[ Pobierz całość w formacie PDF ]
//-->.pos {position:absolute; z-index: 0; left: 0px; top: 0px;}Infection with 2009 H1N1 Influenza VirusPrimes for Immunological Memory inHuman Nose-Associated Lymphoid Tissue,Offering Cross-Reactive Immunity to H1N1and Avian H5N1 VirusesWaleed H. Mahallawi, Anand V. Kasbekar, Maxwell S.McCormick, Katja Hoschler, Nigel Temperton, Samuel C.Leong, Helen Beer, Francesca Ferrara, Paul S. McNamaraand Qibo ZhangJ. Virol.2013, 87(10):5331. DOI: 10.1128/JVI.03547-12.Published Ahead of Print 6 March 2013.Downloaded from http://jvi.asm.org/ on May 20, 2014 by guestDownloaded from http://jvi.asm.org/ on May 20, 2014 by guestUpdated information and services can be found at:http://jvi.asm.org/content/87/10/5331These include:REFERENCESThis article cites 43 articles, 19 of which can be accessed freeat:http://jvi.asm.org/content/87/10/5331#ref-list-1Receive: RSS Feeds, eTOCs, free email alerts (when newarticles cite this article),more»CONTENT ALERTSInformation about commercial reprint orders:http://journals.asm.org/site/misc/reprints.xhtmlTo subscribe to to another ASM Journal go to:http://journals.asm.org/site/subscriptions/Infection with 2009 H1N1 Influenza Virus Primes for ImmunologicalMemory in Human Nose-Associated Lymphoid Tissue, OfferingCross-Reactive Immunity to H1N1 and Avian H5N1 VirusesWaleed H. Mahallawi,aAnand V. Kasbekar,bMaxwell S. McCormick,cKatja Hoschler,dNigel Temperton,eSamuel C. Leong,cHelen Beer,cFrancesca Ferrara,ePaul S. McNamara,fQibo ZhangaDepartment of Clinical Infection, Microbiology and Immunology, Institute of Infection and Global Health, University of Liverpool, Liverpool, United Kingdoma; ENTDepartment, Alder Hey Children’s Hospital, Liverpool, United Kingdomb; ENT Department, Royal Liverpool University Hospital, Liverpool, United Kingdomc; RespiratoryVirus Unit, Health Protection Agency, London, United Kingdomd; Viral Pseudotype Unit, School of Pharmacy, University of Kent, Kent, United Kingdome; Institute of ChildHealth, University of Liverpool, Liverpool, United KingdomfInfluenza is a highly contagious mucosal infection in the respiratory tract. The 2009 pandemic H1N1 (pH1N1) influenza virusinfection resulted in substantial morbidity and mortality in humans. Little is known on whether immunological memory devel-ops following pH1N1 infection and whether it provides protection against other virus subtypes. An enzyme-linked immunosor-bent spot assay was used to analyze hemagglutinin (HA)-specific memory B cell responses after virus antigen stimulation innose-associated lymphoid tissues (NALT) from children and adults. Individuals with serological evidence of previous exposureto pH1N1 showed significant cross-reactive HA-specific memory B cell responses to pH1N1, seasonal H1N1 (sH1N1), and avianH5N1 (aH5N1) viruses upon pH1N1 virus stimulation. pH1N1 virus antigen elicited stronger cross-reactive memory B cell re-sponses than sH1N1 virus. Intriguingly, aH5N1 virus also activated cross-reactive memory responses to sH1N1 and pH1N1 HAsin those who had previous pH1N1 exposure, and that correlated well with the memory response stimulated by pH1N1 virus anti-gen. These memory B cell responses resulted in cross-reactive neutralizing antibodies against sH1N1, 1918 H1N1, and aH5N1viruses. The 2009 pH1N1 infection appeared to have primed human host with B cell memory in NALT that offers cross-protec-tive mucosal immunity to not only H1N1 but also aH5N1 viruses. These findings may have important implications for futurevaccination strategies against influenza. It will be important to induce and/or enhance such cross-protective mucosal memory Bcells.Downloaded from http://jvi.asm.org/ on May 20, 2014 by guestnfluenza is a highly contagious and acute respiratory infectioncaused by influenza virus in the mucosa of the respiratory tract(1). Both seasonal and pandemic influenza virus infections con-tinue to cause substantial morbidity and mortality in humans. The2009 pandemic H1N1 (pH1N1) influenza virus and the potentialof a highly pathogenic pandemic avian H5N1 (aH5N1) influenzavirus highlighted the need for effective preventative strategies.Understanding the development of natural immunity followingthe pH1N1 pandemic may provide important information onhost protective immunity in humans, which could inform futurevaccination strategies against influenza.The pH1N1 virus was antigenically different from seasonalH1N1 (sH1N1) viruses and affected large population groups whowere immunologically naïve to the virus (2–4). Little is known onthe development of immunological memory following thepH1N1 virus infection, how it interacts with other influenza vi-ruses, and whether this memory provides any protective immu-nity to aH5N1 virus, a pathogen with considerable potential tocause a future pandemic.Surface hemagglutinin (HA) is a major virulence factor crucialfor virus binding to host cell membrane and essential in the in-duction of host protective immunity. HA-specific antibodies playa key role in protection against influenza (5,6).During the 2009pH1N1 pandemic, older people ( 65 years) were protected be-cause they had existing anti-HA antibodies induced by previousexposure to antigenically related H1N1 strains, e.g., pandemicA/H1N1 1918 virus or strains circulating before 1957 (4,7, 8).Structurally, HA consists of two domains: a globular head, com-Iposed of part of HA1, and a stalk structure, composed of portionsof HA1 and all of HA2 (9). The globular head contains the variableregion of HA and is the major target for neutralizing antibodiesthat inhibit virus binding to target cells. These neutralizing anti-bodies are traditionally detected by hemagglutination inhibitionassay (HAI). The stalk domain is more conserved. Recent studieshave suggested that antibodies targeting the stalk region may alsohave neutralizing activity and may contribute to the cross-reactiveimmunity to different influenza viruses induced by either infec-tion or vaccination (10–13). There are 16 different influenza virussubtypes of HA, and they are clustered into two groups based onthe molecular relatedness of the HA sequences, group 1 (H1, H2,H5, H6, H8, H9, H11, H12, H13, and H16) and group 2 (H3, H4,H7, H10, H14, and H15) (14).Influenza virus is transmitted through airborne droplets andinfects human nasopharyngeal mucosa. Human adenoids andtonsils are major components of nose-associated lymphoid tissues(NALT) which are considered to be an important part of the mu-cosal immune system (15–17). However, studies have shown thereReceived25 December 2012Accepted28 February 2013Published ahead of print6 March 2013Address correspondence to Qibo Zhang, qibo.zhang@liv.ac.uk.Copyright © 2013, American Society for Microbiology. All Rights Reserved.doi:10.1128/JVI.03547-12The authors have paid a fee to allow immediate free access to this article.May 2013 Volume 87 Number 10Journal of Virologyp. 5331–5339jvi.asm.org5331Mahallawi et al.are some major differences between human NALT in the naso-pharynx and other mucosal compartments such as Peyer’s patchesin the intestine. B cells in the former predominantly produce IgG,whereas the majority of B cells in the latter produce IgA (18,19).We demonstrated previously that pneumococcal protein antigenselicited a predominantly IgG memory B cell response in humanNALT presumably primed by previous colonization (20,21).TheNALTs are considered to be important induction sites for bothmucosal and systemic immunity to upper respiratory pathogens,including influenza virus (16,22–24).The induction of immuno-logical memory against influenza virus most likely involves theseimmunocompetent NALTs, where antigen-specific memory Bcells are primed. However, limited data exist on the developmentand function of such memory B cells in humans. Recent studiesusing monoclonal antibodies from B cells isolated from patientsinfected with either the 1918 or 2009 pandemic H1N1 virusessuggest the presence of memory B cells (25–27). It was also re-ported that some HA-specific monoclonal antibodies isolatedfrom these patients were cross-reactive with the stalk regions ofHAs of a number of different influenza virus strains (13,28).In this study, we investigated the HA-specific memory B cellresponses in human NALT to pH1N1, sH1N1, sH3N2, andaH5N1 viruses. We demonstrated that patients who had serolog-ical evidence of previous exposure to pH1N1 virus showed mem-ory B cell response in NALTs that produce cross-reactive neutral-izing antibodies against a number of influenza virus subtypesupon pH1N1 virus antigen stimulation. The result suggests thatthe 2009 pH1N1 infection primed human host with cross-reactivemucosal memory responses to other H1N1 and the highly patho-genic aH5N1 virus strains. These findings may have importantimplications in future vaccination strategies against influenza.(This study was presented in part at the European Conferenceof Immunology, 5 to 8 September 2012, Glasgow, United King-dom [abstract P-0445] [29]).MATERIALS AND METHODSPatients and samples.Adenoids and tonsils were obtained from childrenand adults (3 to 30 years of age) undergoing adenoidectomy and/or ton-sillectomy between March 2011 and March 2012. A venous blood samplewas obtained. Patients who were previously vaccinated against influenzaor who were immunocompromised in any way were excluded. The studywas approved by the local ethics committee (Liverpool Pediatric ResearchEthics Committee) and written, informed consent obtained from eachpatient/parent as appropriate.Influenza virus antigens.Influenza virus antigens for cell stimulationexperiments were -propiolactone-inactivated, partially purified whole-virus antigens from the National Institute for Biological Standards andControl (NIBSC, United Kingdom) and were used following a standardprocedure as described previously (30). The pH1N1, sH1N1, sH3N2,and aH5N1 virus antigens were derived from A/California/04/2009,A/Brisbane/59/2007, A/Brisbane/10/2007, and A/Vietnam/1203/2004virus strains, respectively.Recombinant HA.Purified recombinant HA proteins of pH1N1 (A/California/04/2009), sH1N1 (A/Brisbane/59/2007), sH3N2 (A/Brisbane/10/2007), aH5N1 (A/Vietnam/1203/2004), H2N2 (A/Singapore/1/57),and H7N3 (A/Canada/RV444/04) virus were from the Biodefense andEmerging Infections Research Resources Repository, ATCC (Manassas,VA). The recombinant HAs of pH1N1 and sH1N1 contain a C-terminalhistidine tag and were produced in High Five insect cells using a baculo-virus expression vector system (31). The HAs were purified from cellculture supernatant by immobilized-metal affinity chromatography(IMAC) and contain a trimerizing (foldon) domain (31). The recombi-nant HAs of sH3N2, aH5N1, H2N2, and H7N3 viruses were full-lengthglycosylated HAs that were produced in Sf9 insect cells using a baculovirusexpression vector system, membrane extracted from infected cells, andpurified by affinity chromatography under native conditions that pre-served their biological activity and tertiary structure. The purified HAforms trimers (32).Cell separation.Adenoidal and tonsillar tissues were transported tothe laboratory in Hanks buffered salt solution supplemented with glu-tamine and antibiotics (penicillin, 100 U/ml; streptomycin, 100 g/ml).Mononuclear cells (MNC) from adenoids and tonsils were isolated usingFicoll density centrifugation following methods described previously (20,33).In some experiments, memory T cells (CD45RO ) or memory B cells(CD27 ) were depleted from adenotonsillar MNC using magnetic-acti-vated cell sorting (MACS) with magnetic microbeads (Miltinyi) beforecell stimulation (20,33).Cell culture and stimulation by influenza virus antigens.Adenoton-sillar MNC were cultured at 4106/ml in RPMI medium containingglutamine, penicillin, streptomycin, and 10% fetal bovine serum (FBS),with and without a predetermined optimal concentration of influenzavirus antigens. For enumerating antibody-secreting cells (ASC) by en-zyme-linked immunospot (ELISpot) assay, adenotonsillar MNC werecultured for 5 days before being transferred to ELISpot plates. Cell culturesupernatants were collected at day 7 and stored at 70°C until assay formeasuring antibodies by enzyme-linked immunosorbent assay (ELISA).Paired experiments in adenoidal and tonsillar MNC revealed no dif-ference in memory B cell responses activated by influenza virus antigens(data not shown). Therefore, data derived from tonsillar MNC only arepresented in this paper.Measurement of memory B cell response by ELISpot assay.HA-spe-cific memory B cell responses following individual virus antigen stimula-tions were analyzed using an ELISpot assay to enumerate HA-specific ASCas described previously (34). Briefly, ELISpot plates (Millipore, UnitedKingdom) were coated overnight with optimized concentrations of re-combinant HAs in phosphate-buffered saline (PBS). Plates were washedand blocked by incubation with RPMI medium containing 10% FBS at37°C for 2 h. Antigen-stimulated MNC were added to the plates andincubated overnight at 37°C. Plates were washed and incubated with bio-tinylated anti-human IgG/IgA antibody (Invitrogen, United Kingdom)for 30 min at room temperature. After washing, avidin D-horseradishperoxidase (HRP) conjugate (Vector Laboratories) was added and themixture was incubated. Colored spots were developed with the addition ofsubstrate (3-amino-9-ethylcarbazole; Sigma) and counted using an auto-mated ELISpot reader (AID; Autoimmune Diagnostika GmbH, Ger-many). The ELISpot assay shows the predominance of HA-specific IgGmemory B cell responses following stimulation by influenza virus anti-gens, so only IgG ASC results are shown.Measurement of HA-specific antibodies by ELISA.HA-specific IgGantibodies were analyzed following a standard ELISA procedure as previ-ously described (35). In brief, ELISA plates were coated with recombinantHAs and incubated overnight at 4°C. After washing, plates were blockedwith 10% FBS followed by incubation of cell culture supernatants at pre-determined optimized dilutions for 1.5 h. Alkaline phosphatase-conju-gated anti-human IgG (Sigma) was then incubated for 1.5 h. After wash-ing, p-nitrophenyl phosphate substrate was applied. Optical density wasmeasured at 405 nm, and data were analyzed using DeltaSoft microplateanalysis software (BioMetallics Inc.). Sandoglobulin (Sandoz, UnitedKingdom), which contains high titers of antibodies to sH1N1 and sH3N2HA, was used as a reference standard for measurement of antibodies tosH1N1 and sH3N2. A human convalescent-phase serum sample from asubject with confirmed pH1N1 infection (BEI Resources, ATCC) wasused as a standard for measurement of anti-pH1N1 HA antibodies. Bothreference standards were arbitrarily assigned an antibody titer of 5,000U/ml.HAI assay.Hemagglutination inhibition (HAI) assays were per-formed following standard methods (8) at the Microbiology Services—Downloaded from http://jvi.asm.org/ on May 20, 2014 by guest5332jvi.asm.orgJournal of VirologypH1N1 Primes Cross-Reactive Memory B Cell ResponseFIG 1pH1N1 virus antigen induces a strong HA-specific memory B cell response. (a) Panels A and B show representative samples from patients with HAI 40and HAI 40, respectively. (b) Numbers of HA-specific IgG ASC in tonsillar MNC enumerated by ELISpot assay after stimulation by pH1N1 virus antigen insubjects with serum anti-pH1N1 HAI titer 40 (n 20) and low HAI titers ( 40,n14). Horizontal lines represent the means of the numbers of HA-specificASC (P 0.001 compared with those with HAI 40).Downloaded from http://jvi.asm.org/ on May 20, 2014 by guestColindale, Health Protection Agency (London, United Kingdom). Thevirus strains used included the following. For the pandemic H1N1 virus,NIBRG122 virus is a reassortant prepared from A/England/195/2009(H1N1v), the prototype United Kingdom isolate antigenically and genet-ically closely related to A/California/4/2009; for the seasonal H1N1 virus,the A/H1N1/Brisbane/59/2007 strain was used; for the seasonal H3N2virus, the A/H3N2/Brisbane/10/2007 strain was used; and for the avianH5N1 virus, NIBRG-14 virus, a reassortant prepared from A/H5N1/Viet-nam/1194/2004 virus, was used.Influenza pseudotype virus production and neutralization assay.The construction of lentiviral pseudotypes with an HA envelope glyco-protein derived from the highly pathogenic H5N1 avian influenza virus(A/Viet Nam/1194/04) has been described previously (36). H1N1 HA-expressing plasmids were constructed for A/Brisbane/59/2007 (H1N1)and A/South Carolina/1/18 (H1N1) virus using analogous methodolo-gies. Pseudotype viruses were produced by cotransfection of HEK293T/17cells with the respective HA plasmids, the HIV gag-pol plasmid p8.91, andthe reporter plasmid pCSFLW (expressing firefly luciferase) usingFugene-6 transfection reagent (Roche, United Kingdom). For the produc-tion of the H1N1 pseudotypes, protease-expressing plasmid was alsoadded to the transfection mixture as described previously (11). The HAcontent was normalized via a surrogate readout of the firefly relative lightunits (RLU)/ml for each virus. For the virus neutralization assays, cellculture supernatant samples were 2-fold serially diluted in culture me-dium and mixed with each pseudotype virus (1 106RLU firefly lucifer-ase input) at a 1:1 (vol/vol) ratio. After incubation at 37°C for 1 h, 1 104HEK293T cells were added to each well of a white 96-well flat-bottomedtissue culture plate. Firefly RLU values were determined 72 h later byluminometry using a Bright-Glo assay system (Promega, United King-dom).Statistical analysis.Differences in memory response or antibody ti-ters between different groups were analyzed by analysis of variance andStudent’sttest. Association between two factors was analyzed by Pear-son’s correlation. APvalue of 0.05 was considered statistically signifi-cant. Statistical analysis was performed using SPSS software (version 16).RESULTSThe 2009 pH1N1 virus induces memory B cell responses thatcross-react with sH1N1 and aH5N1 viruses.To analyze pH1N1HA-specific memory B cell responses in tonsillar MNC, anELISpot assay was performed to enumerate numbers of HA anti-gen-specific ASC after stimulation with pH1N1 virus antigen.Large numbers (mean ASC/106MNC, 50.0) of HA-specific IgGASC responding to pH1N1 virus were found after pH1N1 virusantigen stimulation in subjects with a serum anti-pH1N1 HAItiter 40, whereas minimal numbers (5.5) of ASC were seen inthose with low ( 40) HAI titers (Fig.1;P0.01).In contrast, when the subjects were divided into two groupswith serum HAI titers 40 and 40 against either sH1N1 orsH3N2 viruses, there was no difference between the two groups inthe numbers of pH1N1 HA-specific IgG ASC after pH1N1 antigenstimulation (P 0.05; data not shown).To determine whether this pH1N1 HA-specific memory B cellresponse cross-reacted with other influenza A virus subtypes,memory B cell responses to sH1N1, sH3N2, and aH5N1 HAs werealso analyzed following tonsillar MNC stimulation with pH1N1virus antigen. Numbers of HA-specific IgG ASC responding tosH1N1 and aH5N1 virus after the antigen stimulation in subjectswith serum anti-pH1N1 HAI titer 40 were significantly higherthan in those who had anti-pH1N1 HAI titer 40 (36.2 versus 8.7and 35.0 versus 6.1 for anti-sH1N1 and -aH5N1 ASC, respec-tively) (Fig.2aandb;P0.01). Further analysis revealed a goodcorrelation (r0.73,P0.001) between the number of HA-specific ASC responding to pH1N1 and that responding to sH1N1virus (Fig.2c).However, no difference between subjects with anti-pH1N1 HAI titer 40 and those with HAI titer 40 was found inthe numbers of specific IgG ASC responding to sH3N2 HA afterpH1N1 virus antigen stimulation (Fig.2b).pH1N1 virus elicits stronger cross-reactive memory B cellresponses than sH1N1 and sH3N2 virus antigens.To comparepandemic and seasonal influenza A virus-induced memory B cellresponses and their cross-reactivities, HA-specific memory B cellresponses in tonsillar MNC following stimulation with sH1N1andsH3N2 virus antigens were analyzed. Stimulation with the sH1N1virus antigen (A/Brisbane/59/2007) elicited a modest increase inthe number of HA-specific ASC responding to sH1N1 (meanASC/106MNC, 24.8) and pH1N1 (26.5), but no increase was seenin the number of specific ASC responding to sH3N2 and aH5N1HAs (Fig.3a)either in patients with anti-sH1N1 HAI 40 or inthose with anti-sH1N1 HAI 40 (data not shown). Stimulationswith the A/New Caledonia/20/99 H1N1 virus antigen resulted insimilar numbers of HA-specific ASC responding to sH1N1 (mean,22.8) and pH1N1 (21.0). This was in contrast to the strongercross-reactive memory B cell responses elicited by pH1N1 antigenstimulation not only in response to pH1N1 (50.8) but also inresponse to sH1N1 (38.0) and aH5N1 (32.7) viruses in patientsMay 2013 Volume 87 Number 10jvi.asm.org5333Mahallawi et al.Downloaded from http://jvi.asm.org/ on May 20, 2014 by guestFIG 2pH1N1 virus antigen elicits memory B cell responses that cross-react with sH1N1 and avian H5N1 viruses. (a) The magnitudes of HA-specific IgGmemory B cell responses to sH1N1 (Œ), pH1N1 (o), aH5N1 ( ), and sH3N2 ( ) in tonsillar MNC were analyzed after pH1N1 virus antigen stimulation andcompared between subjects with serum anti-pH1N1 HAI titer 40 (filled symbols) and those with HAI 40 (open symbols) (*,P0.01). (b) ELISpot imagesof HA-specific ASC responding to sH1N1, pH1N1, aH5N1, and sH3N2 in tonsillar MNC from one representative patient after pH1N1 antigen stimulation.Control, negative control with no specific HA antigen coating in ELISpot assay. (c) There was a good correlation between the numbers of HA-specific ASCresponding to pH1N1 and that responding to sH1N1 after pH1N1 antigen stimulation (r 0.73,P0.001).with anti-pH1N1 HAI titer 40 (Fig.3a).Nevertheless, there wasa positive correlation between the pH1N1 HA-specific memory Bcell response activated by the pH1N1 antigen and the sH1N1 HA-specific memory response elicited by the sH1N1 antigen stimula-tion (Fig.3b;r0.88;P0.001). In comparison, stimulationwith the sH3N2 antigen did not induce an increase in the numberof HA-specific ASC responding to the sH1N1, pH1N1, andaH5N1 virus, although it did induce a strong increase in the num-ber of ASC responding to sH3N2 HA (49.6) (Fig.3c).Avian H5N1 virus antigen elicits a cross-reactive memory Bcell response similar to that seen with pH1N1 antigen.We rea-soned that if pH1N1 infection in patients induced memory B cellscross-reactive to aH5N1, these cells should mount a memory re-sponse upon an antigenic challenge by aH5N1 virus. TonsillarMNC were stimulated with aH5N1 virus antigen followed by anal-ysis of HA-specific ASC. Indeed, this stimulation elicited memoryB cell responses, with mean numbers of IgG ASC responding tosH1N1 (20.8) and pH1N1 virus (50.8) HAs similar to the numbersinduced by pH1N1 antigen in patients who had an anti-pH1N1HAI titer 40 (Fig.4a).A moderate response to aH5N1 (14.6) butnot sH3N2 virus HA was also observed in these patients (Fig.4a).Figure 4bshows that the numbers of pH1N1 HA-specific ASCelicited by aH5N1 antigen stimulation correlated well with thatelicited by pH1N1 antigen stimulation (r 0.85;P0.01). Nosignificant ASC response to HA of any of the four viruses wasfound after aH5N1 antigen stimulation in subjects with an anti-pH1N1 HAI titer 40.Further analysis using memory T cell (CD45RO )- or mem-ory B cell (CD27 )-depleted tonsillar MNC failed to show anysignificant numbers of HA-specific ASC by ELISpot assay aftereach virus antigen stimulation (data not shown). This suggeststhat the HA-specific ASCs detected were derived from memoryrather than naive B cells in tonsillar MNC.pH1N1 virus activates memory B cell responses that producecross-reactive neutralizing antibodies.As expected, there was agood correlation between the numbers of pH1N1 HA-specific IgGASC in tonsillar MNC and anti-HA IgG antibody titers in cellculture supernatants after pH1N1 antigen stimulation (Fig.5a;5334jvi.asm.orgJournal of Virology[ Pobierz całość w formacie PDF ]

zanotowane.pl

doc.pisz.pl

pdf.pisz.pl

slaveofficial.keep.pl