Ethylene diamine tetraaceticacidERK Extracellular signal-regulatedkinaseGBS Guillain-Barré syndromeGGT gamma glutamyl transferasegln L-glutamineGPCR G-protein coupled receptorH pylon Helicobacter pylonHRP Horseradish peroxideIL interleukin1NT407 human intestinal 407Jnk c-Jun NH2-terminal kinaseKan kanamycinkDa kilodaltonLB Luria-BertaniLOS Lipo-oligosaccharideLPS Lipo-oligosaccharideMAPK Mitogen activated protein kinaseMEM Minimal Essential MediaMH Muller HintonMIC Minimum Inhibitory ConcentrationMOl multiplicity of infectionNEAA non-essential amino acidNF-icB Nuclear factor kappa BOD600 Optical density at 600nmPBS phosphate buffered salineviiir.,p.m.revolutions per minuteRTK ReceptorTyrosine kinaseSDS-PAGE Sodium DodecylSulphate Polyacrylamide Gelt-BOOH tert-butyl hydroperoxideTBSTTris-Buffered Saline Tween-20TCAtricholoracetic acidTLR toll-like receptorUI uninfectedWI wild-typeixACKNOWLEDGEMENTSI offer my enduring gratitudeand sincere thanks to my supervisor,Dr. Erin Gaynor, forher constant encouragement andsupport. Her positive and cheerfulattitude as well as helpfulguidance throughout my studieshave allowed me to work productivelyyet pleasantly in the labduring the past two years. I verymuch appreciate her incredible enthusiasmand patience inoffering direction and insightsat each stage of my project. Thishas truly been a wonderfullearning experience for me.I would also like to thank members of theCvitkovitch lab from the Universityof Torontofor their kindness and hospitality duringthe 2 weeks when I was in Toronto performingtheamino acid transport assay experiment.I am especially grateful to Kirsten Krastel andDr. DennisCvitkovitch, not only for their extremeenthusiasm and hospitality, but alsofor helping me withthe assays and data interpretations.I also thank Rhonda Hobb andDr. Stuart Thompson from theMedical College of Georgia for performingthe in vivo mouse infections. Ithank Dr. MichaelGold from our department for kindly providingthe MAPK antibodies.Sincere thanks to my thesis committee members,Dr. Bruce Vallance and Dr. Bill Mohn.A huge thank you to each and everyone inthe Gaynor lab, past and present,for creating apositive, energetic and entertainingwork environment, which madeeveryday at the lab fun andenjoyable. Particularly, I would like tothank Mizue Naito, my partnerin fighting off the worldof diarrheal diseases, for being a great friendin and out of the lab. I also thankDr. EmilisaFrirdich, who I always sharea good laugh with discussing variouskinds of issues, lab related ornot. Grad school has never been dullfor a day with their presence.I have to offer my deepest thanks tomy family, particularly my momand dad, as well asmy friends, for their continuous support.Especially, I have to thankDr. Mm Liu, for being agreat listener, and being like a sisteranyone would only dream to have. Hercompassion, wisdomadvice and constant encouragementhave helped me tremendouslythrough the past few years ofmy life. I would also like to thankKaren Chang, Karen To and MilaChou, my dearest and loyalfriends, for being supportivesince our undergraduate and highschool years; I would not havemade it through without them.Finally, I would like to thank the CanadianInstitutes of Health Research (CIHR)andMichael Smith Foundation for HealthResearch (MSFHR) for fundingmy research stipend, aswell as CIHR, the MSFHR, andthe Burroughs Welicome Fund forsupporting my researchactivities.xDEDICATION¶Th my unck creng - my inspirerin science, wño is no longer wit/ius.xi1.0 INTRODUCTION11.1 CAMPYLOBACTERJEJUNI- HISTORY AND EPIDEMIOLOGYCampyiobacterjejuni isa highly prevalent bacterial pathogenthat causes humangastroenteritis (Younget al., 2007). Campylobacterspp. were first observed in 1886byEscherich in the colonsof neonates. However,it was not until 1968 that Dekeyser,Buzier andcolleagues first isolatedCampylobacter spp. fromthe feces of a human diarrheicpatient (Butzler,2004) . By the mid- 1 970s, Campylobacterspp. were recognized as importanthuman pathogens.Currently, Campylobacterspp. are ranked as the most commonbacterial cause ofdiarrheal illnesses in developedcountries (Blaser, 1997). Withan estimated 400 million casesper year worldwide (Girardet al., 2006), Campylobacterspp. affect more individualsthan E. coil0157, Saimoneiia spp, andShigelia spp. combined (Blaseret ai., 1983). For instance,of thepopulation in North Americaand Western Europe are infectedeach year, with approximately3million people infectedin the United States alone (Meadet al., 1999). The genusCampylobactercontains 17 species,including C. hyointestinaiis,C. ianienae, C. sputorum,C. mucosalis, C.concisus, C. curvus,C. rectus, C. gracilis,C. showae, C. hominis, C. lan,C. insuiaenigrae, C.canadensis, C. upsaliensis,C. heiveticus, C. coli, andC. jejuni (Debruyne et al., 2008).Althoughseveral of these causediarrheal disease, Campyiobacterjejuniis estimated to accountfor 90% ofreported cases of diarrhealcampylobacteriosis (Meadet ai., 1999).C. jejuni is a zoonotic organismthat exists as a commensalin chickens and otheravianspecies (Beery et al.,1988); hence, infection of chickenswith C. jejuni typicallydoes not lead todiarrhea. This commensalrelationship isthought to account for thefact that the mainsource ofhuman C. jejuni infectionis via consumption of contaminatedpoultry or cross-contaminationofother food with raw poultryjuice. In addition,high carriage rates ofC. jejuni have beenidentified in Canadian geeseand migratory ducks, suggestingthat wild birdsmay play a roleinspreading C. jejuni in theenvironment (Pachaet ai., 1988). Other routesof human C. jejuniinfection are alsoof a primary zoonoticnature and include ingestionof water contaminatedby2fecal runoff and ingestionof unpasteurized milk (Young etal., 2007; Blaser et al., 1983). Thelatter two also accountfor the majority of large-scaleCampylobacteriosis outbreaks.1.2 CAMPYLOBACTERIOSIS1.2.1 SymptomsUpon infection, C. jejuni predominantlycolonizes the jejunum, ileumand thecolon in humans (Allos and Blaser,1995). The infectious dosefor C. jejuni can be quitelow, with some strains colonizingwith an approximate 200 bacteria/dose(Blaser, 1997).Clinical symptoms oftenappear within 4 days andare characterized by profuse,oftenbloody diarrhea, acute abdominalpain, fever, malaise, nausea,and vomiting (Alloset al.,1998; Blaser et al., 1983). The acuteillness usually lastsfor 2-5 days, and patientstypically recover fully. However,approximately 25%of patients experience a relapse.Complications of infectioncan also include intestinalhemorrhage, haemolytic uraemicsyndrome, or inflammationof abdominal lymphnodes, also known as mesentericadenitis(Allos et al., 1998).1.2.2 Medical SequelaeAcute C. jejuni infectionsusually resolve by 10-14days. In some cases, longer-term medical sequelae canoccur, including reactivearthritis, Guillain-BarréSyndrome(GBS), which is the primarycause of acute ascending bilateralparalysis, and MillerFisher Syndrome,a non-paralytic rare variantof GBS (Humphrey et al.,2007). GBS isanautoimmune disorderin which antibodies areproduced against themyelin sheath thatsurrounds the axons of theperipheral nerves.This inhibits effective signaltransmissionresulting in acute paralysis.Although the exact causeof GBS still remainsunknown, a3large majority of GI3S cases have been associated with prior C. jejuni infection (13-39%of cases) (Yu et iii., 2006). C. jejuni Iipo-oligosaccharide (LOS) closely resembles theperipheral nerve ganglioside, GM1 (Yu et al., 2006), and duringC. jejuni infection,antibodies produced against C. jejuni LOS may cross-react with GM1, thereby inducingGBS. GBS symptoms usually manifest 1-3 weeks after the resolutionof enteritissymptoms (Butzler, 2004).1.2.3 TreatmentAlthough acute C. jejuni infection is often self-limiting and typically resolveswithin 2 weeks, antimicrobial agents are sometimes required for moresevere,complicated or systemic infection. Currently, macrolides such as erythromycinandfluororquinolones such as ciprofloxacin are the most common antibioticsprescribed foruncomplicated C. jejuni infection (Nachamkin, 2001). Furthermore,antibiotics have alsobeen incorporated to feedings of chicks to lower C. jejuni colonization.Unfortunately,reports describing the emergence of antimicrobial-resistant C. jejunihave increased sincethe late l980s (Moore et al., 2001). Although a few candidatevaccines have beendeveloped (Girard et al., 2006), an effective vaccine is stillunavailable. Vaccinedevelopment has been challenging due to the complex natureof C. jejuni surface antigens,including several polysaccharide structures conferring multipleserotype combinations.1.3 MORPHOLOGY AND CHARACTERISTICSC. jejuni is a spiral, Gram-negative bacillus that is 0.2-0.9 .tm wide,0.5-5.0 im long. Itbelongs to the epsilon class of proteobacteria, in the order of Campylobacteriaceae,togetherwith the genera Helicobacter, Wolinella, Sulfurospirillum, Acrobacterand Dehalospirillum.These species typically have small genomes of 1.6-2.0 megabases(Young et al., 2007; Thomas4et a!., 1999). The completegenome sequences havebeen deduced for theC. jejuni strains NCTC11168, RM1221,8 1-176 and 269.97 isolated fromhuman blood culture. Thegenome size mayvary slightly dependingon the strain; for instance,C. jejuni NCTC 11168 has1633 proteinencoding genes, andC. jejuni 81-176 has 1748protein encoding genes accordingto the NationalMicrobial Pathogen DataResource Center( a naturally competentspecies, C. jejuni isable to take up DNA fromthe environment. Witha flagellum at each end,C.jejuni is highly motile andoften exhibits rapid dartingand spinning motions(Young et a!., 2007).Campylobacter is a difficultorganism to culture, dueto its unique growth requirements.As afastidious microaerophilic organism,it survives optimally under12% CO2 and6% 02 conditionsand requires complex nutrientsfor optimal survival (Garenauxet a!., 2008; Park, 2002).Because C. jejuni is asaccharolyticand lacks a unidirectionalenzyme, 6-phospho-fructokinase,in the glycolytic pathway,it cannot metabolizeglucose and instead dependson amino acidsforboth carbon and energy sources(Guccione eta!., 2008;Muller eta!., 2005).1.4 METABOLISM1.4.1 Amino acid metabolismin C. jejuniAmino acids (AAs) are essentialnutrients requiredby C. jejuni for growthandsurvival (Guccione eta!., 2008; Muller et a!.,2005; Velayudhanand Kelly, 2002).As themost important carbon andenergy resource, AAsthus contributeto the synthesisofnumerous enzymes andmetabolites includingfatty acids, nucleicacids, proteins andother amino acids as wellas organic acids, whichcan serve as metabolicsubstrates forthe TCA cycle. Previousstudies identified asparagine,aspartate, glutamine,glutamate,serine and proline asthe major energy sourcesin C. jejuni strainNCTC 11168(Guccione5et al., 2008; Leon-Kempis Mdel et al., 2006). It is clear that as these AAs can be readilycatabolized and act as precursors for the formation of organic acids such as pyruvate,fumarate, and succinate, which directly feed into the trichioracetic (TCA) cycle, servingas electron donors and/or acceptors essential for respiration. It was previously shown thatglutamine, also the major nitrogen donor in C. jejuni, is rapidly utilized at a significantlyhigher rate than glutamate (Westfall et al., 1986).1.4.2 Glutamine metabolism and pathogenesisAside from being primary energy sources for growth and survival, AAs and theirderivate also contribute to pathogenesis in several bacterial pathogens (Guccione et al.,2008; Barnes et al., 2007; Shibayama et al., 2007; Tullius et al., 2003; Smirnova et al.,2001). Recently, an enzyme involved in glutamine metabolism was associated withbacterial stress responses and host-pathogen interactions in both Helicobacterpylori andC. jejuni (Barnes et al., 2007; Shibayama et al., 2007). Gamma-glutamyl transpeptidase(GGT) in H pylon is involved in the hydrolysis of extracellular glutamine or glutathione(GSH) into glutamate for subsequent uptake and is involved in the induction of host cellapoptosis and in vivo colonization (Shibayama et al., 2007). In C. jejuni, GOT is involvedin H20resistance, host cell apoptosis, intracellular survival, and colonization of chicksand mice (Barnes et al., 2007; Hofreuter et al., 2006). GSH also functions as anantioxidant that protects E. coli from oxidative damage and osmotic shock (Smirnova etal., 2001). In addition, glutamine synthetase, GinA, in Mycobacterium tuberculosis, hasalso been shown to be essential for virulence, as a AglnA mutant is avirulent in vivo andunable to survive within human macrophages in vitro (Tullius et al., 2003). These studieshighlight the importance of AA homeostasis and its potential role in bacterial6pathogenesis. However, details describingthe molecular mechanisms of how AAmetabolism is linked to pathogenesis remain largelyunknown and await investigation.1.4.3 Glutamine contributes to host cellsurvivalWhile little is known about glutamine metabolismin bacteria and how itcontributes to pathogenesis, the effect of glutamineon signaling processes in eukaryoticcells has been characterized. Intracellular glutaminemodulates heat shock protein (Hsp)expression and reactive oxygen species (ROS) levelsin cells such as enterocytes, orintestinal epithelilal cells, and promoteslymphocyte cell proliferation as well assuppressing cell death (Phanvijhitsiri etal., 2006; Wischmeyer et al., 2003; Changet al.,2002; Chang et aL, 1999). Moreover, apoptosisin intestinal epithelial cells is preventedby the presence of intracellular L-glutamine (Larsonet al., 2007; Nakamura and Hagen,2002). Specifically, a study showed that L-glutamineis able to inhibit H pylon-inducedapoptosis in enterocytes, suggesting an anti-cytotoxicactivity of glutamine in thestomach (Nakamura and Hagen, 2002).1.5 ATP BINDING CASSETTE(ABC) TRANSPORTERS1.5.1 AA-ABC Transporter StructuresTo effectively utilize AAs for energyproduction, AAs must be activelytransported from the external environmentinto the C.jejuni cytosol acrossthecytoplasmic membrane barrier.This process is predominantly mediatedby ABC(ATPBinding Cassette) transporters,a superfamily composed of proteinsfound in bothprokaryotes and eukaryotes. ABC uptaketransporters are typically comprisedof threefunctional domains:a permease, an ATPase, and a substrate-bindingprotein (Fig. 1.1).7While the permease and ATPasecomprise the core of an ABC transporterand, in bacteria,are typically encodedby genes in the same operon, the substrate-binding domainisusually encoded by a gene encoded elsewhereon the genome (Muller et al., 2005;Hekstra and Tommassen, 1993). In bacteria,a functional transporter usually contains ahomodimer of a single transmembrane polypeptideor a heterodimer of two distinctpolypeptides (Saurin et al., 1999). Eachof the hydrophobic permease componentstypically forms 6 putative alpha-helical integraltransmembrane polypeptide segments,which shuttle the substrate across thebacterial membrane. The hydrophilic ATPasecomponent is typically formed by two monomersand resides in either the periplasmorthe cytoplasm. The catalytic pocket for ATPhydrolysis to provide energy fortransfer ofthe substrate into the bacterial cytoplasmis only produced by cooperationbetween thetwo monomers. In contrast to thepermease domain, the ATPasesshow a high degree ofsequence similarity and identity acrossthe ABC family, implying a conservedstructureand function for these domains(Garmory and Titball, 2004;Jones and George, 1999).Specifically, these proteins contain threehighly conserved motifs: (1)Walker A, aglycine-rich loop; (2) Walker B,a hydrophobic domain, at the ATPbinding site, and (3)C-motif, (LSGGWW/RJKWR), also knownas the peptide linker, whichis found in allABC transporters. This motif is located immediatelyN-terminal to the WalkerB motifand is the site of mutations which severelyimpair function in manyABC transporters.It is also interesting to note thatthe peripheral substrate bindingprotein and ATPbinding protein can be promiscuousand therefore free to interact withother transportersystems. Previous studies haveshown that in the absence of abinding protein, certaintransporter systems may couple withother periplasmic-binding proteinsor ATP bindingproteins to achieve high efficiency uptake(Forward et al., 1997;Schlosser et al., 1997;8Wilken et al., 1996; Hekstra and Tommassen, 1993), suggesting that these ABCtransporter components are not always confined ex1usive1y to a single system.9PeriplasmCytoplasmFigure 1.1. Schematic diagram of a bacterial ATP Binding Casette (ABC) transporter. AnABC transporter is composed of two alpha-helical transmembrane proteins and an ATP-bindingprotein, which are encoded by genes in the same operon. A periplasmic substrate binding proteinencoded by a gene outside of this operon is often coupled to this systemPeriplasmicbinding proteinIntegral‘—transmembraneATP bindingdomains101.5.2 ABC Transporter FunctionsABC transporters play an essential role in most bacteriadue to their ability torapidly mediate nutrient uptake for energy and survival (Higgins,1992). Through ATPdependent active transport, nutrients such as sugars or aminoacids can be accumulatedwithin bacteria against a concentration gradient (Dippel andBoos, 2005). Although theexact mechanism remains unclear, it is proposed that once loadedwith a specificsubstrate, the substrate binding protein is able todock on the external domain of thepermease protein to initiate ATP binding to the ATPase. Thistriggers simultaneousconformational change between the permease and ATPase toopen the channel, and ATPhydrolysis to facilitate active transportof the substrate into the cytoplasm against theconcentration gradient. In addition to importing nutrientsfor consumption, ABCtransporters may also export substances from the bacterialcytoplasm (Nikaido, 2002).Previous studies have shown that ABCtransporters participate in virulence proteinsecretion (Holland et al., 2005) as wellas the efflux of hydrophobic compounds,some ofwhich exert antimicrobial activities(Quinn et al., 2007; Elkins and Beenken,2005).Bacterial ABC transporters have alsobeen proposed to regulate other processes,such asmodifying signal transduction pathways ( 2003). For instance, theoligopeptide permease of the Bacillus Subtilius ABCtransporter stimulates competencedevelopment and the initiation ofsporulation by importing signaling peptides.Therefore,ABC transporters have been shown toaffect bacterial virulence by several differentmeans.111.5.3 AA-ABC transportersserve as pathogenesis determinantsin C. jejuniIt is estimated that C. jejuni harbours30 complete ABC transporter systems, bothimporters and exporters (Kelly,2008). Although it is difficult to deducethe substrates foreach transporter system, it is predictedthat most of these systems areresponsible for AAand organic acid transport, sinceC. jejuni utilizes AAs as a primary carbonsource(Guccione et al., 2008). SeveralAA-ABC transporter system componentsalreadyidentified in C. jejuni includea glutamate/aspartate-binding protein(Peb 1) (LeonKempis Mdel et al., 2006), a cysteine-bindingprotein (CjaA) (Muller et al., 2005),and aserine transporter protein (SdaC) (Velayudhanet al., 2004).Peb 1, CjaA and SdaC all participatein C. jejuni virulence (Leon-KempisMdel etal., 2006; Muller et al., 2005; Velayudhanet al., 2004). Pebi is currently thebestcharacterized AA-ABC transportercomponent and has been suspectedto function as asurface antigen and adhesin. It containstwo predicted sequencesfor signal peptidases Iand II binding, and therefore can serveas both a surface exposed adhesinand aperiplasmic solute. Absenceof Pebi attenuates infectionboth in vitro and in vivo (Mulleret al., 2007; Leon-Kempis Mdel etal., 2006; Muller et al., 2005).Similarly, CjaA mayact as a surface antigen involvedin host cell adhesion (Muller etal., 2005). SdaCparticipates in serine uptakeand is also important for virulencein vivo (Velayudhan et al.,2004).1.5.4 AA-ABC transporters serveas pathogenesis determinantsin other pathogensABC transporters in other bacterial pathogensact as virulence determinants.Forinstance, the ATP bindingprotein (G1nQ) of a glutaminetransporter in Group BStreptococcus is involved inhost cell surface adhesion andin vivo virulence (Tamuraetal., 2002). A putative polypeptide ABCtransporter system inSalmonella, encodedby the12yefABEF operon, also serves as a virulence factor,since deletion of the ATPasecomponent inhibits the ability of these bacteria to proliferatewithin host cells (Eswarappaet al., 2008). Several ABC transporter componentsin Brucella are also important forvirulence both in vitro and in vivo (Castaneda-Roldanet al., 2006; Rosinha et al., 2002).1.6 PHYSIOLOGY AND STRESS RESPONSEC. jejuni needs to traverse unfavourable external environments andovercome variouschallenges posed during the transmission process to establishinfection and cause disease.Several stress response proteins enable C.jejuni to surviveand respond to external stimuli suchas oxidative, aerobic, osmotic, heat and acid stresses.1.6.1 Heat stressC. jejuni replicates within a temperature range of approximately32°C- 47°C(Garenaux et al., 2008; Park, 2002). However, its optimaltemperature is between 3 7°C-42°C, which spans the average body temperature of humans(37°C) and avian species(42°C). C. jejuni is often acquired through ingestionof undercooked poultry, where it haslikely survived both heating and cooling processes duringhandling. C. jejuni harboursseveral heat shock proteins (Hsps) which are often inducedas a result of a thermal stressresponse (Murphy et al., 2006). DnaJ, DnaK, GroEL,and C1pB are the major Hspsassociated with thermo-tolerance inC. jejuni. Expression of some Hsps can also increasein response to other stresses such as aerobic and oxidativeor osmotic stresses (Murphyetal., 2006). For instance, the endoproteaseHtrA and the acyltransferase HtrB areHsps ofC. jejuni induced during aerobic, osmotic and oxidativeshocks in addition to heat stress(Phongsisay et al., 2007; Brondsted et al., 2005).Other stress regulators such as RacRS,atwo-compoenent signal transduction regulatorsystem, and orthologues of HrcA and13HspR, have also been proposed to associate with thermo-regulation (Alter and Scherer,2006). More recently, polyphosphate (poly P) and the two-component system sensorkinase Cj 1226c, have also been identified to participate in hyper-osmotic stress resistance(Candon et al., 2007; Svensson and Gaynor, manuscript submitted).1.6.2 Aerobic stressAs a capnophilic and microaerophilic organism, C. jejuni is sensitive to highlevels of oxygen and has an optimal growth environment of 12% CO2 and 6%02(Murphy et al., 2006). Hence, it is generally sensitive to the level of oxygen in theatmosphere, which is approximately 21%. However, C. jejuni uses several strategies totolerate aerobic stress (Jones et aL, 1993). In addition to inducing heat shock proteinssuch as HtrA and HtrB as described above, C. jejuni utilizes the stringent response tocounter high02/low CO2 stress (Gaynor et al., 2005) and also produces proteins tofacilitate the removal of reactive oxygen species (ROS), which are generated as a resultof aerobic metabolism.1.6.3 Oxidative (reactive oxygen species) stressAs previously mentioned, C. jejuni encounters ROS stresses as a result of aerobicrespiration. However, C. jejuni is also challenged by high levels of oxidative stressinsidehosts and inside epithelial cells and macrophages, since ROS such as superoxide,hydrogen peroxide and other organic hydroperoxides are frequently produced byhostcells as a defense mechanism to cause bacterial cell injury (van Vliet et al., 2002; Dayetal., 2000).C.jejuni expresses several oxidative stress response enzymes to remove ROS.Superoxide dismutase (SodB), catalase (KatA) and alkyl hydroxide reductase(AhpC) are14the three major proteins identified in C.jejuni to inactivate ROS. The iron-containingSodB is proposed to provide the first line of defense during exposure of C. jejuni to airsince it has been shown to degrade superoxide anions and oxygen radicals such asdioxygens into hydrogen peroxide and oxygen (Purdy eta!., 1999; Pesci eta!., 1994;Purdy and Park, 1994). KatA is essential for hydrogen peroxide resistance as it convertsperoxide into water and oxygen (Day et a!., 2000; Mongkolsuk eta!., 1998). The iron-regulated AhpC is important specifically for resistance to toxic hydroperoxideintermediates such as cumene and tert-butyl hydroperoxide, but not inorganic peroxidessuch as hydrogen peroxide (Stead and Park, 2000; Baillon eta!., 1999).1.6.4 Osmotic stressSensitivity to hyper-osmotic stress often arises during transmission processes suchas food processing, desiccation and salt water survival. C. jejuni exhibits a lowertolerance to osmotic stress compared to other food-borne bacterial pathogens (Alter andScherer, 2006). Factors contributing to osmotic stress responses have not beeninvestigated in detail, and only three genes to date have been identified as beingassociated with hyper-osmotic stress resistance: htrB (Phongsisay et a!., 2007), ppkl(encodes polyphosphate kinase 1) (Candon et a!., 2007) and cj]226c (encodes a twocomponent system sensor kinase) (Svensson and Gaynor, manuscript submitted). Each ofthese genes is important for survival in hyper-osmotic conditions.151.7 HOST-PATHOGEN INTERACTIONS1.7.1 Host epithelial cell colonization, invasion, and intracellular survival.To establish an infection, C. jejuni must first colonize the intestinal epithelium.This involves traversing the mucus layer of the gastrointestinal tract and attaching tointestinal epithelial cells. C. jejuni harbours several surface adhesion factors that enableattachment to host cell surface receptors, leading to colonization. F1aA (flagellin), Peb 1(an ABC transporter periplasmic substrate binding protein component that also serves asa surface adhesin) (Pei et a!., 1998), CadF (Campylobacter adhesion to fibronectin)(Konkel eta!., 2005), and J1pA (a lipoprotein) (Jin eta!., 2001) interact with host cellsurface proteins during infection. Lipooligosaccharide (LOS), another maj or surfacemolecule found on the outer leaflet of the Cj outer membrane, has an endotoxic propertyand is involved in host cell adhesion (Fry et a!., 2000).Although C. jejuni is often considered an extracellular pathogen, it may invadehost cells via a caveolae and microtubule dependent but actin-independent processfollowing initial colonization (Watson and Galan, 2008; Oelschlaeger et a!., 1993). Bothtranscellular and paracellular translocation are evident during Cf infection (Montevilleand Konkel, 2002; Bras and Ketley, 1999; Oelschlaeger eta!., 1993; Konkel eta!., 1992),and the detailed mechanism of the internalization process remains an area of debate.Once C. jejuni invades host cells, it is able to survive within the cells. Although theintracellular fate of C. jejuni remains unclear, this bacterial pathogen utilizes mechanismsthat allow it to survive within host cells (Candon et a!., 2007; Mihaljevic et a!., 2007;Naikare eta!., 2006; Gaynor eta!., 2005; Day eta!., 2000; Oelschlaeger eta!., 1993;Kiehibauch et a!., 1985). While an early study suggested that C. jejuni survives in avacuole mononuclear phagocytes (Kiehlbauch et a!., 1985), a more recent study using16intestinal epithelial cells showed that C. jejuni sustains intracellular survival bypreventing fusion of this vacuole, subsequently named the Campylobacter-containingvacuole (CCV), with the lysosome (Watson and Galan, 2008).While C. jejuni can survive intracellularly within intestinal epithelial cells,survival within macrophages has continued to remain controversial. Using electronmicroscopy and colony-forming unit (CFU) assays, previous studies have shown thatfollowing phagocytosis, C. jejuni is capable of both short- and long-term intracellularsurvival within cultured J77A. 1 murine macrophages, 28SC human monocytes, andperitoneal macrophages from BALB/c mice (up to 1 0 CFU by day 6) (Hickey et a?.,2005; Day et a?., 2000; Kiehlbauch et al., 1985). However, other studies using C57BL/6murine bone marrow derived macrophages and macrophages derived from humanperipheral monocytes have made contradicting observations, suggesting C. jejuni israpidly killed in 24 hours following phagocytosis by these macrophages as it is beingdelivered to the lysosome (Watson and Galan, 2008; Wassenaar et a?., 1997; Myszewskiand Stern, 1991).C. jejuni interactions with dendritic cells (DC5) have not been investigated indetail. DCs are important in both the innate and adaptive immune responses to microbialpathogens. DCs play an important role during infection of enteric pathogens such asSalmonella enteric serovar Typhimurium, H pylon and Shigellaflexineri, where releaseof various cytokines is induced (Kranzer et a?., 2004; Edgeworth et al., 2002; Marriott etal., 1999; Medzhitov and Janeway, 1997). To date, only one study has described C. jejuniinteraction with DCs, where it was found that 99% of internalized C. jejuni were killedwithin 24 hr, and no cytotoxicity effect was induced (Hu et a?., 2006).171.7.2 ToxinCampylobacter produces several virulence factors that alter host cellular activities,one of which is a multi-subunit toxin termed cytolethal distending toxin (CDT). CDT isalso produced by several other bacterial pathogens such as Salmonella enterica serovarTyphi, Shigeila dystenteriae, E. coil, Actinobacilius actinomycetemcomitans, andenterohepatic Helicobacter spp. (Smith and Bayles, 2006). It is encoded by 3 highlyconserved genes, cdtA, cdtB, cdtC: CdtA and CdtC function as dimeric subunits forminga complex with CdtB, and CdtB functions as a DNase I like protein that triggers DNAdouble-strand breaks (Lara-Tejero and Galan, 2000). Using purified CDT protein, theCDT was shown to cause apoptosis and cell death in eukaryotic cells, preventdephosphorylation of the cyclin B1/cdc2 protein kinase complex, and elicit eukaryoticcell cycle arrest in the G2/M transition phase prior to mitosis (Lara-Tejero and Galan,2000).1.7.3 ApoptosisApoptosis, or programmed cell death, is characterized by chromatin condensationand DNA fragmentation. Since CDT is produced by C. jejuni, and C. jejuni causesapoptosis in host cells, it has been postulated that CDT is the primary factor responsiblefor inducing apoptosis in C. jejuni-infected cells (Ceelen et al., 2006; Smith and Bayles,2006). However, a recent study suggested that apoptosis is not dependent upon thepresence of CDT, since infection of T84 intestinal epithelial cells with a CDT mutant didnot yield a reduced cytotoxic effect compared to WT infection (Kalischuk et al., 2007).Additionally, it was proposed that C. jejuni may also cause necrotic cell death viaa CDT-independent pathway (Kalischuk et ai., 2007). Necrotic cells, unlike apoptotic cells,rapidly lose plasma membrane integrity after the death stimulus.18Although C. jejuni CDT has been shown to induce apoptosis in the humanmonocytic cell line THP-1 (Hickey and Guerry, 2005 IAI), another study showed that aproteinase K- and heat-stable component of C. jejuni is also capable of stimulatingapoptosis (Siegesmund et a!., 2004). Experiments by the same group further suggestedthat Cia (Campylobacter invasion antigens) proteins contribute to apoptosis in THP-lmacrophages (Siegesmund et al., 2004). Detailed mechanisms underlying apoptosis inboth epithelial cells and macrophages during C. jejuni infection remain to be investigated.1.8 HOST CELL SIGNALING1.8.1 MAPK signaling pathwayOne feature of C. jejuni pathogenesis is the stimulation of host cell signaltransduction events to trigger an inflammatory response in addition to apoptosis(Borrmann eta?., 2007; Ru eta?., 2006; Rickey eta?., 2005; MacCallum eta?., 2005;Watson and Galan, 2005; Siegesmund et a?., 2004). The MAPK (mitogen-activatedprotein kinase) family plays an important role in mediating signal transduction and isactivated by a wide range of environmental stimuli via different cell surface receptorssuch as toll-like receptors (TLR), G-protein coupled receptors (GPCR), integrin, receptortyrosine kinases (RTK), and calcium ion channels (Schorey and Cooper, 2003). There arethree maj or MAPK families in mammals, the extracellular signal-regulated kinase 1/2(ERK1/2), c-Jun NH2-terminal kinase (JNK) and the p38 kinases, all of which lead tosubsequent gene expression changes important for regulating diverse cellular activitiessuch as proliferation, differentiation and apoptosis in response to external stimuli (Shan eta?., 2007; Galindo eta?., 2004).191.8.2 MAPK dependent signaling activation in C. jejuni infected cellsC. jejuni infection in 1NT407 human epithelial cells as well as Caco-2 and T84human intestinal epithelial cells has been shown to trigger ERK 1/2, JNK and p38 kinasephosphorylation (Chen et a!., 2006; Hu et a!., 2006; MacCallum et a!., 2005; Watson andGalan, 2005). Furthermore, several lines of evidence from work on other organisms haveshown that pathogen-induced ERK activation in macrophages is important for apoptosisinduction (Fettucciari et a!., 2003; Tang eta!., 1998; Xia et a!., 1995), andlipopolysaccharide (LPS) has been identified as a virulence factor responsible for MAPKstimulation in monocytes and macrophages (Thomas eta!., 2006; MacCallum et al., 2005;Guha and Mackman, 2001; Ruckdeschel et a!., 1997).1.8.3 Inflammatory responseCampylobacter infection often stimulates a series of immune responses in hostcells, resulting in inflammation. Activation of a proinflammatory cytokine, interleukin(IL)-8, is a hallmark of C. jejuni pathogenesis. IL-8 is important for the recruitment ofother immune cells including DCs, macrophages and neutrophils, which interact with C.jejuni to cause the host mucosal inflammatory response, which is critical for thegeneration of diarrhea. IL-8 production in host epithelial cells may be stimulated via aCDT-dependent or -independent process. While CDT contributes to IL-8 secretion in the1NT407 cell line (Hickey, 2000), it was found that both IL-8 and NF-KB (nuclearfactor- KB) activation are initiated through an ERK-dependent MAPK signaling pathwayin cultured T84 epithelial cells 8 hours post-infection (Watson and Galan, 2005). Otherpro-inflammatory cytokines such as IL-lcL/f3, IL-6, IL-8 and tumor necrosis factor alpha(TNF-a) are also induced in monocytic and epithelial cells (Young et a!., 2007; Hickey et20al., 2005; Siegesmund et al., 2004; Jones et al., 2003), and IL-1, IL-6, IL-8, IL-b, IL-12, interferon-gamma (IFN-’y) and TNF-cL, are also activated in infected DCs (Hu et al.,2006).1.9 OBJECTIVE AND HYPOTHESISIn a previous study, we found that transcription of the putative C. jejuni AA-ABCtransporter system encoded by cj0467-9 was induced during in vitro infection of 1NT407 cells(Gaynor et al., 2005), suggesting a role for these genes in pathogenesis. Our objective in thisstudy was to investigate the biological function(s) of this system. We hypothesize that if thecj0467-9 encoded AA-ABC transporter expression level is increased during cellular infection,then absence of this system will alter C. jejuni physiology and pathogenesis.21I2.1 BACTERIAL STRAINS, CELL LINES, MEDIA AND GROWTH CONDITIONSCampylobacterjejuni 81-176 wild type (WT) and mutants were cultured in MuellerHinton (MH) broth (Oxoid Ltd, Hampshire, England) or agar supplemented with 1 Ojig/ml ofvancomycin and 5 ig/ml of trimethoprim (MHTV) in a microaerobic and capnophilic (hereafterreferred to as microaerobic for simplicity) environment at 37°C. Microaerobic environmentswere generated using Oxoid CampyGen gas packs in an enclosed container or a tn-gas incubatorwith 6%02 and 12% CO2.C. jejuni 81-176 mutants zXpaqP (cj0467) and zXpaqQ (cj0469) werecultured in the same condition with the addition of 50 ig/m1 of kanamycin (kan). For overnightbroth cultures, freshly growing plates of bacteria were inoculated into MH broth to a startingoptical density at 600 nm (0D600)of approximately 0.004. The flasks were placed in an enclosedjar under microaerobic conditions and grown at 37°C shaking at 200 r.p.m. All E. coli DH5astrains were grown in Luria Bertani (LB) agar or broth at 37°C. RAW264.7 and 1NT407 cellswere cultured at 37°C in humidified air with 5% CO2.RAW264.7 cells were maintained inDulbecco modified Eagle Media (DMEM) (Gibco, Grand Island, NY) supplemented with 10%fetal bovine serum (FBS) (Gibco) while 1NT407 were maintained in Minimal Essential Media(MEM) (Gibco) supplemented with 10% FBS and Caco-2 cells were maintained in DMEM with1% non-essential amino acid (NEAA) and 10% FBS.2.2 CONSTRUCTION OF CAMPYLOBACTER JEJUNI 81-176 APAQP AND APAQQ TARGETEDDELETION MUTANTSTo construct the ApaqP (cj0467) and ApaqQ (cj0469) mutant strains in C. jejuni 81-176,the target genes were amplified by PCR from C. jejuni chromosomal DNA preparedusing aDNA extraction kit (Promega, Nepean, CA), with primers paqPFl (5’-TCTAGAGAAGATGGAGAAATTTTG-3’) and paqPRl (5’-23TCTAGAACACCACAAAAAGCCAT-3’),as well as paqQFl (5’-TCTAGATCCTTGCAGAGTATTC-3’) and paqQRl(5’-TCTAGATACCAACTGAGCTAAACC-3’), yielding 1.3 kb and 1.2kb fragments, respectively,with XbaI sites at the flanking regions. ThePCR products were purified (Qiagen, Mississauga,ON) and cloned into the pGEM®-T vector (Promega); theseconstructs were designated pGEMpaqF and pGEM-paqQ. Restriction digestion was performedusing the enzymes HindIII forpaqP and MscI with BamHI for paqQ, to remove approximately 400bp from paqP and 290 bpfrom paqQ. A non-polar aphA-3 cassette encoding kanamycin resistance(KanR)was digestedfrom the pUC18K2 plasmid (Menard et al., 1993) usingthe enzymes HindIII or MscI andBamHI, and ligated to the digested pGEM-paqP and pGEM-paqQ vectors,creating suicidevectors carrying the ApaqP: .aphA-3 and ApaqQ: .aphA-3deletion constructs. Followingselection and amplification in E. coli DH5ct, the mutagenicplasmids were purified from DH5ausing a Qiagen Midi-prep kit and transformedinto C. jejuni 81-176 WT by naturaltransformation or by electroporation as previously described (Candonet al., 2007). As pGEM®-Tis a suicide vector in C. jejuni, colonies recovered from MHTV+ kanplates should representstable chromosomal integrants resulting from double cross-overhomologous recombination.Genomic DNA from several mutant clones was prepared using theWizard Genomic DNA kit(Promega). The resulting mutants were designated ApaqPand ApaqQ. Insertional inactivation ofthe paqP and paqQ genes viaKanRcassette insertion was verified by PCR and sequencinganalysis as well as Southern blot analysis.2.3 COMPLEMENTATION OF APAQP DELETIONMUTANTGeneration of a re-constituted WT strain of C. jejuni, designatedApaqPc, was achievedby natural transformation of the t\paqP mutant with thepRRC-paqF plasmid, carrying a24chloramphenicol resistance gene. To create the pRRC-paqP construct,the paqP fragment wasobtained from pGEM-paqP by XbaI digestion. This fragment was subsequentlyinserted into thepRRC vector carrying a chioramphenicol resistant cassette, resulting inthe pRRC-paqPconstruct (paqF: . CAT). The pRRC delivery vector contains the C. jejuni1 6S and 28S rRNAgenes (Karlyshev and Wren, 2005) and provides a means to express genes,driven by the CATpromoter, at a heterologous chromosomal location in C. jejuni. This vector was deliveredto thezXpaqP train via natural transformation, as previously described (Candonet al., 2007). Coloniescarrying the paqP: CAT fragment were tested for sensitivity to chloramphenicolby plating onMR agar plates containing 20 ig/m1 of chloramphenicol and50 of kanamycin.Sequencing and PCR amplification using the paqP primer set confirmedthe selected colonies tobe reconstituted WT strains carrying the recombinant paqF gene.2.4 RNA EXTRACTION AND REVERSE TRANSCRIPT- PCR (RT-PCR)ANALYSISC. jejuni RNA isolation was performed as previously described (Gaynor et al., 2005).Reverse transcription of the purified RNA was performed using SuperScript II Mix and RandomPrimer (Invitrogen, Burlington, ON) followed by purification usingthe Qiagen PCR purificationkit. The purified cDNA products were PCR amplified for nssR using primersnssRFl(5’ AGAACTTTTATCTAGTGTAGG-3’) and nssRR 1 (5’ -CGTCCTTAAATCTAATGC-3’),and tuf using primers tufFi (5’- GCGTGGTATTACTATTGCTAC -3’)and tufRi. (5’-TCGAAGTCAGTGTGTGGAG-3’). RNA was confirmed as DNA-freeby RT-PCR.2.5 SOUTHERN BLOT HYBRIDIZATIONGenomic DNA was isolated using a DNA isolation kitas described above. For eachSouthern blot, 100 ng of DNA from each strain was digestedby EcoRV, which cuts in themiddle of theKanRcassette but does not cut either paqP or paqQ. DNA was separatedon a250.75% agarose gel and directly blottedonto PVDF membranes using 0.25 N NaOH and 0.75 MNaCl as the transfer solution accordingto the manufacturer’s instruction in the DIG High PrimeDNA Labeling and Detection StarterKit II (Roche Applied Science, Mannheim, Germany). Allprobes were non-radioactively labeled using this kit. ThepaqP fragment was obtained bydigesting the pGEM-paqP plasmid withXbaI while the paqQ fragment was obtained bydigesting the pGEM-paqQ plasmid with XbaI andMscI. The resulting fragments were denaturedby heating at 99°C for 10 mm and placed on ice. Hexanucleotides,DIG and Kienow were addedto the denatured fragment and incubated overnight at37°C as described in the DIG-kit. The blotwas hybridized with these probes and visualized usingenhanced chemiluminescence (ECL)(Perkin Elmer, Waltham, MA). Insertion of theKanRcassette into paqP and paqQ should resultin the generation of two smaller probe-reactivefragments (due to EcoRV digestion of theKanRcassette) rather than the larger WT-sized fragment. In thecase of /paqQ, only one smallerfragment is readily visible at the exposure shown.2.6 AMINo ACID TRANSPORT ASSAYSC. jejuni cells were grown for 15 hr to mid-log growthphase in 15 ml of MH broth.Cultures were harvested by centrifugation,washed twice in M9 minimal media(90.2 mMNa2HPO4,22.0 mM KH2PO4,8.56 mM NaCl, 18.7mM NH4C1, 2 mM MgSO4,22.2 mMglucose,10 iM CaC12,pH7.4), and re-suspended in the samemedium, to an approximate 0D600 of1.0.The re-suspension was kept on ice for no longer than4 hr. An aliquot of 150 ilof the cellsuspension was added to 1.5 ml of M9 minimal mediumcontaining 0.5% (v/v) lactic acid(Sigma-Aldrich, Oakville, ON), and was allowedto equilibrate by incubating at 37°Cfor 3 mm.The assay was then initiated by the addition of 5 Mof the[‘4C]-labelled amino acid (6.9-9.36GBq mmoF’) (Perkin Elmer). Samples (0.1 ml) werewithdrawn at the indicated time points,26collected by vacuum filtration through 0.22 jim membrane filters (Millipore Corp., Billerica,MA), and washed twice with M9 minimal media. Sample filters were then immersed in Filter-Count scintillation cocktail (Fisher Scientific, Ontario, Canada) and counted in a BeckmanCoulterTMscintillation counter. Transporter activity was expressed as nMoles of amino acidtransported, mg dry cells, min12.7 CELL INFECTION ASSAY FOR COLONIZATION, INVASION, INTRACELLULAR SURVIVALAND GENTAMICIN MIC DETERMINATION[NT407 cells (a human intestinal epithelial cell line), RAW264.7 cells (a murinemacrophage cell line) or Caco-2 cells (human intestinal epithelial cell lines) were seeded into 24well tissue culture plates at semi-confluency(.-.5x105cells/mi) and allowed to grow forapproximately 20 hr prior to infection. Mid-log phase WT and C. jejuni mutants from overnightshaking cultures were added to pre-warmed MEM or DMEM, which was used to infect the cellsat an MOP-- 200 for 3 hr. The cells were then washed 3 times with PBS to remove any unboundbacteria. To assay adhesion, 1 ml of 1% Triton X-100 in PBS was added to some of the wellsfor5 mm to disrupt the cells; the samples were plated on MH agar, and grown for 48 hr at 37°C in amicroaerobic condition. To assay invasion, fresh media supplemented with 10% FBS and 150jig/ml of gentamicin were added to the cells and incubated at 37°C for 2 hr before washing andtreating with Triton as described. Next, to assay short-term intracellular survival, freshmediasupplemented with 10% FBS containing 10 jig /ml gentamicin for an additional4 hr beforewashing and treating with Triton X-l00 (the 9 hr post-infection timepoint). Finally, to assaylong-term intracellular survival, cells were left in 10 jig/mI gentamicinfor an additional 15 hr (or24 hr after initial infection) before being subjected to washesand Triton X-1 00 treatment.Gentamicin susceptibility of each strain was tested by determining the minimum inhibitoryconcentration (MIC) using an E-test® strip (AB Biodisc).272.8 IN VIVO COLONIZATION USING A MOUSE MODELThe in vivo study was performed by Rhonda Hobbs from our collaborator, Dr. StuartThompson’s laboratory in the Medical College of Georgia. BALB/cByJ mice from Jackson.Laboratories (Bar Harbor, ME) were housed at the animal care centre at the Medical College ofGeorgia, with seven mice per experimental group. Each mouse was infected with 5 x i09 CFUWT or ApaqP C. jejuni via oral gavage as previously described (Pajaniappan et al., 2008).C.jejuni shed in fecal pellets from each mouse at 7, 14,19,28 and 35 days post-infection, werehomogenized and enumerated on MR agar containing 5% (v/v) sheep’s blood and 20ig/mlcefoperazone, 10 jig/mi vancomycin and 2 jig/mI amphotericin B (CVA). The level ofdetectionwas 1 x102CFU/g fecal pellet. All animal treatments were carried out in accordance with NIHguidelines for the care and use of laboratory animals, using procedures approvedby the MedicalCollege of Georgia Institutional Care and Use Committee.2.9 OXIDATIVE, AEROTOLERANCE, HEAT STRESS AND OSMOTIC STRESSSURVIVAL ASSAYSTo assay oxidative stress, C. jejuni from an overnight culture in MR broth wasinoculatedinto fresh MR broth to an 0D600 0.6; 1 ml of this culture was subsequentlyadded to each wellof a 24-well plate. Tert-butyl hydroperoxide, also known as t-BOOH (SigmaAldrich) andhydrogen peroxide, H20,or paraquat (Sigma Aldrich) were made in MR brothat variousconcentrations. 1 ml of the oxidative agent was added to C. jejuni cultures andincubated for 30mm at 37°C in microaerobic conditions before harvesting for CFU enumeration.Aerotolerancewas examined by diluting C. jejuni grown to mid-log phasein MR broth to an initial 0D6000.004 in fresh MR broth and incubated in shaking culture at 200 r.p.m and37°C in atmosphericconditions. To assess limited CO2 stress and heat stress, mid-log phase bacteriawere seriallydiluted, spotted onto MH agar plates, and incubated at 37°C ina 5% CO2 incubator, or 45°C in28microaerobic conditions. Osmotic stress was examined by spotting the dilutions onto MH agarplates supplemented with 0.17 M NaC1 and incubated at 37°C ina microaerobic environment.2.10 CELLULAR APOPTOSIS AND VIABILITY DETECTIONAnnexin-Vstaining. RAW264.7 cells were plated at semi-confluency on cover-slipsin24-well plates and cultured for 24 hr before infecting with C. jejuni followed by gentamicintreatment as described above. At the end of 9 hr infection (with gentamicin treatmentasdescribed previously in intracellular survival assay), cells were washed 3 times withPBS andstained with Annexin-V fluorescein and propidium iodide according to manufacturer’sinstructions (Roche Applied Sciences). Annexin-V labeling was visualized usinga Nikon eclipseTE2000 microscope (Nikon Instruments Inc., Mississauga, ON) fitted with appropriatefilter setsfor detecting fluorescence. The total cell number was approximatedby counting cells at 6different fields under DIC (Differential Interference Contrast). Percentage of celldeath wascalculated by dividing the number of annexin-stained cells by total numberof cells.DI4PI staining and immunoflorescence. Following infection, RAW264.7 cellswerewashed and fixed as described by Guttman et. al. (Guttman et al., 2007a). Cellularnuclei werelabeled with 4’,6-Diamidino-2-phenylindole (DAPI) andmounted using Vectashield (VectorLabs, Burlington, ON). Relative cell viability was fluorescently quantifiedby counting DAPIstained cell nuclei in multiple randomly selected fields in a similar manner as previouslydescribed by Barnes et. al. (Barnes et al., 2007). Cells exhibit fragmented and condensednucleior have lifted off the plate are considered non-viable therefore were notenumerated for viability.2.11 CELL LYSATE PREPARATION AND WESTERN BLOTTINGRAW264.7 cells were grown on 150 mm tissue culturedishes and infected at an MOl200 for 9 hr. Cells were washed 3 times with PBS containing1mM CaC12and 1mM MgC1229followed by treatment with RIPA lysis buffer (150 mM NaC1, 50 mM Tris pH 7.4, 5 mM EDTA,1% Nonidet P-40, 1% deoxycholic acid, 1% SDS) for 10 mm on ice. Western blotting wasperformed according to Guttman et. al. (Guttman et al., 2007b). Briefly, equal amounts of totalproteins were loaded and separated on 10% SDS-polyacrylamide gels and transferred to PVDFmembranes (Bio-Rad Laboratories, Mississauga, ON). Membranes were blocked with 4% skimmilk and washed in Tris-buffered saline with 0.1% Tween-20 (TBST) 3 times for 5 mm each.Primary mouse anti-phospho-ERK1/2 antibody (Cell Signaling Technology, Beverly, MA) wasused at a dilution of 1:2000 and rabbit anti-ERK antibody (Santa Cruz Biotechnology, SantaCruz, CA) was used at a 1:2000 dilution (0.1g/m1). Rabbit anti-phospho-JNK (Thr183/Tyr185)antibody (Cell Signaling Technology) and rabbit anti-INK1 antibody (Santa Cruz Biotechnology)were used at a 1:1000 dilution. After washing, a horseradish peroxidase (HRP)-conjugatedsecondary antibody (Cell Signaling Technology) was used at 1:5000 for labeling. Signals weredetected by ECL (Perkin Elmer).2.12 ADDITIONAL PHENOTYPIC ASSAYSMicroaerobic growth in MH broth. C. jejuni were grown microaerobically at 37°C in MR brothto mid-log phase overnight. Bacteria were diluted to an 0D600 of 0.0 land placed undermicroaerobic conditions at 37°C, shaking at 200 r.p.m.; serial dilutions of cultures were madeand colony forming units (CFUs) were measured over time by plating on MR agar plates.Minimal media survivaL Strains were grown as described above and diluted to an 0D600 of 0.02.Cultures were in MEM at 37°C shaking at 200 r.p.m. under microaerobic condition and CFUswere measured over time by plating serially diluted cultures onto MH agar plates.- 30Anaerobic survivaL Log phase bacteria grown as described above were diluted to an 0D600 of0.01. Cultures were placed under anaerobic conditions using an Anaero-GasPak (Oxoid), at 37°C,shaking at 200 r.p.m. CFUs were measured as described.Motility. Log phase C. jejuni were diluted to 0D600 0.02 and 1il stabbed of each strain intoMH agar plates that contained 0.4% agar. Migration of the cells from the point of inoculationwas analyzed following 24 hr of incubation at 37°C under microaerobic conditions.Low iron conditions. Log phase bacteria were diluted to an 0D600 of 0.1, and 100 jil of eachstrain was evenly spread on a MH agar plate. Disks containing desferal (Fe3chelator) anddipyridyl (Fe2chelator) at 40 mM were placed on the cell lawns and incubated at 37°C for 24 hrbefore measuring the zone of inhibition.Antimicrobial susceptibility test. Log phase bacteria were diluted to 0D600 of 0.1, and 100 jiL ofeach strain was evenly spread on a MH agar plate. Disks containing erythromycin, nalidixic acidand amikacin were placed on the cell lawns and incubated at 37°C for 24 hr before measuring thezone of inhibition.Heat Shock. C. jejuni from an overnight culture was inoculated into MR media to an 0D600 of0.2 and placed at 55°C for 5 mm or 15 mm before cooling on ice. Serial dilutions of cultureswere made and spotted onto agar plates. CFUs were measured followed by 48 hr of incubationunder 37°C microaerobic conditions.pH sensitivity. C. jejuni from an overnight culture was diluted to 0D600 of 0.0005 and 0.002, andserial dilutions were inoculated onto MH agar plates at different pHs: pH 4.0,5.0, 6.0, 7.0, 8.0and 9.0. CFUs were measured after 48 hours of incubation under 37°C microaerobic conditions.Nutrient Deprivation assay. Overnight C. jejuni cultures were resuspended to 0D600 of 0.015 in1 Oml MEM supplemented with 1% non-essential amino acids and incubated in 37°C shaking31cultures in the absence or presence of 40 mM L-glutamine.Bacterial growth was assessed atvarious time intervals by optical density at 600 nm analyses and CFU enumeration.Glutamine rescue assay. RAW264.7 and 1NT407 cells were grown to semi-confluency oncover-slips in 24 well plates as described above. Log-phase C. jejuni were diluted to OD600 0.02in DMEM or MEM with 10% FBS in the absence or presence of 20mM L-glutamine. The cellswere infected with C. jejuni (2 ml per well) for 9 hours. At the end of the infection, cells werewashed with three times with PBS, fixed with 3% paraformaldehyde for 15 minutes at roomtemperature, washed once with PBS, and permealized with 0.2% Triton X-l00 for 5 minutes atroom temperature. Finally, the cells were washed and stained with DAPI for visualization ofcellular nuclei on the microscope.Gamma-glutamyl transferase (GGT) assay. The GGT assay protocol was adapted from (Barnes et al., 2007). Briefly, log-phase C. jejuni from overnight cultures were centrifugedat lO,000xg for 5 mm. Pellets were resuspended in 3 ml of 1M Tris-EDTA buffer (pH 8) andcells were disrupted using an ultra-sonicator (Ultrasonic Processor SL, Misonix Incorporated)atsetting 3 for 10 times at 10 sec each. Lysed bacteria were centrifuged at 13,000 r.p.m, for 5 mmat 4°C and supernatants were collected for a Bradford assay to determine protein concentration.GGT reagent (Pointe Scientific, Inc. Canton, MI) was reconstituted in 10 ml dH2O accordingtothe instruction protocol. Bacterial lysates were diluted by 1/10 in dH2O and 10jilwas added to190 p.! pre-warmed GGT reagent in a 96 well microtire plate. The assay was incubatedat 37°Cand absorbance was measured at 405 nm at various time points.Protein secretion profile. 100 ml of log-phase C. jejuni from overnight culturewas centrifugedat 10,000 x g for 5 mm. The supernatant was filtered using 0.22p.m filter, and a 1/10 volume of100% trichloracetic acid (TCA) (w/v) (Sigma Aldrich) was addedto the supernatant forovernight precipitation at 4°C. The suspension was centrifuged at9800xg for 30 mm and the32pellets were re-suspended in 200 pi of lx SDS-PAGEsample buffer and subjected to 16.5%SDS-PAGE. Briefly, the gel was fixed for 1 hour in fixing solution (40: 5:55 solution of 95%ethanol/acetic acid! distilled water (v/v)), treated with oxidizing solution (0.7% periodic acid infixing solution (v/v)) for 5 mill, followed by three washes with dH2O, 10 mm each. Next, the gelwas stained with fresh silver staining reagent (20 mM NaOH, 0.67% AgNO3,1.33% NH4OH indH2O (v/v)) with gentle shaking. Finally, the gel was washed with distilled water 3 times for 10minutes, and developing solution (Bio-Rad Silver Stain Developer) was added to the gel for 10minutes before adding the stop solution (5% acetic acid and 10% ammonium pesulfate in dH2O).333.1 OverviewAA-ABC transporter systems in C. jejuni and other pathogenic organisms act asvirulence factors during cellular infections (Leon-Kempis Mdcl et al., 2006; Muller et al., 2005;Tamura et al., 2002). In a previous study, we observed increased transcription of cj0467-9,which encodes a putative AA-ABC transporter, during in vitro infection of INT407 cells,suggesting a role for these genes in pathogenesis. This led us to explore the role of the Ci 0467-9AA-ABC transporter system and its connection with host cells during the infection process. Asboth the integral membrane protein, or permease, and ATP binding protein/ATPase componentsare frequently required to produce a functional AA-ABC transporter (Nikaido, 2002), wegenerated non-polar mutations in both the Cj0467 (PaqP) permease and the Cj0469 (PaqQ)ATPase, using a highly invasive C. jejuni 81-176 strain, to ensure a comprehensive analysis ofthe system’s functions (rationale for the “Paq” designation is described below). Both mutantswere tested for their tolerance in environmental stresses which can be often encountered in vitroor in vivo, amino acid transport to explore the system’s biochemical function, and its roles in thepathogen-host cell interaction.353.2 cj0467-9 encodes a putative amino acid (AA)ABC transporterThe putative AA-ABC transporter system encoded by cj0467-9 is highly conserved andoccurs in the same genomic context among C. jejuni strains,as described in CampyDB(!campydb). For simplicity, we have referredto the cj0467-9 genenumbers as annotated for strain NCTC 11168, although our studies wereperformed using theinvasive strain 81-176. Cj0467-9 is annotated as Cjj0492-4 in the 81-176strain (campyDB).cj0467 and cj0468 are predicted to encode integral membrane proteins, withCj0467 exhibiting33% identity to the E. coil glutamine ABC transporter permease GlnP, and Cj0468 exhibiting26% identity to the E. coil G1tJ permease (campyDB). cj0469 is predicted to encode an ATPbinding protein with 55% identity to the Bacillus subtiiis glutamineABC transporter ATPbinding protein G1nQ (campyDB). National Centre for BiotechnologyInformation BLAST andArchael and Bacterial ABC transporter database searches also showedthat Cj0467 and Cj0468are similar to GlnP (approximately 43% identity), whileCj0469 is highly similar to GInQ(approximately 60% identity) in Streptococcus pneumoniae, Pseudomonasspp., andHelicobacter spp. Cj0467 and Cj0469 also exhibit homology to otherlikely C. jejuni AA-ABCtransport components, including two uncharacterized proteins putativelyannotated as G1nP(Cj0940c) and GInQ (Cj0902) based on slightly higher initial BLASTsearch homologies(Cj0940c: 37.5% identity to E. coli GlnP; Cj0902: 56.2% identityto Bacillus stearothermophilusG1nQ). To avoid confusion with these previously annotatedgenes, and as our data (describedbelow) indicate that Cj0467 and Cj0469 participate not onlyin glutamine transport but also inother important biological processes, we have designatedthem PaqP and PaqQ, respectively[pathogenesis-associated glutamine(q) ABC transporter permease(E)and ATPase(Q)].363.3 Construction of targeted, non-polar IpaqP and tpaqQ disruption strainsTo investigate the role of this AA-ABC transporter system in C. jejuni, paqF and paqQwere individually disrupted using a non-polar kanamycin resistance(KanR)cassette, aphA-3(Figure 3.1 A). PCR, sequencing, and Southern blot analyses confirmed that the resultant mutants,hereafter referred to as /paqP and ApaqQ, harboured disruptions in the appropriate genes(Figure 3.1 B, and data not shown). To verify that Kan’ insertion into both paqP and paqQ wasnon-polar, RT-PCR was used to confirm that nssR and tufwere transcribed in both mutant strains(Figure 3. 1C), and that paqQ was transcribed in the ApaqP strain (data not shown).37A BFigure 3.1. Generation of non-polar, single insert ApaqP and zXpaqQ disruptionstrains. (A)Genomic organization of the cj0467, cj0468 and cj0469 genes, encoding aputative AA-ABCtransporter system. The aphA-3 cassette encoding kanamycinresistance was used to createinsertion-deletions in cj0467 (paqF) and cj0469 (paqQ) genes bydouble crossover homologousrecombination. The resultant mutant strains are designated ApaqPand ApaqQ. (B) Southernblots of EcoRV-digested genomic DNA using paqP and paqQprobes confirm that the targetgene in each mutant was disrupted. (C) Reverse transcriptionPCR (RT-PCR) was performed toassay transcription of genes upstream of ApaqF (nssR) anddownstream of zXpaqQ (tuf) in thedeletion strains.paqP paqQCL7-+aphA-3 aphA-3CPCRStrainRT1.5kb10kb0,5 kbnssRWT ApaqP+- + -tufWT ApaqP t.paqQ+- +- +-— —ProbepaqP paqQStrain WT e.paqP WT z.paqQ8.0kb—6.0 kb——-—5.0 kb—4.0kb—3.0kb—383.4 The ApaqP and ApaqQ C.jejuni mutantsare defective for L-glutamine uptake andmoderately defective for uptake of other amino acidsTo investigate the functionality of PaqP and PaqQ, transport assays were performed usingdifferent radioactively labeled AAs. Based on strong homology to G1nP and GlnQ andbioinformatics analyses described above, together with a previous report suggesting that thisAA-ABC system may participate in cysteine transport (Muller et al., 2005), WT C. jejuni and thetwo mutants were grown to mid-log phase and assayed for rates of uptake of[‘4C]-L-glutamine,[‘4C]-L-glutamate,[14C] -L-aspartate and [‘4C]-L-cysteine. Both mutants exhibited a significantdefect for glutamine uptake: at 1 mi glutamine transport levels of both /spaqF (-4.97nmol/minlmg dry weight) and /xpaqQ (-4.66 nmol/minlmg dry weight) mutants were <50% thatof the WT strain (—4.12 nmol/minlmg dry weight) (Figure 3.2A). An intermediate defect wasobserved for glutamate uptake (Figure 3.2B), while much smaller decreases in cysteineandaspartate uptake were observed (Figure 3.2C, 3.2D). This suggests that this AA-ABC transportersystem has high affinity for glutamine but also appears to transport other amino acids to a lesserextent. The complemented ApaqPc strain partially restored glutamine uptake, asa two tailed ttest did not show a statistically significant difference between WT and z\ paqFc (Figure 3 .2E).39A-t•1>-UU)E0E•1)U)B— EpaqPApaqQglutamate0 1 2Time (minutes)3Cglutamine uptake at 1 minute**WT• zpaqP--. tpaqQcysteine5 6-.- WT• ApaqPtpaqQaspartateDTime (minutes)U)-oU)E0CU)I:1.tEI I11’U).,1.0’0.9’0.8____0U)E0,6_____E 0.4’_____WT tpaqP ApaqPc tpaqQ__Strainsj 4Time (minutes)Figure 3.2. zXpaqP and zspaqQ mutants exhibit reduced levels of glutamine and other aminoacid uptake relative to WT. C. jejuni 81-176 WT (square, solid line), ApaqP (circle, dashedline) and ApaqQ (triangle, dotted line) mutants were grown microaerobically, shaking in MHbroth, for 1 5hr to early log phase(—0.3 0D600/ml), then harvested and assayed for high affinitytransport of (A) [‘4C] L- glutamine, (B) [‘4C] L-glutamate, (C) [‘4C] L- cysteine, or (D) [‘4C] Laspartate at a final concentration of 5 1iM. Samples were taken either every 30 sec, 1 mm or 2mm. (E) ApaqP and ApaqQ both showed statistically significant differences in glutamine uptakecompared to WT, while the ApaqP complemented strain ApaqPc showed no statisticallysignificant difference in glutamine uptake compared to WT. Rate of uptakes were determinedfrom three separate biological replicate cultures, with duplicate samples harvested for each strainat each time point. The asterisk(*)represents statistical significance (p < 0.05) using a two-tailedt-test. Each experiment was performed in triplicates (n=3) and the result is a representation ofthree separate experiments.—a—o403.5 C. jejuni z\paqP and tpaqQ exhibit increased short-term intracellular survival inRAW264.7 macrophages, 1NT407 epithelial cells, and Caco-2 intestinal epithelial cellsIn a previous microarray analysis, we observed an increase in levels of cj0467-9 mRNAin C. jejuni during 1NT407 cell infections (Gaynor et al., 2005). Thus, we were interested inexploring whether this system might influence the interaction of C. jejuni with host cells.RAW264.7 murine macrophages as well as 1NT407 and Caco-2 human epithelial cells wereinfected with WT, zXpaqF, and /xpaqQ strains and assayed for adherence, invasion, andintracellular survival using a gentamicin protection assay. No significant differences in celladherence, invasion or long-term intracellular survival at 24 hr were observed for the mutantscompared to the WT strain (Figure 3.3). However, short-term intracellular survival (9 hr post-infection) of both ApaqP and ApaqQ mutants was up to 20-fold higher than WT in RAW264.7cells (Figure 3.3A) and approximately 10-fold higher than WT in 1NT407 cells (Figure 3.3B).Human intestinal Caco-2 epithelial cells were also tested, and similar results were observed(Figure 3.3 C). Both WT and mutant strains were equally susceptible to gentamicin by E-test stripanalyses (MIC 1.5 ig/ml), suggesting that the elevated short-term intracellular survival level ofthe mutants was nOt due to differences in gentamicin resistance. The level of intracellularsurvival at 9 hours post-infection was also examined for the ApaqPc complement strain, and wasfound to be similar to the ApaqP mutant (data not shown).41A RAW2641‘ 1.0x109D1.Oxl08tOxiQ71.Oxl1,0x106i.OxlG4B 1NT407ApaqPN1EpaqQi.0x109t0x101.0x1071.0x106C,1.0x105(V10x104C,1,OxlO’1.0x101.0x10C,a,CaI—a)C)1.OxlQ4*— IVT%%*- ApaqP-iSpaqQ1.0x1081.0x1080‘:1.0x106rx1.0x105tOxlO41.OxlO’tOxlO6DCi01.Ox10a,C,1.0x103CaiOxlO’e*Time (h)‘4a a*--Time (h)‘AEpaqQVJT ApaqPStrainsCWT ApaqQlspaqPStrainsCaco-2EpaqP-tpaqQI I7 8 9 10*1.0x1061.o1oi.0x106I 1 I I I I123456Time (h)*F,AI1pqP ApaqQStrains42Figure 3.3. ApaqP and z\paqQmutants exhibit a statistically significant increase in short-term intracellular survival in macrophages and epithelialcells 9 hr post-infection. WT(filled), ApaqP (striped) and ApaqQ (white) C. jejuniwere grown overnight in MR shaking brothto log phase. (A) A murine macrophage cell line, RAW264.7,and (B) a human epithelial cellline, 1NT407, and (C) Caco-2, a human intestinal epithelialcell line, were infected with bacteriaat an MOT 200. After 3 hr, the cells were washed withPBS and treated with 150 jig/mi ofgentamicin for an additional 2 hr to kill extracellularbacteria before washing the cells andadding fresh media with 10 jig/ml gentamicin; after an additional 4hr, intracellular bacteria wererecovered and plated for CFU enumeration. Allexperiments included triplicate infections foreach strain for each time point. The asterisk(*)represents statistical significance(p<0.05) usinga two-tailed t-test. This result is a representative experiment of three separaterepeats.433.6 No significant differences between the 4paqP mutant and WTC. jejuni colonizationwere observed in vivo using a mouse infection model.To evaluate whether the enhanced in vitro intracellular survival of the AA-ABC mutantsmight also reflect a survival difference in vivo, we infected BALB/cByJ mice with WT andz\paqP mutant strains according to a previously established mouse colonization model(Pajaniappan et al., 2008; Pei et al., 1998). Colonization was monitored for seven to 35 days.zXpaqP mutant C. jejuni colonization levels did not significantly differ from the WT at any giventime point (Figure 3.4). It should be noted that although WT C. jejuni colonized the mouseintestinal tract at a significant and high level up to 28 days, this and other tractable animalmodels do not consistently trigger inflammation and thus can only assay colonization, notvirulence (Chang and Miller, 2006; Hendrixson and DiRita, 2004).441 xl012Day7 Dayl4 Dayl9 Day28Day35.1 x1011_____•• A •.•cop••Co- 1x1010°“D•01X1009. :.00••.00lxlQ°80. .lxlO°700lxlO°6-. 0050iXi’.i•. 0>.0.1x10°-0.0 0• lxiO°3 • 0c 1 xlO°2limit of detectionlxi0011 xlO°° i I I IIWT LpaqP WT bpaqP WT EipaqP WTzpaqP WT ApaqPStrainsFigure 3.4. IpaqP is not defective for mouse colonizationin vivo. WT (solid circle) andtheApaqP mutant (open circle)C. jejuni colonized BALB/cByJ mice at a similar levelfrom day 7up to day 35 post-infection. The dashed line indicatesthe level of detection, which was1 x102CFU/g fecal pellet. Values on the x-axis representmice with no detectable colonization.Thisexperiment was performed by Rhonda Hobbsfrom the Thompson Lab in the MedicalCollege ofGeorgia.453.7 z\paqP and 4paqQC. jejuni mutants exhibit increased resistance to limited CO2 andaerobic conditions.To survive successfully throughout the pathogenesis cycle, C. jejuni must overcome amultitude of environmental stresses in both extracellular and intracellular environments(Mihaljevic et al., 2007). As a microaerophilic organism, C. jejuni requires elevated levels ofCO2 for normal growth and is sensitive to atmospheric levels of02. To examinewhethermutation of this AA-ABC transporter influenced the growth of C. jejuni under sub-optimal gasconditions, mid-log phase bacteria were serially diluted, spotted onto MH agar, and allowed togrow in a 5% (vs. ideal 12%) CO2 environment. Under these conditions, boththe z\paqP andApaqQ mutants grew better than the WT strain (Figure 3.5A). To specifically evaluate aerobicsensitivity, WT and mutant C. jejuni were grown microaerobically in broth culture, shifted tonormal aerobic atmospheric conditions, and assayed for survival after 4-6 hrsby CFUenumeration. Consistent with the CO2 growth observations, both mutants exhibited sustainedsurvival under aerobic conditions compared to the WT strain (Figure3.5B).46A 5% COWTApaqPApaqQB—E. DLLDC..)Cl)5c:0aerobic survivaJWTtpaqPzXpaqQFigure 3.5. zXpaqP and zXpaqQ are more resistantto aerobic stresses than WT.(A) Log phase cultures grown microaerobically were serially diluted,plated and incubated in a5% CO2 incubator at 37°C overnight. (B) Log phasecultures WT (solid), ApaqP (striped), andApaqQ (white), grown microaerobically, were shiftedto normal atmospheric oxygen conditionsfor 4 and 6 hr before being serially diluted, plated andincubated under microaerobic conditionsfor 2 days for CFU enumeration. Error bar representsassay replicates (n=3) of bacteria recoveredat each time point. This result is a representativeexperiment of three separate repeats.Theasterisk(*)indicates a statistically significant difference betweenthe mutants and WT (p<O.O5).serial lO4old dilutions* ** r4. 6Time (hour)473.8 i\paqP and tpaqQ C. jejuni mutants exhibitan increase in resistance to organichydroperoxide tBOOH, but not inorganic peroxides H20and paraquat.02- and C02-related stress responses are often closely associated with other oxidativestress responses in C. jejuni (Alter and Scherer, 2006;Baillon et al., 1999). We thus investigatedwhether loss of this AA-ABC transporter system in C. jejuni mightalso influence responses toother oxygen derivatives such as reactive oxygen species (ROS). WT,ApaqP and ApaqQ mutantC. jejuni were treated for 30 mm with hydrogen peroxide (H20),an inorganic peroxidefrequently generated by macrophages as a defense mechanism to eradicateintramacrophagepathogens (Alter and Scherer, 2006), or tert-butyl hydroperoxide (t-BOOH),an organic peroxidealso generated by host cells (Baillon et al., 1999). While neither mutantexhibited alteredsensitivity to H20 or paraquat (Figure 3.6A, 3.6B),both displayed increased resistance to tBOOH (Figure 3.6C). ApaqP in particular was found toexhibit higher t-BOOH resistance thanWT at a statistically significant level. Although ApaqQ did not exhibita statistically significantdifference relative to WT, experimental repeats consistentlyshowed a higher t-BOOH toleranceof zxpaqQ relative to WT. Together, our data indicate that disruptionof this AA-ABC systemimpacts aerobic and ROS sensitivity in C. jejuni.48ABParaq uat_E—.-tpaqP-.ApaqQConcentration (mM)10tert-butyl hydroperoxide—.-ApaqP“r.tpaqQ0.010.00 01)1 0.02 01)3 1104Concntratien (mFFigure 3.6. The ApaqP mutant exhibits a significant increase in resistance to tert-butylhydroperoxide (t-BOOH) but not to hydrogen peroxide (11202) or paraquat. WT (square),ApaqP (circle) and ApaqQ (triangle) were treated with (A) H20,(B) paraquat, or (C) t-BOOH atvarious concentrations and incubated microaerobically at 37°C for 30 minutes before beingharvested for CFU enumeration. All samples were taken in triplicates at each concentration. Theasterisk(*)represents statistical significance (p-value <0.05) using a two-tailed t-test.Hydrogen peroxide10ApaqP>LL.a00.10.010.0010.00010.0 0.5 1.0 1.5 2.0Concentration (mM)C493.9 Tolerance to heat and osmotic stressis differentially altered in .‘ipaqP and zipaqQmutantsTo examine if the ApaqP and /ipaqQ mutants exhibit other stress tolerance alterations,the mutants were also subjected to heat and osmotic stress assays.To test heat tolerance, mid-logphase bacteria were diluted to 0D600 0.2 in MH broth, and serial dilutionswere spotted ontoMR agar and grown under microaerobic conditions at 45°C, a mild heat stress conditionfor C.jejuni (Phongsisay et al., 2007; Brondsted et at., 2005; Park, 2002; Konkelet at., 1998). As acontrol, cultures were harvested and plated at 37°C (Figure 3.7A). While ApaqPexhibited anapproximate 10-fold increase in heat stress resistance comparedto WT, ApaqQ wasapproximately 10-fold more sensitive to heat stress than WT (Figure 3 .7B).Similar results wereobtained when the strains were subjected to growth under osmotic stressusing MR agar with0.17 M NaC1: the ApaqP mutant exhibited increased resistanceto osmotic stress compared toWT, while the ApaqQ mutant was more sensitive to NaC1 (Figure3.7C).50A serial 10-fold dilutions37°CB 45°CC 37°Cwith O.17M NaCIWTApaqPApaqQWTApaqPApaqQLpaqPcFigure 3.7. The zXpaqP mutant is more resistantto hçat and osmotic stress while thetxpaqQmutant is more sensitive than WT. Overnightbacterial cultures were serially dilutedfrom aninitial concentration of 0D600 O.2. (equivalent toapproximately 1x109 cfu!ml) andgrownmicroaerobically for 2 days at (A) 37°C and(B) 45°C on MR agar to assess the responseto heatstress, or at (C) 37°C on MR agar supplementedwith O.17M NaCl to assess the responsetoosmotic stress. This result is a representationof more than three separate repeats.513.10 Macrophages infected withC. jejuni txpaqP and ApaqQ exhibit reduced apoptosisan( ERK activation compared to macrophagesinfected with WT bacteria.The observations described above suggested that the enhanced intracellular survivalofthe zXpaqP and zXpaqQ mutants can at least partly be attributedto enhanced or altered bacterialstress tolerance. As host cell death results in detachment of the cells fromthe tissue culturesurface and a resultant loss of C. jejuni inside those cells from the intracellularsurvival assaycounts, we hypothesized that the observed increase in intracellular survivalof the mutants mightalso reflect increased host cell survival. To investigate this, we separatelymonitored levels ofhost cell apoptosis and cell viability using Annexin-V Fluor and DAPIstaining, followed byvisualization using fluorescent microscopy. Cells were also stainedwith anti-Campylobacterjejuni antibody to ensure infection (data not shown). Cell viability levelswere assessed bycounting the number of cell nuclei remainingon the cover-slip that were neither condensed norfragmented following infection. Consistent with previous findingsin THP- I human macrophages(Siegesmund et al., 2004), apoptosis was induced in WT C. jejuni-infectedRAW264.7 cells(Figure 3.8A). However, apoptosis was significantly reduced in zXpaqP-and z\paqQ-infectedcells compared to WT-infected cells (Figure 3.8A),by 66% and 55%, respectively (Figure 3.8B).The ratio of viable cells in WT-infected RAW264.7 cellsrelative to uninfected cells wasapproximately 30% less than that of the zXpaqF- andApaqQ- infected cells (Figure 3.8C).Finally, to explore if reduced levels of apoptosis observed inApaqP- and ApaqQinfected RAW264.7 macrophages correlated with alteredhost cell signal transduction pathways,we examined phosphorylation of severalMAP kinases known or suspected to be induced uponC.jejuni infection (Chen et al., 2006; MacCallum et al.,2005; Watson and Galan, 2005). Wefoundthat ERK phosphorylation levels in RAW264.7 macrophagesinfected with either ApaqP orApaqQ were severely diminished compared to WT-infectedcells at 9 hr post-infection (Figure3.8D). In contrast, iNK and p38 phosphorylation levelswere similar in macrophages infected52with WT and mutant strains, as were p-ERK levels at time points where the mutants did notexhibit invasion or intracellular survival differences from WT (Figure 3.8D, and data not shown).Levels of phospho-ERK INT407 or Caco-2 epithelial cells infected with either ApaqP or ApaqQwere also similar to those observed for WT-infected cells (Figure 3.9). These data indicatethatthis AA-ABC transporter system participates in ERK activation in C. jejuni-infectedmacrophages.53kDa605040605040kDaDIC Annexin-VB*3.0%V>,1.0%xaC0.0%CAUIWTbpaqPApaqQD*ImlApaqP ApaqQStrains*I*ICli,‘— l_LO.a,—.DøUI WTApaqPApaqQUI WT ApaqP ApaqQStrainsUI WTApaqPApaqQ_,__p-ERK1/2ERKF*p-JNK2/3[•*p-JNK140JNK54Figure 3.8. ApaqP and ApaqQ infectedcells showed decreased levels of apoptosis and ERKactivation compared to WT infected cells.(A) C. jejuni infected RAW264.7 cells were stainedwith Armexin-V fluorescein to visualize apoptoticcells at 9 hr post-infection with gentamcintreatment as mentioned in materialsand methods. Frames shown are one representative view ofseveral different frames. Scale bar=5 jiM. (B) Percentage of Annexin-V stained cells per totalnumber of cells at i OX magnification. Errorbar represents counts from several random fields. (C)C. jejuni infected RAW264.7 cells were stained with DAPI and cellularviability was assessed bycounting cellular nuclei per random field. Error barrepresents numbers of DAPI counts from amultiple random fields. (D) Phospho-ERK levels wereinduced at the 9 hr time point in WT butnot ApaqP and zpaqQ-infected RAW264.7 cells.ERK1/2 phosphorylation (42 and 44 kDa) wasdetected by Western blot using anti-phospho-p44/42MAPK (Thr202/Tyr204) antibody. Themembrane was stripped and reprobed with anti-ERKantibody to assess total ERK protein levelsas a control. JNK phosphorylations (46k Da and 54 kDa)were detected using Western blot usinganti-phospho-SAPKIJNK (Thri 83/Tyr 185) antibody. The membrane wasstripped and re-probedwith anti-JNK1 antibody to assess total iNK protein level as a control. Thisis one representativeexperiment of three separate repeats.55AB1NT407 Caco2UI Wt ApaqP ApaqQUI WT ApaqP tpaqQkDa kDa60—505040—p-ERK1/230—20—kDa kDa60—i50—_____40—RK40_3030—_____________20—_____________C1NT407UI WT ipaqP zpaqQkDa60—5040——p-JNK30—kDa40JNK30_Figure 3.9. ApaqP and zXpaqQ infectedCaco-2 and 1NT407 epithelial cells show similarpERK and p-JNK levels as WT infected cells. 1NT407and Caco-2 cells were infected with WTand mutant C. jejuni for 9 hr before being harvestedfor cell lysate preparation.563.11 A summary of phenotypes for which ApaqP and ApaqQ mutants werenotsinificantIy different from the WT strain.In order to understand the role of the AA-ABC system in bacterialstress tolerance,several additional phenotypic tests related to physiological stresseswere examined in the ApaqPand ApaqQ strains. These included bacterial growthunder microaerobic conditions, survival inminimal media or minimal media supplemented with serum, survivalunder anaerobic and lowiron conditions, as well as antimicrobial susceptibility, pH stresses,heat shock, motility andbioflim formation. To assess if the mutants exhibitedany difference in their ability to survive inminimal media +1- additional glutamine, glutamine was added to MEMand survival wasmonitored over time. In addition, to test if glutamine supplementationrescued host cell viabilityduring C. jejuni infection, a glutamine rescue assay was performed by supplementingthe culturemedium with excess glutamine during infection, and cells werestained with DAPI to assess hostcell viability. Finally, levels of protein secretion and GGT activity in WTand mutants were alsoexamined (Table 3.1)57Table 3.1. Phenotypes that were indistinguishablebetween 4paqP, ApaqQ mutants and WTC. jejuni8l-176 -PHENOTYPE COMMENTS FIGURE/TABLEMicroaerobic growth No statistical significant difference(MH broth) between WT and mutants Figure 3.1 OAMinimal media survival No statistical significant difference(MEM) between WT and mutants Figure 3.1 OBRich media survival(DMEM+ 10% heat-inactivatedNo statistical significant differenceFBS) between WT and mutants Figure 3.IOCAnaerobic growth curve (MR No statistical significant differencebroth) between WT and mutants Figure 3.1 1Low iron conditions (desferal No significant difference betweenand dipyridyl) WT and mutants Table 3.2No significant difference betweenAntimicrobial agents WT and mutants Table3.2Lethal at pH4, pH5, pH9. Optimalgrowth atpH6 andpH7.No statistical significant differencepH sensitivity test (MR agar) between WT and mutantsNot shown5 minute treatment showed identicalresults from untreated samples shown55°C Heat shock in Figure 3.7A Not shownNo difference between WT andMotility mutants Figure3.12No statistical significant differenceGOT assay between WT and mutantsFigure 3.13Supplementation of glutamine toRAW264.7 and 1NT407 cells duringWT C. jejuni infection did not showGlutamine rescue assay any difference in DAPI staining from(Infection in DMEM+10% FBS cells without glutamine-1+ 40mM glutamine) supplementation.Not shownProtein secretion No significant difference between(Culture supernatant) WT and mutantsFigure 3.14Nutrient deprivation assay No significant differencebetween(Survival in MEM-/+ 40mM WT and mutantsin optical density orglutamine) CFU measurement.Figure 3.1558Al.0x10°— WT80BTime(h)1.0c10’:.L0x101.0x101.0x10) 5 10 15 20 25Time (h)C1.0x101°- wr- .paqPApaqQDUTime (h)Figure 3.10. ApaqP and ApaqQ mutants survive similarly to WT C. jejuni in rich orminimal media. (A) Shaking culture in MH broth at a starting optical density of 0D600 0.006and (B) MEM at a starting optical density of OD600 0.02 (C) DMEM with 10% heat-inactivatedFBS at a starting optical density ofOD600=0.02.591.0x108Figure 3.11. LspaqP and ApaqQ mutants exhibit similar survival as the WT strain underanaerobic conditions over time. Log phase C. jejuni were diluted to 0D600 -0.03 in MH broth.Cultures were placed in a sealed jar with an Anaero-GasPak, and incubated at 37°C, shaking at200 r.p.m. Serial dilutions of cultures were made at each time point and spotted onto agar platesfor CFU enumeration.Table 3.2. Antimicrobial Susceptibility test. Using diffusion discs, the zone of inhibitiondiameter was recorded (mm).AntimicrobialType agent Amount WT 4pagP ApagQIron chelator Dipyridyl 0.4j.imol 10 11 10Iron chelator Desferal 0.4pmolR*R RAminoglycoside Amikacin 30ig 27 25 30Fluoroquinolone Nalidixic Acid 30ig 38 38 38Macrolide Erythromycin l5jig 40 40 40Resistant, no zone of inhibition.DUC.)cuU)CuIC.)Cu1.0x1071.0x1061.0x1051.0x1041.0x1031.0x1021.0x1011.OxlO°1.0x1011.0x102—.— WT— ApaqPe•ApaqQ0 10 20 30 40 50 60 70 80Time (h)603.12 WT and mutant C. jejuni exhibit similar toleranceto heat shock and pH stresses.To test tolerance to heat shock, log phase bacteria were diluted in MH broth and heated at55°C for 5 mm or 15 mm before harvesting the bacteria for serial dilutions and microaerobicgrowth for 2 days at 37°C for CFU enumeration. While no bacteria were recovered from the 15mm time point, the number of CFU recovered from the bacteria treated for 5 mm was identical tothe untreated sample (Figure 3.7A). No differences were observed among the strains. To assesstolerance to a range of pH (pH 4-. pH 9), log phase bacteria were diluted and spotted onto MRagar at different pHs. All C. jejuni strains exhibited optimal growth under neutral, or slightlyacidic environments (pH 6-7), and were sensitive to acidic (pH 4, 5) and basic (pH 9) conditions.No differences were observed between WT and mutants.61Figure 3.12. ApaqP and ApaqQ mutants do not exhibit a motility defect in 0.4% agar. Logphase C. jejuni were diluted to 0D600 - 0.02 and 1 l of each strain stabbed into MR agar platescontaining 0.4% agar. Migration of the cells from the point of inoculation was analyzedfollowing 24 hr of incubation at 37°C under microaerobic conditions. The experiment wasperformed in 3 replicates.2.001.751.00•0 10 5;,Figure 3.13. Rates of GGT activity in ApaqP and tipaqQ mutants do not significantlydeviate from WT C. jejuni. Absorbance was measured at 405nm at a 10 mm interval.International unit (U/ml) measurements were obtained by multiplying the absorbance reading by2.211 according to the manufacture’s protocol. Protein concentrations (mg/ml) of each bacteriallysate sample were measured using a Bradford assay. The experiment was performed inreplicates (n=10), and this is one representative experiment of three separate repeats.—.--WT—.- iXpaqP--h-- EpaqQ00•CC.,C0(31.25CCuC0.75CIC)CI I I10 20 30Time (Minutes)62kDa9572-5536—17—Figure 3.14. ApaqP and t’paqQ mutants exhibit a similar protein secretion profileas WT C.jejuni. Culture supernatants were collected from overnight C. jejuni cultures, and proteins wereprepared for TCA precipitation and SDS-PAGE silver staining.ApaqP ApaqQ28 —63ADLI..C)ci)0C)Cua)C)cuBE00Cø>%0Ca)CuU0.0Time (hr)— WT (MEM+gln)EpaqP (MEM+gln)‘tpaqQ (MEM+gln)—a--WT(MEM)-a-- zpaqP(MEM).-e EpaqQ (MEM)CTime (hr)MEM (-Gin) MEM (+GIn)O.Q.WTFiI.,paqP ApaqQ paqPc WT ApaqP ApaqQ ApaqPcStrainsFigure 3.15. zXpaqP and ApaqQ mutants and WT C. jejuni survive similarly in minimalessential media supplemented with L-glutamine. Log phase C. jejuni were grown in 1 OmlMEM with 1% NEAA supplement in the absence or presence of 40 mM L-glutamine. Bacteriawere grown microaerobically under shaking conditions and cultures were harvested at each timepoint and diluted for (A) CFU enumeration or (B) Absorbance reading at 600nm. (C) Final cellyields after 24 hr microaerobic growth in MEM minimal medium only or with glutamine. Cellswere measured at 600 nm. No significant difference was observed between the WT and mutantstrains. The experiment was performed in triplicate and repeated 3 times.64O (l CONC riD CC. jejuni pathogenesis depends on the organism’s ability to survive diverse conditionsencountered in the external environment, in the intestinal tract, and inside epithelial and immunesystem cells. External stresses frequently challenging C. jejuni include heat, osmotic, oxidative,and aerobic stress as well as nutrient limitation. Growing evidence also indicates that C. jejunican survive intracellularly within host enterocytes and macrophages (Watson and Galan, 2008;Young et al., 2007; Day et al., 2000), suggesting that C. jejuni must also tolerate intracellularstresses for sustained survival. However, specific factors promoting intracellular bacterialsurvival and triggering host cell signal transduction pathways remain largely unknown.In this study, we have investigated the biological and pathogenesis-related roles of aputative C. jejuni AA-ABC transporter originally annotated as Cj0467-9. As both the integralmembrane protein, or permease, and ATP binding proteinlATPase components are frequentlyrequired to produce a functional AA-ABC transporter (Nikaido, 2002), we generated non-polarmutations in both the Cj0467 (PaqP) permease and the Cj0469 (PaqQ) ATPase to ensure acomprehensive analysis of the system’s functions. Bioinformatics analyses predictedthis AAABC system to be a glutamine ABC transporter with high homology to glutamine ABCtransporters in Helicobacter, Streptococcus and Pseudomonas spp.; a previous study alsosuggested that this system may be a putative cysteine ABC transporter coupledto CjaA, acysteine substrate binding protein (Muller et al., 2005). Ouramino acid transport assays indicatethat the transporter does not exhibit specificity exclusivelyfor a single amino acid, since bothApaqF and i\paqQ mutants were defective for the uptakeof glutamine, glutamate, cysteine andasparate. Nevertheless, glutamine appeared to be a preferredsubstrate, since both mutantsexhibited a more severe decrease in glutamine uptake comparedto the other amino acids. Weobserved some residual glutamine transport in theApaqP and ApaqQ mutants (Figure 3.2), andpreliminary glutamine uptake competition assays performedon the z\paqF and ApaqQ mutants(unpublished observations) likewise indicate that that thereis likely a redundant mechanism for66glutamine transport in C. jejuni. This notion is supported by the previous annotation of cjO94Ocand cj0902 as glnP and glnQ which, fike paqP and paqQ, are present in all sequenced strains ofC. jejuni (Pearson et al., 2007; Hofreuter et al., 2006; Parkhill et al., 2000)Our previously published microarray analysis identifying a significant increase in levelsof cj0467-9 mRNA during 1NT407 cell infection (Gaynor et at., 2005) suggested a potential rolefor this AA-ABC transporter in C. jejuni pathogenesis. Infection assays using 1NT407 cells,Caco-2 cells and RAW264.7 macrophages revealed no significant differences in adhesion,invasion, or long-term intracellular survival at 24 hr for the ApaqP and ApaqQ mutantscompared to WT. Mouse colonization levels were likewise similar for C. jejuni WT and mutant-infected animals, at both shorter- and longer-term time points. Surprisingly, however, anapparent increase in short-term intracellular survival was observed for the ApaqP and ApaqQmutants in 1NT407, Caco-2 and RAW264.7 infected cells 9 hr post-infection. This is in contrastto observations for other genes identified in the aforementioned microarray cluster as up-regulated during cell infection: both AspoT and pVir mutants are diminished for invasion(Gaynor et al., 2005; Bacon et al., 2002; Bacon et at., 2000), and AspoT is also defective forintracellular survival (Gaynor et at., 2005). Our findings also deviate from previous studies onthe C. jejuni Peb 1 asparatate/glutamate binding protein mutant (Leon-Kempis Mdcl et al., 2006;Pei et al., 1998), as well as a Group B Streptococcus glutamine ABC transporter AglnQ mutant(Tamura et al., 2002), where significant infection defects were observed both in vitro and in vivo.One initial hypothesis to explain the enhanced intracellular survival of the ApaqP andApaqQ mutants was that they may be more resilient to stresses occurring inside cells. Forinstance, the intracellular environment will have different 02 and CO2 levels than the intestinaltract or the external environment. Furthermore, as a defense mechanism, host epithelial cells andmacrophages produce oxygen derivatives and reactive oxygen species (ROS) to generate a67highly oxidative environment (Baillon et a!., 1999; Purdy et at., 1999; Shepherd, 1986).Consistent with this hypothesis, the zXpaqP and ApaqQ mutants exhibited an increasedability togrow in a sub-optimal CO2 environment, as well as an increased resistance to aerobic and tBOOH organic peroxide stresses. Interestingly, it was recently shown that a gamma glutamyltranspeptidase (GGT)-deficient C. jejuni mutant (Aggt) exhibited increased invasionefficiencyconcomitant with increased H20 resistance (Barnes et at., 2007). GGT is involved in thedegradation of glutathione to amino acids or peptides in the gamma glutamylcycle, andglutathione has been previously implicated in resistance to oxidant-mediated killing ineukaryotes (Shi et at., 1993; Tate and Meister, 1981). Although a connection between ourobservations and GGT is plausible, particularly given the involvement of related molecules (i.e.,amino acids such as glutamine) in both pathways, several lines of evidence suggest that our AAABC transporter system affects stress responses independent of GGT: our mutants were nothyper-resistant to H20 (Figure 3.6), a Aggt mutant was previously found to be defective formouse colonization (Hofreuter et a!., 2006), and there was no apparent alteration of GOT activityin our mutant strains (Figure 3.14). The aerobic, CO2.and oxidative stress phenotypes of /ipaqPand ApaqQ are also distinct from those observed in an AnssR mutant, even though nssR (cj0466),which is involved in nitrosative stress response regulation (Elvers et a!., 2005), is directlyupstream of and likely co-transcribed with paqP and paqQ. Although little is known aboutaerotolerance and CO2 stress survival in C. jejuni, the stringent response, which is modulated bythe spoT gene, has also been shown to be up-regulated during cell infections, and was found tobe important for both low CO2 and high 02 stress survival (Gaynor et a!., 2005). C. jejuni alsoharbours a multitude of oxidative stress regulators to target a specific type of ROS. For instance,while H20 is readily degraded by catalase (Day et a!., 2000), organic peroxides such as t-BOOHare degraded by alkyl hydroperoxide reductase, known as AhpC (Klancnik et at., 2006). It is68possible that PaqP and PaqQ are involved in this organic peroxide degradation pathwaythroughmeans yet to be identified, or that this AA-ABC transporter system participates inoxidativestress tolerance by a novel mechanism.Previous studies in Salmonella typhimurium and C. jejuni suggested that cross-talkexistsbetween oxidative and heat or osmotic stress responses (Phongsisay et al., 2007;Andersen et al.,2005; Brondsted et al., 2005; Rychlik and Barrow, 2005; Stead and Park, 2000; Leeetal., 1995).Since an increase in oxidative stress resistance was observed in both z\paqPand ApaqQ mutants,we suspected that these mutants may also exhibit an increased resistance to heatand salttreatments. As predicted, the zXpaqP permease mutant likewise tolerated heat andosmoticstresses better than WT; however, the paqQ ATPase mutant was unexpectedlymore sensitivethan WT to both heat and osmotic stresses. Thus, although both components are importantforamino acid uptake, and participate in host cell interactions and oxidative/aerobicstress responses,they appear to exhibit differential functions in thermoregulation and osmoregulation. Althoughsimilar observations have not previously been reported, and thus no literature is currentlyavailable to help account for this phenomenon, several hypotheses may explain why loss of apermease or ATPase could have different consequences on specific aspects of bacterialphysiology. For instance, as soluble proteins, ATPases harbor a certain level of promiscuity(Wyckoff et al., 1999; Schlosser et al., 1997; Wilken et al., 1996; Hekstra and Tommassen,1993). Thus, an ATPase may couple with other ABC transporter systems in theabsence of afunctional permease, while a system that has lost ATPase function butretains functionalpermease proteins may be at least partially rescued by another ATP binding protein. Thishypothesis is also consistent with the high degree of homology (52.9% identity) observedbetween PaqQ and Cj0902/GlnQ. Alternatively, absence of an ATPase may result in reducedAlP consumption, causing a general disruption in energy balance. Another possibleinterpretation is that the PaqP permease and PaqQ ATPase may exert distinct effects on gating69small solutes or metabolites important for balancing the membrane electropotential gradient,which has been associated with osmoregulation and shock resistance in E. coli (Berger andHeppel, 1974). Extensive future studies will be required to distinguish between these possibilities.In addition to characterizing physiological differences between the C. jejuni mutants, wealso investigated if this AA-ABC transporter system might alter host cell viability, thereby alsoaffecting the number of viable bacteria (CFU5) recovered in the intracellular survival assay.Interestingly, RAW264.7 cells (but not 1NT407 cells) infected with mutants exhibitedapproximately 30% higher viability than cells infected with WT bacteria at the short-termintracellular survival time point. Consistent with the cell viability data, ApaqF- and A.paqQinfected RAW264.7 macrophages also displayed a significantly lower level of apoptosis, orprogrammed cell death, as assayed by Annexin-V staining. Several previous reports have alsomade tangible connections between host cell cytotoxicity and specific C. jejuni virulence factors,although the literature varies considerably depending on C. jejuni strain and host cell type used.For instance, one study reported that apoptosis of 28SC monocytes induced by infection with C.jejuni strain 81-176 was dependent on cytolethal distending toxin (CDT) (Hickey et a!., 2005), aDNase-like molecule that causes host cell cycle arrest (Lara-Tejero and Galan, 2000;Whitehouse et a!., 1998). However, another study reported that apoptosis of THP- 1 monocytesby C. jejuni strain F3801 1 was CDT- and lipooligosaccharide (LOS)-independent [in contrast toLPS observations for other pathogens (Guiney, 2005; Navarre and Zychlinsky, 2000)] andinstead involved secreted C. jejuni Cia proteins (Siegesmund et a!., 2004; Konkel et a!., 1999). Athird study reported that T84 enterocytes underwent oncosis (not apoptosis) in responseto C.jejuni infection that was dependent on the C. jejuni strain used (i.e., more invasivestrains weremore oncotic) as well as the F1aAB flagellins, but was independent of CDT (Kalischuk et a!.,2007). We have not observed any obvious differences in secreted protein profiles between ourWT and ApaqP or ApaqQ mutant strains (Figure 3.15). Nonetheless, it will be very interesting to70explore, in future work, potential connections between our findings and those described in theabove studies.To explore mechanisms underlying the effect of this AA-ABC transporter system onmacrophage apoptosis, we next investigated whether several host cell MAP kinase proteinsexhibited altered phosphorylation profiles in WT compared to mutant-infected cells. EachMAPK signaling molecule investigated is essential for eukaryotic cell growth and survival (Shanet al., 2007; Schorey and Cooper, 2003). C. jejuni has been shown to activate these signalingpathways, which in turn leads to significant downstream effects, including cytokine productionand host cell damage (Chen et al., 2006; MacCallum et al., 2005; Watson and Galan, 2005).Interestingly, deletion ofpaqP orpaqQ in C. jejuni abrogated ERK phosphorylation inRAW264.7 cell infections at the same short-term intracellular survival time point assayed for cellviability (Figure 7C). In contrast, WT, /xpaqP, and zXpaqQ elicited similar ERK activationprofiles during TNT4O7 and Caco-2 epithelial cell infections, and during later infectiontimepoints in RAW264.7 cells (Figure 3.9 and data not shown). INK and p38 kinase, whichparticipate in different MAPK signaling pathways, also displayed no significant differences inphosphorylation between WT and mutant-infected RAW264.7 cells.Our observation that this AA-ABC transporter system participates in ERK activationduring macrophage infection was an unexpected and novel finding. It is also interesting tohypothesize that this may at least partially contribute to our observed PaqP- and PaqQ-inducedmacrophage apoptosis. The role of ERK activity in apoptosis has been controversial. Somestudies have proposed that ERK is important for cell proliferation, survival and anti-apoptosis(Huang et al., 2007; Larson et al., 2007; Marin-Kuan et al., 2007). However, other studies haverevealed that ERK activation contributes to cisplatin-induced apoptosis in HeLa and A549 cellsand hydrochloric-inomenine-induced apoptosis in RAW264. 7 cells (Xaus et al., 2001; Wang etal., 2000) as well asH20-induced apoptosis in intestinal epithelial cells (Zhou et al., 2005).71Additionally, ERK activation in macrophages has been suggested to participate in caspaseindependent apoptotic signal transduction during Group B Streptococcus infection (Fettucciari etal., 2003). Activation of ERK and pro-inflammatory cytokines is also often stimulated byvirulence determinants such as LPS via toll-like receptor (TLR) signaling (Thomas et al., 2006;Guha and Mackman, 2001). Future work will be required to determine if this is also the caseduring C. jejuni infection, However, to date there have been no connections made between C.jejuni and host cell TLR4 activation, and as noted above, at least one study suggested that C.jejuni-induced macrophage apoptosis may be LOS-independent (Siegesmund et al., 2004).Interestingly, glutamine has also been implicated as an anti-apoptotic factor. For instance,glutamine supplementation has been shown to reduce apoptosis of human intestinal epithelialcells in a maimer that may interface with the ERK activation pathway (Larson et al., 2007; Evanset al., 2005; Evans et al., 2003). Furthermore, glutamine depletion was found to elicit apoptosisin epithelial cells and immune cells such as T-cells and neutrophils (Chang et al., 2002). Thus,one working hypothesis to explain the increased host cell death observed for WT vs. ApaqF andApaqQ mutant-infected cells is that to acquire nutrients, intracellular C. jejuni may hijackglutamine (and likely other amino acids) from host cells, and that this AA-ABC transportersystem, which is up-regulated during cell infection, participates in this process. Preliminaryexperiments exploring glutamine supplementation during cell infections have not to dateidentified an effect on our C. jejuni-induced macrophage apoptosis (Figure 3.16). However, thishypothesis is consistent with our previous observations that C. jejuni requires the stringentresponse (spoT) and polyphosphate kinase 1 (ppkl) for both survival in low-nutrientenvironments and in an intracellular environment (Candon et al., 2007; Gaynor et al., 2005), aswell as recent work suggesting that C. jejuni resides intracellularly in an anaerobic and likelynutrient-poor vacuole (Watson and Galan, 2008).72In summary, we have provided evidence that the AA-ABC transporter systemcomponents PaqP and PaqQ participate in several key aspects of C. jejuni stress survival andpathogenesis, some of which are novel not only for C. jejuni but for other bacteria as well. Ourwork has also created additional links between C. jejuni disease etiology, physiology, and basicmetabolic processes (i.e., amino acid uptake), an increasingly emerging theme in the study of thisprevalent pathogen. Future work aimed at addressing the hypotheses above as well as identifyingpotentially unpredicted functions for this system should yield additional insight into C. jejunibiology and pathogenesis and potentially uncover new roles for AA-ABC transporters in otherbacteria as well.73IWe have shown that PaqP and PaqQ are involved in several key aspects ofC. jejuniphysiology and biology. For instance, we found that in the absence ofPaqP or PaqQ, C. jejuniexhibited an increased tolerance to CO2 and aerobic stress. This isin direct contrast toobservations with the AspoT stringent response mutant,which exhibits an increase in sensitivityto aerobic and CO2 stresses (Gaynor et al., 2005).Thus, one ensuing hypothesis is thatdiminished nutrient (i.e., glutamine) uptake may causeApaqP and ApaqQ to up-regulate thestringent response in order to survive the unfavourablenutrient-limited condition. To test this,we will first examine levels of spoT transcription in themutants vs. WT by Real-Timequantitative PCR. Second, we will assess levels of thestringent response effector moleculeguanosine tetraphosphate (ppGpp), which is synthesized bySpoT during times of nutrient stressand binds RNA polymerase to elicit globaltranscriptional changes to the stress condition. Thiswill be accomplished by labeling WT and mutant strainswith 32P and detecting ppGpp and itsprecursor pppGpp by Thin Layer Chromatography. If a deficiency in AA uptakein mutantsdefective for our AA-ABC transporter system activates the stringent response,then lack of PaqPor PaqQ should increase spoT transcription and/or ppGppsynthesis.We have also demonstrated that PaqP and PaqQ impact intracellular survival,host celldeath, and the ERK-MAPK signaling pathway. However, it is not completely understoodwhether PaqP and PaqQ induce macrophage apoptosis via ERK-MAPK dependent signalingindependent of or dependent on other bacterial apoptosis-inducing factors, such as the CDT.Since the CDT has been postulated as a major inducer of apoptosis in C. jejuni infected cells(Hickey et al., 2000), one future question is whether PaqP and PaqQ enhance apoptosis in aCDT-dependent or -independent manner. One approach to explore this will be to generateAcdtBtxpaqP and AcdtBzXpaqQ double knockout mutants and compare levels of apoptosis inRAW264.7 macrophages infected with WT, zXpaqP, zipaqQ, AcdtB, /cdtBApaqP, and75AcdtBApaqQ mutants. If CDT inducesapoptosis in RAW264.7 cells in a PaqP- and PaqQdependent manner, or vice versa, levelsof apoptosis induced by all mutant strainsshould besimilar to each other. Conversely,if PaqP and PaqQ induce apoptosis in RAW264.7 cells in aCDT-independent manner, we would expect infectionwith AcdtBzXpaqF or z\cdtBApaqQ doublemutants to result in significantly reducedapoptosis compared to single mutant ApaqF,ApaqQand z\cdtB-infected cells. Regardlessof the outcome, this study will introduce a novel paradigmfurther associating this ABC transporter withpathogenesis during bacterial infection.To date, nothing has been explored regarding the correlationof PaqP and PaqQ withMAPK signaling pathways, nor is much understoodabout C. jejuni induction of MAP kinaseactivation in general. Hence, we are alsointerested in investigating the upstream moleculareffects of PaqP and PaqQ on the activationof ERK signaling in macrophages duringC. jejuniinfection. One of our hypotheses is that thisABC transporter system, or anotherC. jejuni factor(i.e., an outer membrane protein or othercell surface component) that may be directlyinfluencedby the transporter system, may interact withhost cells and activate ERKby triggering a signaltransduction cascade upon internalization(or while surviving intracellularly). Aninitial approachwill include examining signaling moleculesupstream of ERK in the MAPKpathway. MEK(Raf-MAPKIERK kinase), immediatelyupstream of ERK, will be the first logicalcandidate toexamine. This will be accomplishedby assessing levels of phosphorylated MEKin WT andzXpaqP and ApaqQ infected RAW264.7cells by Western blot analysis.Next, to investigate ifPaqP and PaqQ are involved in ERK-MAPKactivation in a Ras-dependent mannersimilar to thereceptor tyrosine kinases (RTKs), levelsof Raf-l activation will be examined.It is understoodthat RTK is activated by extracellularligands, often growth factors(such as the epidermalgrowth factor), and undergoes autophosphorylation.The phosphorylated tyrosinekinase receptorthen recruits RasGRP to activateRas from Ras-GDP (an inactive form)into Ras-GTP (an active76form). Ras is a small G protein that recruits Raf- 1, an intermediate signaling molecule thattriggers a series of downstream cascades to activate ERK (Hagemanri and Blank, 2001). If PaqPand PaqQ activate ERK via a Ras and Raf- 1 dependent mechanism similar to the RTKs, it willbe interesting to identify signaling molecules that may interact with PaqP and PaqQ to triggersubsequent signal cascades in the MAPK pathway. If PaqP and PaqQ are involved in theRas/Raf- 1 dependent pathway, we will next investigate whether PaqP and PaqQ are involved inthe activation of RasGRP. In addition, we will use MEKIERK inhibitor to investigate whetherPaqP and PaqQ contribute to apoptosis in host cells in an ERK-dependent manner. It is expectedthat if PaqP or PaqQ contribute to apoptosis via ERK activation, we should observe a decrease inlevels of apoptosis (or increased levels of cell viability) in the MEK!ERK inhibitor treated WTinfected cells.Finally, we will examine an alternate putative glutamine ABC transporter permease GInPand ATPase G1nQ to investigate whether the observed phenotypes in ApaqP and zXpaqQ are dueto shifts in AA (specifically glutamine) metabolism or by other potential effects of PaqP andPaqQ. It is hypothesized that if increased levels of intracellular survival and stress resistance aredue to changes in glutamine metabolism, then AglnP and zXglnQ mutants with reduced glutamineuptake will behave similarly to the /paqP and ApaqQ strains. Furthermore, double knockoutmutants AglnFApaqF and AglnQApaqQ should show an augmented phenotype in comparison tothe single knockout mutants.These future experiments will enhance our knowledge of PaqP and PaqQ and yield newideas in understanding the role of ABC transporters in bacterial pathogenesis.77REFERENCESAllos, B.M. and Biaser, M.J.(1995) Campylobacterjejuni and the expanding spectrum of relatedinfections. Clin Infect Dis. 20: 1092-1099;quiz 1100-1091.Allos, B.M., Lippy, F.T., Carisen,A., Washburn, R.G. and Blaser, M.J. (1998) Campylobacterjejuni strains from patients with Guillain-Barresyndrome. Emerg Infect Dis. 4: 263-268.Alter, I. and Scherer, K. (2006) Stressresponse of Campylobacter spp. and its role in foodprocessing. J Vet Med B Infect Dis Vet Public Health.53: 351-357.Andersen, M.T., Brondsted, L., Pearson,B.M., Muiholland, F., Parker, M., Pin, C., et al (2005)Diverse roles for HspR in Campylobacterjejuni revealedby the proteome, transcriptome andphenotypic characterization of an hspR mutant. Microbiology.151: 905-915.Bacon, D.J., Aim, R.A., Burr, D.H., Hu, L., Kopecko,D.J., Ewing, C.P., et al (2000)Involvement of a plasmid in virulence ofCampylobacterjejuni81-176. Infect Immun. 68: 4384-4390.Bacon, D.J., Aim, R.A., Hu, L., Hickey, T.E., Ewing,C.P., Batchelor, R.A., et al (2002) DNAsequence and mutational analyses of thepVir plasmid of Campylobacterjejuni 8 1-176. InfectImmun. 70: 6242-6250.Baillon, M.L., van Vliet, A.H., Ketiey, J.M,, Constantinidou,C. and Penn, C.W. (1999) An iron-regulated alkyl hydroperoxide reductase(AhpC) confers aerotolerance and oxidativestressresistance to the microaerophilic pathogen Campylobacterjejuni.JBacteriol. 181: 4798-4804.Barnes, I.H., Bagnall, M.C., Browning,D.D., Thompson, S.A., Manning,G. and Newell, D.G.(2007) Gamma-glutamyl transpeptidasehas a role in the persistent colonizationof the avian gutby Campylobacterjejuni. Microb Pathog. 43: 198-207.Beery, J.T., Hugdahl, M.B. and Doyle, M.P.(1988) Colonization of gastrointestinal tractsofchicks by Campylobacterjejuni. ApplEnviron Microbiol. 54: 2365-2370.Berger, E.A. and Heppel, L.A. (1974) Differentmechanisms of energy coupling forthe shock-sensitive and shock-resistant amino acidpermeases of Escherichia coli.J Biol Chem. 249: 7747-7755.Blaser, M.J. (1997) Epidemiologic andclinical features of Campylobacterjejuniinfections. JInfect Dis. 176 Suppl 2: S103-105.Blaser, M.J., Taylor, D.N. and Feldman,R.A. (1983) Epidemiology of Campylobacterjejuniinfections. Epidemiol Rev. 5: 157-176.Borrmann, E., Berndt, A., Hanel, I. andKohler, H. (2007) Campylobacter-inducedinterleukin-8responses in human intestinal epithelial cellsand primary intestinal chick cells. VetMicrobiol.124: 115-124.78Bras, A.M. and Ketley, J.M. (1999) Transcellular translocation ofCampylobacterjejuniacrosshuman polarised epithelial monolayers. FEMS Microbiol Lett. 179: 209-2 15.Brondsted, L., Andersen, M.T., Parker, M., Jorgensen, K. and Ingmer, H. (2005) The HtrAprotease of Campylobacterjejuni is required for heat and oxygen tolerance and for optimalinteraction with human epithelial cells. Appi Environ Microbiol. 71: 3205-32 12.Butzler, J.P. (2004) Campylobacter, from obscurity to celebrity. Clin Microbiol Infect. 10: 868-876.Candon, H.L., Allan, B.J., Fraley, C.D. and Gaynor, E.C. (2007) Polyphosphate kinase 1 is apathogenesis determinant in Campylobacterjejuni. JBacteriol. 189: 8099-8 108.Castaneda-Roldan, E.I., Ouahrani-Bettache, S., Saldana, Z., Avelino, F., Rendon, M.A., Dornand,J. and Giron, J.A. (2006) Characterization of SP41, a surface protein of Brucella associated withadherence and invasion of host epithelial cells. Cell Microbiol. 8: 1877-1887.Ceelen, L.M., Decostere, A., Ducatelle, R. and Haesebrouck, F. (2006) Cytolethal distendingtoxin generates cell death by inducing a bottleneck in the cell cycle. Microbiol Res. 161:109-120.Chang, C. and Miller, J.F. (2006) Campylobacterjejuni colonization of mice with limitedentericflora. Infect Immun. 74: 5261-5271.Chang, W.K., Yang, K.D. and Shaio, M.F. (1999) Lymphocyte proliferation modulatedbyglutamine: involved in the endogenous redox reaction. Clin Exp Immunol. 117:482-488.Chang, W.K., Yang, K.D., Chuang, H., Jan, J.T. and Shaio, M.F. (2002) Glutamineprotectsactivated human T cells from apoptosis by up-regulating glutathione and Bcl-2levels. ClinImmunol. 104: 151-160.Chen, M.L., Ge, Z., Fox, J.G. and Schauer, D.B. (2006) Disruptionof tight junctions andinduction of proinflammatory cytokine responses in colonicepithelial cells by Campylobacterjejuni.Infectlmmun. 74: 6581-6589.Day, W.A., Jr., Sajecki, J.L., Pitts, T.M. and Joens, L.A.(2000) Role of catalase inCampylobacterjejuni intracellular survival. Infect Immun. 68: 6337-6345.Debruyne, L., Samyn, F., De Brandt, E., Vandenberg,0., Heyndrickx, M. and Vandamme, P.(2008) Comparative performance of different PCR assays forthe identification ofCampylobacter jejuni and Campylobacter coli. Res Microbiol.159: 88-93.Dippel, R. and Boos, W. (2005) The maltodextrin systemof Escherichia coli: metabolism andtransport. JBacteriol. 187: 8322-833 1.Edgeworth, J.D., Spencer, J., Phalipon, A., Griffin, G.E. andSansonetti, P.J. (2002) Cytotoxicityand interleukin- 1 beta processing following Shigella flexneriinfection of human monocytederived dendritic cells. EurJlmmunol. 32: 1464-1471.79Elkins, C.A. and Beenken, K.E. (2005) Modeling thetripartite drug efflux pump archetype:structural and functional studies of the macromolecularconstituents reveal more than their namesimply. JChemother. 17: 581-592.Elvers, K.T., Turner, S.M., Wainwright, L.M.,Marsden, G., Hinds, J., Cole, J.A., et al (2005)NssR, a member of the Crp-Fnr superfamilyfrom Campylobacterjejuni, regulates anitrosativestress-responsive regulon that includes both asingle-domain and a truncated haemoglobin. MolMicrobiol. 57: 735-750.Eswarappa, S.M., Panguluri, K.K., Hensel, M.and Chakravortty, D. (2008) The yejABEFoperon of Salmonella confers resistanceto antimicrobial peptides and contributes toits virulence.Microbiology. 154: 666-678.Evans, M.E., Jones, D.P. and Ziegler, T.R. (2003)Glutamine prevents cytokine-inducedapoptosis in human colonic epithelial cells. JNutr.133: 3065-3071.Evans, M.E., Jones, D.P. and Ziegler, T.R. (2005) Glutamineinhibits cytokine-induced apoptosisin human colonic epithelial cells viathe pyrimidine pathway. Am JFhysiol GastrointestLiverPhysiol. 289: G388-396.Fettucciari, K., Fetriconi, I., Bartoli, A.,Rossi, R. and Marconi, P. (2003) Involvement ofmitogen-activated protein kinases in Group BStreptococcus-induced macrophage apoptosis.Pharmacol Res. 47: 355-362.Forward, J.A., Behrendt, M.C., Wyborn, N.R.,Cross, R. and Kelly, D.J. (1997) TRAPtransporters: a new family of periplasmicsolute transport systems encoded by the dctPQM genesof Rhodobacter capsulatus and by homologsin diverse gram-negative bacteria. JBacteriol. 179:5482-5493.Fry, B.N., Feng, S., Chen, Y.Y., Newell, D.G., Coloe,P.J. and Korolik, V. (2000) The galE geneofCampylobacterjejuni is involved in lipopolysaccharide synthesisand virulence. Infect Immun.68: 2594-2601.Galindo, C.L., Fadi, A.A., Sha, J., Gutierrez, C.,Jr., Popov, V.L., Boldogh, I., et al (2004)Aeromonas hydrophila cytotoxic enterotoxin activatesmitogen-activated protein kinases andinduces apoptosis in murine macrophages and humanintestinal epithelial cells. J Biol Chem. 279:37597-376 12.Garenaux, A., Jugiau, F., Rama, F., de Jonge, R., Denis,M., Federighi, M. and Ritz, M. (2008)Survival of Campylobacterjejuni Strains from DifferentOrigins Under Oxidative StressConditions: Effect of Temperature. Curr Microbiol. 56: 293-297.Garmory, H.S. and Titball, R.W. (2004) ATP-bindingcassette transporters are targets for thedevelopment of antibacterial vaccinesand therapies. Infect Immun. 72: 6757-6763.Gaynor, E.C., Wells, D.H., MacKichan, J.K. andFalkow, S. (2005) The Campylobacterjejunistringent response controls specific stresssurvival and virulence-associated phenotypes. MolMicrobiol. 56: 8-27.80Girard, M.P., Steele, D., Chaignat, C.L. and Kieny, M.P.(2006) A review of vaccineresearchand development: human enteric infections. Vaccine.24: 2732-2750.Guccione, E., Del Rocio Leon-Kempis, M., Pearson,B.M., Hitchin, E., Muiholland, F.,vanDiemen, P.M., et al (2008) Amino-acid dependentgrowth of Campylobacter jejuni: Keyrolesfor aspartase (AspA) under microaerobic andoxygen-limited conditions and identificationofAspB (Cj0762), essential for growth onglutamate. Mol Microbiol.Guha, M. and Mackman, N. (2001) LPSinduction of gene expression in humanmonocytes. CellSignal. 13: 85-94.Guiney, D.G. (2005) The role of host cell deathin Salmonella infections. Curr Top MicrobiolImmunol. 289: 13 1-150.Guttman, J.A., Kazemi, P., Lin, A.E., Vogi,A.W. and Finlay, B.B. (2007a)Desmosomes areunaltered during infections by attaching and effacingpathogens. Anat Rec (Hoboken). 290:199-205.Guttman, J.A., Samji, F.N., Li, Y.,Deng, W., Lin, A. and Finlay, B.B. (2007b)Aquaporinscontribute to diarrhoea caused by attachingand effacing bacterial pathogens. Cell Microbiol. 9:13 1-141.Hagemann, C. and Blank, J.L. (2001) The upsand downs of MEK kinase interactions. CellSignal. 13: 863-875.Hekstra, D. and Tommassen, J. (1993)Functional exchangeability of the ABC proteins of theperiplasmic binding protein-dependenttransport systems Ugp and Mal of Escherichia coli. JBacteriol. 175: 6546-6552.Hendrixson, D.R. and DiRita, V.J. (2004)Identification of Campylobacter jejuni genes involvedin conimensal colonization of the chick gastrointestinaltract. Mol Microbiol. 52: 47 1-484.Hickey, T.E., Majam, G. and Guerry,P. (2005) Intracellular survival of Campylobacterjejuni inhuman monocytic cells and induction of apoptoticdeath by cytholethal distending toxin. InfectImmun. 73: 5 194-5197.Hickey, T.E., McVeigh, A.L., Scott,D.A., Michielutti, R.E., Bixby, A., Carroll, S.A., et al (2000)Campylobacter jejuni cytolethal distending toxinmediates release of interleukin-8 from intestinalepithelial cells. Infect Immun. 68: 6535-6541.Higgins, C.F. (1992) ABC transporters: frommicroorganisms to man. Annu Rev Cell Biol. 8: 67-113.Hofreuter, D., Tsai, J., Watson, R.O., Novik, V.,Altman, B., Benitez, M., et al (2006) Uniquefeatures of a highly pathogenic Campylobacter jejunistrain. Infect Immun. 74: 4694-4707.Holland, I.B., Schmitt, L. and Young, J. (2005) Type 1protein secretion in bacteria, the ABC-transporter dependent pathway (review). Mol Membr Biol.22: 29-39.81Hu, L., Bray, M.D., Osorio, M. and Kopecko, D.J. (2006) Campylobacterjejuni inducesmaturation and cytokine production in human dendritic cells. InfectImmun. 74: 2697-2705.Huang, D., Khoe, M., Befekadu, M., Chung, S., Takata, Y., Ilic, D. andBryer-Ash, M. (2007)Focal adhesion kinase mediates cell survival via NF-kappaBand ERK signaling pathways. Am JPhysiol Cell Physiol. 292: Cl 339-1352.Humphrey, T., O’Brien, S. and Madsen, M. (2007) Campylobacters as zoonotic pathogens:afood production perspective. IntJFoodMicrobiol. 117: 237-257.Jin, S., Joe, A., Lynett, J., Hani, E.K., Sherman, P. and Chan, V.L. (2001)J1pA, a novel surface-exposed lipoprotein specific to Campylobacter jejuni, mediates adherenceto host epithelial cells.Mol Microbiol. 39: 1225-1236.Jones, D.M., Sutcliffe, E.M., Rios, R., Fox, A.J. and Curry, A. (1993) Campylobacterjejuniadapts to aerobic metabolism in the environment. JMedMicrobiol.38: 145-150.Jones, M.A., Totemeyer., S., Maskell, D.J., Bryant, C.E. and Barrow,P.A. (2003) Induction ofproinflammatory responses in the human monocytic cell line THP-1by Campylobacterjejuni.Infect Immun. 71: 2626-2633.Jones, P.M. and George, A.M. (1999) Subunit interactions in ABC transporters:towards afunctional architecture. FEMS Microbiol Lett. 179: 187-202.Kalischuk, L.D., Inglis, G.D. and Buret, A.G. (2007) Strain-dependent inductionof epithelial celloncosis by Campylobacterjejuni is correlated with invasion abilityand is independent ofcytolethal distending toxin. Microbiology. 153: 2952-2963.Karlyshev, A.V. and Wren, B.W. (2005) Development and applicationof an insertional systemfor gene delivery and expression in Campylobacter jejuni. Appi EnvironMicrobiol. 71: 4004-4013.Kelly, D.J. 2008, Complexity and versatility in the metabolismand physiology ofCampylobacterjejuni (Chapter 3). In Campylobacter 3.Nachamkin I., Szymanski, C., and Blaser,M. Washington, DC: American Society for Microbiology Press,pp.41-61 (in press).Kiehibauch, J.A., Albach, R.A., Baum, L.L.and Chang, K.P. (1985) Phagocytosis ofCampylobacterjejuni and its intracellular survival inmononuclear phagocytes. Infect Immun. 48:446-45 1.Klancnik, A., Botteldoorn, N., Herman, L.and Mozina, S.S. (2006) Survival and stress inducedexpression of groEL and rpoD of Campylobacterjejuni from different growth phases. mtJ FoodMicrobiol. 112: 200-207.Konkel, M.E., Mead, D.J., Hayes, S.F. and Cieplak,W., Jr. (1992) Translocation ofCampylobacterjejuni across human polarized epithelialcell monolayer cultures. Jlnfect Dis.166: 308-3 15.82Konkel, M.E., Kim, B.J., Rivera-Amill, V.and Garvis, S.G. (1999) Identification of proteinsrequired for the internalization of Campylobacter jejuni into cultured mammalian cells. AdvExpMedBiol. 473: 2 15-224.Konkel, M.E., Kim, B.J., Kiena, J.D., Young, C.R. and Ziprin, R. (1998) Characterizationof thethermal stress response of Campylobacterjejuni. Infect Immun. 66: 3666-3672.Konkel, M.E., Christensen, J.E., Keech, A.M., Monteville, M.R., Kiena, J.D. and Garvis,S.G.(2005) Identification of a fibronectin-binding domain within the Campylobacterjejuni CadFprotein. Mol Microbiol. 57: 1022-1035.Kranzer, K., Eckhardt, A., Aigner, M., Knoll, G., Deml, L., Speth, C., et al (2004) Induction ofmaturation and cytokine release of human dendritic cells by Helicobacter pylon. Infect Immun.72: 4416-4423.Lara-Tejero, M. and Galan, J.E. (2000) A bacterial toxin that controls cell cycle progressionas adeoxyribonuclease I-like protein. Science. 290: 354-357.Larson, S.D., Li, J., Chung, D.H. and Evers, B.M. (2007) Molecular Mechanisms ContributingtoGlutamine-Mediated Intestinal Cell Survival. Am JFhysiol Gastrointest Liver Physiol.Lee, I.S., Lin, J., Hall., H.K., Bearson, B. and Foster, J.W. (1995) The stationary-phasesigmafactor sigma S (RpoS) is required for a sustained acid tolerance response in virulentSalmonellatyphimurium. Mol Microbiol. 17: 155-167.Leon-Kempis Mdel. R., Guccione, E., Mulholland, F., Williamson, M.P.and Kelly, D.J. (2006)The Campylobacterjejuni PEB1a adhesin is an aspartate/glutamate-bindingprotein of an ABCtransporter essential for microaerobic growthon dicarboxylic amino acids. Mol Microbiol. 60:1262-1275.MacCallum, A., Haddock, G. and Everest, P.H. (2005) Campylobacterjejuniactivates mitogenactivated protein kinases in Caco-2 cell monolayers and in vitro infectedprimary human colonictissue. Microbiology. 151: 2765-2772.Marin-Kuan, M., Nestler, S., Verguet, C., Bezencon, C., Piguet,D., Delatour, T., et al (2007)MAPK-ERK activation in kidney of male rats chronically fed ochratoxinA at a dose causing asignificant incidence of renal carcinoma. Toxicol Appi Pharmacol.224: 174-181.Marriott, I., Hammond, T.G., Thomas, E.K. and Bost,K.L. (1999) Salmonella efficiently enterand survive within cultured CDI 1 c+ dendritic cells initiating cytokineexpression. Eur JImmunol. 29: 1107-1115.Matsuo, M, Ueda K., Ryder, T. and Ashcroft, F. 2003, The sulfonylureareceptor: an ABCtransporter that acts as an ion channel regulator, in ABCproteins form bacteria to man, Holland,I.B., Coles, SPC, Kuchler, K and Higggins, CF eds. AcademicPress, London, UK, pp55l-575.Mead, P.S., Slutsker, L., Griffin, P.M. and Tauxe, R.V. (1999) Food-relatedillness and death inthe united states reply to dr. hedberg. Emerg Infect Dis. 5: 84 1-842.83Medzhitov, R. and Janeway, C.A., Jr. (1997) Innate immunity:the virtues of a nonclonal systemof recognition. Cell. 91: 295-298.Menard, R., Sansonetti, P.J. and Parsot, C. (1993)Nonpolar mutagenesis of the ipa genes definesIpaB, IpaC, and IpaD as effectors of Shigella flexneri entryinto epithelial cells. JBacteriol. 175:5899-5906.Mihaijevic, R.R., Sikic, M., Klancnik, A., Brumini, G.,Mozina, S.S. and Abram, M. (2007)Environmental stress factors affecting survival and virulenceof Campylobacter jejuni. MicrobPathog. 43: 120-125.Mongkolsuk, S., Praituan, W., Loprasert, S., Fuangthong,M. and Chamnongpol, S. (1998)Identification and characterization of a new organichydroperoxide resistance (ohr) gene with anovel pattern of oxidative stress regulation fromXanthomonas campestris pv. phaseoli. JBacteriol. 180: 2636-2643.Monteville, M.R. and Konkel, M.E. (2002) Fibronectin-facilitatedinvasion of T84 eukaryoticcells by Campylobacterjejuni occurs preferentially atthe basolateral cell surface. Infect Immun.70: 6665-6671.Moore, J.E., Crowe, M., Heaney, N. and Crothers,E. (2001) Antibiotic resistance inCampylobacter spp. isolated from human faeces(1980-2000) and foods (1997-2000) in NorthernIreland: an update. JAntimicrob Chemother. 48:455-457.Muller, A., Leon-Kempis Mdel, R., Dodson, E., Wilson, K.S.,Wilkinson, A.J. and Kelly, D,J.(2007) A bacterial virulence factor with adual role as an adhesin and a solute-binding protein:the crystal structure at 1.5 A resolution of the PEB 1 aprotein from the food-borne humanpathogen Campylobacterjejuni. JMol Biol. 372: 160-171.Muller, A., Thomas, G.H., Horler, R., Brannigan, J.A.,Blagova, E., Levdikov, V.M., et al (2005)An ATP-binding cassette-type cysteine transporterin Campylobacterjejuni inferred from thestructure of an extracytoplasmic solute receptor protein. Mol Microbiol.57: 143-15 5.Murphy, C., Carroll, C. and Jordan, K.N. (2006) Environmentalsurvival mechanisms of thefoodborne pathogen Campylobacterjejuni. JAppi Microbiol. 100: 623-632.Myszewski, M.A. and Stern, N.J. (1991) Phagocytosis and intracellular killing of Campylobacterjejuni by elicited chicken peritonealmacrophages. Avian Dis. 35: 750-755.Nachamkin, I. (2001) Campylobacter Enteritis and the Guillain-Barre Syndrome.Curr Infect DisRep.3: 116-122.Naikare, H., Palyada, K., Panciera, R., Marlow, D. and Stintzi, A. (2006) Major role for FeoB inCampylobacterjejuni ferrous iron acquisition, gut colonization, andintracellular survival. InfectImmun. 74: 5433-5444.Nakamura, E. and Hagen, S.J. (2002) Role of glutamine and arginase in protection againstammonia-induced cell death in gastric epithelial cells. Am JPhysiol Gastrointest Liver Physiol.283: G1264-1275.84Navarre, W.W. and Zychlinsky, A. (2000)Pathogen-induced apoptosis ofmacrophages: acommon end for different pathogenic strategies.Cell Microbiol. 2: 265-273.Nikaido, H. (2002) How are theABC transporters energized?Proc Nati Acad Sci USA. 99:9609-9610.Oelschlaeger, T.A., Guerry, P. andKopecko, D.J. (1993) Unusualmicrotubule-dependentendocytosis mechanisms triggered byCampylobacterjejuni and Citrobacterfreundii. Proc NatlAcadSci USA. 90: 6884-6888.Pacha, R.E., Clark, G.W.,Williams, E.A. and Carter, A.M. (1988)Migratory birds of centralWashington as reservoirs of Campylobacterjejuni.Can JMicrobiol. 34: 80-82.Pajaniappan, M., Hall, J.E., Cawthraw, S.A.,Newell, D.G., Gaynor, E.C., Fields, J.A.,et al(2008) A temperature-regulatedCampylobacterjejuni gluconate dehydrogenaseis involved inrespiration-dependent energy conservationand chicken colonization. MolMicrobiol. 68: 474-491.Park, S.F. (2002) The physiologyof Campylobacter species and itsrelevance to their role asfoodborne pathogens. IntJFoodMicrobiol. 74: 177-188.Parkhill, J., Wren, B.W., Mungall,K., Ketley, J.M., Churcher, C., Basham, D.,et a! (2000) Thegenome sequence of the food-bornepathogen Campylobacterjejuni reveals hypervariablesequences. Nature. 403: 665-668.Pearson, B.M., Gaskin, D.J., Segers, R.P.,Wells, J.M., Nuijten, P.J. and van Vliet, A.H.(2007)The complete genome sequence ofCampylobacterjejuni strain 81116 (NCTCI 1828).J Bacteriol.189: 8402-8403.Pei, Z., Burucoa, C., Grignon, B.,Baqar, S., Huang, X.Z., Kopecko, D.J., etal (1998) Mutationin the peb 1 A locus of Campylobacterjejuni reduces interactions with epithelialcells andintestinal colonization of mice. InfectImmun. 66: 93 8-943.Pesci, E.C., Cottle, D.L. andPickett, C.L. (1994) Genetic, enzymatic, and pathogenicstudies ofthe iron superoxide dismutase ofCampylobacterjejuni. Infect Immun. 62:2687-2694.Phanvijhitsiri, K., Musch, M.W., Ropeleski,M.J. and Chang, E.B. (2006) Heat induction of heatshock protein 25 requires cellular glutaminein intestinal epithelial cells. Am JPhysiol CellPhysiol. 291: C290-299.Phongsisay, V., Perera, V.N. and Fry, B.N.(2007) Expression of the htrB gene is essential forresponsiveness of Salmonella typhimuriumand Campylobacterjejuni to harsh environments.Microbiology. 153: 254-262.Purdy, D. and Park, S.F. (1994)Cloning, nucleotide sequence and characterization ofa geneencoding superoxide dismutase from Campylobacterjejuni and Campylobacter coli.Microbiology. 140 (Pt5): 1203-1208.85Purdy, D., Cawthraw, S., Dickinson, J.H.,Newell, D.G. and Park, S.F. (1999)Generation of asuperoxide dismutase (SOD)-deficient mutantof Campylobacter coli: evidencefor thesignificance of SOD in Campylobactersurvival and colonization. AppiEnviron Microbiol. 65:2540-2546.Quinn, T., Bolla, J.M., Pages, J.M. and Fanning, S.(2007) Antibiotic-resistant Campylobacter:could efflux pump inhibitors controlinfection? JAntimicrob Chemother. 59:1230-1236.Rosinha, G.M., Freitas, D.A., Miyoshi,A., Azevedo, V., Campos, E., Cravero,S.L., et al (2002)Identification and characterizationof a Brucella abortus ATP-bindingcassette transporterhomolog to Rhizobium meliloti ExsA andits role in virulence and protectionin mice. InfectImmun. 70: 5036-5044.Ruckdeschel, K., Machold, J., Roggenkamp,A., Schubert, S., Pierre, J., Zumbihl,R., et al (1997)Yersinia enterocolitica promotes deactivationof macrophage mitogen-activatedprotein kinasesextracellular signal-regulated kinase-1/2, p38,and c-Jun NH2-terminal kinase. Correlationwithits inhibitory effect on tumornecrosis factor-alpha production. JBiol Chem.272: 15920-15927.Rychlik, I. and Barrow, P.A. (2005) Salmonellastress management and its relevancetobehaviour during intestinal colonisationand infection. FEMSMicrobiol Rev. 29:1021-1040.Saurin, W., Hofnung, M. and Dassa, E.(1999) Getting in or out: earlysegregation betweenimporters and exporters in the evolutionof ATP-binding cassette (ABC) transporters.J Mol Evol.48: 22-41.Schiosser, A., Kampers, T. and Schrempf,H. (1997) The StreptomycesAlP-binding componentMsiK assists in cellobiose and maltosetransport. JBacteriol. 179: 2092-2095.Schorey, J.S. and Cooper, A.M. (2003)Macrophage signalling upon mycobacterial infection:theMAP kinases lead the way. Cell Microbiol.5: 133-142.Shan, L., He, P. and Sheen, J. (2007)Intercepting host MAPK signaling cascades by bacterialtype III effectors. Cell Host Microbe. 1:167-174.Shepherd, V.L. (1986) The role of the respiratoryburst of phagocytes in host defense. SeminRespir infect. 1:99-106.Shi, M., Gozal, E., Choy, H.A. andForman, H.J. (1993) Extracellular glutathione and gammaglutamyl transpeptidase prevent H202-inducedinjury by 2,3-dimethoxy-1,4-naphthoquinone.Free Radic Biol Med. 15: 57-67.Shibayama, K., Wachino, J., Arakawa, Y., Saidijam,M., Rutherford, N.G. and Henderson, P.J.(2007) Metabolism of glutamine and glutathionevia gamma-glutamyltranspeptidase andglutamate transport in Helicobacter pylon:possible significance in the pathophysiology oftheorganism. Mol Microbiol. 64: 3 96-406.Siegesmund, A.M., Konkel, M.E., Kiena,J.D. and Mixter, P.F. (2004) Campylobacterjejuniinfection of differentiated THP- 1 macrophages resultsin interleukin 1 beta release and caspase1-independent apoptosis. Microbiology. 150: 56 1-569.86Smirnova, G.V., Krasnykh, T.A. and Oktyabrsky,O.N. (2001) Role ofglutathione in theresponse of Escherichia coli to osmotic stress.Biochemistry (Mosc). 66: 973-978.Smith, J.L. and Bayles, D.O. (2006)The contribution of cytolethaldistending toxin to bacterialpathogenesis. Crit Rev Microbiol.32: 227-248.Stead, D. and Park, S.F. (2000)Roles of Fe superoxide dismutaseand catalase in resistance ofCampylobacter coli to freeze-thawstress. Appl Environ Microbiol.66: 3110-3112.Tamura, G.S., Nittayajarn, A.and Schoentag, D.L. (2002) A glutaminetransport gene, glnQ, isrequired for fibronectin adherenceand virulence of group B streptococci.Infect Immun. 70:2877-2885.Tang, P., Sutherland, C.L., Gold,M.R. and Finlay, B.B. (1998) Listeriamonocytogenes invasionof epithelial cells requires theMEK- I /ERK-2 mitogen-activatedprotein kinase pathway. InfectImmun. 66: 1106-1112.Tate, S. S. and Meister, A. (1981)gamma-Glutamyl transpeptidase : catalytic,structural andfunctional aspects. Mol CellBiochem. 39: 357-368.Thomas, C., Hill, D.J. and Mabey,M. (1999) Morphological changes ofsynchronizedCampylobacterjejuni populationsduring growth in single phaseliquid culture. Lett AppiMicrobiol. 28: 194-198.Thomas, K.E., Galligan, C.L.,Newman, R.D., Fish, EN. and Vogel, S.N.(2006) Contribution ofinterferon-beta to the murinemacrophage response to the toll-like receptor4 agonist,lipopolysaccharide. JBiol Chem.281: 31119-31130.Tullius, M.V., Harth, G. and Horwitz, M.A.(2003) Glutamine synthetase GlnAl is essential forgrowth of Mycobacterium tuberculosis inhuman THP- 1 macrophages and guinea pigs.InfectImmun. 71: 3927-3936.van Vliet, A.H., Ketley, J.M., Park, S.F.and Penn, C.W. (2002) The role of iron inCampylobacter gene regulation, metabolismand oxidative stress defense. FEMS MicrobiolRev.26: 173-186.Velayudhan, J. and Kelly, D.J. (2002)Analysis of gluconeogenic and anaplerotic enzymes inCampylobacterjejuni: an essentialrole for phosphoenolpyruvate carboxykinase. Microbiology.148: 685-694.Velayudhan, J., Jones, M.A.,Barrow, P.A. and Kelly, D.J. (2004) L-serine catabolism via anoxygen-labile L-serine dehydratase is essentialfor colonization of the avian gut byCampylobacterjejuni. Infect Immun.72: 260-268.Wang, X., Martindale, J.L. and Holbrook,N.J. (2000) Requirement for ERK activation incisplatin-induced apoptosis. JBiol Chem. 275:3943 5-39443.87Wassenaar, T.M., Engeiskirchen, M., Park, S.and Lastovica, A. (1997) Differentialuptake andkilling potential of Campylobacterjejuni byhuman peripheral monocytes/macrophages.MedMicrobiol Immunol. 186: 139-144.Watson, R.O. and Galan, J.E. (2005) Signaltransduction in Campylobacterjejuni-inducedcytokine production. Cell Microbiol. 7: 655-665.Watson, R.O. and Galan, J.E. (2008)Campylobacter jejuni survives withinepithelial cells byavoiding delivery to lysosomes. FLoS Fathog.4: e14.Westfall, H.N., Rollins, D.M. and Weiss,E. (1986) Substrate utilization byCampylobacterjejuniand Campylobacter coli. AppiEnviron Microbiol. 52: 700-705.Whitehouse, C.A., Balbo, P.B., Pesci, E.C.,Cottle, D.L., Mirabito, P.M. and Pickett,C.L. (1998)Campylobacter jejuni cytolethaldistending toxin causes a G2-phase cellcycle block. InfectImmun. 66: 1934-1940.Wilken, S., Schmees, G. and Schneider,E. (1996) A putative helicaldomain in the Ma1K subunitof the ATP-binding-cassette transportsystem for maltose of Salmonella typhimurium(Ma1FGK2)is crucial for interactionwith Ma1F and Ma1G. A study usingthe LacK protein of Agrobacteriumradiobacter as a tool. Mol Microbiol.22: 655-666.Wischmeyer, P.E., Vanden Hoek, T.L.,Li, C., Shao, Z., Ren, H., Riehm, J. andBecker, L.B.(2003) Glutamine preservescardiomyocyte viability and enhances recovery of contractilefunction after ischemia-reperfusioninjury. JPENJFarenter Enteral Nutr. 27: 116-122.Wyckoff, E.E., Valle, A.M., Smith,S.L. and Payne, S.M. (1999) A multifunctional ATP-bindingcassette transporter system from Vibriocholerae transports vibriobactin and enterobactin. JBacteriol. 181: 7588-7596.Xaus, J., Comalada, M., Valledor, A.F.,Cardo, M., Herrero, C., Soler, C., et al (2001) Molecularmechanisms involved in macrophagesurvival, proliferation, activation or apoptosis.Immunobiology. 204: 543-550.Xia, Z., Dickens, M., Raingeaud, J.,Davis, R.J. and Greenberg, M.E. (1995) Opposing effects ofERK and JNK-p38 MAP kinaseson apoptosis. Science. 270: 1326-133 1.Young, K.T., Davis, L.M. and Dirita,V.J. (2007) Campylobacterjejuni: molecular biology andpathogenesis. Nat Rev Microbiol. 5: 665-679.Yu, R.K., Usuki, S. and Ariga, T. (2006) Gangliosidemolecular mimicry and its pathologicalroles in Guillain-Barre syndromeand related diseases. Infect Immun. 74: 65 17-6527.Zhou, Y., Wang,Q.,Evers, B.M. and Chung, D.H. (2005) Signal transductionpathwaysinvolved in oxidative stress-inducedintestinal epithelial cell apoptosis. Pediatr Res. 58: 1192-1197.88


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items