hopaci-db: host-pseudomonas and coxiella interaction database
TRANSCRIPT
HoPaCI-DB host-Pseudomonas and Coxiellainteraction databaseSophie Bleves1 Irmtraud Dunger2 Mathias C Walter23 Dimitrios Frangoulidis4
Gabi Kastenmuller2 Rome Voulhoux1 and Andreas Ruepp2
1CNRSAix-Marseille University Laboratoire drsquoIngenierie des Systemes Macromoleculaires (UMR7255) Institut deMicrobiologie de la Mediterranee (IMM) 31 Chemin Joseph Aiguier 13402 Marseille cedex 20 France 2Institutefor Bioinformatics and Systems Biology (MIPS) Helmholtz Zentrum Munchen - German Research Center forEnvironmental Health (GmbH) Ingolstadter Landstr 1 D-85764 Neuherberg Germany 3Department of Genome-Oriented Bioinformatics Center of Life and Food Science Weihenstephan Technische Universitat MunchenFreising Germany and 4Bundeswehr Institute of Microbiology Neuherbergstrasse 11 80937 Munich Germany
Received July 17 2013 Revised September 19 2013 Accepted September 21 2013
ABSTRACT
Bacterial infectious diseases are the result of multi-factorial processes affected by the interplay betweenvirulence factors and host targets The host-Pseudomonas and Coxiella interaction database(HoPaCI-DB) is a publicly available manuallycurated integrative database (httpmipshelmholtz-muenchendeHoPaCI) of hostndashpathogen interactiondata from Pseudomonas aeruginosa and Coxiellaburnetii The resource provides structured informa-tion on 3585 experimentally validated interactionsbetween molecules bioprocesses and cellular struc-tures extracted from the scientific literatureSystematic annotation and interactive graphical rep-resentation of disease networks make HoPaCI-DB aversatile knowledge base for biologists and networkbiology approaches
INTRODUCTION
After the discovery of penicillin in 1928 by AlexanderFleming and its use in medicine in the 1940s antibioticswere considered as the most powerful weapon in thewar against bacteria and referred as lsquomagic bulletsrsquoAntibiotics rapidly revolutionized the treatment of infec-tious diseases but within a few years penicillin-resistantbacteria began to appear The enemy was fighting backtill the appearance of lsquosuperbugsrsquo that have become resist-ant to not just one but to many of the commonly usedantibiotics It became evident that to disarm bacteria wemust first learn more about them Perhaps the mostexciting recent advance has been the development of the
interface discipline named lsquocellular microbiologyrsquo (1)which reveals how pathogenic bacteria interact with hostcells in what is turning out to be a complex evolutionarybattle of competing gene products Pathogenesis is amultifactorial process which depends on the immunestatus of the host the nature of the species or strains(virulence factors) and the number of organisms in theinitial exposure This communication occurs at manylevels colonization (mainly adherence) penetration andspread survival in the host and tissue injury A successfulinfection depends on the interplay between virulencefactors and host targets as cellular components (eg cyto-skeleton) and signalling pathways leading to inflamma-tion or apoptosis It is also a multidimensional processas the programme of events is organized in time and spaceand the virulence factors can be switched on and off bycomplex regulatory networks This cross-talk between thepathogen and its host is tightly regulated on both sidesBoth players in the disease game have developed a varietyof mechanisms to counter the othersrsquo defencesExisting information resources in the field of hostndash
pathogen interactions cover a broad spectrum of organ-isms including virus-specific databases (2) bacteria-centricdatabases (3) fungi (4) and databases with a mixedspectrum of organisms (56) However there is a lack ofresources that collect experimentally verified results toannotate comprehensive biological networks on the levelof cellular systems that are at the centre of the hostndashpathogen interactions Here we present HoPaCI-DB adatabase with manually curated information from hostndashpathogen interactions HoPaCI-DB has been developed tofacilitate systems-level analyses for providing betterinsight into the complex networks of pathways and inter-actions that govern epidemiology in Pseudomonas
To whom correspondence should be addressed Tel +49 3187 3189 Fax +49 3187 3585 Email andreasruepphelmholtz-muenchendeCorrespondence may also be addressed to Rome Voulhoux Tel +33 4 91 16 41 26 Fax +33 4 91 71 21 24 Email voulhouximmcnrsfr
The authors wish it to be known that in their opinion the first three authors should be regarded as Joint First Authors
Published online 16 October 2013 Nucleic Acids Research 2014 Vol 42 Database issue D671ndashD676doi101093nargkt925
The Author(s) 2013 Published by Oxford University PressThis is an Open Access article distributed under the terms of the Creative Commons Attribution License (httpcreativecommonsorglicensesby30) whichpermits unrestricted reuse distribution and reproduction in any medium provided the original work is properly cited
at UB
der TU
Muenchen on O
ctober 20 2016httpnaroxfordjournalsorg
Dow
nloaded from
aeruginosa and Coxiella burnetii Multiple search op-tions and interactive graphical presentation of networks(Figures 1 and 2) enable inspection of the manifold inter-relations between heterogeneous disease factors that arerequired for understanding the pathobiology of acute andchronic infections
Curation of hostndashpathogen interactions
HoPaCI-DB covers hostndashpathogen interaction data fromthe pathogenic bacteria P aeruginosa and C burnetiiP aeruginosa is an efficient Gram-negative opportunisticpathogen causing serious acute infections in patients whoare immunocompromised (chemotherapy HIV infection)or who are mechanically ventilated for instance One inten hospital-acquired infections is from PseudomonasP aeruginosa is also responsible for fatal chronic respira-tory disease of patients with cystic fibrosis C burnetti isan obligate intracellular Gram-negative zoonoticpathogen which persists in a large reservoir amongmultiple species (eg small ruminants like sheep cattlegoats) regularly leading to disease outbreaks (7) Inhumans the agent causes Q (Query) fever which mainlyis an acute disease (pneumonia hepatitis) but in up to 2a chronic course of the disease is seen (endocarditis) thatcould be fatal The infection results from inhalation orfrom direct contact with milk urine faeces or birthproducts of infected animals Because of its highly infec-tious nature its high stability in the environment and itsinhalational route of transmission C burnetii isrecognized as a potential agent of bioterrorismUnderstanding of the pathogenic mechanisms com-
prises not only interrelations between pathogenicbacteria and host organisms but also intra-bacterialprocesses such as signalling transduction and hostinternal processes like inflammation The vast majorityof experimental results that have been gained in the fieldin recent decades are hidden in the prose of scientificliterature A comprehensive understanding of thedisease-related processes requires the compilation of thedistributed information in a single resource Furthermoregeneration of a resource that allows further processing ofthe data to perform systems biology analyses bioinfor-matics analyses or graphical representation requires sys-tematic presentation of the information and biocurationby using established biological vocabularies Systematictransformation of complex information such as functionalannotation from free text into biological vocabularies is anon-trivial task and it has been demonstrated that currenttext mining methods are not yet able to produce satisfac-tory results for the extraction of biological information(8) The term lsquoinduction of endothelial cell gapsrsquo (9) forexample cannot easily be transferred into the respectiveGene Ontology term lsquoestablishment of endothelial barrierrsquo(GO0061028) by automated methodsComprehensive information extraction from the bio-
medical publications with high quality of the completedatabase content of HoPaCI-DB is obtained byexperienced biocurators who manually annotate thecomplete articles from peer-reviewed scientific literatureThe detailed manual curation permitted us to richly
annotate the interactions and to place them in theirrelevant context This contextual annotation includesdetails like the bacterial strains used in the experimentsuse of host-model organisms supporting publication celltype cell line and tissue In contrast to genome-centricresources HoPaCI-DB pursues a network-orientedapproach Hostndashpathogen interactions depend on anumber of complex processes such as two-componentsignal transduction systems quorum sensing and iron ac-quisition To provide users an instructive overview aboutthe most important mechanisms we compiled 25 focustopics on the homepage so far Focus topics includenetworks of the responsible disease-relevant factors suchas proteins protein complexes cellular processeschemical compounds or cellular compartments Focustopics (see later in text) are hyperlinked to respectiveweb pages where users can inspect lists of the involvedinteractions statistics about the involved componentsand links to interactive graphical diagrams (Figure 2)
The biological contents offer a meaningful synopsis ofthe pathobiological interaction network The focus topiclsquoAcyl-HSL quorum-sensingrsquo (QS) for example illustratesseveral disease-associated processes in P aeruginosa(i) types II III and VI secretion systems (T236SS) areinversely controlled by QS T2SS and T6SS (loci 2 and 3)are activated while T3SS and T6SS (locus 1) are repressed(ii) Additional interactions show the enzymatic activitiesthat regulate the amount of Acyl-HSL QS signalling mol-ecules (iii) QS regulates biofilm formation and architec-ture and many other virulence factors like siderophoresinvolved in iron acquisition
The C burnetii focus topic type IVB secretion system(T4BSS) demonstrates the current model of the IcmDotT4BSS shown to translocate a large number of bacterialeffector proteins into the host cell during infection(Supplementary Figure S1) The graphical networkfigure reveals the following (i) the constituting genes ofthe secretion machinery itself (ii) the translocated effectorproteins and their localization (iii) the contribution of thephagosome acidification on the effector secretion and (iv)the impact of the T4BSS on phagosome maturationC burnetii replication and host cell death
It should be noted that focus topics are notencapsulated entities of information but can be extendedwith tools offered by the graph viewer (see later in text) Itis a conceptual decision to preferentially annotate litera-ture information that can be extended to larger networkstructures As of July 2013 we have reviewed 218 publi-cations and curated 3585 disease-relevant interactions
Data structure of HoPaCI-DB
For transformation of the biomedical information into adata structure fulfilling the needs of wet lab scientists aswell as for bioinformatics applications information inHoPaCI-DB is structured as three types of information(10) (i) structured information (ii) textual comment and(iii) general information
(i) Core element for biocuration of hostndashpathogen inter-actions is the structured information describing the inter-action between two elements for example between the
D672 Nucleic Acids Research 2014 Vol 42 Database issue
at UB
der TU
Muenchen on O
ctober 20 2016httpnaroxfordjournalsorg
Dow
nloaded from
compound Psl polysaccharide and the phagocytosis-associated bioprocess opsonization (see interaction-ID30776) Subjects and objects are molecules such asproteins nucleic acids or chemical compounds and otherelements like cellular processes phenotypes or environ-mental factors To provide the content of HoPaCI-DBin a standardized format we use names and identifiersfrom established resources like EntrezGene (11) KEGG(12) or CORUM (13) for annotation (ii) The very basic
information of the structural part is complemented by thetextual comment This part provides information concern-ing experimental conditions details about the infectionprocess or exact cellular localization of a process As anexample the modus operandi how P aeruginosa injects asoluble adenylate cyclase ExoY into the cytosol of pul-monary microvascular endothelial cells generating acAMP signal that disrupts the endothelial cell barrier isnot representable by a structured information but requires
Figure 1 The HoPaCI-DB home page and curation of an interaction (a) The HoPaCI-DB home page contains statistics search options and linksto focus topics (b) A manually curated interaction containing general information textual information (comment) and structured information(formal description)
Nucleic Acids Research 2014 Vol 42 Database issue D673
at UB
der TU
Muenchen on O
ctober 20 2016httpnaroxfordjournalsorg
Dow
nloaded from
a textual comment (Interaction-ID 19684) (iii) Thegeneral information includes basic contents such as litera-ture reference or host and pathogen organisms and alsodetails like bacterial strain organism model tissue and cellline Most studies use the Pseudomonas strains PAO1 orPA14 The increased virulence of PA14 is mainly due toExoU a type III secreted potent cytotoxin absent inPAO1 and a mutation in the ladS gene that leads to anelevated T3SS activity and increased cytotoxicity towardsmammalian cells (14)Hostndashpathogen interaction information is enriched with
data from external resources Whenever possible elementsfrom HoPaCI-DB are hyperlinked to information-richdatabase entries from PubMed (15) EntrezGene (11)KEGG (16) Gene Ontology (17) and other resourcesThe complete database information or results from
database searches can be downloaded as flat files or inSystems Biology Markup Language (SBML) a free andopen interchange XML format (18) Files in the SBMLformat can be visualized and analysed with networkanalysis tools such as Cytoscape (19)
Other publicly available resources that provide hostndashpathogen interaction information are rather gene orgenome oriented PATRIC is a bacterial bioinformaticsresource with a focus on human pathogenic species (3)PATRIC includes a built-in system for predicting genesassigning gene functions and reconstructing metabolicpathways The resource links the database informationto a variety of external resources such as KEGG GeneOntology and PDB and offers various analysis tools PHI-base has an even broader scope by providing manuallycurated information of pathogenicity virulence and
Figure 2 Graphical presentation of P aeruginosa T1SS protein secretion systemcomplex focus topic The graph shows the T1SS protein secretionsystemcomplex with functional interactions between proteinsprotein complexes (beige) chemical compounds (green) and biological processes(orange) The T1SS secretion system consists of the Apr Has and Bap systems T1SS (Apr) was found to be specific for the alkaline proteaseAprA and an uncharacterized protein (AprX) The T1SS (Has) is associated with the heme uptake system (Has) playing a role in iron utilizationT1SS (Bap) manages the transport of BapA an adhesin involved in the two-component system PprAB triggered hyper-biofilm phenotype
D674 Nucleic Acids Research 2014 Vol 42 Database issue
at UB
der TU
Muenchen on O
ctober 20 2016httpnaroxfordjournalsorg
Dow
nloaded from
effector genes from fungal oomycete and bacterial patho-gens which infect animal plant fungal and insect hosts(5) PHI-base catalogues validated genes that are requiredfor the disease-causing ability of a microbe genes ofthe host response and verified targets of known bio-active compounds which either kill pathogens or arrestpathogen growthdevelopment Resources such asPATRIC or PHI-base provide a wealth of detailsthat complement the network-oriented approach ofHoPaCI-DB
Search options and visualization
In addition to the predefined focus topics flexible web-based interface of HoPaCI-DB allows to investigate anarea of interest by various search options (Figure 1)The basic search covers all annotated information fromHoPaCI-DB or can be restricted for specific types of in-formation such as lsquogeneproteincomplexrsquo lsquobiologicalprocessrsquo or lsquochemical compoundrsquo To make the search op-eration intuitive all of the information types are listed in alsquodropdown boxrsquo If results of an initial search need refine-ment HoPaCI-DB offers addition of further searches byusing the lsquorefine queryrsquo option Extension or contractionof the search space can be achieved by using one of thethree operators lsquoandrsquo lsquoorrsquo and lsquonotrsquo The search resultsappear as a list that is linked to entries providing thedetails of the manually curated information
In addition search results are linked to a graphical toolthat dynamically generates a graph from the search results(Figure 2) Within the interactive graph interaction infor-mation is shown as colour-coded nodes (nodes are objectssuch as a protein chemical compound or bioprocesses)The nodes are linked via edges defining the mode of inter-action (eg protein A increases_activity of protein B) Thegraph software offers tools for convenient retrieval of theannotated information (i) While moving the mouse cursorover edges pop-up windows appear that present the userimportant information of the interaction such as thecomment and the literature reference (ii) The graph toolallows choosing between two different options organic orhierarchical layout (iii) Other functionalities includeoptions to move nodes within the graph or to extend thegraph with all interaction information about a node ofinterest A description of the different functionalities canbe found on the help pages of HoPaCI-DB
Application of HoPaCI-DB
As said before P aeruginosa causes acute and chronicinfections The choice between those two modes of viru-lence and life styles is under the control of a complexregulatory network involving notably two small RNAsRsmY and RsmZ Through their effect on the post-tran-scriptional RsmA regulator they are controlling key viru-lence factors as flagellum type IV pili biofilm T3- andT6SS The expression of these two small RNAs is crucialfor the attenuated persistence of P aerugniosa in lungs ofinfected people The fine interconnection between thiscascade of regulators and P aeruginosa virulence factorsis now easily accessible to a broader audience thanks tothe graphical view of the HoPaCI-DB (Supplementary
Figure S2) This representation which highlights thecentral role of these small RNAs constitutes a schematicpicture and reveals new targets for the development ofantibacterials Looking for inhibitors of RsmY andRsmZ expression can be an alternative strategy to theuse of antibiotics to fight the infectionMoreover the HoPaCI-DB graphical view presentation
provides a comprehensive overview on the architecturalorganization of the different Pseudomonas nanomachinesinvolved in protein secretion The T2SS for example isconstituted by at least 12 different Xcp proteins The Xcpproteins are recovered in both bacterial membranes andinteract with each other in three sub-complexes an innermembrane platform an outer membrane pore and atransperiplasmic pilus-like structure By clicking on theXcp T2SS graphical view representation in HoPaCI-DBwe have immediate access to all the interactions so fardiscovered among the different components and theirsub-organization in the bacterial envelope This unprece-dented representation opens the search for new proteinndashprotein interactions among those components Moreinterestingly this can be extended to the secreted sub-strates to explore the sequential interactions of the sub-strates with the machinery during the transportDeciphering the assembly mechanism and the T2SSsubstrate interactions may make it possible to identifynew targets for the development of antimicrobialdisruptorsTo date little is known about C burnetii virulence
factors One established observation is that Coxiella ispresent in a specific cellular niche of the host thephagolysosome Several studies in the recent yearsrevealed different parts of the regulatory network whichfacilitates the survival of Coxiella under the harsh condi-tions of the phagolysosome Using the HoPaCI-DB it ispossible to present these parts and its (transitive) connec-tions in an interactive manner As an example weexecuted the simple query lsquophagosome maturationrsquo togive an overview of the different cellular componentsand biological processes involved in this process(Supplementary Figure S3) The complex structure is pre-sented comprehensibly in a graphical manner (based on 39interactions extracted from 15 articles) Phagosome mat-uration triggered by C burnetii infection is increased byseveral pathway structures like acidification endocytosisactin cytoskeleton reorganization and DotIcm type IVBT4BSS It finally ends up in the generation of a specialphagolysosome the parasitophorous vacuole Inside thisstructure both variants of C burnetii (phase I and phaseII) are able to replicate demonstrating that the onlyknown virulence factor LPS is therefore not responsiblefor the survival inside the parasitophorous vacuoleUsing the HoPaCI-DB it is also possible to show inter-
actions of drugs and chemical compounds which offer theopportunity to identify new aspects in the treatment andprophylaxis of Q fever For example in SupplementaryFigure S3 chloramphenicol is involved in the generationof the phagolysosome because it is able to block the fusionof endocytic vesicles and lysosomes and the formation ofthe characteristic large spacious vacuole Anothercompound the antiviral substance and protein inhibitor
Nucleic Acids Research 2014 Vol 42 Database issue D675
at UB
der TU
Muenchen on O
ctober 20 2016httpnaroxfordjournalsorg
Dow
nloaded from
Brefeldin A is able to decrease the size of the Coxiellareplicative vacuoles Even the information presented hereis based on a limited number of publications it shows theadvantages and strength of such an integrative view ofnumerous and heterogeneous data In the future we willupdate the database annually and provide novel hostndashpathogen interactions for HoPaCI-DB
SUPPLEMENTARY DATA
Supplementary Data are available at NAR Online
FUNDING
Funding for open access charge The lsquolsquoPathomicsrsquorsquo ERA-net PATHO Grant [ANR-08-PATH-004-01] supportedthis work (in part) Helmholtz Zentrum Munchen -German Research Center for Environmental Health(GmbH) The German Ministry of Education andResearch (BMBF) under contract No [01KI1001]supported this work (in part)
Conflict of interest statement None declared
REFERENCES
1 CossartP BoquetP NormarkS and RappuoliR (1996)Cellular microbiology emerging Science 271 315ndash316
2 SquiresB MackenC Garcia-SastreA GodboleS NoronhaJHuntV ChangR LarsenCN KlemE BiersackK et al(2008) BioHealthBase informatics support in the elucidation ofinfluenza virus host pathogen interactions and virulence NucleicAcids Res 36 D497ndashD503
3 GillespieJJ WattamAR CammerSA GabbardJLShuklaMP DalayO DriscollT HixD ManeSP MaoCet al (2011) PATRIC the comprehensive bacterial bioinformaticsresource with a focus on human pathogenic species InfectImmun 79 4286ndash4298
4 VialasV Nogales-CadenasR NombelaC Pascual-MontanoAand GilC (2009) Proteopathogen a protein database forstudying Candida albicansmdashhost interaction Proteomics 94664ndash4668
5 WinnenburgR UrbanM BeachamA BaldwinTKHollandS LindebergM HansenH RawlingsC Hammond-KosackKE and KohlerJ (2008) PHI-base update additions tothe pathogen host interaction database Nucleic Acids Res 36D572ndashD576
6 XiangZ TianY and HeY (2007) PHIDIAS a pathogen-hostinteraction data integration and analysis system Genome Biol 8R150
7 CutlerSJ BouzidM and CutlerRR (2007) Q fever J Infect54 313ndash318
8 HirschmanL YehA BlaschkeC and ValenciaA (2005)Overview of BioCreAtIvE critical assessment of informationextraction for biology BMC Bioinformatics 6(Suppl 1) S1
9 SaynerSL BalczonR FrankDW CooperDM andStevensT (2011) Filamin A is a phosphorylation target ofmembrane but not cytosolic adenylyl cyclase activity AmJ Physiol Lung Cell Mol Physiol 301 L117ndashL124
10 LechnerM HohnV BraunerB DungerI FoboGFrishmanG MontroneC KastenmullerG WaegeleB andRueppA (2012) CIDeR multifactorial interaction networks inhuman diseases Genome Biol 13 R62
11 MaglottD OstellJ PruittKD and TatusovaT (2011) EntrezGene gene-centered information at NCBI Nucleic Acids Res 39D52ndashD57
12 KanehisaM GotoS FurumichiM TanabeM andHirakawaM (2010) KEGG for representation and analysis ofmolecular networks involving diseases and drugs Nucleic AcidsRes 38 D355ndashD360
13 RueppA WaegeleB LechnerM BraunerB Dunger-KaltenbachI FoboG FrishmanG MontroneC andMewesHW (2010) CORUM the comprehensive resource ofmammalian protein complexesndash2009 Nucleic Acids Res 38D497ndashD501
14 MikkelsenH McMullanR and FillouxA (2011) ThePseudomonas aeruginosa reference strain PA14 displaysincreased virulence due to a mutation in ladS PLoS One 6e29113
15 NCBI Resource Coordinators (2013) Database resources of theNational Center for Biotechnology Information Nucleic AcidsRes 41 D8ndashD20
16 NakayaA KatayamaT ItohM HiranukaK KawashimaSMoriyaY OkudaS TanakaM TokimatsuT YamanishiYet al (2013) KEGG OC a large-scale automatic construction oftaxonomy-based ortholog clusters Nucleic Acids Res 41D353ndashD357
17 Gene Ontology Consortium (2013) Gene Ontology annotationsand resources Nucleic Acids Res 41 D530ndashD535
18 HuckaM FinneyA SauroHM BolouriH DoyleJCKitanoH ArkinAP BornsteinBJ BrayD Cornish-BowdenA et al (2003) The systems biology markup language(SBML) a medium for representation and exchange ofbiochemical network models Bioinformatics 19 524ndash531
19 ShannonP MarkielA OzierO BaligaNS WangJTRamageD AminN SchwikowskiB and IdekerT (2003)Cytoscape a software environment for integrated modelsof biomolecular interaction networks Genome Res 132498ndash2504
D676 Nucleic Acids Research 2014 Vol 42 Database issue
at UB
der TU
Muenchen on O
ctober 20 2016httpnaroxfordjournalsorg
Dow
nloaded from
aeruginosa and Coxiella burnetii Multiple search op-tions and interactive graphical presentation of networks(Figures 1 and 2) enable inspection of the manifold inter-relations between heterogeneous disease factors that arerequired for understanding the pathobiology of acute andchronic infections
Curation of hostndashpathogen interactions
HoPaCI-DB covers hostndashpathogen interaction data fromthe pathogenic bacteria P aeruginosa and C burnetiiP aeruginosa is an efficient Gram-negative opportunisticpathogen causing serious acute infections in patients whoare immunocompromised (chemotherapy HIV infection)or who are mechanically ventilated for instance One inten hospital-acquired infections is from PseudomonasP aeruginosa is also responsible for fatal chronic respira-tory disease of patients with cystic fibrosis C burnetti isan obligate intracellular Gram-negative zoonoticpathogen which persists in a large reservoir amongmultiple species (eg small ruminants like sheep cattlegoats) regularly leading to disease outbreaks (7) Inhumans the agent causes Q (Query) fever which mainlyis an acute disease (pneumonia hepatitis) but in up to 2a chronic course of the disease is seen (endocarditis) thatcould be fatal The infection results from inhalation orfrom direct contact with milk urine faeces or birthproducts of infected animals Because of its highly infec-tious nature its high stability in the environment and itsinhalational route of transmission C burnetii isrecognized as a potential agent of bioterrorismUnderstanding of the pathogenic mechanisms com-
prises not only interrelations between pathogenicbacteria and host organisms but also intra-bacterialprocesses such as signalling transduction and hostinternal processes like inflammation The vast majorityof experimental results that have been gained in the fieldin recent decades are hidden in the prose of scientificliterature A comprehensive understanding of thedisease-related processes requires the compilation of thedistributed information in a single resource Furthermoregeneration of a resource that allows further processing ofthe data to perform systems biology analyses bioinfor-matics analyses or graphical representation requires sys-tematic presentation of the information and biocurationby using established biological vocabularies Systematictransformation of complex information such as functionalannotation from free text into biological vocabularies is anon-trivial task and it has been demonstrated that currenttext mining methods are not yet able to produce satisfac-tory results for the extraction of biological information(8) The term lsquoinduction of endothelial cell gapsrsquo (9) forexample cannot easily be transferred into the respectiveGene Ontology term lsquoestablishment of endothelial barrierrsquo(GO0061028) by automated methodsComprehensive information extraction from the bio-
medical publications with high quality of the completedatabase content of HoPaCI-DB is obtained byexperienced biocurators who manually annotate thecomplete articles from peer-reviewed scientific literatureThe detailed manual curation permitted us to richly
annotate the interactions and to place them in theirrelevant context This contextual annotation includesdetails like the bacterial strains used in the experimentsuse of host-model organisms supporting publication celltype cell line and tissue In contrast to genome-centricresources HoPaCI-DB pursues a network-orientedapproach Hostndashpathogen interactions depend on anumber of complex processes such as two-componentsignal transduction systems quorum sensing and iron ac-quisition To provide users an instructive overview aboutthe most important mechanisms we compiled 25 focustopics on the homepage so far Focus topics includenetworks of the responsible disease-relevant factors suchas proteins protein complexes cellular processeschemical compounds or cellular compartments Focustopics (see later in text) are hyperlinked to respectiveweb pages where users can inspect lists of the involvedinteractions statistics about the involved componentsand links to interactive graphical diagrams (Figure 2)
The biological contents offer a meaningful synopsis ofthe pathobiological interaction network The focus topiclsquoAcyl-HSL quorum-sensingrsquo (QS) for example illustratesseveral disease-associated processes in P aeruginosa(i) types II III and VI secretion systems (T236SS) areinversely controlled by QS T2SS and T6SS (loci 2 and 3)are activated while T3SS and T6SS (locus 1) are repressed(ii) Additional interactions show the enzymatic activitiesthat regulate the amount of Acyl-HSL QS signalling mol-ecules (iii) QS regulates biofilm formation and architec-ture and many other virulence factors like siderophoresinvolved in iron acquisition
The C burnetii focus topic type IVB secretion system(T4BSS) demonstrates the current model of the IcmDotT4BSS shown to translocate a large number of bacterialeffector proteins into the host cell during infection(Supplementary Figure S1) The graphical networkfigure reveals the following (i) the constituting genes ofthe secretion machinery itself (ii) the translocated effectorproteins and their localization (iii) the contribution of thephagosome acidification on the effector secretion and (iv)the impact of the T4BSS on phagosome maturationC burnetii replication and host cell death
It should be noted that focus topics are notencapsulated entities of information but can be extendedwith tools offered by the graph viewer (see later in text) Itis a conceptual decision to preferentially annotate litera-ture information that can be extended to larger networkstructures As of July 2013 we have reviewed 218 publi-cations and curated 3585 disease-relevant interactions
Data structure of HoPaCI-DB
For transformation of the biomedical information into adata structure fulfilling the needs of wet lab scientists aswell as for bioinformatics applications information inHoPaCI-DB is structured as three types of information(10) (i) structured information (ii) textual comment and(iii) general information
(i) Core element for biocuration of hostndashpathogen inter-actions is the structured information describing the inter-action between two elements for example between the
D672 Nucleic Acids Research 2014 Vol 42 Database issue
at UB
der TU
Muenchen on O
ctober 20 2016httpnaroxfordjournalsorg
Dow
nloaded from
compound Psl polysaccharide and the phagocytosis-associated bioprocess opsonization (see interaction-ID30776) Subjects and objects are molecules such asproteins nucleic acids or chemical compounds and otherelements like cellular processes phenotypes or environ-mental factors To provide the content of HoPaCI-DBin a standardized format we use names and identifiersfrom established resources like EntrezGene (11) KEGG(12) or CORUM (13) for annotation (ii) The very basic
information of the structural part is complemented by thetextual comment This part provides information concern-ing experimental conditions details about the infectionprocess or exact cellular localization of a process As anexample the modus operandi how P aeruginosa injects asoluble adenylate cyclase ExoY into the cytosol of pul-monary microvascular endothelial cells generating acAMP signal that disrupts the endothelial cell barrier isnot representable by a structured information but requires
Figure 1 The HoPaCI-DB home page and curation of an interaction (a) The HoPaCI-DB home page contains statistics search options and linksto focus topics (b) A manually curated interaction containing general information textual information (comment) and structured information(formal description)
Nucleic Acids Research 2014 Vol 42 Database issue D673
at UB
der TU
Muenchen on O
ctober 20 2016httpnaroxfordjournalsorg
Dow
nloaded from
a textual comment (Interaction-ID 19684) (iii) Thegeneral information includes basic contents such as litera-ture reference or host and pathogen organisms and alsodetails like bacterial strain organism model tissue and cellline Most studies use the Pseudomonas strains PAO1 orPA14 The increased virulence of PA14 is mainly due toExoU a type III secreted potent cytotoxin absent inPAO1 and a mutation in the ladS gene that leads to anelevated T3SS activity and increased cytotoxicity towardsmammalian cells (14)Hostndashpathogen interaction information is enriched with
data from external resources Whenever possible elementsfrom HoPaCI-DB are hyperlinked to information-richdatabase entries from PubMed (15) EntrezGene (11)KEGG (16) Gene Ontology (17) and other resourcesThe complete database information or results from
database searches can be downloaded as flat files or inSystems Biology Markup Language (SBML) a free andopen interchange XML format (18) Files in the SBMLformat can be visualized and analysed with networkanalysis tools such as Cytoscape (19)
Other publicly available resources that provide hostndashpathogen interaction information are rather gene orgenome oriented PATRIC is a bacterial bioinformaticsresource with a focus on human pathogenic species (3)PATRIC includes a built-in system for predicting genesassigning gene functions and reconstructing metabolicpathways The resource links the database informationto a variety of external resources such as KEGG GeneOntology and PDB and offers various analysis tools PHI-base has an even broader scope by providing manuallycurated information of pathogenicity virulence and
Figure 2 Graphical presentation of P aeruginosa T1SS protein secretion systemcomplex focus topic The graph shows the T1SS protein secretionsystemcomplex with functional interactions between proteinsprotein complexes (beige) chemical compounds (green) and biological processes(orange) The T1SS secretion system consists of the Apr Has and Bap systems T1SS (Apr) was found to be specific for the alkaline proteaseAprA and an uncharacterized protein (AprX) The T1SS (Has) is associated with the heme uptake system (Has) playing a role in iron utilizationT1SS (Bap) manages the transport of BapA an adhesin involved in the two-component system PprAB triggered hyper-biofilm phenotype
D674 Nucleic Acids Research 2014 Vol 42 Database issue
at UB
der TU
Muenchen on O
ctober 20 2016httpnaroxfordjournalsorg
Dow
nloaded from
effector genes from fungal oomycete and bacterial patho-gens which infect animal plant fungal and insect hosts(5) PHI-base catalogues validated genes that are requiredfor the disease-causing ability of a microbe genes ofthe host response and verified targets of known bio-active compounds which either kill pathogens or arrestpathogen growthdevelopment Resources such asPATRIC or PHI-base provide a wealth of detailsthat complement the network-oriented approach ofHoPaCI-DB
Search options and visualization
In addition to the predefined focus topics flexible web-based interface of HoPaCI-DB allows to investigate anarea of interest by various search options (Figure 1)The basic search covers all annotated information fromHoPaCI-DB or can be restricted for specific types of in-formation such as lsquogeneproteincomplexrsquo lsquobiologicalprocessrsquo or lsquochemical compoundrsquo To make the search op-eration intuitive all of the information types are listed in alsquodropdown boxrsquo If results of an initial search need refine-ment HoPaCI-DB offers addition of further searches byusing the lsquorefine queryrsquo option Extension or contractionof the search space can be achieved by using one of thethree operators lsquoandrsquo lsquoorrsquo and lsquonotrsquo The search resultsappear as a list that is linked to entries providing thedetails of the manually curated information
In addition search results are linked to a graphical toolthat dynamically generates a graph from the search results(Figure 2) Within the interactive graph interaction infor-mation is shown as colour-coded nodes (nodes are objectssuch as a protein chemical compound or bioprocesses)The nodes are linked via edges defining the mode of inter-action (eg protein A increases_activity of protein B) Thegraph software offers tools for convenient retrieval of theannotated information (i) While moving the mouse cursorover edges pop-up windows appear that present the userimportant information of the interaction such as thecomment and the literature reference (ii) The graph toolallows choosing between two different options organic orhierarchical layout (iii) Other functionalities includeoptions to move nodes within the graph or to extend thegraph with all interaction information about a node ofinterest A description of the different functionalities canbe found on the help pages of HoPaCI-DB
Application of HoPaCI-DB
As said before P aeruginosa causes acute and chronicinfections The choice between those two modes of viru-lence and life styles is under the control of a complexregulatory network involving notably two small RNAsRsmY and RsmZ Through their effect on the post-tran-scriptional RsmA regulator they are controlling key viru-lence factors as flagellum type IV pili biofilm T3- andT6SS The expression of these two small RNAs is crucialfor the attenuated persistence of P aerugniosa in lungs ofinfected people The fine interconnection between thiscascade of regulators and P aeruginosa virulence factorsis now easily accessible to a broader audience thanks tothe graphical view of the HoPaCI-DB (Supplementary
Figure S2) This representation which highlights thecentral role of these small RNAs constitutes a schematicpicture and reveals new targets for the development ofantibacterials Looking for inhibitors of RsmY andRsmZ expression can be an alternative strategy to theuse of antibiotics to fight the infectionMoreover the HoPaCI-DB graphical view presentation
provides a comprehensive overview on the architecturalorganization of the different Pseudomonas nanomachinesinvolved in protein secretion The T2SS for example isconstituted by at least 12 different Xcp proteins The Xcpproteins are recovered in both bacterial membranes andinteract with each other in three sub-complexes an innermembrane platform an outer membrane pore and atransperiplasmic pilus-like structure By clicking on theXcp T2SS graphical view representation in HoPaCI-DBwe have immediate access to all the interactions so fardiscovered among the different components and theirsub-organization in the bacterial envelope This unprece-dented representation opens the search for new proteinndashprotein interactions among those components Moreinterestingly this can be extended to the secreted sub-strates to explore the sequential interactions of the sub-strates with the machinery during the transportDeciphering the assembly mechanism and the T2SSsubstrate interactions may make it possible to identifynew targets for the development of antimicrobialdisruptorsTo date little is known about C burnetii virulence
factors One established observation is that Coxiella ispresent in a specific cellular niche of the host thephagolysosome Several studies in the recent yearsrevealed different parts of the regulatory network whichfacilitates the survival of Coxiella under the harsh condi-tions of the phagolysosome Using the HoPaCI-DB it ispossible to present these parts and its (transitive) connec-tions in an interactive manner As an example weexecuted the simple query lsquophagosome maturationrsquo togive an overview of the different cellular componentsand biological processes involved in this process(Supplementary Figure S3) The complex structure is pre-sented comprehensibly in a graphical manner (based on 39interactions extracted from 15 articles) Phagosome mat-uration triggered by C burnetii infection is increased byseveral pathway structures like acidification endocytosisactin cytoskeleton reorganization and DotIcm type IVBT4BSS It finally ends up in the generation of a specialphagolysosome the parasitophorous vacuole Inside thisstructure both variants of C burnetii (phase I and phaseII) are able to replicate demonstrating that the onlyknown virulence factor LPS is therefore not responsiblefor the survival inside the parasitophorous vacuoleUsing the HoPaCI-DB it is also possible to show inter-
actions of drugs and chemical compounds which offer theopportunity to identify new aspects in the treatment andprophylaxis of Q fever For example in SupplementaryFigure S3 chloramphenicol is involved in the generationof the phagolysosome because it is able to block the fusionof endocytic vesicles and lysosomes and the formation ofthe characteristic large spacious vacuole Anothercompound the antiviral substance and protein inhibitor
Nucleic Acids Research 2014 Vol 42 Database issue D675
at UB
der TU
Muenchen on O
ctober 20 2016httpnaroxfordjournalsorg
Dow
nloaded from
Brefeldin A is able to decrease the size of the Coxiellareplicative vacuoles Even the information presented hereis based on a limited number of publications it shows theadvantages and strength of such an integrative view ofnumerous and heterogeneous data In the future we willupdate the database annually and provide novel hostndashpathogen interactions for HoPaCI-DB
SUPPLEMENTARY DATA
Supplementary Data are available at NAR Online
FUNDING
Funding for open access charge The lsquolsquoPathomicsrsquorsquo ERA-net PATHO Grant [ANR-08-PATH-004-01] supportedthis work (in part) Helmholtz Zentrum Munchen -German Research Center for Environmental Health(GmbH) The German Ministry of Education andResearch (BMBF) under contract No [01KI1001]supported this work (in part)
Conflict of interest statement None declared
REFERENCES
1 CossartP BoquetP NormarkS and RappuoliR (1996)Cellular microbiology emerging Science 271 315ndash316
2 SquiresB MackenC Garcia-SastreA GodboleS NoronhaJHuntV ChangR LarsenCN KlemE BiersackK et al(2008) BioHealthBase informatics support in the elucidation ofinfluenza virus host pathogen interactions and virulence NucleicAcids Res 36 D497ndashD503
3 GillespieJJ WattamAR CammerSA GabbardJLShuklaMP DalayO DriscollT HixD ManeSP MaoCet al (2011) PATRIC the comprehensive bacterial bioinformaticsresource with a focus on human pathogenic species InfectImmun 79 4286ndash4298
4 VialasV Nogales-CadenasR NombelaC Pascual-MontanoAand GilC (2009) Proteopathogen a protein database forstudying Candida albicansmdashhost interaction Proteomics 94664ndash4668
5 WinnenburgR UrbanM BeachamA BaldwinTKHollandS LindebergM HansenH RawlingsC Hammond-KosackKE and KohlerJ (2008) PHI-base update additions tothe pathogen host interaction database Nucleic Acids Res 36D572ndashD576
6 XiangZ TianY and HeY (2007) PHIDIAS a pathogen-hostinteraction data integration and analysis system Genome Biol 8R150
7 CutlerSJ BouzidM and CutlerRR (2007) Q fever J Infect54 313ndash318
8 HirschmanL YehA BlaschkeC and ValenciaA (2005)Overview of BioCreAtIvE critical assessment of informationextraction for biology BMC Bioinformatics 6(Suppl 1) S1
9 SaynerSL BalczonR FrankDW CooperDM andStevensT (2011) Filamin A is a phosphorylation target ofmembrane but not cytosolic adenylyl cyclase activity AmJ Physiol Lung Cell Mol Physiol 301 L117ndashL124
10 LechnerM HohnV BraunerB DungerI FoboGFrishmanG MontroneC KastenmullerG WaegeleB andRueppA (2012) CIDeR multifactorial interaction networks inhuman diseases Genome Biol 13 R62
11 MaglottD OstellJ PruittKD and TatusovaT (2011) EntrezGene gene-centered information at NCBI Nucleic Acids Res 39D52ndashD57
12 KanehisaM GotoS FurumichiM TanabeM andHirakawaM (2010) KEGG for representation and analysis ofmolecular networks involving diseases and drugs Nucleic AcidsRes 38 D355ndashD360
13 RueppA WaegeleB LechnerM BraunerB Dunger-KaltenbachI FoboG FrishmanG MontroneC andMewesHW (2010) CORUM the comprehensive resource ofmammalian protein complexesndash2009 Nucleic Acids Res 38D497ndashD501
14 MikkelsenH McMullanR and FillouxA (2011) ThePseudomonas aeruginosa reference strain PA14 displaysincreased virulence due to a mutation in ladS PLoS One 6e29113
15 NCBI Resource Coordinators (2013) Database resources of theNational Center for Biotechnology Information Nucleic AcidsRes 41 D8ndashD20
16 NakayaA KatayamaT ItohM HiranukaK KawashimaSMoriyaY OkudaS TanakaM TokimatsuT YamanishiYet al (2013) KEGG OC a large-scale automatic construction oftaxonomy-based ortholog clusters Nucleic Acids Res 41D353ndashD357
17 Gene Ontology Consortium (2013) Gene Ontology annotationsand resources Nucleic Acids Res 41 D530ndashD535
18 HuckaM FinneyA SauroHM BolouriH DoyleJCKitanoH ArkinAP BornsteinBJ BrayD Cornish-BowdenA et al (2003) The systems biology markup language(SBML) a medium for representation and exchange ofbiochemical network models Bioinformatics 19 524ndash531
19 ShannonP MarkielA OzierO BaligaNS WangJTRamageD AminN SchwikowskiB and IdekerT (2003)Cytoscape a software environment for integrated modelsof biomolecular interaction networks Genome Res 132498ndash2504
D676 Nucleic Acids Research 2014 Vol 42 Database issue
at UB
der TU
Muenchen on O
ctober 20 2016httpnaroxfordjournalsorg
Dow
nloaded from
compound Psl polysaccharide and the phagocytosis-associated bioprocess opsonization (see interaction-ID30776) Subjects and objects are molecules such asproteins nucleic acids or chemical compounds and otherelements like cellular processes phenotypes or environ-mental factors To provide the content of HoPaCI-DBin a standardized format we use names and identifiersfrom established resources like EntrezGene (11) KEGG(12) or CORUM (13) for annotation (ii) The very basic
information of the structural part is complemented by thetextual comment This part provides information concern-ing experimental conditions details about the infectionprocess or exact cellular localization of a process As anexample the modus operandi how P aeruginosa injects asoluble adenylate cyclase ExoY into the cytosol of pul-monary microvascular endothelial cells generating acAMP signal that disrupts the endothelial cell barrier isnot representable by a structured information but requires
Figure 1 The HoPaCI-DB home page and curation of an interaction (a) The HoPaCI-DB home page contains statistics search options and linksto focus topics (b) A manually curated interaction containing general information textual information (comment) and structured information(formal description)
Nucleic Acids Research 2014 Vol 42 Database issue D673
at UB
der TU
Muenchen on O
ctober 20 2016httpnaroxfordjournalsorg
Dow
nloaded from
a textual comment (Interaction-ID 19684) (iii) Thegeneral information includes basic contents such as litera-ture reference or host and pathogen organisms and alsodetails like bacterial strain organism model tissue and cellline Most studies use the Pseudomonas strains PAO1 orPA14 The increased virulence of PA14 is mainly due toExoU a type III secreted potent cytotoxin absent inPAO1 and a mutation in the ladS gene that leads to anelevated T3SS activity and increased cytotoxicity towardsmammalian cells (14)Hostndashpathogen interaction information is enriched with
data from external resources Whenever possible elementsfrom HoPaCI-DB are hyperlinked to information-richdatabase entries from PubMed (15) EntrezGene (11)KEGG (16) Gene Ontology (17) and other resourcesThe complete database information or results from
database searches can be downloaded as flat files or inSystems Biology Markup Language (SBML) a free andopen interchange XML format (18) Files in the SBMLformat can be visualized and analysed with networkanalysis tools such as Cytoscape (19)
Other publicly available resources that provide hostndashpathogen interaction information are rather gene orgenome oriented PATRIC is a bacterial bioinformaticsresource with a focus on human pathogenic species (3)PATRIC includes a built-in system for predicting genesassigning gene functions and reconstructing metabolicpathways The resource links the database informationto a variety of external resources such as KEGG GeneOntology and PDB and offers various analysis tools PHI-base has an even broader scope by providing manuallycurated information of pathogenicity virulence and
Figure 2 Graphical presentation of P aeruginosa T1SS protein secretion systemcomplex focus topic The graph shows the T1SS protein secretionsystemcomplex with functional interactions between proteinsprotein complexes (beige) chemical compounds (green) and biological processes(orange) The T1SS secretion system consists of the Apr Has and Bap systems T1SS (Apr) was found to be specific for the alkaline proteaseAprA and an uncharacterized protein (AprX) The T1SS (Has) is associated with the heme uptake system (Has) playing a role in iron utilizationT1SS (Bap) manages the transport of BapA an adhesin involved in the two-component system PprAB triggered hyper-biofilm phenotype
D674 Nucleic Acids Research 2014 Vol 42 Database issue
at UB
der TU
Muenchen on O
ctober 20 2016httpnaroxfordjournalsorg
Dow
nloaded from
effector genes from fungal oomycete and bacterial patho-gens which infect animal plant fungal and insect hosts(5) PHI-base catalogues validated genes that are requiredfor the disease-causing ability of a microbe genes ofthe host response and verified targets of known bio-active compounds which either kill pathogens or arrestpathogen growthdevelopment Resources such asPATRIC or PHI-base provide a wealth of detailsthat complement the network-oriented approach ofHoPaCI-DB
Search options and visualization
In addition to the predefined focus topics flexible web-based interface of HoPaCI-DB allows to investigate anarea of interest by various search options (Figure 1)The basic search covers all annotated information fromHoPaCI-DB or can be restricted for specific types of in-formation such as lsquogeneproteincomplexrsquo lsquobiologicalprocessrsquo or lsquochemical compoundrsquo To make the search op-eration intuitive all of the information types are listed in alsquodropdown boxrsquo If results of an initial search need refine-ment HoPaCI-DB offers addition of further searches byusing the lsquorefine queryrsquo option Extension or contractionof the search space can be achieved by using one of thethree operators lsquoandrsquo lsquoorrsquo and lsquonotrsquo The search resultsappear as a list that is linked to entries providing thedetails of the manually curated information
In addition search results are linked to a graphical toolthat dynamically generates a graph from the search results(Figure 2) Within the interactive graph interaction infor-mation is shown as colour-coded nodes (nodes are objectssuch as a protein chemical compound or bioprocesses)The nodes are linked via edges defining the mode of inter-action (eg protein A increases_activity of protein B) Thegraph software offers tools for convenient retrieval of theannotated information (i) While moving the mouse cursorover edges pop-up windows appear that present the userimportant information of the interaction such as thecomment and the literature reference (ii) The graph toolallows choosing between two different options organic orhierarchical layout (iii) Other functionalities includeoptions to move nodes within the graph or to extend thegraph with all interaction information about a node ofinterest A description of the different functionalities canbe found on the help pages of HoPaCI-DB
Application of HoPaCI-DB
As said before P aeruginosa causes acute and chronicinfections The choice between those two modes of viru-lence and life styles is under the control of a complexregulatory network involving notably two small RNAsRsmY and RsmZ Through their effect on the post-tran-scriptional RsmA regulator they are controlling key viru-lence factors as flagellum type IV pili biofilm T3- andT6SS The expression of these two small RNAs is crucialfor the attenuated persistence of P aerugniosa in lungs ofinfected people The fine interconnection between thiscascade of regulators and P aeruginosa virulence factorsis now easily accessible to a broader audience thanks tothe graphical view of the HoPaCI-DB (Supplementary
Figure S2) This representation which highlights thecentral role of these small RNAs constitutes a schematicpicture and reveals new targets for the development ofantibacterials Looking for inhibitors of RsmY andRsmZ expression can be an alternative strategy to theuse of antibiotics to fight the infectionMoreover the HoPaCI-DB graphical view presentation
provides a comprehensive overview on the architecturalorganization of the different Pseudomonas nanomachinesinvolved in protein secretion The T2SS for example isconstituted by at least 12 different Xcp proteins The Xcpproteins are recovered in both bacterial membranes andinteract with each other in three sub-complexes an innermembrane platform an outer membrane pore and atransperiplasmic pilus-like structure By clicking on theXcp T2SS graphical view representation in HoPaCI-DBwe have immediate access to all the interactions so fardiscovered among the different components and theirsub-organization in the bacterial envelope This unprece-dented representation opens the search for new proteinndashprotein interactions among those components Moreinterestingly this can be extended to the secreted sub-strates to explore the sequential interactions of the sub-strates with the machinery during the transportDeciphering the assembly mechanism and the T2SSsubstrate interactions may make it possible to identifynew targets for the development of antimicrobialdisruptorsTo date little is known about C burnetii virulence
factors One established observation is that Coxiella ispresent in a specific cellular niche of the host thephagolysosome Several studies in the recent yearsrevealed different parts of the regulatory network whichfacilitates the survival of Coxiella under the harsh condi-tions of the phagolysosome Using the HoPaCI-DB it ispossible to present these parts and its (transitive) connec-tions in an interactive manner As an example weexecuted the simple query lsquophagosome maturationrsquo togive an overview of the different cellular componentsand biological processes involved in this process(Supplementary Figure S3) The complex structure is pre-sented comprehensibly in a graphical manner (based on 39interactions extracted from 15 articles) Phagosome mat-uration triggered by C burnetii infection is increased byseveral pathway structures like acidification endocytosisactin cytoskeleton reorganization and DotIcm type IVBT4BSS It finally ends up in the generation of a specialphagolysosome the parasitophorous vacuole Inside thisstructure both variants of C burnetii (phase I and phaseII) are able to replicate demonstrating that the onlyknown virulence factor LPS is therefore not responsiblefor the survival inside the parasitophorous vacuoleUsing the HoPaCI-DB it is also possible to show inter-
actions of drugs and chemical compounds which offer theopportunity to identify new aspects in the treatment andprophylaxis of Q fever For example in SupplementaryFigure S3 chloramphenicol is involved in the generationof the phagolysosome because it is able to block the fusionof endocytic vesicles and lysosomes and the formation ofthe characteristic large spacious vacuole Anothercompound the antiviral substance and protein inhibitor
Nucleic Acids Research 2014 Vol 42 Database issue D675
at UB
der TU
Muenchen on O
ctober 20 2016httpnaroxfordjournalsorg
Dow
nloaded from
Brefeldin A is able to decrease the size of the Coxiellareplicative vacuoles Even the information presented hereis based on a limited number of publications it shows theadvantages and strength of such an integrative view ofnumerous and heterogeneous data In the future we willupdate the database annually and provide novel hostndashpathogen interactions for HoPaCI-DB
SUPPLEMENTARY DATA
Supplementary Data are available at NAR Online
FUNDING
Funding for open access charge The lsquolsquoPathomicsrsquorsquo ERA-net PATHO Grant [ANR-08-PATH-004-01] supportedthis work (in part) Helmholtz Zentrum Munchen -German Research Center for Environmental Health(GmbH) The German Ministry of Education andResearch (BMBF) under contract No [01KI1001]supported this work (in part)
Conflict of interest statement None declared
REFERENCES
1 CossartP BoquetP NormarkS and RappuoliR (1996)Cellular microbiology emerging Science 271 315ndash316
2 SquiresB MackenC Garcia-SastreA GodboleS NoronhaJHuntV ChangR LarsenCN KlemE BiersackK et al(2008) BioHealthBase informatics support in the elucidation ofinfluenza virus host pathogen interactions and virulence NucleicAcids Res 36 D497ndashD503
3 GillespieJJ WattamAR CammerSA GabbardJLShuklaMP DalayO DriscollT HixD ManeSP MaoCet al (2011) PATRIC the comprehensive bacterial bioinformaticsresource with a focus on human pathogenic species InfectImmun 79 4286ndash4298
4 VialasV Nogales-CadenasR NombelaC Pascual-MontanoAand GilC (2009) Proteopathogen a protein database forstudying Candida albicansmdashhost interaction Proteomics 94664ndash4668
5 WinnenburgR UrbanM BeachamA BaldwinTKHollandS LindebergM HansenH RawlingsC Hammond-KosackKE and KohlerJ (2008) PHI-base update additions tothe pathogen host interaction database Nucleic Acids Res 36D572ndashD576
6 XiangZ TianY and HeY (2007) PHIDIAS a pathogen-hostinteraction data integration and analysis system Genome Biol 8R150
7 CutlerSJ BouzidM and CutlerRR (2007) Q fever J Infect54 313ndash318
8 HirschmanL YehA BlaschkeC and ValenciaA (2005)Overview of BioCreAtIvE critical assessment of informationextraction for biology BMC Bioinformatics 6(Suppl 1) S1
9 SaynerSL BalczonR FrankDW CooperDM andStevensT (2011) Filamin A is a phosphorylation target ofmembrane but not cytosolic adenylyl cyclase activity AmJ Physiol Lung Cell Mol Physiol 301 L117ndashL124
10 LechnerM HohnV BraunerB DungerI FoboGFrishmanG MontroneC KastenmullerG WaegeleB andRueppA (2012) CIDeR multifactorial interaction networks inhuman diseases Genome Biol 13 R62
11 MaglottD OstellJ PruittKD and TatusovaT (2011) EntrezGene gene-centered information at NCBI Nucleic Acids Res 39D52ndashD57
12 KanehisaM GotoS FurumichiM TanabeM andHirakawaM (2010) KEGG for representation and analysis ofmolecular networks involving diseases and drugs Nucleic AcidsRes 38 D355ndashD360
13 RueppA WaegeleB LechnerM BraunerB Dunger-KaltenbachI FoboG FrishmanG MontroneC andMewesHW (2010) CORUM the comprehensive resource ofmammalian protein complexesndash2009 Nucleic Acids Res 38D497ndashD501
14 MikkelsenH McMullanR and FillouxA (2011) ThePseudomonas aeruginosa reference strain PA14 displaysincreased virulence due to a mutation in ladS PLoS One 6e29113
15 NCBI Resource Coordinators (2013) Database resources of theNational Center for Biotechnology Information Nucleic AcidsRes 41 D8ndashD20
16 NakayaA KatayamaT ItohM HiranukaK KawashimaSMoriyaY OkudaS TanakaM TokimatsuT YamanishiYet al (2013) KEGG OC a large-scale automatic construction oftaxonomy-based ortholog clusters Nucleic Acids Res 41D353ndashD357
17 Gene Ontology Consortium (2013) Gene Ontology annotationsand resources Nucleic Acids Res 41 D530ndashD535
18 HuckaM FinneyA SauroHM BolouriH DoyleJCKitanoH ArkinAP BornsteinBJ BrayD Cornish-BowdenA et al (2003) The systems biology markup language(SBML) a medium for representation and exchange ofbiochemical network models Bioinformatics 19 524ndash531
19 ShannonP MarkielA OzierO BaligaNS WangJTRamageD AminN SchwikowskiB and IdekerT (2003)Cytoscape a software environment for integrated modelsof biomolecular interaction networks Genome Res 132498ndash2504
D676 Nucleic Acids Research 2014 Vol 42 Database issue
at UB
der TU
Muenchen on O
ctober 20 2016httpnaroxfordjournalsorg
Dow
nloaded from
a textual comment (Interaction-ID 19684) (iii) Thegeneral information includes basic contents such as litera-ture reference or host and pathogen organisms and alsodetails like bacterial strain organism model tissue and cellline Most studies use the Pseudomonas strains PAO1 orPA14 The increased virulence of PA14 is mainly due toExoU a type III secreted potent cytotoxin absent inPAO1 and a mutation in the ladS gene that leads to anelevated T3SS activity and increased cytotoxicity towardsmammalian cells (14)Hostndashpathogen interaction information is enriched with
data from external resources Whenever possible elementsfrom HoPaCI-DB are hyperlinked to information-richdatabase entries from PubMed (15) EntrezGene (11)KEGG (16) Gene Ontology (17) and other resourcesThe complete database information or results from
database searches can be downloaded as flat files or inSystems Biology Markup Language (SBML) a free andopen interchange XML format (18) Files in the SBMLformat can be visualized and analysed with networkanalysis tools such as Cytoscape (19)
Other publicly available resources that provide hostndashpathogen interaction information are rather gene orgenome oriented PATRIC is a bacterial bioinformaticsresource with a focus on human pathogenic species (3)PATRIC includes a built-in system for predicting genesassigning gene functions and reconstructing metabolicpathways The resource links the database informationto a variety of external resources such as KEGG GeneOntology and PDB and offers various analysis tools PHI-base has an even broader scope by providing manuallycurated information of pathogenicity virulence and
Figure 2 Graphical presentation of P aeruginosa T1SS protein secretion systemcomplex focus topic The graph shows the T1SS protein secretionsystemcomplex with functional interactions between proteinsprotein complexes (beige) chemical compounds (green) and biological processes(orange) The T1SS secretion system consists of the Apr Has and Bap systems T1SS (Apr) was found to be specific for the alkaline proteaseAprA and an uncharacterized protein (AprX) The T1SS (Has) is associated with the heme uptake system (Has) playing a role in iron utilizationT1SS (Bap) manages the transport of BapA an adhesin involved in the two-component system PprAB triggered hyper-biofilm phenotype
D674 Nucleic Acids Research 2014 Vol 42 Database issue
at UB
der TU
Muenchen on O
ctober 20 2016httpnaroxfordjournalsorg
Dow
nloaded from
effector genes from fungal oomycete and bacterial patho-gens which infect animal plant fungal and insect hosts(5) PHI-base catalogues validated genes that are requiredfor the disease-causing ability of a microbe genes ofthe host response and verified targets of known bio-active compounds which either kill pathogens or arrestpathogen growthdevelopment Resources such asPATRIC or PHI-base provide a wealth of detailsthat complement the network-oriented approach ofHoPaCI-DB
Search options and visualization
In addition to the predefined focus topics flexible web-based interface of HoPaCI-DB allows to investigate anarea of interest by various search options (Figure 1)The basic search covers all annotated information fromHoPaCI-DB or can be restricted for specific types of in-formation such as lsquogeneproteincomplexrsquo lsquobiologicalprocessrsquo or lsquochemical compoundrsquo To make the search op-eration intuitive all of the information types are listed in alsquodropdown boxrsquo If results of an initial search need refine-ment HoPaCI-DB offers addition of further searches byusing the lsquorefine queryrsquo option Extension or contractionof the search space can be achieved by using one of thethree operators lsquoandrsquo lsquoorrsquo and lsquonotrsquo The search resultsappear as a list that is linked to entries providing thedetails of the manually curated information
In addition search results are linked to a graphical toolthat dynamically generates a graph from the search results(Figure 2) Within the interactive graph interaction infor-mation is shown as colour-coded nodes (nodes are objectssuch as a protein chemical compound or bioprocesses)The nodes are linked via edges defining the mode of inter-action (eg protein A increases_activity of protein B) Thegraph software offers tools for convenient retrieval of theannotated information (i) While moving the mouse cursorover edges pop-up windows appear that present the userimportant information of the interaction such as thecomment and the literature reference (ii) The graph toolallows choosing between two different options organic orhierarchical layout (iii) Other functionalities includeoptions to move nodes within the graph or to extend thegraph with all interaction information about a node ofinterest A description of the different functionalities canbe found on the help pages of HoPaCI-DB
Application of HoPaCI-DB
As said before P aeruginosa causes acute and chronicinfections The choice between those two modes of viru-lence and life styles is under the control of a complexregulatory network involving notably two small RNAsRsmY and RsmZ Through their effect on the post-tran-scriptional RsmA regulator they are controlling key viru-lence factors as flagellum type IV pili biofilm T3- andT6SS The expression of these two small RNAs is crucialfor the attenuated persistence of P aerugniosa in lungs ofinfected people The fine interconnection between thiscascade of regulators and P aeruginosa virulence factorsis now easily accessible to a broader audience thanks tothe graphical view of the HoPaCI-DB (Supplementary
Figure S2) This representation which highlights thecentral role of these small RNAs constitutes a schematicpicture and reveals new targets for the development ofantibacterials Looking for inhibitors of RsmY andRsmZ expression can be an alternative strategy to theuse of antibiotics to fight the infectionMoreover the HoPaCI-DB graphical view presentation
provides a comprehensive overview on the architecturalorganization of the different Pseudomonas nanomachinesinvolved in protein secretion The T2SS for example isconstituted by at least 12 different Xcp proteins The Xcpproteins are recovered in both bacterial membranes andinteract with each other in three sub-complexes an innermembrane platform an outer membrane pore and atransperiplasmic pilus-like structure By clicking on theXcp T2SS graphical view representation in HoPaCI-DBwe have immediate access to all the interactions so fardiscovered among the different components and theirsub-organization in the bacterial envelope This unprece-dented representation opens the search for new proteinndashprotein interactions among those components Moreinterestingly this can be extended to the secreted sub-strates to explore the sequential interactions of the sub-strates with the machinery during the transportDeciphering the assembly mechanism and the T2SSsubstrate interactions may make it possible to identifynew targets for the development of antimicrobialdisruptorsTo date little is known about C burnetii virulence
factors One established observation is that Coxiella ispresent in a specific cellular niche of the host thephagolysosome Several studies in the recent yearsrevealed different parts of the regulatory network whichfacilitates the survival of Coxiella under the harsh condi-tions of the phagolysosome Using the HoPaCI-DB it ispossible to present these parts and its (transitive) connec-tions in an interactive manner As an example weexecuted the simple query lsquophagosome maturationrsquo togive an overview of the different cellular componentsand biological processes involved in this process(Supplementary Figure S3) The complex structure is pre-sented comprehensibly in a graphical manner (based on 39interactions extracted from 15 articles) Phagosome mat-uration triggered by C burnetii infection is increased byseveral pathway structures like acidification endocytosisactin cytoskeleton reorganization and DotIcm type IVBT4BSS It finally ends up in the generation of a specialphagolysosome the parasitophorous vacuole Inside thisstructure both variants of C burnetii (phase I and phaseII) are able to replicate demonstrating that the onlyknown virulence factor LPS is therefore not responsiblefor the survival inside the parasitophorous vacuoleUsing the HoPaCI-DB it is also possible to show inter-
actions of drugs and chemical compounds which offer theopportunity to identify new aspects in the treatment andprophylaxis of Q fever For example in SupplementaryFigure S3 chloramphenicol is involved in the generationof the phagolysosome because it is able to block the fusionof endocytic vesicles and lysosomes and the formation ofthe characteristic large spacious vacuole Anothercompound the antiviral substance and protein inhibitor
Nucleic Acids Research 2014 Vol 42 Database issue D675
at UB
der TU
Muenchen on O
ctober 20 2016httpnaroxfordjournalsorg
Dow
nloaded from
Brefeldin A is able to decrease the size of the Coxiellareplicative vacuoles Even the information presented hereis based on a limited number of publications it shows theadvantages and strength of such an integrative view ofnumerous and heterogeneous data In the future we willupdate the database annually and provide novel hostndashpathogen interactions for HoPaCI-DB
SUPPLEMENTARY DATA
Supplementary Data are available at NAR Online
FUNDING
Funding for open access charge The lsquolsquoPathomicsrsquorsquo ERA-net PATHO Grant [ANR-08-PATH-004-01] supportedthis work (in part) Helmholtz Zentrum Munchen -German Research Center for Environmental Health(GmbH) The German Ministry of Education andResearch (BMBF) under contract No [01KI1001]supported this work (in part)
Conflict of interest statement None declared
REFERENCES
1 CossartP BoquetP NormarkS and RappuoliR (1996)Cellular microbiology emerging Science 271 315ndash316
2 SquiresB MackenC Garcia-SastreA GodboleS NoronhaJHuntV ChangR LarsenCN KlemE BiersackK et al(2008) BioHealthBase informatics support in the elucidation ofinfluenza virus host pathogen interactions and virulence NucleicAcids Res 36 D497ndashD503
3 GillespieJJ WattamAR CammerSA GabbardJLShuklaMP DalayO DriscollT HixD ManeSP MaoCet al (2011) PATRIC the comprehensive bacterial bioinformaticsresource with a focus on human pathogenic species InfectImmun 79 4286ndash4298
4 VialasV Nogales-CadenasR NombelaC Pascual-MontanoAand GilC (2009) Proteopathogen a protein database forstudying Candida albicansmdashhost interaction Proteomics 94664ndash4668
5 WinnenburgR UrbanM BeachamA BaldwinTKHollandS LindebergM HansenH RawlingsC Hammond-KosackKE and KohlerJ (2008) PHI-base update additions tothe pathogen host interaction database Nucleic Acids Res 36D572ndashD576
6 XiangZ TianY and HeY (2007) PHIDIAS a pathogen-hostinteraction data integration and analysis system Genome Biol 8R150
7 CutlerSJ BouzidM and CutlerRR (2007) Q fever J Infect54 313ndash318
8 HirschmanL YehA BlaschkeC and ValenciaA (2005)Overview of BioCreAtIvE critical assessment of informationextraction for biology BMC Bioinformatics 6(Suppl 1) S1
9 SaynerSL BalczonR FrankDW CooperDM andStevensT (2011) Filamin A is a phosphorylation target ofmembrane but not cytosolic adenylyl cyclase activity AmJ Physiol Lung Cell Mol Physiol 301 L117ndashL124
10 LechnerM HohnV BraunerB DungerI FoboGFrishmanG MontroneC KastenmullerG WaegeleB andRueppA (2012) CIDeR multifactorial interaction networks inhuman diseases Genome Biol 13 R62
11 MaglottD OstellJ PruittKD and TatusovaT (2011) EntrezGene gene-centered information at NCBI Nucleic Acids Res 39D52ndashD57
12 KanehisaM GotoS FurumichiM TanabeM andHirakawaM (2010) KEGG for representation and analysis ofmolecular networks involving diseases and drugs Nucleic AcidsRes 38 D355ndashD360
13 RueppA WaegeleB LechnerM BraunerB Dunger-KaltenbachI FoboG FrishmanG MontroneC andMewesHW (2010) CORUM the comprehensive resource ofmammalian protein complexesndash2009 Nucleic Acids Res 38D497ndashD501
14 MikkelsenH McMullanR and FillouxA (2011) ThePseudomonas aeruginosa reference strain PA14 displaysincreased virulence due to a mutation in ladS PLoS One 6e29113
15 NCBI Resource Coordinators (2013) Database resources of theNational Center for Biotechnology Information Nucleic AcidsRes 41 D8ndashD20
16 NakayaA KatayamaT ItohM HiranukaK KawashimaSMoriyaY OkudaS TanakaM TokimatsuT YamanishiYet al (2013) KEGG OC a large-scale automatic construction oftaxonomy-based ortholog clusters Nucleic Acids Res 41D353ndashD357
17 Gene Ontology Consortium (2013) Gene Ontology annotationsand resources Nucleic Acids Res 41 D530ndashD535
18 HuckaM FinneyA SauroHM BolouriH DoyleJCKitanoH ArkinAP BornsteinBJ BrayD Cornish-BowdenA et al (2003) The systems biology markup language(SBML) a medium for representation and exchange ofbiochemical network models Bioinformatics 19 524ndash531
19 ShannonP MarkielA OzierO BaligaNS WangJTRamageD AminN SchwikowskiB and IdekerT (2003)Cytoscape a software environment for integrated modelsof biomolecular interaction networks Genome Res 132498ndash2504
D676 Nucleic Acids Research 2014 Vol 42 Database issue
at UB
der TU
Muenchen on O
ctober 20 2016httpnaroxfordjournalsorg
Dow
nloaded from
effector genes from fungal oomycete and bacterial patho-gens which infect animal plant fungal and insect hosts(5) PHI-base catalogues validated genes that are requiredfor the disease-causing ability of a microbe genes ofthe host response and verified targets of known bio-active compounds which either kill pathogens or arrestpathogen growthdevelopment Resources such asPATRIC or PHI-base provide a wealth of detailsthat complement the network-oriented approach ofHoPaCI-DB
Search options and visualization
In addition to the predefined focus topics flexible web-based interface of HoPaCI-DB allows to investigate anarea of interest by various search options (Figure 1)The basic search covers all annotated information fromHoPaCI-DB or can be restricted for specific types of in-formation such as lsquogeneproteincomplexrsquo lsquobiologicalprocessrsquo or lsquochemical compoundrsquo To make the search op-eration intuitive all of the information types are listed in alsquodropdown boxrsquo If results of an initial search need refine-ment HoPaCI-DB offers addition of further searches byusing the lsquorefine queryrsquo option Extension or contractionof the search space can be achieved by using one of thethree operators lsquoandrsquo lsquoorrsquo and lsquonotrsquo The search resultsappear as a list that is linked to entries providing thedetails of the manually curated information
In addition search results are linked to a graphical toolthat dynamically generates a graph from the search results(Figure 2) Within the interactive graph interaction infor-mation is shown as colour-coded nodes (nodes are objectssuch as a protein chemical compound or bioprocesses)The nodes are linked via edges defining the mode of inter-action (eg protein A increases_activity of protein B) Thegraph software offers tools for convenient retrieval of theannotated information (i) While moving the mouse cursorover edges pop-up windows appear that present the userimportant information of the interaction such as thecomment and the literature reference (ii) The graph toolallows choosing between two different options organic orhierarchical layout (iii) Other functionalities includeoptions to move nodes within the graph or to extend thegraph with all interaction information about a node ofinterest A description of the different functionalities canbe found on the help pages of HoPaCI-DB
Application of HoPaCI-DB
As said before P aeruginosa causes acute and chronicinfections The choice between those two modes of viru-lence and life styles is under the control of a complexregulatory network involving notably two small RNAsRsmY and RsmZ Through their effect on the post-tran-scriptional RsmA regulator they are controlling key viru-lence factors as flagellum type IV pili biofilm T3- andT6SS The expression of these two small RNAs is crucialfor the attenuated persistence of P aerugniosa in lungs ofinfected people The fine interconnection between thiscascade of regulators and P aeruginosa virulence factorsis now easily accessible to a broader audience thanks tothe graphical view of the HoPaCI-DB (Supplementary
Figure S2) This representation which highlights thecentral role of these small RNAs constitutes a schematicpicture and reveals new targets for the development ofantibacterials Looking for inhibitors of RsmY andRsmZ expression can be an alternative strategy to theuse of antibiotics to fight the infectionMoreover the HoPaCI-DB graphical view presentation
provides a comprehensive overview on the architecturalorganization of the different Pseudomonas nanomachinesinvolved in protein secretion The T2SS for example isconstituted by at least 12 different Xcp proteins The Xcpproteins are recovered in both bacterial membranes andinteract with each other in three sub-complexes an innermembrane platform an outer membrane pore and atransperiplasmic pilus-like structure By clicking on theXcp T2SS graphical view representation in HoPaCI-DBwe have immediate access to all the interactions so fardiscovered among the different components and theirsub-organization in the bacterial envelope This unprece-dented representation opens the search for new proteinndashprotein interactions among those components Moreinterestingly this can be extended to the secreted sub-strates to explore the sequential interactions of the sub-strates with the machinery during the transportDeciphering the assembly mechanism and the T2SSsubstrate interactions may make it possible to identifynew targets for the development of antimicrobialdisruptorsTo date little is known about C burnetii virulence
factors One established observation is that Coxiella ispresent in a specific cellular niche of the host thephagolysosome Several studies in the recent yearsrevealed different parts of the regulatory network whichfacilitates the survival of Coxiella under the harsh condi-tions of the phagolysosome Using the HoPaCI-DB it ispossible to present these parts and its (transitive) connec-tions in an interactive manner As an example weexecuted the simple query lsquophagosome maturationrsquo togive an overview of the different cellular componentsand biological processes involved in this process(Supplementary Figure S3) The complex structure is pre-sented comprehensibly in a graphical manner (based on 39interactions extracted from 15 articles) Phagosome mat-uration triggered by C burnetii infection is increased byseveral pathway structures like acidification endocytosisactin cytoskeleton reorganization and DotIcm type IVBT4BSS It finally ends up in the generation of a specialphagolysosome the parasitophorous vacuole Inside thisstructure both variants of C burnetii (phase I and phaseII) are able to replicate demonstrating that the onlyknown virulence factor LPS is therefore not responsiblefor the survival inside the parasitophorous vacuoleUsing the HoPaCI-DB it is also possible to show inter-
actions of drugs and chemical compounds which offer theopportunity to identify new aspects in the treatment andprophylaxis of Q fever For example in SupplementaryFigure S3 chloramphenicol is involved in the generationof the phagolysosome because it is able to block the fusionof endocytic vesicles and lysosomes and the formation ofthe characteristic large spacious vacuole Anothercompound the antiviral substance and protein inhibitor
Nucleic Acids Research 2014 Vol 42 Database issue D675
at UB
der TU
Muenchen on O
ctober 20 2016httpnaroxfordjournalsorg
Dow
nloaded from
Brefeldin A is able to decrease the size of the Coxiellareplicative vacuoles Even the information presented hereis based on a limited number of publications it shows theadvantages and strength of such an integrative view ofnumerous and heterogeneous data In the future we willupdate the database annually and provide novel hostndashpathogen interactions for HoPaCI-DB
SUPPLEMENTARY DATA
Supplementary Data are available at NAR Online
FUNDING
Funding for open access charge The lsquolsquoPathomicsrsquorsquo ERA-net PATHO Grant [ANR-08-PATH-004-01] supportedthis work (in part) Helmholtz Zentrum Munchen -German Research Center for Environmental Health(GmbH) The German Ministry of Education andResearch (BMBF) under contract No [01KI1001]supported this work (in part)
Conflict of interest statement None declared
REFERENCES
1 CossartP BoquetP NormarkS and RappuoliR (1996)Cellular microbiology emerging Science 271 315ndash316
2 SquiresB MackenC Garcia-SastreA GodboleS NoronhaJHuntV ChangR LarsenCN KlemE BiersackK et al(2008) BioHealthBase informatics support in the elucidation ofinfluenza virus host pathogen interactions and virulence NucleicAcids Res 36 D497ndashD503
3 GillespieJJ WattamAR CammerSA GabbardJLShuklaMP DalayO DriscollT HixD ManeSP MaoCet al (2011) PATRIC the comprehensive bacterial bioinformaticsresource with a focus on human pathogenic species InfectImmun 79 4286ndash4298
4 VialasV Nogales-CadenasR NombelaC Pascual-MontanoAand GilC (2009) Proteopathogen a protein database forstudying Candida albicansmdashhost interaction Proteomics 94664ndash4668
5 WinnenburgR UrbanM BeachamA BaldwinTKHollandS LindebergM HansenH RawlingsC Hammond-KosackKE and KohlerJ (2008) PHI-base update additions tothe pathogen host interaction database Nucleic Acids Res 36D572ndashD576
6 XiangZ TianY and HeY (2007) PHIDIAS a pathogen-hostinteraction data integration and analysis system Genome Biol 8R150
7 CutlerSJ BouzidM and CutlerRR (2007) Q fever J Infect54 313ndash318
8 HirschmanL YehA BlaschkeC and ValenciaA (2005)Overview of BioCreAtIvE critical assessment of informationextraction for biology BMC Bioinformatics 6(Suppl 1) S1
9 SaynerSL BalczonR FrankDW CooperDM andStevensT (2011) Filamin A is a phosphorylation target ofmembrane but not cytosolic adenylyl cyclase activity AmJ Physiol Lung Cell Mol Physiol 301 L117ndashL124
10 LechnerM HohnV BraunerB DungerI FoboGFrishmanG MontroneC KastenmullerG WaegeleB andRueppA (2012) CIDeR multifactorial interaction networks inhuman diseases Genome Biol 13 R62
11 MaglottD OstellJ PruittKD and TatusovaT (2011) EntrezGene gene-centered information at NCBI Nucleic Acids Res 39D52ndashD57
12 KanehisaM GotoS FurumichiM TanabeM andHirakawaM (2010) KEGG for representation and analysis ofmolecular networks involving diseases and drugs Nucleic AcidsRes 38 D355ndashD360
13 RueppA WaegeleB LechnerM BraunerB Dunger-KaltenbachI FoboG FrishmanG MontroneC andMewesHW (2010) CORUM the comprehensive resource ofmammalian protein complexesndash2009 Nucleic Acids Res 38D497ndashD501
14 MikkelsenH McMullanR and FillouxA (2011) ThePseudomonas aeruginosa reference strain PA14 displaysincreased virulence due to a mutation in ladS PLoS One 6e29113
15 NCBI Resource Coordinators (2013) Database resources of theNational Center for Biotechnology Information Nucleic AcidsRes 41 D8ndashD20
16 NakayaA KatayamaT ItohM HiranukaK KawashimaSMoriyaY OkudaS TanakaM TokimatsuT YamanishiYet al (2013) KEGG OC a large-scale automatic construction oftaxonomy-based ortholog clusters Nucleic Acids Res 41D353ndashD357
17 Gene Ontology Consortium (2013) Gene Ontology annotationsand resources Nucleic Acids Res 41 D530ndashD535
18 HuckaM FinneyA SauroHM BolouriH DoyleJCKitanoH ArkinAP BornsteinBJ BrayD Cornish-BowdenA et al (2003) The systems biology markup language(SBML) a medium for representation and exchange ofbiochemical network models Bioinformatics 19 524ndash531
19 ShannonP MarkielA OzierO BaligaNS WangJTRamageD AminN SchwikowskiB and IdekerT (2003)Cytoscape a software environment for integrated modelsof biomolecular interaction networks Genome Res 132498ndash2504
D676 Nucleic Acids Research 2014 Vol 42 Database issue
at UB
der TU
Muenchen on O
ctober 20 2016httpnaroxfordjournalsorg
Dow
nloaded from
Brefeldin A is able to decrease the size of the Coxiellareplicative vacuoles Even the information presented hereis based on a limited number of publications it shows theadvantages and strength of such an integrative view ofnumerous and heterogeneous data In the future we willupdate the database annually and provide novel hostndashpathogen interactions for HoPaCI-DB
SUPPLEMENTARY DATA
Supplementary Data are available at NAR Online
FUNDING
Funding for open access charge The lsquolsquoPathomicsrsquorsquo ERA-net PATHO Grant [ANR-08-PATH-004-01] supportedthis work (in part) Helmholtz Zentrum Munchen -German Research Center for Environmental Health(GmbH) The German Ministry of Education andResearch (BMBF) under contract No [01KI1001]supported this work (in part)
Conflict of interest statement None declared
REFERENCES
1 CossartP BoquetP NormarkS and RappuoliR (1996)Cellular microbiology emerging Science 271 315ndash316
2 SquiresB MackenC Garcia-SastreA GodboleS NoronhaJHuntV ChangR LarsenCN KlemE BiersackK et al(2008) BioHealthBase informatics support in the elucidation ofinfluenza virus host pathogen interactions and virulence NucleicAcids Res 36 D497ndashD503
3 GillespieJJ WattamAR CammerSA GabbardJLShuklaMP DalayO DriscollT HixD ManeSP MaoCet al (2011) PATRIC the comprehensive bacterial bioinformaticsresource with a focus on human pathogenic species InfectImmun 79 4286ndash4298
4 VialasV Nogales-CadenasR NombelaC Pascual-MontanoAand GilC (2009) Proteopathogen a protein database forstudying Candida albicansmdashhost interaction Proteomics 94664ndash4668
5 WinnenburgR UrbanM BeachamA BaldwinTKHollandS LindebergM HansenH RawlingsC Hammond-KosackKE and KohlerJ (2008) PHI-base update additions tothe pathogen host interaction database Nucleic Acids Res 36D572ndashD576
6 XiangZ TianY and HeY (2007) PHIDIAS a pathogen-hostinteraction data integration and analysis system Genome Biol 8R150
7 CutlerSJ BouzidM and CutlerRR (2007) Q fever J Infect54 313ndash318
8 HirschmanL YehA BlaschkeC and ValenciaA (2005)Overview of BioCreAtIvE critical assessment of informationextraction for biology BMC Bioinformatics 6(Suppl 1) S1
9 SaynerSL BalczonR FrankDW CooperDM andStevensT (2011) Filamin A is a phosphorylation target ofmembrane but not cytosolic adenylyl cyclase activity AmJ Physiol Lung Cell Mol Physiol 301 L117ndashL124
10 LechnerM HohnV BraunerB DungerI FoboGFrishmanG MontroneC KastenmullerG WaegeleB andRueppA (2012) CIDeR multifactorial interaction networks inhuman diseases Genome Biol 13 R62
11 MaglottD OstellJ PruittKD and TatusovaT (2011) EntrezGene gene-centered information at NCBI Nucleic Acids Res 39D52ndashD57
12 KanehisaM GotoS FurumichiM TanabeM andHirakawaM (2010) KEGG for representation and analysis ofmolecular networks involving diseases and drugs Nucleic AcidsRes 38 D355ndashD360
13 RueppA WaegeleB LechnerM BraunerB Dunger-KaltenbachI FoboG FrishmanG MontroneC andMewesHW (2010) CORUM the comprehensive resource ofmammalian protein complexesndash2009 Nucleic Acids Res 38D497ndashD501
14 MikkelsenH McMullanR and FillouxA (2011) ThePseudomonas aeruginosa reference strain PA14 displaysincreased virulence due to a mutation in ladS PLoS One 6e29113
15 NCBI Resource Coordinators (2013) Database resources of theNational Center for Biotechnology Information Nucleic AcidsRes 41 D8ndashD20
16 NakayaA KatayamaT ItohM HiranukaK KawashimaSMoriyaY OkudaS TanakaM TokimatsuT YamanishiYet al (2013) KEGG OC a large-scale automatic construction oftaxonomy-based ortholog clusters Nucleic Acids Res 41D353ndashD357
17 Gene Ontology Consortium (2013) Gene Ontology annotationsand resources Nucleic Acids Res 41 D530ndashD535
18 HuckaM FinneyA SauroHM BolouriH DoyleJCKitanoH ArkinAP BornsteinBJ BrayD Cornish-BowdenA et al (2003) The systems biology markup language(SBML) a medium for representation and exchange ofbiochemical network models Bioinformatics 19 524ndash531
19 ShannonP MarkielA OzierO BaligaNS WangJTRamageD AminN SchwikowskiB and IdekerT (2003)Cytoscape a software environment for integrated modelsof biomolecular interaction networks Genome Res 132498ndash2504
D676 Nucleic Acids Research 2014 Vol 42 Database issue
at UB
der TU
Muenchen on O
ctober 20 2016httpnaroxfordjournalsorg
Dow
nloaded from