ava i l ab l e a t www.sc i enced i r ec t . com

C l i n i ca l Immuno logy

www.e l sev i e r . com/ l o ca t e / y c l im

Clinical Immunology (2014) 152, 77–89

Identifying functional anti-Staphylococcusaureus antibodies by sequencing antibodyrepertoires of patient plasmablasts

Daniel R. Lua,b,c,1, Yann-Chong Tana,b,c,1, Sarah Kongpachith a,b,c,Xiaoyong Cai a,b, Emily A. Steina,b, Tamsin M. Lindstroma,Jeremy Sokolovea,b, William H. Robinsona,b,c,⁎

a Division of Immunology and Rheumatology, Stanford University, CCSR 4135, 269 Campus Dr., Stanford, CA 94305, USAb VA Palo Alto Health Care System, 3801 Miranda Ave, Palo Alto, CA 94304, USAc Stanford Immunology Program, Stanford University School of Medicine, Stanford, CA 94305, USA

Received 9 November 2013; accepted with revision 15 February 2014Available online 1 March 2014

Abbreviations: VH, Heavy-chain varsegment and J gene segment); VL, Lighsegment); S. aureus, Staphylococcus au⁎ Corresponding author at: Division o

Stanford, CA 94305, USA.E-mail address: wrobins@stanford.

1 These authors contributed equally

http://dx.doi.org/10.1016/j.clim.2011521-6616/Published by Elsevier Inc.

KEYWORDSStaphylococcus aureus;Antibody;Infection

Abstract Infection by Staphylococcus aureus is on the rise, and there is a need for a betterunderstanding of host immune responses that combat S. aureus. Here we use DNA barcoding toenable deep sequencing of the paired heavy- and light-chain immunoglobulin genes expressed byindividual plasmablasts derived from S. aureus-infected humans. Bioinformatic analysis of theantibody repertoires revealed clonal families of heavy-chain sequences and enabled rational

selection of antibodies for recombinant expression. Of the ten recombinant antibodiesproduced, seven bound to S. aureus, of which four promoted opsonophagocytosis of S. aureus.Five of the antibodies bound to known S. aureus cell-surface antigens, including fibronectin-bindingprotein A. Fibronectin-binding protein A-specific antibodies were isolated from two independentS. aureus-infected patients and mediated neutrophil killing of S. aureus in in vitro assays. Thus,our DNA barcoding approach enabled efficient identification of antibodies involved in protectivehost antibody responses against S. aureus.Published by Elsevier Inc.

iable region domain (comprisedt-chain variable region domainreusf Immunology and Rheumatolo

edu (W.H. Robinson).to the studies.

4.02.010

of the recombined immunoglobulin heavy-chain V gene segment, D gene(comprised of the recombined light-chain V gene segment and J gene

gy, Stanford University School of Medicine, CCSR 4135, 269 Campus Dr.,

78 D.R. Lu et al.

1. Introduction

Staphylococcus aureus, the second-most common bacterialpathogen isolated from humans, is estimated to cause overhalf a million cases of invasive infection and over 18,000deaths annually in the United States [1–3]. The emergenceof multidrug-resistant strains, such as methicillin-resistantS. aureus (MRSA), has complicated treatment procedures,prompting research into vaccines and novel immunother-apies. Still, how S. aureus interacts with the human immunesystem is unclear, and no vaccines are currently approvedfor preventing or treating S. aureus infections [4]. Antibodytiters against S. aureus proteins increase during acute infec-tion [5], and patients with immunoglobulin deficienciesare more susceptible to S. aureus infections [6], suggestingthat the host antibody response is important for containingS. aureus. However, the critical S. aureus antigens that elicitprotective host immunity remain undefined, with little or nocorrelation observed between disease severity and levels ofserological antibodies against known S. aureus virulencefactors [7].

Antibodies circulating in the blood are produced byplasma cells and their transient B-cell precursors (termedplasmablasts). After successfully completing antigen-dependentaffinity maturation, naïve and memory B cells differentiateinto plasmablasts and proliferate rapidly in the blood beforetrafficking to infected tissues. Although plasmablast fre-quencies in peripheral blood are typically low before andafter antigen stimulation of an immune response (~0.2% ofcirculating B cells), they can account for over 30% ofcirculating B cells during an infection [8,9]. Thus, theexpansion, specificity and accessibility of circulating plasma-blasts provide a powerful approach to characterize the activeB-cell response.

Accurate identification of the affinity-matured andexpanding B-cell lineages requires methods that can bothapproximate the frequency of B cells expressing similarvariable-region genes and recover the endogenous heavy-and light-chain variable-region pairings from individualB cells. Previously published high-throughput approachesthat sequence only the heavy-chain or heavy- and light-chaincomplementarity-determining-region 3 (CDR3) from bulkB cells and/or do not implement any methods to normalizethe relative abundance of sequencing reads to the number ofcells expressing a given sequence [10–13], making it difficultto determine whether highly represented sequences arethe result of antigen-driven B-cell clonal proliferation orof sequence biases introduced during PCR amplification.Additionally, the inability to recover heavy- and light-chainpairs precludes the functional characterization of theendogenous antibodies.

To overcome these limitations, we use a novel DNAbarcode system to sequence the paired heavy- andlight-chain antibody genes expressed by individual plasma-blasts, and in doing so to accurately determine theproportion of plasmablasts that express specific germlinesequences and somatic mutations. Herein we sequence theendogenous antibodies expressed by individual plasma-blasts circulating in the blood of individuals with S. aureusbacteremia, bioinformatically identify clonal families of

antibodies sharing heavy and light chain VJ sequences, andrecombinantly produce and characterize of the binding andfunctional properties of representative antibodies.

2. Material and methods

2.1. Single-cell sorting of peripheral blood plasmablasts

Blood was collected from individuals with culture-confirmedS. aureus bacteremia after after informed consent wasobtained and under human subject protocols approved bythe Investigational Review Board at Stanford University. Forall individuals, blood specimens were obtained within 24 hafter the initiation of standard-of-care antibiotic treat-ment. PBMCs were isolated using a Ficoll layer andstained with CD3-V450 (BD 560365), IgA-FITC (AbD SerotecSTAR142F or Miltenyi #130-093-071), IgM-FITC (AbD SerotecSTAR146F), IgM-APC (BD 551062) or IgM-PE (AbD SerotecSTAR146PE), CD20-PerCP-Cy5.5 (BD 340955), CD38-PE-Cy7(BD 335808), CD19-APC (BD 340437) or CD19-BrilliantViolet 421 (Biolegend 302233), and CD27-APC-H7 (BD560222). IgG+ plasmablasts were gated on CD19+/intCD20−CD27++CD38++IgA− IgM− cells and individuallysorted using a BD FACSAria III into a 96-well PCR platecontaining a hypotonic lysis buffer (10 mM Tris–HClpH 7.6) containing 2 mM dNTPs (NEB), 5 μM oligo(dT)20VN, and 1 unit/μl of Ribolock (Fermentas), an RNaseinhibitor.

2.2. Reverse transcription (RT) and PolymeraseChain Reaction (PCR) with well-ID and plate-IDadaptors

We added 6 mM MgCl2 with Ribolock, Superscript III(Life Technologies), and 1 μM final concentration of theappropriate well-ID oligonucleotide barcode to sortedplasmablast plates and performed RT at 42 °C for 120 min.RT products from each plate were pooled, extracted withphenol-chloroform-isoamyl alcohol, and concentrated withAmicon Ultra-0.5 30 kDa (Millipore) units.

We used Phusion Hot Start II DNA polymerase (NEB/Fermentas) for both the first PCR (PCR1) and the second PCR(PCR2). The following PCR reaction conditions were used:200 μM of dNTPs, 0.2 μM of primers, 0.2 U of Phusionpolymerase, 4% DMSO, and 2 μl of RT product. PCR1 wasperformed with forward (FW) primers containing, at their 5′end, a plate-ID barcode oligonucleotide, as well as a 454Titanium adaptor, and with gene-specific reverse primers(GSP) for amplifying the gamma, kappa, and lambda chains.PCR2 was performed with FW primers containing a 454Titanium adaptor at their 5′ end and reverse GSP containinga plate-ID barcode oligonucleotide and a 454 Titaniumadaptor at their 3′ end.

We pooled the amplified DNA, gel purified them,and purified them with Ampure XP beads (Beckman Coulter).Amplicon concentrations were determined by using Pico-green DNA assay kits (Invitrogen), and amplicons were sentto Roche for 454 sequencing.

79Identification of functional anti-S. aureus antibodies

2.3. Bioinformatic analysis and generation ofdendrograms

Sff output files from 454 sequencing, containing sequencesand quality scores for each nucleotide, were read intoPython by using the Biopython package, and sequences weregrouped and parsed into separate sff files on the basis oftheir plate- and well-ID combinations. Consensus sequencescorresponding to plasmablast heavy- and light-chain se-quences were generated using Newbler 2.6. Where multipleassemblies occurred per well, an assembly was accepted ifit represented N50% of all reads in the given well.Otherwise, we disregarded all reads from that well topreclude instances where a well contained more than onecell and to account for sequencing errors introduced duringpyrosequencing.

To assign V(D)J families for consensus sequences, weanalyzed heavy- and light-chain sequences with the IMGTHighV-QUEST15 database [14] to predict germline alleleusage, germline sequence recombination, and shared muta-tions relative to the germline sequence.

To generate dendrograms, we first binned heavy- andlight-chain sequences according to their V-gene usage andaligned them with Muscle28 [15]. They were clustered withPhyML maximum-likelihood clustering and rooted by theirgermline V gene [16]. Each V-gene phylogenetic tree wasthen arranged by heavy-chain V-gene families and drawn byusing ETE30 [17].

2.4. Cloning and expression of recombinant antibodies

Antibodies representative of the heavy-chain families wereselected for expression on the basis of (1) having acorresponding light-chain plate- and well-ID combinationand (2) being the heavy- and light-chain pair most represen-tative of the consensus sequence of the family. Gammaheavy chains were inserted into vector pEE6.4 (Lonza), andkappa and lambda light chains were inserted into vectorpEE12.4 (Lonza). We used well-ID-specific FW primers andGSP reverse primers to selectively amplify cDNA for insertioninto the expression vectors. Heavy- and light-chain se-quences were cloned by first inserting the sequences of theantibody constant regions into the vector and inserting theleader and V(D)J sequences by using standard molecularbiology techniques.

Expression of recombinant antibodies was performed bydual transfection of heavy-chain-containing pEE6.4 con-structs and light-chain-containing pEE12.4 constructs in293T cells, using Lipofectamine 2000 (Life Technologies).Antibodies were purified from culture supernatants by usingProtein A Plus agarose beads (Pierce), eluted with IgGelution buffer (Pierce), and quantified by BCA Protein Assay(Thermo Scientific).

2.5. Bacterial culture and preparation

S. aureus Wood46 (ATCC 10832) were grown in LB toearly-logarithmic (0.25 optical density units (ODU)),mid-logarithmic (0.5 ODU), or stationary (2.0 ODU) growthphases, as determined by spectrophotometry at 550 nm. Wepelleted bacteria by spinning at 5000 g for 10 min, and

resuspended them in 0.1% sodium azide in PBS to haltbacterial metabolism and to capture snapshots of theprotein expression.

For experiments in which S. aureus were grown in thepresence of fibroblasts, NIH/3T3 mouse embryonic fibroblasts(ATCC CRL-1658) were cultured in DMEM (ATCC 30-2002)supplemented with 10% FBS, 0.6 μg/ml L-glutamine, 0.2 U/mlpenicillin, and 0.2 μg/ml streptomycin and stimulated with300 pM recombinant mouse TGF-β1 (R&D Systems) in culturemedia minus antibiotics for 72 h. S. aureus Wood46 weregrown to stationary phase in LB resuspended in DMEM, andincubated with NIH/3T3 cells for 2 h. NIH/3T3 cells werethen lysed with hypotonic buffer and washed, and bacterialcells were resuspended in 0.1% sodium azide in PBS.

For experiments with Escher coli, DH10B E. coli weregrown in LB for 14 h, pelleted, and resuspended in 0.1%sodium azide in PBS.

2.6. S. aureus lysate dot-blot analysis

To prepare S. aureus lysates for dot-blot, 0.25 g of growthphase-specific bacteria in sodium azide (dry weight) wasincubated with B-PER lysis solution (Thermo Scientific), 1×Halt protease inhibitor, and 0.05 μg/ml freshly preparedlysostaphin (Sigma) for 35 min at 37 °C, then for 15 min atambient temperature with shaking. Samples were centri-fuged at 15,000 g for 15 min at 4 °C, and the supernatantswere collected and quantitated by BCA assay.

Dot-blot was performed on 0.45 μm nitrocellulose trans-fer membrane with the Bio-Dot Microfiltration Apparatus(Bio-Rad) as per manufacturer's instructions. 500 ng lysate in50 μl TBS was spotted per well and blocked with 3% BSA–TBS. Recombinant test antibodies were diluted to 2 μg/ml in0.1% TBST with 3% BSA and added to appropriate wells.1:15,000 goat anti-human IgG-Fc HRP (Bethyl Laboratories)secondary antibody was used for detection, and dot-blotswere visualized with West Femto Substrate (ThermoScientific).

2.7. Flow cytometric analysis of antibody binding

We blocked sodium-azide-treated S. aureus Wood46 orDH10B E. coli cells with 1% BSA–PBS for 45 min at 4 °C,then added 10 μg/ml of recombinant test antibodies in 1%BSA–PBS for 60 min at 4 °C, washed three times with 1%BSA–PBS, and incubated for 20 min at 4 °C in the dark with1:5 mouse anti-human-IgG PE secondary antibody (BDPharmingen) diluted 1:5. Cells were washed three timeswith PBS, and antibody surface binding was detected with anLSR Fortessa.

2.8. S. aureus antigen array analysis

A full-proteome microarray of the S. aureus USA300 strain—astrain prevalent in community-acquired S. aureus [18]—containing approximately 2700 S. aureus proteins per array(Antigen Discovery Inc., Irvine, CA) was used for determiningthe specificity of recombinant antibodies derived fromS. aureus-infected humans. Conditions used for preparing,probing, and analyzing the microarray chips were performedas described in [19].

80 D.R. Lu et al.

2.9. S. aureus antigen ELISA

We used ELISA to determine whether the recombinantantibodies derived from bacteremic individuals bind topurified S. aureus antigens. We coated Nunc Maxisorp plateswith 1.5 μg/ml lipoteichoic acid or peptidoglycan purifiedfrom S. aureus (Sigma Aldrich) in pH 7.4 PBS and blocked theplates overnight with PBS containing 1% BSA. Recombinantantibodies were added at 1 μg/ml and diluted serially 3×.After washing with PBST (PBS with 0.05% Tween20),antibody binding was detected by using an HRP-conjugatedgoat anti-human IgG-Fc antibody and TMB substrate.Absorbance was measured with a SpectroMax M5 spectro-photometer at 450 nm.

2.10. Flow cytometric analysis of S. aureusopsonophagocytosis

To label sodium-azide-treated S. aureus for detection ofbacterial internalization, we washed and resuspended thebacteria in PBS, and then incubated them with 2 μl ofCellTrace CFSE (Invitrogen) for 30 min at 37 °C withconstant shaking at 250 rpm. CFSE-staining was stopped byincubating the cells for 5 min with 500 μl ice-cold PBS andwashing them three times in PBS.

We cultured THP-1 human monocyte-like cells (ATCCTIB-202) in RPMI-1640 supplemented with 10% FBS, 0.6 μg/mlL-glutamine, 0.2 U/ml penicillin, and 0.2 μg/ml streptomycinat a density below 5 × 105 cells/ml to minimize macrophagedifferentiation [20].

To assess opsonophagocytosis, we incubated 2 × 106

CFSE-labeled S. aureus for 60 min with 10 μg/ml antibodyin 1% BSA–PBS, washed them three times, and resuspendedthem in pre-warmed 25 μl RPMI-1640 with 1% FBS. 100 μlTHP1 cells were then added to the CFSE-labeled S. aureusand incubated for 40 min at 37 °C with shaking at 200 rpm.Phagocytosis was stopped by adding 1 ml ice-cold PBS to thecell mixture and centrifuging it at 350 g for 3 min at 4 °C.Cell-surface and extracellular fluorescence were quenchedby resuspending the cell pellet with 200 μl ice-cold TrypanBlue (2 mg/ml in PBS) in 2% paraformaldehyde [21]. N10,000THP1 cellular events were counted with an LSR Fortessa.

2.11. Neutrophil bactericidal assay

HL-60 cells were cultured with RPMI-1640 supplementedwith 10% FBS, 0.6 μg/ml L-glutamine, 0.2 U/ml penicillin,0.2 μg/ml streptomycin, and 50 μM β-mercaptoethanol.HL-60 cells were differentiated for 96 h in culture mediacontaining 1.3% DMSO and 2.5 μM all-trans retinoic acid aspreviously described [22] and resuspended in HBSS supple-mented with 10% human AB serum at 106 cells/ml.

S. aureus Wood46 were grown in LB overnight at 37 °C,and resuspended in HBSS with 10% human AB serum at106 bacteria/ml, and 10 μg/ml recombinant test antibodywas added to the bacteria. HL-60 cells and bacteria werethen added at a 1:1 volume ratio and incubated for 1 h at37 °C with shaking. Three 1:10 serial dilutions of the HL-60:bacteria mixture were performed in PBS. Bacteria wereplated on TSA plates (BD), grown at 37 °C for 16 h, andbacterial colony-forming units were counted.

3. Results

3.1. Plasmablast levels are elevated during S. aureusacute bacteremia

Previous studies detected transient elevations in levels ofperipheral blood plasmablasts that peak approximately7 days following an immune challenge [9]. We investigatedwhether patients with community-acquired S. aureus bac-teremia and acute clinical manifestations exhibit elevatedplasmablast levels (CD19+/int/CD20−/CD27++/CD38++) in pe-ripheral blood. All individuals with S. aureus bacteremiahad greater than 1% plasmablasts per total peripheral bloodB cells, with a mean of 3.2%, indicating a ten-fold increaseabove normal plasmablast frequencies (Figs. 1A and B).

3.2. IgG plasmablast single-cell sequencing revealsexpression of antibody families with shared genesegments

To obtain paired heavy- and light-chain antibody sequencesfrom individual single-cell sorted IgG+ plasmablasts, weutilized a DNA barcoding method that our laboratorydeveloped [23]. This method uses the dual function ofSuperscript III reverse transcriptase to synthesize cDNA fromthe mRNA of single cells, and to thereby enable tagging ofthe newly generated cDNA with well-specific (and thuscell-specific) oligonucleotide barcodes at the 5′ end ofmRNA via template switching [24]. Primers containingplate-specific barcodes were then added and hybridized toa universal linker sequence flanking the 5′ end of thewell-IDs. The resulting double-tagged amplicons weresequenced by 454 GS FLX+, generating ~800,000 reads perrun, with a mean read length of 600 base-pairs (bp) forheavy chains and 480 bp for light chains. These lengths weresufficient for the recovery of the entire V(D)J region ofheavy and light chains spanning the Framework 1 andFramework 4 conserved regions.

After discarding any incomplete VH and VL sequenceassemblies, we generated dendrograms of each patient'splasmablast heavy-chain (VH dendrograms) and paired heavy-and light-chain (VH:VL dendrograms) repertoire to visualizethe number of antibodies using identical germline IGHV genes,which we define as an IGHV family (Fig. 2 and SupplementalFig. S1). Because the VH dendrograms contained moresequences than the paired-chain dendrograms, we focusedon plasmablast heavy-chain dendrograms in determiningwhich immunoglobulin rearrangements were preferentiallyselected for in response to S. aureus infection.

For patients 1 and 2, who mounted robust plasmablastresponses and successfully cleared their clinical infections(and from whom we obtained 120 and 125 full-length VHsequences, respectively), we analyzed the extent to whichIGHV families containing more than five VH sequenceassemblies also used the same IGHJ genes (Fig. 2). In pa-tient 1, nine IGHV families contained more than five re-covered VH sequences, and five of these IGHV families usedthe same IGHJ gene in at least 50% of the sequences(Fig. 2A). Of the nine IGHV families, IGHV3-7 had the highestpercentage of shared germline IGHV and IGHJ usage (64%IGHJ4). Notably, five of the seven VH sequences in the

ALi

ve/D

ead

CD19C

D20

CD19

CD

38

CD27

IgM

+ Ig

A

SSC-A

7.2

13 9.8

20

B

Health

y

S. aur

eus b

acte

rem

ia

% p

Iasm

abla

sts

of to

tal C

D19

+ B

cel

ls

***

0

2

4

6

8

Figure 1 Plasmablast levels are elevated during acute S. aureus bacteremia. (A) Fluorescence-activated gating strategy forsingle-cell sorting IgG-secreting plasmablasts from PBMCs of S. aureus-infected individuals. Plasmablasts were gated on CD19+/intCD20−

CD27++CD38++IgA−IgM− live B cells. (B).

81Identification of functional anti-S. aureus antibodies

IGHV5-a family in patient 1 used IGHJ6 and had highly similarsomatic mutations in both the heavy- and light-chain variableregions, suggesting that the five plasmablasts containing theseVH sequences may have arisen from the same B-cell lineage(Supplemental Fig. S2). Patient 2 displayed greater heteroge-neity in the combination of IGHV and IGHJ genes used inthe rearranged sequences. Six IGHV families contained morethan five VH sequences and family IGHV3-30 had the highestpercentage of shared germline IGHJ usage (40% IGHJ4) (Fig. 2B).

3.3. Rational selection of highly representedantibody sequences for recombinant antibodyexpression

Because we observed that certain IGHV and IGHJ alleleswere combinatorially selected for during the response toS. aureus infection and thus appeared more frequently in the

antibody repertoire, we asked whether antibodies encodedby these sequences can bind to S. aureus. We selectedrepresentative antibodies from IGHV families with highlyshared IGHJ for recombinant human monoclonal IgG expres-sion (S3, S4, S5, S8, S10) (Supplemental Fig. S1 and Table 1).For comparison, we also expressed five VH:VL pairs that didnot contain shared intra-patient IGHV and IGHJ usage (S1,S2, S6, S7, S9), which we term singletons.

3.4. Recombinant antibodies from familieswith shared IGHV–IGHJ usage are enriched foranti-S. aureus binding

To determine whether the recombinant antibodies weexpressed could bind to S. aureus, we first performeddot-blots using crude lysates of the Wood46 strain ofS. aureus. The Wood46 strain of S. aureus contains only low

29%

57%

14%

57%

57%

14%

29%

14%

43%29%

14%

33%

33%

17%

17%17%

50%

17%

17%

17%

17%

33%

17%

17%

33%

33%

17%

17%

22%

50%

17%

11%

IGH

V1

1-3 1-8

1-46 1-69

1-18

IGH

V2

2-5

2-26

IGHV3

3-53

3-30

3-33

3-15

3-72

3-74

3-11

3-23

3-73-9

3-733-30-33-48

3-20

3-21

3-663-64

IGHV4

4-59

4-4

4-61

4-34

4-31

4-39

IGHV5

5-51

5-a

0.25

5-a

3-30

4-59

3-23

3-74

9%9%

64%

9%9%

3-7

1-181-69

3-21

%

13

A B

S10

S8S9S3 S2

5

J1

J2

J3

J4

J5

J6

not known IGHV families (>5 recoverable heavy chains)

IGHV families with high % of shared IGHJ genes (most frequent IGHJ is >2x greater than second-most frequent IGHJ)

Staph Patient 1 Staph Patient 2

17%

17%

17%

50%

10%

40%

20%

30%

38%

38%

13%

13%

9%9%

9%

18%18%

9%

27%

8%8%

23%

38%

8%

15%

25%

38%

13%

IGH

V1

1-3 1-

2 8-1 1-

46

1-69

1-24

1-18

IGH

V22-

702-

5

IGH

V3

3-53

3-30

3-33

3-15

3-72

3-74

3-11

3-7

3-93-73

3-4303-

33-

3-48

3-49

3-23

3-20

3-21

3-64

IGHV4

4-30-2

4-59

4-4

4-b4-61

4-34

4-39

IGHV6

6-1

0.25

3-15

3-30

1-2

3-23

4-59

3-7

S5

S4

S1

S7

S6

25%

Figure 2 Heavy-chain dendrograms of the plasmablast antibody repertoires of S. aureus-bacteremic individuals and shared HC V–Janalysis. Each peripheral node depicts a sequenced VH region with a unique plate- and well-ID barcode combination. Plasmablastantibody repertoires of two individuals with S. aureus bacteremia: (A) patient 1 and (B) patient 2. Red lines indicate the IGHV familieswith greater than 5 recovered full-length VH sequences, from which analysis of shared IGHJ was performed. Pie charts of the IGHJgene usage for each IGHV family are displayed adjacent to the branch representing each IGHV. For presentation, red bolded linesindicate the IGHV families where a prominent IGHJ gene was used. S1, S2, S3, S4, S5, S6, S7, S8, S9, and S10 denote the recombinantantibodies that were selected for recombinant expression. Circles denote antibodies that bind S. aureus, and squares denoteantibodies that do not. Recombinant antibodies from shared gene families are denoted in blue, and antibodies not sharing generearrangements are denoted in green.

82 D.R. Lu et al.

levels of protein A, thereby minimizing non-specific bindingto immunoglobulin. Because bacterial gene expressiondiffers significantly between early logarithmic (early-log)and stationary growth phases [25], we assessed antibodybinding to lysates from bacteria at both growth phases.Quantitative dot-blot analysis revealed that the antibodiesS2, S4, S5, S6, S8, and S10 bound to S. aureus in the early-loggrowth phase with at least three-fold greater intensity ascompared to the lysis buffer alone, whereas only S3 boundstrongly to S. aureus in the stationary phase (Fig. 3A).

We next used flow cytometry to determine whether anyof the recombinant antibodies could bind to the surface ofthe S. aureus Wood46 strain at three different growthphases. We found that antibody S3 bound to bacteria at allgrowth phases, whereas S8 bound to bacteria at early-logand mid-log phases, but not at the stationary phase(Supplemental Fig. S3). Antibodies S2, S4, S5, S6 and S10,which bound S. aureus in the dot-blot assay, did not bind

S. aureus in any of the growth phases using the standardgrowth medium alone (Fig. 3B).

During infection, S. aureus often adhere to connectivetissues and extracellular matrix proteins [26]. Fibroblastsare the most abundant cell type in soft connective tissue andsecrete many of the factors that comprise the extracellularmatrix [27]. Therefore, we sought to determine whetherculturing S. aureus in the presence of fibroblasts moreclosely resembles the environment of S. aureus colonizationand thereby induces differential regulation of S. aureussurface antigens. Flow cytometric analysis of S. aureusWood46 grown in the presence of TGF-β-stimulated NIH/3T3fibroblasts revealed that not only the S3, but also the S6and S10 recombinant antibodies can bind robustly to thebacterial surface (Fig. 3C and Supplemental Fig. S3). None ofthese antibodies bind to E. coli (Supplemental Fig. S3).Importantly, in all surface-binding experiments, two recom-binant antibodies that recognize the trivalent inactivated flu

B

A

C

S1 S2 S3 S4 S5 S6 S7 S8 S9S10

0

10

20

30

40early-log phase

mid-log phase

stationary phase

S1 S2 S3 S4 S5 S6 S7 S8 S9S10

F14 n

egco

ntrol

F26 n

egco

ntrol

Secon

dary

only

0

2

4

6

8

10

F14 n

egco

ntrol

F26 n

egco

ntrol

Seconda

ryon

ly

Sur

face

-bin

ding

eve

nts

(%

of c

ells

)S

urfa

ce-b

indi

ng e

vent

s (

% o

f cel

ls)

Dot

blo

t int

ensi

ty/ly

sis

buffe

r on

ly

S1 S2 S3 S4 S5 S6 S7 S8 S9S10

early-log phase lysate

0

1

2

3

4

5

6

F26 n

eg co

ntro

l

Figure 3 Screening the binding activity of recombinant antibodies generated from the plasmablasts of S. aureus-infectedindividuals. (A) Dot-blot analysis of antibody binding to lysates prepared from S. aureus Wood46 at early-log and stationary growthphases. Data are dot-blot densitometry values shown as a fold increase relative to control (lysis buffer only). (B, C) Flow cytometricdetection of antibody binding to the surface of S. aureus Wood46 at different growth phases in LB (B) or in the presence of NIH/3T3fibroblasts (C). Recombinant antibodies from shared gene families are denoted in blue, and antibodies not sharing generearrangements are denoted in green. Influenza-specific antibodies F14 and F26 were used as isotype controls. Data in (B) and (C) areshown as the mean ± s.e.m. of triplicates and are representative of three or more independent experiments *P b 0.05 by two-tailedunpaired t-test, comparing binding of S. aureus recombinant antibodies to that of the isotype control antibodies.

83Identification of functional anti-S. aureus antibodies

84 D.R. Lu et al.

vaccine did not bind to S. aureus, indicating the specificityof the antibody binding.

Collectively, our antibody-binding analyses show that allfive antibodies from clonal families with sharedintra-patient IGHV and IGHJ usage (i.e., S3, S4, S5, S8, andS10) bind to S. aureus, whereas only two of the fiveantibodies from families lacking common IGHV and IGHJusage (i.e., S2 and S6) bind to S. aureus (Table 1). Further,our results demonstrate differential binding of individualantibodies to S. aureus at different growth phases.

3.5. Recombinant antibodies derived fromperipheralblood plasmablasts bind to S. aureus virulence factorsand surface antigens

We used a full-proteome microarray of the S. aureus USA300strain (Antigen Discovery, Inc.)—a strain prevalent incommunity-acquired S. aureus outbreaks [18]—to identifythe targets of these S. aureus-specific recombinant antibod-ies. Antibodies S6 and S10 bound to SAUSA300_2441 andSAUSA300_2441-s1 with four-fold intensity over background(Fig. 4A). These two hits correspond to different segments ofthe same protein, fibronectin-binding protein A (FnBPA).FnBPA is a MSCRAMM (microbial surface components recog-nizing adhesive matrix molecules) expressed on the bacte-rial membrane that allows the bacteria to attach to theextracellular-matrix proteins fibronectin and fibrinogenand thereby promotes invasion of host tissues [28–30].Subsequent binding analyses showed that S6 and S10 bind tofull-length FnBPA and to an N-terminal fragment of FnBPA butnot a C-terminal fragment of FnBPA (Figs. 4B–C). Furthermore,antibody S4 bound moderately to superantigen-like protein5 (SSL5), a secreted virulence factor that inhibits neutrophilmigration (Fig. 4A) [31,32].

A large majority of healthy and infected individualscontain serum reactive to S. aureus peptidoglycan (PGN)and lipoteichoic acid (LTA) [7,33]. ELISAs using purifiedS. aureus PGN and LTA revealed that S6 bound robustly to

Table 1 Summary of recombinant antibodies derived from S. au

Monoclonalantibody a

Selectionrationale

HC + LC (HC) V–J mutcount b

S3 Shared HC V–J 29 (21)S4 Shared HC V–J 33 (26)S5 Shared HC V–J 45 (34)S8 Shared HC V–J 60 (38)S10 Shared HC V–J 21 (16)S1 Singleton 28 (22)S2 Singleton 60 (32)S6 Singleton 28 (18)

S7 Singleton 34 (19)S9 Singleton 54 (35)a Recombinant monoclonal IgG antibodies were generated from two

derived from patient 1. S1, S4, S5, S6, and S7 were derived from patieb Mutations of the heavy- and light-chain V and J segments were obtc LTA = lipoteichoic acid; SSL5 = superantigen-like protein 5; FnBPA

PGN, while antibodies S2 and S5 bound weakly (Fig. 5A), andS3 bound strongly to LTA (Fig. 5B).

3.6. Antibodies that bind to S. aureus differentiallypromote opsonization and killing of S. aureus

A major way in which antibodies protect against pathogens isthrough opsonization, a process in which antibodies coat thepathogen and thus mark it for ingestion and destructionby phagocytes. We used undifferentiated THP1 (humanmonocyte-like) cells and CFSE-labeled inactivated S. aureusto test whether the recombinant antibodies that bound tosurface S. aureus proteins could function as opsonins. Usingpositive and negative control antibodies and halting theassay at 30-minute intervals allowed us to optimize theduration of the assay (Supplemental Fig. S4A). We found thatS3 and S8 promoted phagocytosis of S. aureus at theearly-log growth phase (Fig. 6A), whereas only S3 promotedphagocytosis of S. aureus at the stationary phase (Fig. 6C).Antibodies S6 and S10 promoted phagocytosis of S. aureusgrown in the presence of fibroblasts (Fig. 6E and Supple-mental Fig. S4B). Blocking FcγRs with FcR Block beforeco-incubation of monocytes and bacteria abrogated theantibody-induced increase in phagocytosis, confirming thatS3, S6, S8, and S10 function as opsonins (Figs. 6B, D, F,Supplemental Fig. S4B). Notably, antibody S2, which boundto early-log phase bacterial lysate with the second highestintensity in the dot-blot screen, did not enhance internal-ization of S. aureus cells (Fig. 6E).

To test the antibacterial activity of the recombinantantibodies, we incubated them with live S. aureus anddifferentiated HL-60 (human neutrophil-like) cells andevaluated any antibody-associated decreases in formationof bacterial colonies, as a measure of the antibodies'antibacterial activity. The S. aureus-binding antibody S6reduced colony formation, whereas the S. aureus-bindingantibody S2 and a flu-specific control antibody did not(Fig. 6G).

reus bacteremic patients.

ation Binds S.aureus?

Antigen(s) c Enhances phagocytosis/killing

Yes LTA YesYes SSL5 NoYes Unknown –Yes Unknown YesYes FnBPA YesNo – –Yes PGN NoYes FnBPA,

PGNYes

No – –No – –

patients with S. aureus infections. S2, S3, S8, S9, and S10 werent 2.ained from IMGT.= fibronectin-binding protein A; PGN = peptidoglycan.

A

B

C

Figure 4 Identification of protein antigens targeted by plasmablast-derived recombinant antibodies. (A) Screen of recombinantantibody binding to a S. aureus USA300 proteome microarray. Recombinant antibodies from shared gene families are denoted in blue,and antibodies not sharing gene rearrangements are denoted in green. Measurement of antibody binding ranged from 5000 units(low-affinity) to 20,000 units (high-affinity). Mean fluorescence intensity is displayed and the scale bar represents the range offluorescence units measured. Only antigens with a binding intensity greater than two standard deviations of signal above backgroundare shown. (B) Schematic representation of the recombinant full-length and truncated N-terminal and C-terminal fragments of FnBPAwith domains annotated; numbers indicate the amino-acid residue position. (C) Antibody-binding assay plotting S6 and S10 bindingagainst recombinant full-length, truncated N-terminal, and truncated C-terminal fragments of FnBPA.

85Identification of functional anti-S. aureus antibodies

4. Discussion

In this study, we use a novel barcoding method that wedeveloped for the full-length sequencing of the paired VH andVL antibody sequences expressed by individual plasmablasts.This method, through cell-specific barcoding of mRNA,further enables us to bioinformatically index the readsrecovered from high-throughput sequencing to a given cellto accurately determine the proportion of B cells withshared V(D)J sequences and somatic mutations, therebycorrecting for biases in amplification that may occur duringlibrary amplification, emulsion PCR or sequencing. This

accurate determination of frequently used antibody rear-rangements and similar somatic mutations enables accurateidentification of clonal plasmablast populations as well ashighly selected somaticmutations. In addition, template-switchaddition of well-ID barcodes and the hybridization of plate-IDbarcodes onto universal adapter oligonucleotide at the 5′ endof the well-ID barcodes avoids the need for degenerateV-region primers which may fail to amplify certain allelicvariants or highly mutated V-regions, and thereby provideunbiased hybridization of PCR primers.

Barcode-enabled pairing of entire VH and VL sequencesfrom single cells allows us not only to profile the expansion

0

0.1

0.2

0.3

0.4

0.5

S1 S2 S3 S4 S5 S6 S7 S8 S9S10 F26

Abs

orba

nce

(OD

450)

0

0.5

1.0

1.5

2.0

S1 S2 S3 S4 S5 S6 S7 S8 S9S10 F26

Abs

orba

nce

(OD

450)

A

B

Figure 5 Identification of S. aureus cell wall componentstargeted by plasmablast-derived recombinant antibodies. (A)ELISA of recombinant antibody binding to S. aureus peptidogly-can. (B) ELISA of recombinant antibody binding to S. aureuslipoteichoic acid. Recombinant antibodies from shared genefamilies are denoted in blue, and antibodies not sharing generearrangements are denoted in green. Influenza-specific anti-body F26 was used as an isotype control. ELISAs were performedin triplicate and are presented as mean values of all repli-cates ± s.e.m.

86 D.R. Lu et al.

of plasmablast clonal families, but also to recombinantlyproduce relevant monoclonal antibodies secreted byplasmablasts and thus to characterize the binding andfunctional properties of endogenous antibodies generatedin human immune responses to S. aureus infection. Previoushigh-throughput methods for evaluating antibody reper-toires have largely focused on the sequence diversity of VHor heavy-chain CDR3 regions [10–12], precluding character-ization of the binding and functional properties of thespecific individual antibodies generated by each cell. Othermethods that sequence paired VH and VL [12,34] do notsequence through the full-length V-region, and as a resultnecessitate surrogate, non-native V-region sequences forrecombinant antibody expression. Our method, through

barcoding of all the cDNA generated by each individual Bcell, enables identification and matching of the cognateheavy- and light-chain antibody sequences expressed by thesame cell, and thereby the recombinant production of thebona fide antibody expressed by that cell. Recombinantantibody production enables binding and functional analysesof the expressed antibodies in the absence of other bloodantibodies, an advantage over approaches based on analysisof serum antibodies [35].

We used this approach to identify the antibodiesexpressed by plasmablasts generated in S. aureus bacter-emic individuals, and to use the recombinant antibodyproducts to determine whether they bind and/or killS. aureus. Seven of the ten (70%) recombinant antibodiesderived from two S. aureus-infected individuals bound toS. aureus. This percentage is similar to the findings ofprevious studies that investigated plasmablast specificityagainst vaccine protein antigens and viral peptides [9,8].The high percentage of S. aureus-specific antibodies fromfamilies with shared IGHV and IGHJ usage suggests that theanalysis of antibody families can facilitate identificationof candidate immunoglobulin rearrangements and somatichypermutations preferentially selected for during infec-tion. This also suggests that characterization of antibodiesexpressed by plasmablasts, the population of activatedB cells present in an immune response, may provide astrategy to identify antibodies targeting the critical anti-genic determinants in an anti-microbial immune response.Further studies with additional S. aureus-infected patientsor other infectious diseases will be necessary to elucidatewhether antibodies produced by rapidly expanding B-cellpopulations are most relevant to the containment ofinfection.

Of the seven antibodies that bound to S. aureus,four enhanced S. aureus opsonization. Pro-opsonophagocyticantibodies S6 and S10, which were derived from separatepatients, both bound to an extracellular region of FnBPAaway from its fibronectin-binding domains and C-terminalmembrane-spanning domain [36], suggesting that theseantibodies can recognize FnBPA and mediate opsonizationeven when it is bound to fibronectin. These antibodiesalso bind away from the FnBPA fibronectin-binding domainrepeats, which are known to exhibit inter-strain variance[37]. The finding that two antibodies containing uniquegermline gene segments from different patients bothbound to the FnBPA N-terminus suggests that the FnBPAN-terminus may be an immunodominant epitope duringS. aureus bacteremia. Further analysis is necessary todetermine whether the epitopes recognized by theseantibodies are conserved across multiple bacterial strains.Antigens targeted robustly by multiple patients withfavorable clinical outcomes may be useful for vaccinedesign.

In contrast, antibody S2, which also bound S. aureus,didn't promote opsonophagocytosis or bacterial killing.Emerging evidence reveals that antibodies that bind toS. aureus antigens can fail to promote opsonophagocytickilling independent of staphylococcal evasion mechanisms[38,39], which may be due to an antibody binding orientationthatmasks the accessibility of the Fc domain. This underscoresthe utility and need for recombinant antibody-enabledfunctional analyses.

S.a. e

arly

logon

ly

F14 1

0 µg

F26 1

0 µg

S3 10

µg

S3 5

µg

S3 1

µg

S8 10 µ

g

S8 5 µ

g

S8 1 µ

g

op 1

0 µg

op 5

µg

op 1

µg

0

20

40

60

S.a. e

arly

log o

nly

F14 1

0 µg

F26 1

0 µg

S3 10

µg

FcR b

lock S

3 10

µg

S810

µg

FcR b

lock S

8 10 µ

g0

20

40

60

S.a. s

tatio

nary

only

F14 1

0 µg

F26 1

0 µg

S3 10

µg

FcR b

lock S

310

µg

0

10

20

30

40

50

S.a. s

tatio

nary

only

F1410

µg

F2610

µg

S3 10

µg

S3 5

µg

S3 1 µg

S810

µg

S85 µ

g

S8 1 µg

op10

µg

op 5

µg

op 1

µg

0

10

20

30

40

50

***

**

n.s.

***

******

S.a. o

nly

F14 1

0 µg

F26 1

0 µg

S2 10

µg

S6 10

µg

S6 5

µg

S6 1

µg

S10 1

0 µg

S10 5

µg

S101

µg

op 1

0 µg

op 5

µg

op1

µg0

20

40

60

80

100

S.a. o

nly

F14 1

0µg

S210

µg

S6 10

µg

FcR b

lock S

6 10

µg

S10 1

0 µg

FcR b

lock S

1010

µg

0

20

40

60

80

100

F14 S2 S6

0

20

40

60

n.s.

**

* ** *

**

A B

C D

E F G

Ant

ibac

teria

l act

ivity

(m

ean

% d

ecre

ase

in C

FU

/ml)

(% in

tern

aliz

atio

n)(%

inte

rnal

izat

ion)

(% in

tern

aliz

atio

n)

(% in

tern

aliz

atio

n)(%

inte

rnal

izat

ion)

(% in

tern

aliz

atio

n)

S. a

ureu

s ph

agoc

ytos

is

S. a

ureu

s ph

agoc

ytos

is

S. a

ureu

s ph

agoc

ytos

is

S. a

ureu

s ph

agoc

ytos

is

S. a

ureu

s ph

agoc

ytos

is

S. a

ureu

s ph

agoc

ytos

is

***

Figure 6 Opsonophagocytic and bactericidal capability of recombinant antibodies generated from S. aureus-infected individuals.(A, C, E) Flow cytometric analysis of antibody-mediated phagocytosis. CFSE-labeled S. aureusWood46 fixed in early-log growth phase(A), fixed in stationary growth phase (C), or grown with NIH/3T3 fibroblasts (E), were incubated with undifferentiated THP1monocyte-like cells and 1 μg/ml, 5 μg/ml, or 10 μg/ml of recombinant antibody. (B, D, F) Flow cytometric analysis of FcRdependence of antibody-mediated phagocytosis. Early-log phase (B), stationary phase (D), and NIH/3T3-co-cultured (F),antibody-opsonized, CFSE-labeled S. aureus Wood46 were incubated with undifferentiated THP1 cells pre-treated withFcR-blocking human IgG. (G) Evaluation of the antibodies' antibacterial activity. Live S. aureus Wood46 were incubated withantibodies and activated HL-60 neutrophils. Bacteria were plated following HL-60 lysis, and mean percent decrease in CFU wasmeasured. Flu-specific antibodies F14 and F26 were used as isotype controls. op = anti-S. aureus polyclonal rabbit IgG antibodies.Data are from three experiments performed on separate days and are presented as mean values of all replicates ± s.e.m. Statisticalsignificance was determined by using two-tailed unpaired t-tests; n.s., not significant (P N 0.05); *P b 0.05; **P b 0.01; ***P b 0.001.

87Identification of functional anti-S. aureus antibodies

88 D.R. Lu et al.

Multiple recombinant plasmablast antibodies bound tonon-protein S. aureus antigens PGN and LTA. The majority ofplasmablasts in peripheral blood are believed to differenti-ate from B-cell precursors that have undergone conventionalCD4+ T-cell-dependent activation and somatic hypermutationin B-cell follicles, and are therefore specific for protein antigensand zwitterionic polysaccharides. Marginal zone and B1 B cellsare important in the immune response against bloodbornepathogens and can circulate in the blood as plasmablasts;however, their immunoglobulin genes are believed to undergolimited somatic hypermutation and antibodies secretedby these B-cell subsets show limited evidence of affinitymaturation. Interestingly, the sequence analysis revealedthat all antibodies had high mutation numbers (Table 1),suggesting that the plasmablasts expressing these antibod-ies were activated by uncharacterized mechanisms ofT-cell-dependent activation or that some extrafollicularB cells may undergo considerable affinity maturation thathas not previously been appreciated.

As anticipated, all recombinant antibodies bound toIgG-binding proteins A and Sbi in the antigen microarrays(data not shown), which are believed to have a role inbacterial evasion of the host immune system by disruptingopsonization and complement fixation [36,40]. Protein Sbihas also been implicated in immune evasion through thedegradation of C3 [41]. Further tests with Fab fragments arerequired to determine whether the antibodies bind specif-ically to protein Sbi. That S2, S5, and S8 did not bind to anyproteins on the microarray could be because of inter-strainvariation. Also, the cell-free Rapid Translation System usedfor expressing the S. aureus proteins for the microarray mayhave unexpectedly altered the conformation of certainS. aureus proteins, masking epitopes normally recognizedby these antibodies.

There were several technical limitations to the approachused in this study. Selection of antibodies for recombinantexpression on the basis of shared gene segments did notinclude antibodies with different sequences sharing CDRprotein structures and potentially similar binding properties.In addition, the limited sequencing depth per patient mayhave prevented detection of sequences representing smallclonal families. Future studies sequencing additional pa-tients and plasmablasts per patient will be necessary toelucidate clonal family frequencies and identify clonalfamilies shared between multiple patients.

The efficient recovery of anti-S. aureus reactive antibodiesfrom peripheral blood plasmablasts derived from S. aureusbacteremic humans indicates that this approach can efficientlyand effectively reveal antigens targeted by the anti-microbialB cell response. This strategy can be leveraged to identifyand isolate valuable diagnostic and therapeutic antibodies, tomonitor the status of the humoral immune response over thecourse of infection, and to determine critical epitopes forvaccine design and monitoring.

Supplementary data to this article can be found online athttp://dx.doi.org/10.1016/j.clim.2014.02.010.

Funding

This research was supported by the National Institutes ofHealth (NIH) RC1 AR058713 and NIH National Heart, Lung,

and Blood Institute Proteomics Center N01-HV-00242 toW.H.R.; and an Agency for Science, Technology and ResearchNSS (PhD) fellowship to Y-C.T.

Conflict of interest statement

Y-C.T. is an employee of and owns equity in Atreca, Inc.; andJ.S. and W.H.R. are consultants to and own equity in Atreca,Inc. All other authors have no disclosures.

Contributors

D.R.L. designed and performed experiments, analyzedresults, interpreted results, and wrote the manuscript.Y-C.T. designed and performed experiments, analyzedresults, interpreted results, and wrote the manuscript. S.K.performed bioinformatic analysis of antibody sequencingexperiments. X.C. cloned and expressed antibodies. E.A.S.designed experiments and interpreted results. T.M.L. con-tributed to interpretation of results, and wrote and editedthe manuscript. J.S. contributed to experimental design,data analysis and interpretation, and obtained bloodsamples. W.H.R. designed experiments, analyzed results,interpreted results, and wrote the manuscript.

Sequences assembled from deep sequencing are availableat GenBank (accession numbers KJ486843-KJ487434).

References

[1] D.A. Talan, et al., Comparison of Staphylococcus aureus from skinand soft-tissue infections in US emergency department patients,2004 and 2008, Clin. Infect. Dis. 53 (2) (2011) 144–149.

[2] E. Klein, D.L. Smith, R. Laxminarayan, Hospitalizations anddeaths caused by methicillin-resistant Staphylococcus aureus,United States, 1999–2005, Emerg. Infect. Dis. 13 (12) (2007)1840–1846.

[3] R.M. Klevens, et al., Invasive methicillin-resistant Staphylo-coccus aureus infections in the United States, JAMA 298 (15)(2007) 1763–1771.

[4] R.A. Proctor, Challenges for a universal Staphylococcus aureusvaccine, Clin. Infect. Dis. 54 (8) (2012) 1179–1186.

[5] S. Holtfreter, J. Kolata, B.M. Broker, Towards the immuneproteome of Staphylococcus aureus — the anti-S. aureusantibody response, Int. J. Med. Microbiol. 300 (2–3) (2010)176–192.

[6] P. Wood, et al., Recognition, clinical diagnosis and managementof patients with primary antibody deficiencies: a systematicreview, Clin. Exp. Immunol. 149 (3) (2007) 410–423.

[7] A. Dryla, et al., Comparison of antibody repertoires againstStaphylococcus aureus in healthy individuals and in acutelyinfected patients, Clin. Diagn. Lab. Immunol. 12 (3) (2005)387–398.

[8] J. Wrammert, et al., Rapid and massive virus-specificplasmablast responses during acute dengue virus infection inhumans, J. Virol. 86 (6) (2012) 2911–2918.

[9] J. Wrammert, et al., Rapid cloning of high-affinity humanmonoclonal antibodies against influenza virus, Nature 453(7195) (2008) 667–671.

[10] S.D. Boyd, et al., Measurement and clinical monitoring ofhuman lymphocyte clonality by massively parallel VDJ pyrose-quencing, Sci. Transl. Med. 1 (12) (2009) 12ra23.

89Identification of functional anti-S. aureus antibodies

[11] N. Jiang, et al., Lineage structure of the human antibodyrepertoire in response to influenza vaccination, Sci. Transl.Med. 5 (171) (2013) 171ra19.

[12] B.J. DeKosky, et al., High-throughput sequencing of the pairedhuman immunoglobulin heavy and light chain repertoire, Nat.Biotechnol. 31 (2) (2013) 166–169.

[13] W.C. Cheung, et al., A proteomics approach for the identifi-cation and cloning of monoclonal antibodies from serum, Nat.Biotechnol. 30 (5) (2012) 447–452.

[14] X. Brochet, M.P. Lefranc, V. Giudicelli, IMGT/V-QUEST: thehighly customized and integrated system for IG and TRstandardized V–J and V–D–J sequence analysis, Nucleic AcidsRes. 36 (Web Server issue) (2008) W503–W508.

[15] R.C. Edgar, MUSCLE: multiple sequence alignment with highaccuracy and high throughput, Nucleic Acids Res. 32 (5) (2004)1792–1797.

[16] S. Guindon, et al., New algorithms and methods to estimatemaximum-likelihood phylogenies: assessing the performance ofPhyML 3.0, Syst. Biol. 59 (3) (2010) 307–321.

[17] J. Huerta-Cepas, J. Dopazo, T. Gabaldon, ETE: a pythonEnvironment for Tree Exploration, BMC Bioinforma. 11 (2010) 24.

[18] F.C. Tenover, R.V. Goering, Methicillin-resistant Staphylococcusaureus strain USA300: origin and epidemiology, J. Antimicrob.Chemother. 64 (3) (2009) 441–446.

[19] K. Burnside, et al., Vaccination with a UV-irradiated geneti-cally attenuated mutant of Staphylococcus aureus providesprotection against subsequent systemic infection, J. Infect.Dis. 206 (11) (2012) 1734–1744.

[20] R. Ataide, et al., Using an improved phagocytosis assay toevaluate the effect of HIV on specific antibodies to pregnancy-associated malaria, PLoS One 5 (5) (2010) e10807.

[21] J. Nuutila, E.M. Lilius, Flow cytometric quantitative determi-nation of ingestion by phagocytes needs the distinguishing ofoverlapping populations of binding and ingesting cells, Cytom-etry A 65 (2) (2005) 93–102.

[22] T.R. Breitman, S.E. Selonick, S.J. Collins, Induction ofdifferentiation of the human promyelocytic leukemia cell line(HL-60) by retinoic acid, Proc. Natl. Acad. Sci. U. S. A. 77 (5)(1980) 2936–2940.

[23] Y.C. Tan, et al., High-throughput sequencing of natively pairedantibody chains provides evidence for original antigenic sinshaping the antibody response to influenza vaccination, Clin.Immunol. 151 (1) (2014) 55–65.

[24] Y.Y. Zhu, et al., Reverse transcriptase template switching: aSMART approach for full-length cDNA library construction,Biotechniques 30 (4) (2001) 892–897.

[25] A. Resch, et al., Differential gene expression profiling ofStaphylococcus aureus cultivated under biofilm and planktonicconditions, Appl. Environ. Microbiol. 71 (5) (2005) 2663–2676.

[26] A.H. Bartlett, K.G. Hulten, Staphylococcus aureus pathogene-sis: secretion systems, adhesins, and invasins, Pediatr. Infect.Dis. J. 29 (9) (2010) 860–861.

[27] R. Ross, The fibroblast and wound repair, Biol. Rev. Camb.Philos. Soc. 43 (1) (1968) 51–96.

[28] K. Jonsson, et al., Two different genes encode fibronectinbinding proteins in Staphylococcus aureus. The completenucleotide sequence and characterization of the secondgene, Eur. J. Biochem. 202 (3) (1991) 1041–1048.

[29] U. Schwarz-Linek, M. Hook, J.R. Potts, The molecular basis offibronectin-mediated bacterial adherence to host cells, Mol.Microbiol. 52 (3) (2004) 631–641.

[30] K. Ponnuraj, et al., A “dock, lock, and latch” structural modelfor a staphylococcal adhesin binding to fibrinogen, Cell 115 (2)(2003) 217–228.

[31] J. Bestebroer, et al., Staphylococcal superantigen-like 5 bindsPSGL-1 and inhibits P-selectin-mediated neutrophil rolling,Blood 109 (7) (2007) 2936–2943.

[32] S. Itoh, et al., Staphylococcal superantigen-like protein 5inhibits matrix metalloproteinase 9 from human neutrophils,Infect. Immun. 78 (7) (2010) 3298–3305.

[33] H. Wergeland, et al., Antibodies to Staphylococcus aureuspeptidoglycan and lipoteichoic acid in sera from blood donorsand patients with staphylococcal infections, Acta Pathol.Microbiol. Immunol. Scand. B 92 (5) (1984) 265–269.

[34] C.E. Busse, et al., Single-cell based high-throughput sequenc-ing of full-length immunoglobulin heavy and light chain genes,Eur. J. Immunol. 44 (2) (2014) 597–603.

[35] S. Sato, et al., Proteomics-directed cloning of circulatingantiviral human monoclonal antibodies, Nat. Biotechnol. 30(11) (2012) 1039–1043.

[36] C. Signas, et al., Nucleotide sequence of the gene fora fibronectin-binding protein from Staphylococcus aureus: useof this peptide sequence in the synthesis of biologically activepeptides, Proc. Natl. Acad. Sci. U. S. A. 86 (2) (1989) 699–703.

[37] K. Rice, et al., Variance in fibronectin binding and fnb locuspolymorphisms in Staphylococcus aureus: identification ofantigenic variation in a fibronectin binding protein adhesin ofthe epidemic CMRSA-1 strain of methicillin-resistant S. aureus,Infect. Immun. 69 (6) (2001) 3791–3799.

[38] H.K. Kim, et al., IsdA and IsdB antibodies protect mice againstStaphylococcus aureus abscess formation and lethal challenge,Vaccine 28 (38) (2010) 6382–6392.

[39] D. Skurnik, et al., Natural antibodies in normal human seruminhibit Staphylococcus aureus capsular polysaccharide vaccineefficacy, Clin. Infect. Dis. 55 (9) (2012) 1188–1197.

[40] T.J. Foster, Immune evasion by staphylococci, Nat. Rev.Microbiol. 3 (12) (2005) 948–958.

[41] E.J. Smith, et al., The Sbi protein is a multifunctional immuneevasion factor of Staphylococcus aureus, Infect. Immun. 79 (9)(2011) 3801–3809.