A robust, high-throughput assay to determine the phagocytic activity of clinical antibody samples

A robust, high-throughput assay to determine the phagocytic activity of clinical antibody samples
of 12
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
  Research paper A robust, high-throughput assay to determine the phagocytic activity of clinical antibody samples Margaret E. Ackerman a, ⁎ , Brian Moldt b,1 , Richard T. Wyatt b , Anne-Sophie Dugast a ,Elizabeth McAndrew a , Stephen Tsoukas a , Stephanie Jost a , Christoph T. Berger a ,Gaia Sciaranghella a , Qingquan Liu a , Darrell J. Irvine a,c , Dennis R. Burton a,b , Galit Alter a a Ragon Institute of Massachusetts General Hospital, Massachusetts Institute of Technology and Harvard University(formerly known as Partners AIDS Research Center of Massachusetts General Hospital), Boston, MA, United States b Department of Immunology and Microbial Science and IAVI Neutralizing Antibody Consortium, The Scripps Research Institute, La Jolla, CA, United States c Departments of Biological Engineering and Materials Science and Engineering, Massachusetts Institute of Technology, Cambridge, MA, United States a r t i c l e i n f o a b s t r a c t  Article history: Received 3 August 2010Received in revised form 12 November 2010Accepted 17 December 2010Available online 27 December 2010 Phagocytosis can be induced via the engagement of Fc γ  receptors by antibody-opsonized material.Furthermore,theefficiencyofantibody-inducedeffectorfunctionshasbeenshowntobedramaticallymodulated by changes in antibody glycosylation. Because infection can modulate antibody glycans,which in turn modulate antibody functions, assays capable of determining the induction of effectorfunctionsratherthanneutralizationortiterprovideavaluableopportunitytomorefullycharacterizethe quality of the adaptive immune response. Here we describe a robust and high-throughput flowcytometricassaytodefinethephagocyticactivityofantigen-specificantibodiesfromclinicalsamples.ThisassayemploysamonocyticcelllinethatexpressesnumerousFcreceptors:includinginhibitoryandactivating,andhighandlowaffinityreceptors — allowingcomplexphenotypestobestudied.Wedemonstratetheadaptabilityofthishigh-throughput,flow-basedassaytomeasureantigen-specificantibody-mediated phagocytosis against an array of viruses, including influenza, HIV, and dengue.Thephagocytosisassayformatfurtherallowsforsimultaneousanalysisofcytokinerelease,aswellasdetermination of the role of specific Fc γ -receptor subtypes, making it a highly useful system forparsing differences in the ability of clinical and vaccine induced antibody samples to recruit thiscritical effector function.© 2010 Elsevier B.V. All rights reserved. Keywords: PhagocytosisAntibodyADCCAntibody-dependent phagocytosisMonocytesFc receptorEffector function 1. Introduction Antibodiesarepotentdeterminantsofthehumoralimmuneresponse. Though generated as a result of the interaction of Band T cells, antibodies trigger their cytotoxic effects byinteracting with complement and innate effector cells. Thustheyprovideafunctionallinkbetweentheadaptiveandinnateimmunesystem.Theyconsistoftwoidenticalvariabledomains(Fv) capable of recognizing a target antigen, and a singleconstant domain (Fc) capable of interacting with the effectorcells of the immune system. Traditionally, the epitope recog-nized by the Fv domains has been thought to be of paramountimportance, in that binding to some epitopes can block, orneutralizethenativefunctionofthecognateantigen.However,the neutralizing activity mediated by the Fv domains of theseantibodieshasbeenfoundtobeinsuf  fi cientfortheirprotectiveeffects in numerous settings(Clynes et al., 2000; Johnson andGlennie,2003;SchmidtandGessner,2005;Hesselletal.,2007),and evidence of the importance of the constant domain'seffector function in clinical outcomes has been accumulatingacross  fi elds ranging from cancer immunotherapy (Dall'Ozzoet al., 2004) to autoimmunity(Laszlo et al., 1986) and chronic viralinfection(Shoreetal.,1974).AnalogouslytotheFvescape  Journal of Immunological Methods 366 (2011) 8 – 19 ⁎  Corresponding author. 149 13th St., Room 6616, Charlestown, MA,02129, United States. Tel.: +1 617 643 2837; fax: +1 617 726 5411. E-mail address:  meackerman@partners.org (M.E. Ackerman). 1 Co- fi rst author.0022-1759/$  –  see front matter © 2010 Elsevier B.V. All rights reserved.doi:10.1016/j.jim.2010.12.016 Contents lists available at ScienceDirect  Journal of Immunological Methods  journal homepage: www.elsevier.com/locate/jim  mechanisms such as mutating surface epitopes, several patho-gens evade the Fc-mediated antibody response by expressingproteases that restrict the Fc domain (Shakirova et al., 1985;Berasain et al., 2000; Collin et al., 2002; Vidarsson et al., 2005;Aslam et al., 2008), or glycosidases that remove the sugarresiduesrequiredforinteractionwithFcreceptors(Allhornetal.,2008). Combined, these evasion mechanisms and clinicalcorrelates provide strong evidence as to the importance of Fc-based effector functions inthetherapeutic activityofantibodies.Signi fi cantly, while the primary sequence of the constantdomain is conserved across antibodies of a given isotype, theeffector functions of distinct antibody isotypes are profoundlymodulated by alterations in the glycosylation pro fi le at aspara-gine 297 (Asn297) in the C H 2 domain of the antibody,modulating the range of effector responses a given antibodymay elicit (Boyd et al., 1995). The presence or absence of particular sugar groups on the Fc domain tunes the af  fi nitybetween IgG and Fc receptors (Fc γ Rs) on effector cells, and Fcglycoform represents a potent means of modulating antibodyactivity (Lund et al., 1996; Raju, 2008). This modulation isbidirectional,assomesugarstructuresdramaticallyaffectaf  fi nityto stimulatory FcRs, while others are known to inhibit immuneactivation(Shieldsetal.,2002;Kanekoetal.,2006;Nimmerjahnetal.,2007;Scallonetal.,2007;Raju,2008;AnthonyandRavetch,2010).Similarly, the expression levels of Fc γ R are also able tomodulate antibody activity. Among IgG binding FcRs, multipleisoforms with distinct functions have been identi fi ed: Fc γ gR1(high af  fi nity, activating), Fc γ R2a (low af  fi nity, activating),Fc γ R2b (low af  fi nity, inhibitory), and Fc γ R3 (low af  fi nity,activating). Thus, FcRs for IgG antibodies include both high andlowaf  fi nity,aswellasactivatingandinhibitoryreceptors,eachof whichmayhavedifferentialaf  fi nitiesforvariousIgGglycoforms,and may be expressed at different levels on different cell types.Thus, Fc γ R expression levels combined with Fc glycosylationpatterns represent a highly tunable system for modulating theactivity of antibodies.While numerous classes of innate immune cells express theFc γ R involved in antibody-mediated cytotoxicity, a subset of these are capable of actingas professionalphagocytes, includingmonocytes, macrophages, neutrophils, dendritic cells, and mastcells. Phagocytic mechanisms have a demonstrated importanceinclearance,antigenpresentation,andinnateimmuneactivation.Additionally,antibody-drivenphagocytosisenhancesinfectioninseveral infectious diseases (Halstead and O'Rourke, 1977;Marchette et al., 1979; Bernard et al., 1990; Tamura et al., 1991;Fustetal.,1994;Kozlowskietal.,1995)indicatingarelevanceof phagocytic processes not only on protection from, but insusceptibility to disease, and highlights the signi fi cance of thiseffector function in particular.Overall,asapotentmechanismofantibody-mediatedeffectorfunction, antibody-dependent phagocytosis of immune com-plexes, opsonized pathogens, and host cells represents animportantconnectionbetweentheadaptiveandinnateimmunesystems. Because antibody glycosylation patterns and thereforeFcR af  fi nity, and immune activity differ, a robust means tomeasurethiscriticaleffectorfunctionmayhelpde fi nequalitativedifferences in this effector function during infection, and post-vaccination that may play an important role in protection. Herewe present a high throughput assay capable of measuring thephagocytic activity of antibodies in clinical samples. 2. Materials and methods  2.1. Cells, control antibodies, viral envelope antigens THP-1 cells were purchased from ATCC and cultured asrecommended. Care was taken to keep cultures at cell densitiesbelow0.5×10 6 /mlinordertomaintainconsistentlevelsofFc γ R expression and assay performance.A panel of antibodies to Fc γ R 1, 2, 2A, 2B, and 3 was used(BD Pharmingen, 558592, 555406; Biolegend, 303212; NovusBiologicals,NB100-79947;RDSystems,AF1875)todeterminereceptor expression on THP-1 cells. Quantum Simply Cellularbead standards (Bangs Laboratories, 815) were used toquantify the number of these receptors on THP-1 cellsaccording to the manufacturer's instructions.Fluhemagglutinin(SinoBiological,11085-V08H),dengue-2Epeptide (ProsecBio, DEN-006) and gp120 (Immune Technology,IT-001-0027pYU-2)werebiotinylatedonlysineresiduesusingasulfo-NHSLCbiotinreagent(ThermoScienti fi c,21935)accordingto the manufacturer's instructions. After the reaction, free biotinwas removed by extensive buffer exchange using Amiconcentrifugal concentration units of appropriate molecular weightcutoff.  2.2. Phagocytosis assay Biotinylated antigen was incubated with 1  μ  m  fl uorescentneutravidin beads (Invitrogen, F8776) overnight at 4 °C.Beads were subsequently spun down and washed twice inPBS-BSA in order to remove excess unbound antigen, andthen resuspended at a  fi nal dilution of 1/100 in PBS-BSA.Antigen-coated beads were stored for up to a week at 4 °Cprior to use. Saturation of the beads was determinedexperimentally, via incubation with differing amounts of antigen. Bead coating conditions that gave the maximalphagocytic score in conjunction with a control monoclonalantibody were used. 9×10 5 beads (equivalent of 0.1  μ  l of supplied suspension, or 10  μ  l of the dilution described above)were placed in each well of round bottom 96 well plate.Antibodies were added to each well and the plate wasincubated for a 2 h at 37 °C in order to allow antibodies tobind to the beads. Followingequilibration, 2×10 4 THP-1 cellswere added to each well in a  fi nal volume of 200  μ  l, and theplate was incubated overnight under standard tissue cultureconditions. The next day, half the culture volume wasremoved, saved for subsequent multiplexed analysis of cytokine secretion, and replaced with 100  μ  l of 4% parafor-maldehydebeforeplateswereanalyzedby fl owcytometryona BD LSR II equipped with an HTS plate reader. Samples weremixedthoroughly (100  μ  lmixvolume,repeated threetimes),priortoanalysisof30  μ  lofeachsample,yieldingatleast2000celleventspersample.Aphagocyticscorewasdeterminedbygating the samples on events representing cells, andcalculated as follows: (% bead positive×MFI bead positive,or integrated MFI (Darrah et al., 2007)). For ease of presentation, these scores were then divided by 10 6 .Alternatively, because the  fi rst peak of bead-positive cellsrepresents cells that have phagocytosed a single bead, it ispossible to use the MFI of this peak to determine the averagenumber of beads phagocytosed by each cell, with the addedadvantage of removing instrument variability from the data 9 M.E. Ackerman et al. / Journal of Immunological Methods 366 (2011) 8 – 19  generated.Errorbarsrepresentstandarddeviationsofatleast3 replicates.  2.3. Confocal and video microscopy For experiments involving imaging of the THP-1 cells,1.2×10 6 cells were incubated with 9×10 6 beads (equivalent to1 μ  l of supplied suspension) saturated with biotinylated humanIgG,orleftuncoatedina3mlvolumeina12wellplateovernightat either 4°C, or under standard tissue culture conditions.Cultures were resuspended, 1.5 mls was withdrawn, spundown, aspirated, and  fi xed with paraformaldehyde. Following fi xation,cellswereagainspundown,aspirated,andresuspendedin 500 μ  l of 300mM sucrose in phosphate buffered saline. Thecell membrane was stained by addition of 5  μ  l DiI (Invitrogen,V22885) and incubation at 37°C for 10min. Stained cells werewashed twice before being dropped onto a poly- L  -lysine-coatedmicroscope slide, and then imaged on a Zeiss LSM 510microscope at 63× magni fi cation, Zeiss Plan-Apochromat DIC,1.4NA, oil immersion lens. Microscopy conditions were set suchthat optical sections were not thicker than the bead diameter.Additionally, 3-dimensional stacks were collected in order tocon fi rm the internalization pro fi le of bead particles.Time-lapse images of phagocytosis were acquired on aZeiss Axio Observer at 20× magni fi cation on a Zeiss AxioObserver using a Zeiss Plan-Apochromat, air, 0.8NA lens overa 14-hour period at a rate of 1 image per minute. SeventythousandTHP-1cellswere combinedwitha 30-foldexcess of uncoated, red  fl uorescent beads, and a 30-fold excess of antibody-coated, green  fl uorescent beads, and spun downthrough a collagen matrix at 2000 rpm onto a coverslipchamber (LabTek, 177402) coated with  fi bronectin to aidwithcelladherence.Attheconclusionofthe14-hourimagingperiod, a 63× Zeiss Plan-Apochromat 1.4NA oil immersionlens was used to capture representative images at highermagni fi cation. Image acquisition and analysis was performedusing Metamorph software (Molecular Devices).  2.4. Amnis ImageStreamX   fl ow cytometry Cells were prepared as described for the phagocytosis assaywith the addition of an anti- Fc γ R2-Alexa647 antibody to labelFc γ R2, and DAPI to label nuclei, and run on an AmnisImageStreamX fl owcytometerwith60×magni fi cation.Sampleswere analyzed with the IDEAS software platform in order todetermine an internalization score. Brie fl y, a mask representingthecellmembranewas de fi nedbythebright fi eld image,and aninternal mask was de fi ned by eroding the whole cell mask by 6pixels, resulting in an area signi fi cantly smaller the cellmembrane. An internalization score was calculated as follows:(number of   fl uorescent pixels in internal mask) − (number of  fl uorescent pixels in the entire  fi eld of view).  2.5. Control antibodies and deglycosylation reaction Human IgG (Sigma, I2511) was biotinylated on lysineresidues using a sulfo-NHS LC biotin reagent (Thermo Scienti fi c,21935) as described previously. A fraction of the biotinylatedantibodywassubsequentlydeglycosylatedusingPNGaseF(NEB,P0704) according to the manufacturer's instructions. B12, lalaB12, and double and triple mutant B12 were obtained fromDennis Burton. The anti-dengue antibody 3H5 was purchasedfrom Millipore (Millipore, MAB10226).  2.6. Receptor blocking experiments Receptor blocking antibodies to Fc γ R2 (Abcam, 23336)and Fc γ R3 (Sigma, F3668) were used according to themanufacturer's instructions. Cells were preincubated withblocking antibodies for at least 1 h prior to being mixed withbeads, incubated overnight, and then analyzed by  fl owcytometry.Becausetheblockingantibodiesdonotnecessarilyblock all receptor function under the conditions tested,results are presented as the ratio of phagocytic activity of triple:double mutant for each blocking antibody condition,rather than compared to untreated controls.  2.7. Patient antibodies StudyindividualswererecruitedfromRagonInstitutecohortsand included patients with in fl uenza infection, chronically HIV infected individuals, and healthy individuals. Antibodies wereseparatedfrom otherserum proteinsusingMelonGelaccordingto the manufacturer's instructions (Thermo Scienti fi c, 45206).The study was approved by the Massachusetts General HospitalInstitutional Review Board, and each subject gave writteninformed consent. The anti-dengue 2 antibody, 3H5 (Millipore,MAB10226),amouseisotypecapableofinteractingwithhumanFc γ R2(Littauaetal.,1990;Fleschetal.,1997)wasusedapositivecontrol for dengue-coated beads.  2.8. Luminex assays MilliplexcytokineanalysiswasperformedontheLuminexxMap platform following the manufacturer's instructions(Millipore, custom kit) on supernatants removed fromphagocytosis assay plates. 3. Results and discussion  3.1. The phagocytosis assay In addition to their role in neutralization, antibodies are ableto mediate a number of additional functions including therecruitment of innate immune responses to eliminate antibody-opsonized material. Among these additional antibody mediatedfunctions, antibodies are able to promote phagocytosis, whichmayplayaprofoundroleintherapidcontainmentandclearanceof a pathogen following infection. However, robust assays thatare able to capture differences in the quality of antibody-mediated phagocytosis are lacking. Thus to this end wedeveloped a novel high-throughput assay, using a monocytic-cell line to provide a platform to tease out antigen-speci fi cantibody mediated phagocytosis. Brie fl y, the antibodies of interest are captured on the surface of highly  fl uorescent latexbeads,whicharethenincubatedovernightwithmonocytespriorto analysis by  fl ow cytometry. Because the beads used may becoatedwithanantigenofchoice,thisassayallowsdeterminationofantigen-speci fi cphagocytosiswithoutrequiringpuri fi cationof these antibodies.We employed a widely used monocytic cell line, THP-1(Tsuchiya et al., 1980), which have been extensively used in 10  M.E. Ackerman et al. / Journal of Immunological Methods 366 (2011) 8 – 19  studies of phagocytosis of particles including red blood cells,yeast (Tsuchiya et al., 1982),  E. coli  (Schiff et al., 1997),staphylococcus (Kapetanovic et al., 2007), zymosan (Friedland et al., 1993a), and human cells (Tebo et al., 2002; Beum et al.,2008). They have also been utilized in studies of signalingdownstream of phagocytosis, including cytokine release (Shawet al., 2000), oxidative burst (Gross et al., 1998), and receptor phosphorylation(Lee et al., 2007). Moreoverbecause theygrowin suspension, they are convenient choice for use in a  fl owcytometry-based assay. Additionally, much is known aboutmodulation of Fc γ R expression and maturation in this cell lineupontreatmentwithvariousstimulantsandcytokines(FleitandKobasiuk,1991).Importantly,asdemonstratedinTable1,THP-1 cellsexpressmultipleFc γ receptors,includingFc γ R1,Fc γ R2,andFc γ R3,allowingthiscelllinetocaptureeffectsfromabroadrangeof Fc γ Rs, and re fl ecting the normal Fc γ R2 expression pro fi le of monocytes (data not shown). Signi fi cantly, these cells alsoexpress both subtypes of Fc γ R2: the activating receptorimplicated in phagocytosis, 2A, and the inhibitory, ITIM motif-containing 2B receptor (Fig. 1). When quanti fi ed, we found thatFc γ R2 is expressed at signi fi cantly higher levels than the otherFc γ R. While the phagocytic properties of THP-1 cells areincreased upon exposure to stimulants such as PMA (Tsuchiyaet al., 1982), these treatments tend to stimulate non-Fc-dependent phagocytic processes, and can alter Fc γ R expressionpro fi les (data not shown). Because they therefore tend toincrease baseline phagocytosis levels, and thereby decrease therelative signal change due to antibody-dependent phagocytosis,theassaywedescribedoesnotutilizeanychemicalstimulantstomature the THP-1 cells.Antibodies are captured on the surface of 1.0um  fl uorescentmicrospheres in a 96-well plate. After incubation at 37°C for atleastanhourtoallowopsonizationtooccur,20,000THP-1cellsin200 ul of media are added to each well, and allowed tophagocytose the antibody-coated microspheres overnight at37°C.Thefollowingday,plateswerespundown,andhalfofthesample volume is replaced with 4% paraformaldehyde for fi xation. Samples were then run on a BD LSRII cytometerequipped with a high-throughput sampler. Because the micro-spheres are exceptionally bright, care was taken in determiningacquisition settings in order to ensure both bead-positive andbead-negativepopulationswerewithinthedynamicrangeoftheinstrument.Toprovideaconvenientquantitativemeasureofnetphagocytosis, a phagocytic score was calculated by determiningthe percentage of cells that were bead-positive, and multiplyingbytheirmean fl uorescenceintensity(iMFI)(Darrahetal.,2007).Preliminary experiments were carried out using biotiny-lated bulk human IgG which was trapped directly on thesurfaceofneutravidin-coatedbeads.Arangeofcell(20,000to100,000) and bead (9×10 5 to 9×10 6 ) concentrations wereevaluated, and the ratioof 45 beadsper cell, with20,000 cellsused per well was selected because it yielded the greatestsignal change between antibody coated and non-coatedbeads. While a value of 15 particles per phagocytic cellmight be a more typical ratio, because our assay formatutilizessuspensioncellsanddoesnotcallforcentrifugationtocon fi nebeadsandcellstoa2-dimensionalplaneatthestartof the experiment, we have found that a greater particle to cellratio and a longer incubation time is optimal.  3.2. Microscopy con  fi rmation of phagocytosis In order to demonstrate that the beads were in factphagocytosed, and did not simply associate with the cellmembranes, internalization of the beads was con fi rmed byconfocal microscopy. Fig. 2 presents confocal microscopyimages and corresponding  fl ow cytometry histograms of thephagocytosis of green  fl uorescent microspheres under variousconditions. Bead uptake, as determined by  fl uorescence in the fl ow cytometry readout was maximal when cells wereincubated with antibody-coated beads at 37 °C. In contrast, at0 °C, phagocytosis is impaired, but antibody-coated beads stillassociate with cells due to speci fi c interactions betweenantibodies on the beads and Fc γ R on the cells, leading to alowersignalonFACShistograms,andtheappearanceofsurface-associated rather than phagocytosed beads by confocal  Table 1 Fc γ R quanti fi cation on THP-1 cells. FcgR receptors were quanti fi ed usingbead standards.FcgR Number of receptors ±SD1 66,300 15,0002 171,000 13,0003 18,000 1600 Fig. 1.  Expression of Fc γ R2A and 2B. THP-1 cells were labeled with antibodies speci fi c for the extracellular domain of Fc γ R2A, the Fc receptor implicated inphagocytosis, and the intracellular domain of the inhibitory receptor Fc γ R2B. Cells labeled with secondary only are traced in black, cells labeled with primary andsecondary antibody are traced in gray.11 M.E. Ackerman et al. / Journal of Immunological Methods 366 (2011) 8 – 19  microscopy. While FACS signals from uncoated beads at bothtemperatures are signi fi cantly lower, they are not zero, as thelatex beads tend to associate non-speci fi cally. This level of  fl uorescence represents a negative control in the evaluation of experimental samples.As a second means to verify the internalization of thepresumptively phagocytosed beads, the Amnis ImageStreamXimaging fl owcytometer,whichcapturesanimageofeachcellasit passes through the stream, offers a unique opportunity toperform statistical analysis of bead internalization on a large Fig. 2.  Phagocytosis assay. Confocal microscopy images and  fl ow cytometry histograms of phagocytosis of green  fl uorescent beads with or without antibodycoating, at 37 °C or 0 °C. For confocal microscopy, membrane was stained red with the membrane dye, DiI (red). Histograms represent the fl uorescent signal fromcell-associated beads. Bead uptake is maximal under conditions where Fc γ R-mediated phagocytosis can occur. Fig. 3.  Con fi rmation of phagocytosis. A. Representative images captured by the Amnis ImageStreamX Flow Cytometer of cells without any beads (top), surfacebound beads (middle), or internalized beads (bottom). Nuclei are stained blue, beads green, and Fc γ R2 in red. B. Internalization score calculated by Amnis' IDEASsoftware,indicatinginternalizationofbeadswhencoatedbyantibodyat37 °C,andminimalinternalizationofbeadswithoutantibody-coating, orwhenincubatedat 0 °C. C. Dependence of phagocytic score on the density of antibody coating on the surface of the beads.12  M.E. Ackerman et al. / Journal of Immunological Methods 366 (2011) 8 – 19
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks