ORIGINAL ARTICLE

HIV-specific antibody-dependent phagocytosismatures during HIV infection

Fernanda Ana-Sosa-Batiz1, Angus PR Johnston2,3, Haiyin Liu2, Robert J Center4, Supachai Rerks-Ngarm5,Punnee Pitisuttithum6, Sorachai Nitayaphan7, Jaranit Kaewkungwal8, Jerome H Kim9, Nelson L Michael9,Anthony D Kelleher10, Ivan Stratov1, Stephen J Kent1,11 and Marit Kramski1

Antibody-dependent phagocytosis (ADP) is a potentially important immune mechanism to clear HIV. How HIV-specific ADP

responses mature during HIV infection or in response to vaccinations administered, including the partially successful RV144

HIV vaccine, is not known. We established a modified ADP assay to measure internalisation of HIV antibody (Ab)-opsonised

targets using a specific hybridisation internalisation probe. Labelled beads were coated with both biotinylated HIV gp140

envelope protein and a fluorescent internalisation probe, opsonised with Abs and incubated with a monocytic cell line. The

fluorescence derived from the fluorescent internalisation probe on surface-bound beads, but not from internalised beads, was

quenched by the addition of a complementary quencher probe. HIV Env-specific ADP was measured in 31 subjects during

primary infection and early chronic HIV infection. Although ADP responses were present early during HIV infection, a significant

increase in ADP responses in all 31 subjects studied was detected (Po0.001). However, when we tested 30 HIV-negative

human subjects immunised with the Canarypox/gp120 vaccine regimen (subjects from the RV144 trial) we did not detect

HIV-specific ADP activity. In conclusion, a modified assay was developed to measure HIV-specific ADP. Enhanced ADP responses

early in the course of HIV infection were observed but no ADP activity was detected following the vaccinations administered in

the RV144 trial. Improved vaccine regimens may be needed to capitalise on ADP-mediated immunity against HIV.

Immunology and Cell Biology advance online publication, 10 June 2014; doi:10.1038/icb.2014.42

Antibodies (Abs) are directed against multiple HIV epitopes1 andgenerally develop within a few weeks after infection. Neutralising Absdirected against the HIV envelope protein can prevent HIV infectionin vitro and prevent infection of macaques with SIV/HIV Envchimeric viruses.2–5 The development of neutralising Abs that areable to block infection of a broad range of HIV strains requires a longAb maturation process, which usually takes years to develop.6 HIVbroadly neutralising Abs are, however, difficult to elicit by standardvaccination regimes, as they are unable to recapitulate the prolongedpresence of antigen required for generating highly affinity maturedbroadly neutralising Abs.7

Although neutralising Abs are considered an efficient mode of Ab-mediated protection against HIV, Fc-mediated immune functions alsohave an important role in protection.8,9 In contrast to the prolongeddevelopment of broadly neutralising Abs, all HIV-infected individualsrapidly develop non-neutralising Abs during the primary phase oftheir infection. Abs can mediate different effector functions throughthe binding of their Fc region to the Fcg-receptor (FcgR), which

include antibody-dependent phagocytosis (ADP) and antibody-dependent cellular cytotoxicity (ADCC).10,11

Numerous studies now support the role for ADCC activity in thecontrol of HIV infection where the presence of ADCC correlates witha better disease prognosis and slow HIV progression.12–17 Incomparison with neutralising Abs, ADCC Abs develop muchquicker and require lower levels of Ab maturation.18

In contrast to the growing body of literature on ADCC Abs,comparatively little is known about ADP-mediating Abs in HIVinfection. ADP is a potentially important Ab effector function toeliminate immune complexes (that is, opsonised HIV) and HIV-infected cells. It is unknown when ADP Abs arise during HIVinfection and if they need a similarly long maturation process asneutralising Abs. It was previously shown that ADP-mediating Absfrom controllers and untreated progressors exhibited similar ADPactivity, which was also elevated compared with subjects on anti-retroviral therapy.19 However, little is known about ADP during earlyHIV infection.

1Department of Microbiology and Immunology at the Peter Doherty Institute for Infection and Immunity, The University of Melbourne, Melbourne, Australia; 2Drug Delivery,Disposition and Dynamics, Monash Institute of Pharmaceutical Sciences, Monash University, Parkville, Australia; 3ARC Centre of Excellence in Convergent Bio-Nano Scienceand Technology, Monash University, Parkville, Australia; 4Burnet Institute, Melbourne, Australia; 5Department of Disease Control, Ministry of Public Health, Bangkok, Thailand;6Faculty of Tropical Medicine, Mahidol University, Bangkok, Thailand; 7Armed Forces Research Institute of Medical Sciences, Bangkok, Thailand; 8Data Management Unit,Mahidol University, Bangkok, Thailand; 9USA Military HIV Research Program, Walter Reed Army Institute of Research, Rockville, MD, USA; 10The Kirby Institute for Infection andImmunity in Society, UNSW Medicine, Sydney, Australia and 11ARC Centre of Excellence in Convergent Bio-Nano Science and Technology, The University of Melbourne,Melbourne, AustraliaCorrespondence: Professor SJ Kent, Department of Microbiology and Immunology, at the Peter Doherty Institute for Infection and Immunity, University of Melbourne, 792Elizabeth Street, Parkville, Victoria 3010, Australia.E-mail: skent@unimelb.edu.au

Received 19 February 2014; revised 29 April 2014; accepted 29 April 2014

Immunology and Cell Biology (2014), 1–9& 2014 Australasian Society for Immunology Inc. All rights reserved 0818-9641/14

www.nature.com/icb

The role of ADP Abs in protection from HIV is also of growinginterest. Barouch et al.20 recently studied ADP in a macaquevaccination study using Adenovirus/poxvirus and adenovirus/adenovirus vectors expressing HIV-1 mosaic antigens. Using thisvaccination strategy they demonstrated that protection was correlatedwith functional non-neutralising ADP Abs. However, this is the onlystudy that investigated the potential of vaccine-induced HIV-specificADP Abs in preventing HIV infection. In humans, the partiallyprotective RV144 vaccine elicited minimal HIV-specific neutralisingAbs but did induce non-neutralising Abs.21 Post hoc analysesdemonstrated that ADCC was associated with protection from HIVacquisition in the RV144 trial22 and that serum IgA elicited by thevaccine interfered with binding and functional activity of ADCC-mediating serum IgG.23 However, it is unknown if non-neutralisingAbs with ADP function were also induced by the RV144 vaccineregimen.

To study the role played by Abs on clearance by phagocytosis ofvirus infected cells, it is necessary to use a robust method that is ableto evaluate ADP in vitro. The standard ADP assay developed byAckerman et al.24,25 assesses the phagocytic potential of Abs using Envprotein-coated fluorescent beads that are opsonised with purified IgGand subsequently incubated with the phagocytic cell line THP1, whichexpresses all three main FcgRs. A phagocytic score (% bead-positivecells�mean fluorescence intensity) is calculated as an indicator ofADP. However, this assay measures the sum of binding of beads onthe cell surface in addition to truly phagocytosed (that is, internalised)beads. To specifically assess the ability of Abs to assist internalisationof beads, we adapted a recent published method by Liu et al.26 to

study internalised beads. The assay involves adding a specifichybridisation internalisation probe (SHIP) that allows trulyphagocytised beads to be distinguished from surface-bound beads(Figure 1). We used this assay to investigate the development of ADP-mediating Abs during the early stages of HIV infection and todetermine whether they can be induced by the RV144 preventativevaccination strategy.

RESULTS

ADP-SHIP assay reliably discriminates between surface-bound andinternalised ADP targetsNon-neutralising Ab responses may assist in immunity to HIV andthere is a need to develop ADP assays to specifically measure HIV Ab-specific phagocytosis. The standard ADP assay published24 utilisesbeads with only one colour. Flow cytometry is used to determine the% of cells associated with beads, but it is not possible to distinguishbetween internalised beads and surface-bound beads (Figure 1a).

We developed a modified ADP assay, called ADP-SHIP, whichutilises both FITC and Cy5 (fluorescent internalisation probe, FIPCy5)labelled beads. This dual labelling allows for the discrimination ofsurface-bound and internalised beads after the addition of acomplementary quenching DNA sequence (QPC) at the end of theassay. The QPC cannot diffuse through the membrane of the cell and,therefore, it only binds to FIPCy5 on beads bound to the cell surfaceresulting in the quenching of the Cy5 fluorescence (Figure 1b).Surface-bound and internalised beads are easily detected by flowcytometry as FITCþ single-positive and FITCþCy5þ double-positive cells, respectively. All FITCþ cells are analysed for their

Antibody-Dependent Phagocytosis SHIP Assay

Standard Antibody-Dependent Phagocytosis Assay

Fluorescent bead

Fluorescent internalizationprobe (FIPCy5)

Biotinylated gp140

Quencher probe (QPC)

Immunoglobulin G Monocyte Excitation wavelength

Fc receptor

Figure 1 Antibody-dependent phagocytosis (ADP) assays. (a) Standard ADP assay setup is shown as described in Ackerman et al.24 FITC-labelled beads are

coated with HIV envelope protein (Env) gp140 and incubated with HIV- or HIVþ purified IgG (left panel). The opsonised beads are then incubated with the

monocytic cell line THP1 and ‘phagocytosed’ beads are identified by flow cytometry as FITCþ cells. This assay does not distinguish if beads are actually

internalised or if they only bind to the surface of the cell (middle and right panel). (b) The new ADP-SHIP assay is shown. FITC-labelled beads are coated

similar to the standard ADP assay, with HIV Env gp140 but also with a fluorescent internalisation probe (FIPCy5) and incubated with HIV� or HIVþpurified IgG (left panel). The opsonised beads are then incubated with the monocytic cell line THP1 and ‘phagocytosed’ beads are measured by flow

cytometry as FITCþ Cy5þ double-positive cells (middle panel). The addition of a complementary quenching probe (QPC) specifically blocks the Cy5

fluorescence derived from FIPCy5 on surface-bound beads, but not from internalised beads (right panel). Surface-bound beads are FITCþ single-positive and

internalised beads are double fluorescent (FITCþ Cy5þ ).

ADP in HIVFA-S Batiz et al

2

Immunology and Cell Biology

80.918.4

17.5

33.6

26.2

96.287.4

18.8

17.2

59.740.3

34.6

80.918.4

16.4

7525

Standard ADP ADP-SHIP

PBS

HIV -ve

HIV +ve

FITC+ FITC+ Cy5+ QPc+

FITC+FSC-A

FS

C-H

SS

C-A

FSC-A Cy5+

a bead + cells Beads InsideBeads outside

b

37ºC

4ºC

FITC+

Standard ADP ADP-SHIPc

27.99

10.26

12.3

Figure 2 Validation of the new ADP-SHIP assay. (a) Gating strategy for ADP-SHIP assay. The first two panels show gating strategy on live single cells. The

third panel shows gating on beadþ cells (i.e. cells associated with FITCþ beads, either internalised or on the surface). The fourth panel shows gating on

FITCþCy5þ (internalised beads) and FITCþCy5- (surface associated beads) cells. The filled histogram shows a 4 1C-negative control where minimal

numbers of cells have internalised beads. The open histogram shows a sample incubated at 37 1C where 80.9% of cells associated with beads have at least

one internalised bead. (b) Equal numbers of beadþ cells for both the standard ADP and the ADP-SHIP assay was detected. (c) Deconvolution microscopy

of ADP-mediated uptake. The left panel shows the standard ADP assay using FITC-labelled beads. The arrow for this panel shows beads associating with

cells but does not discriminate whether the beads are surface bound or internalised. The right panel shows the ADP-SHIP assay. Arrows for this panel

indicate either surface-bound beads (FITCþCy5�; green) or internalised beads (FITCþCy5þ ; orange). The cell membrane was stained with wheat germ

agglutinin Alexa Flour 594 (blue). Complementary quenching probe: QPC.

ADP in HIVFA-S Batiz et al

3

Immunology and Cell Biology

FITC+ Cy5+ QPc+

FITC+

3729.7

HIV+ HIV- HIV+ HIV-0

5

10

15

20

0

5

10

15

20

25

30

35

40P

hag

ocy

t ic

sco

r e[(

%b

ead

s+ce

llsx

MF

I)/1

0e4]

%F

I TC

+C

y5 +c e ll s

Standard ADP ADP-SHIP

1 bead per cell

2 beads per cell

> 2 beads per cell

Surface bound beads

HIV+ HIV-0

10

20

30

40

%F

ITC

+C

y5-

cells

15.9

8

5.8

Inte

rnali

sed

Surfa

ce b

ound

Inte

rnali

sed

Surfa

ce b

ound

Inte

rnali

sed

Surfa

ce b

ound

0

5

10

15

20

% F

ITC

+ ce

lls

1 bead / cell 2 beads / cell >2 beads / cell

ADP in HIVFA-S Batiz et al

4

Immunology and Cell Biology

Cy5 fluorescence (gating shown in Figure 2a). Figure 2b shows thatthe additional binding of the FIPCy5 does not diminish the overall %of beads positive cells. This was independent of whether gp140 wasbound before, after, or simultaneously with FIPCy5 (SupplementaryFigure S1A). It further showed that for IgG from healthy donors the% of phagocytosed beads is overestimated as a proportion of thebeads that are only bound to the surface of the cells. One advantage isthat the ADP-SHIP assay directly determines the % of cells that haveinternalised beads (that is, are double-positive FITCþ FIPCy5). Inaddition, the number of beads phagocytosed by each cell (one per cell,two per cell, or more than two per cell) can be reliably determined.Note that there is a low level of internalisation of gp140-coated beadsin the absence of HIV Abs, and thus the ADP-SHIP assay assesses theincrease in phagocytosis above this level (Figure 2b). Discriminationof surface-bound and internalised beads as seen by flow cytometrywas confirmed by deconvolution microscopy (Figure 2c).

Similar to the standard ADP assay, the ADP-SHIP assay needsoptimal conditions for each gp140 protein used including (i) gp140protein concentration used for binding to the beads (SupplementaryFigure S1C), (ii) degree of gp140 biotinylation (SupplementaryFigures S1D and E) and (iii) IgG concentration (SupplementaryFigure S1C). These factors were carefully optimised (shown inSupplementary Figure S1). The optimal levels of gp140 proteincoating were 0.5mgml�1 beads, the degree of biotinylation was20-fold biotin excess and the concentration of IgG employed in theassay was 1mg ml�1 . Coating of the beads with gp140 did not lead toan increase in binding of the beads to the surface of the target cell(Supplementary Figure S1F).

To validate the new ADP-SHIP assay, we assessed ADP activity in20 HIV-infected individuals and compared it with ADP activity of 20healthy donors. Both the standard ADP and ADP-SHIP assays allowedfor discrimination between HIVþ and HIV� ADP activity. It isdifficult to directly compare these two assays because of the differentreadouts; however, the difference in the ratio of phagocytosis of HIV-positive and HIV-negative samples was 2.08 for the standard ADPassay and 2.04 for the ADP-SHIP assay (ratio of either the meanphagocytic score or % FITCþCy5þ cells of HIVþ /HIV� samples)(Figure 3a). However, we did note that the ADP-SHIP assay had awider dynamic range compared with the standard assay. There was nodifference in the levels of surface-bound gp140-coated beads betweenthe HIV� and HIVþ samples in the ADP-SHIP assay (Figure 3b).The coefficient of variation of the ADP-SHIP and standard ADP assaywere similar at 4.56% and 7.24%, respectively across 10 HIVþreplicates (Supplementary Figure S1G). When the number of beadsper cell was analysed, the ADP-SHIP assay revealed that the majorityof cells had only phagocytosed one bead per cell (mean: 14.1% of allcells; range 5.4–19.5%) but also a large proportion of one surface-bound bead per cell (mean: 6.3%; range 4.9–10%) was detected. Asmaller proportion of cells had two (mean 5.9%, range 1.2–9.8%) ormore than two beads per cell (mean 4.1%; range 0.47–8.7%)phagocytosed (Figure 3d). Almost none of the cells had two or morethan two surface-bound beads per cell (mean: 1.1%; range 0.6–3.5%).

Overall, this equates to 24.1% of cells with phagocytosed beads,whereas 7.4% of the cells have surface-bound beads. For this analysisit is not necessary to use microscopy to detect the number ofinternalised beads. Through flow cytometry, it is possible to detect thewhole cell and not only slices through the cell, giving the possibility ofdetermining the number of beads internalised per cell.

ADP Abs during early HIV infection increase over timeLittle is known about the development of ADP Abs during the courseof HIV infection. We, therefore, determined ADP in HIV-infectedpatients (n¼ 31) who were not on treatment during primary andearly chronic infection. Serum samples were paired and medianduration of primary and early chronic infection was 4 months and 25months, respectively. All subjects displayed ADP Abs at the early timepoint, which was well above the ADP signal for HIV� subjects(Figure 4a). Over the 2-year–follow-up, ADP activity increasedsignificantly in 30 of the 31 individuals. For primary and earlychronic infection samples, there was a respective median±interquar-tile range ADP activity of 14.31%±7.20 and 25.19%±6.28% FITCþCy5þ cells. When the data were broken down by the number ofinternalised beads per cell (1, 2 or 42), we found an increase in ADPacross all subgroups in the early chronic infection samples(Figure 4b). There was a significant increase in the number of cellswith one internalised bead for early chronic infection samples(median: 15.61%; range: 11.54–20.55%) compared with primaryinfection samples (median: 9.93%; range: 7.21–18.14%). The numberof cells with more than one bead internalised was also increased(median for primary infection samples: 4.59%; range: 2.85–13.16%;median for early chronic infection samples: 9.82%; range:5.37–19.10%). Cells with 1, 2 or 42 beads only bound to the surfaceof the cell and was independent of primary and early chronicinfection (Figure 4c). As there was a range of time points at whichthe samples were collected (from 1–34 months after the estimatedtime of HIV infection), we plotted the duration of infection againstthe level of ADP. We found ADP Abs were correlated with theduration of infection (Figure 4d).

No detectable ADP Abs in RV144 vaccineesThe ADP-SHIP assay was used to assess ADP activity in HIVvaccinees of the RV144 trial21 in order to gain insight into whetherADP can be elicited by preventative vaccines and to dissect theirpotential role in protection. Thirty samples from vaccine recipientsand 30 samples from placebo recipients were kindly provided by theThai-MHRP, all 2 weeks after the final vaccination (week 26 of thetrial). The IgG from plasma was first tested for ADP activity againstEnv gp140 subtype B (AD8), which was one of the two subtypes usedfor the protein boost used in the vaccination trial. Using the ADP-SHIP assay, no significant difference between vaccinees and placebocould be detected (Figure 5a). Vaccine and placebo samples from theRV144 study did not show ADP activity above the HIV-negativesample and were well below the HIVþ control samples. The RV144vaccine regimen also employed a protein boost with a subtype A/E

Figure 3 Comparison of antibody-dependent phagocytosis with the standard ADP and the ADP-SHIP assays. (a) The phagocytic activity of 20 HIV-infected

and 20 healthy individuals was measured using purified IgG from plasma. Readout is as follows: standard ADP assay: phagocytic score (% beads-positive

cells�mean fluorescence intensity)/104) (left y axis); ADP-SHIP assay: % FITCþ and Cy5þ cells (right y axis). The error bars represent mean ±s.e.m.

****P o0.0001 was determined using Student’s t-test. (b) Levels of surface-bound beads in the ADP-SHIP assay are similar for 20 HIVþ and 20

HIV� IgG samples. (c) Analysis of the number of beads per cell (internalised (FITCþ Cy5þ ) vs surface bound (FITCþ Cy5�) beads) using the ADP-SHIPassay. In this assay, 29.7% of cells have at least one internalised bead (middle panel) and this can be broken down to cells with 1, 2 or 42 beads by

gating on the peaks as shown in the right panel. (d) Analysis of the numbers of internalised beads in the presence of 20 HIVþ sera. The error bars

represent mean ±s.e.m.

ADP in HIVFA-S Batiz et al

5

Immunology and Cell Biology

Env protein. All samples were, therefore, also tested for their ADPactivity against Env gp140 subtype A/E 966, which was the samesubtype used for the vaccine DNA prime. No significant differencecould be detected in ADP activity between placebo and vaccine

recipients using either 966 or AD8 gp140 (Figure 5b). There were alsono differences in levels of surface-bound beads across the threegroups for either the subtype B or A/E gp140 protein-coated beads(Figures 5c and d).

Primar

y

Early

Chron

ic

Primar

y

Early

Chron

ic

Primar

y

Early

Chron

ic0

5

10

15

20

25

Internalised beads per cell

1 bead/cell 2 beads/cell >2 beads/cell

Primary Early Chronic5

10

15

20

25

30

35

40

Duration of infection

% F

ITC

+ C

y5+

cells

% F

ITC

+ C

y5+

cells

0 10 20 30 40 500

10

20

30

40

% internalised beads+ cells

Tim

e af

ter

in fe

ctio

n(m

onth

s)

R2 0.5503p < 0.0001

Primar

y

Early

Chron

ic

Primar

y

Early

Chron

ic

Primar

y

Early

Chron

ic0

2

4

6

8

% F

ITC

+ C

y5-

cells

Surface bound beads per cell

1 bead/cell 2 beads/cell >2 beads/cell

Figure 4 Development of ADP antibodies in HIV-infected patients during primary and early chronic infection. (a) ADP responses were evaluated in HIV-

infected individuals (n¼31) using paired serum samples. Median primary infection was 4 months and median early chronic infection was 2 years 1 month.

****P o0.0001 was determined using Wilcoxon matched-pairs signed rank test. (b) Analysis of the number of beads per cell comparing internalised

(FITCþCy5þ ) and (c) surface-bound (FITCþCy5-) beads in the presence of HIV Abs during primary and early chronic HIV infection (n¼31). The error

bars represent median±IQR. ****Po0.05 was determined using Wilcoxon matched-pairs signed rank test. (d) Correlation of ADP responses with monthsafter HIV infection (n¼31) R2¼0.55 and P o0.0001 determined using Pearson’s correlation.

ADP in HIVFA-S Batiz et al

6

Immunology and Cell Biology

DISCUSSION

There is limited knowledge about the potentially important role ofADP Abs in the control or prevention of HIV infection. Such a role issupported by (i) the correlation of FcgR2a polymorphisms with HIVdisease progression,27 (ii) the correlation of the presence ofADP-mediating IgG2 (binding more efficiently to FcgRIIa) withdelayed progression28,29 and (iii) HIV-associated downregulation ofFcgR2 expression in progressive infection leading to reduced ADPactivity.30

To standardise the measurement of ADP-mediating Abs Ackermanet al.24 developed an ADP assay that employs a widely utilisedmonocytic cell line that expresses a range of FcgRs, including bothinhibitory and activating FcgRs making it a highly useful assay tomeasure ADP function in both HIV infection and vaccine-mediatedprotection. We developed a modified assay, the ADP-SHIP assay,which is able to specifically discriminate between surface-boundand internalised targets by flow cytometry without the need ofemploying imaging techniques. Although the standard ADP and theADP-SHIP assay displayed comparable results, the dynamic range of

responses detected was greater for the ADP-SHIP assay. Further,this assay can reliably detect the number of beads internalisedper cell and the proportion of target cells with only surface-boundbeads. It remains to be shown whether the ADP-SHIP assay offersany improvement in assessing the functional significance of ADP inHIV immunity over the standard ADP. Although we did notfind a difference in levels of surface-bound beads between HIV- andHIVþ samples, it is possible that future studies assessing cells atdifferent time points after exposing opsonised beads to the targetcells might reveal subtle differences. We also note that both thestandard ADP and the ADP-SHIP assays result in a level ofbackground association and internalisation of gp140-coated beadsin the absence of HIV Abs. Thus, both assays measure the Ab-dependent increase in phagocytosis (shown by comparing assays ofHIV�compared with HIVþ IgG samples) rather than the total levelof phagocytosis.

Using the ADP-SHIP assay, we demonstrated that ADP-mediatingAbs develop early during infection and that ADP responses increaseover time in the absence of treatment. Whether this increase in ADP

Vaccinees Placebo HIV+0

5

10

15

%F

ITC

+C

y5-

cells

Vaccinees Placebo HIV+0

5

10

15

%F

I TC

+C

y 5-

cel ls

0

5

10

15

20

30

40

Vaccinees Placebo HIV+

% F

ITC

+ C

y5+

cells

Vaccinees Placebo HIV+0

2

4

6

10

20

30

%F

ITC

+C

y5+

cells

Figure 5 Assessment of ADP in the RV144 HIV vaccine and placebo recipients. ADP was determined for 30 vaccine and 26 placebo recipients at week 26

of the trial. As a positive control ADP for 15 HIV-infected subjects was measured. The grey dotted line indicates ADP for an HIV-negative subject. ADP was

measured against (a) Env gp140 subtype B (strain AD8) and (b) Env gp140 subtype A/E (strain 966). Surface-bound beads are shown for HIV-1AD8 and

HIV-1966 gp140-coated beads in (c) and (d) respectively. The error bars represent mean±s.e.m.

ADP in HIVFA-S Batiz et al

7

Immunology and Cell Biology

activity is due to the overall increase in ADP Ab titres or due to theoverall maturation of the ADP response (specificity and avidity)remains to be elucidated. Ongoing studies from our own laboratoryhave also shown that ADP responses are reduced by anti-retroviraltreatment (data not shown).

Inducing protective Ab responses through vaccination is a goal ofHIV vaccine studies. There is recent evidence on a potential role forvaccine-induced ADP in protecting macaques from SIV infection.20

The only human HIV vaccine trial to date to show some success is theRV144 trial, and we, therefore, analysed plasma samples from a subsetof vaccinees for ADP responses. We could not detect HIV Env-specificADP responses to either B or A/E subtype Env proteins followingvaccination. A recent publication shows a low level of ADP may beinduced by the RV144 regimen using the standard ADP assay.31

However, no ADP assays on sera from placebo-vaccinated or HIV-infected subjects were reported in that study. It will be important toanalyse further samples from RV144 trial vaccinated and placebosubjects with more sensitive ADP assays in future studies. Althoughspeculative, it is possible that improved immunity to HIV may beachievable if vaccine regimens induce a broader range of non-neutralising Ab responses, including ADP.

To conclude, the new ADP-SHIP assay now provides a valuable andspecific tool to further study ADP-mediating Abs in the context ofvaccination and HIV infection.

METHODS

Patients and plasma samplesTo establish the new ADP-SHIP assay, we studied serum samples from HIV-

infected subjects (n¼ 20) at various disease stages and from healthy subjects

(n¼ 20). Further longitudinal serum samples from patients (n¼ 31) during

primary and early chronic infection (4.7 months after infection and 25.9

months after infection) were also assessed for their ADP activity. The list of

patients and their clinical data (VL, CD4 count, time point of infection) is

presented in Table 1. To investigate ADP after vaccination serum samples of

randomly selected vaccine recipients (n¼ 30) and placebo recipients (n¼ 26)

of the RV144 vaccine trial21 were tested at week 26 of the trial. Informed

consent was obtained from all subjects, and the study was approved by the

respective institutional ethics committees.

HIV-IG (NIH AIDS Reagent Program, Division of AIDS, NIAID, NIH:

Catalogue no. 3957 from NABI and NHLBI) was used as a positive control in

all ADP assays.

IgG purification from plasmaTotal IgG was purified from 50–100ml plasma using the Protein G HP

Multitrap and the Antibody Buffer Kit (both GE Healthcare, Buckinghamshire,

UK) according to manufacturer’s instructions. Elution was repeated two times

and pooled elutes were washed twice with PBS using 30k Amicon Ultra-4

Centrifugal Filter Units (Millipore, Bedford, MA, USA). IgG concentration was

determined using the NanDrop 2000.

HIV Env gp140 protein and biotinylationHIV Env gp140 protein (subtype B: AD8 clone of ADA; subtype A/E: 966,

clone 93TH966-89) was secreted by stably transfected HeLa cells and purified

by lentil–lectin affinity chromatography and size exclusion chromatography

using a 16/60 Superdex 200 column as previously described.32 Fractions

enriched for gp140 trimers were pooled and concentrated. Briefly, 20–200mg of

gp140 Env was then biotinylated using EZ-Link Sulfo-NHS-LC-Biotin kit

(Thermo Scientific, Rockford, IL, USA) according to the manufacture’s

instructions using a 20-fold biotin reagent excess. Excess biotin was removed

by washing twice with PBS using the 30 k Amicon Ultra-4 Centrifugal Filter

Units (Millipore).

Standard ADP assayThe standard ADP assay was performed as previously described.24 Briefly,

0.75mg of biontinylated gp140 (unless stated otherwise gp140 AD8 was used)

and 3ml FITC-labelled NeutrAvidin FluoSpheres (beads 1mm) (Invitrogen,

Carlsbad, CA, USA) were incubated together overnight at 4 1C and

subsequently washed with 2% BSA-PBS to remove unbound gp140. Ten

microliter of 1:10 diluted coated beads were opsonised with 1mg ml�1 purified

human IgG, for 2 h at 37 1C before incubated with 1� 10e5 THP1 cells

(monocytic cell line, ATCC TIB-202) in a total volume of 200ml. After an

incubation of 16 h at 37 1C cells were washed with PBS fixed in 1%

formaldehyde before acquired on a LSRII (BD Biosciences, San Jose, CA,

USA) and analysed using FlowJo analysis software version 9.6.2.

ADP-SHIP assayThe ADP-SHIP assay utilises a SHIP. Similar to the standard ADP, 0.75mg

biotinylated HIV Env gp140 together with 3ml of FITC-labelled NeutrAvidin

FluoSpheres (beads 1mm) (Invitrogen) and 3ml of 150mM biotin- and

Cy5-labelled florescent internalisation probe (FIPCy5) (50 Cy5-TCAGTTCAGG

ACCCTCGGCT-N3 30, Integrated DNA Technologies, Coralville, IA, USA)

(26) were incubated overnight at 4 1C. Beads were then washed twice with 2%

BSA-PBS to remove unbound gp140 and FIPCy5. 10ml of a 1:10 dilution of

coated beads were opsonised with 1mg ml�1 purified human IgG for 2 h at

37 1C, and then incubated with 1� 10e5 THP1 cells in a total volume of

200ml. After a 16 h incubation at 37 1C cells were washed with ice cold PBS and

surface-bound beads were quenched by adding 1.01mgml�1 of the comple-

mentary quenching probe (QPC) (50-AGCCGAGGGTCCTGAACTGA-BHQ2-

30 Integrated DNA Technologies) for 10 min at 4 1C.26 Cells were fixed in 1%

formaldehyde before acquired on the LSRII (BD) and analysed using

FlowJo analysis software version 9.6.2. The proportion of cells with surface-

bound or internalised beads was determined by flow cytometry as shown

in Figure 2a. The number of beads internalised (1, 2 or 42) was determined

by gating on the peaks associated with internalised beads as shown in

Figure 3c (right panel). There was a high degree of concordance of this

analysis when we assessed 10 replicates of a HIVþ sera sample, with the

coefficient of variation of 1, 2, and 42 beads per cell being 3.89%, 3.62% and

3.56%, respectively.

Deconvolution microscopySamples used for microscopy were treated the same way as described for the

ADP and ADP-SHIP assay with the exception that a membrane staining step

was performed before fixing. Briefly, 30ml of 5mg ml�1 of Wheat germ

agglutinin (Alexa Fluor594, Life Technologies, Carlsbad, CA, USA) was added

to the cells and incubated for 15 min at 4 1C. Cells were subsequently washed

with PBS and fixed in 30ml 4% formaldehyde before being imaged on an

Applied Precision DeltaVision Deconvolution Microscope. Deconvolution and

image processing was performed using SoftWorx software.

Table 1 Cohort characteristics for primary and early chronic HIV infection

Primary infection Early chronic infection

Median CD4 count (range) 723 cellsml�1 (168–1220) 570 cellsml�1 (76–818)

Median HIV-viral load (range) 4.01 Log10 copies ml�1 plasma (2–5.88) 4.19 Log10 copiesml�1 plasma (1.69–5.32)

Median time after infection (range) 4 months (1–11) 25 months (18–34)

ADP in HIVFA-S Batiz et al

8

Immunology and Cell Biology

ACKNOWLEDGEMENTSThis work was supported by NHMRC program grant no. 510448. The views

expressed in this manuscript are those of the authors and do not reflect the

views of the US Army or the US Department of Defense. We thank the RV144

trial participants, the whole RV144 study team and all patients and blood

donors. We thank Dr Wendy Winnall for statistical advice.

1 Alter G, Moody MA. The humoral response to HIV-1: new insights, renewed focus.J Infect Dis 2010; 202(Suppl 2), S315–S322.

2 Hessell AJ, Rakasz EG, Poignard P, Hangartner L, Landucci G, Forthal DN et al.Broadly neutralizing human anti-HIV antibody 2G12 is effective in protection againstmucosal SHIV challenge even at low serum neutralizing titers. PLoS Pathog 2009; 5:e1000433.

3 Parren PWHI, Marx PA, Hessell AJ, Luckay A, Harouse J, Cheng-Mayer C et al.Antibody protects macaques against vaginal challenge with a pathogenic R5 SIV HIV atserum levels giving complete neutralization in vitro. J Virol 2001; 75: 8340–8347.

4 Mascola JR, Lewis MG, Stiegler G, Harris D, VanCott TC, Hayes D et al. Protection ofMacaques against pathogenic simian/human immunodeficiency virus 89.6PD bypassive transfer of neutralizing antibodies. J Virol 1999; 73: 4009–4018.

5 Moog C, Dereuddre-Bosquet N, Teillaud JL, Biedma ME, Holl V, Van Ham G et al.Protective effect of vaginal application of neutralizing and nonneutralizing inhibitoryantibodies against vaginal SHIV challenge in macaques. Mucosal Immunol 2014; 7:46–56.

6 Stamatatos L, Morris L, Burton DR, Mascola JR. Neutralizing antibodies generatedduring natural HIV-1 infection: good news for an HIV-1 vaccine? Nat Med 2009; 15:866–870.

7 Haynes BF, Kelsoe G, Harrison SC, Kepler TB. B-cell-lineage immunogen design invaccine development with HIV-1 as a case study. Nat Biotechnol 2012; 30: 423–433.

8 Hessell AJ, Hangartner L, Hunter M, Havenith CEG, Beurskens FJ, Bakker JM et al. Fcreceptor but not complement binding is important in antibody protection against HIV.Nature 2007; 449: 101–104.

9 Hezareh M, Hessell AJ, Jensen RC, van de Winkel JG, Parren PW. Effector functionactivities of a panel of mutants of a broadly neutralizing antibody against humanimmunodeficiency virus type 1. J Virol 2001; 75: 12161–12168.

10 Forthal DN, Moog C. Fc receptor-mediated antiviral antibodies. Curr Opin HIV AIDS2009; 4: 388–393.

11 Huber M, Trkola A. Humoral immunity to HIV-1: neutralization and beyond. J InternMed 2007; 262: 5–25.

12 Lambotte O, Ferrari G, Moog C, Yates NL, Liao H-X, Parks RJ et al. Heterogeneousneutralizing antibody and antibody-dependent cell cytotoxicity responses in HIV-1 elitecontrollers. AIDS 2009; 23: 897–906.

13 Baum LL, Cassutt KJ, Knigge K, Khattri R, Margolick J, Rinaldo C et al. HIV-1gp120-specific antibody-dependent cell-mediated cytotoxicity correlates with rate ofdisease progression. J Immunol 1996; 157: 2168–2173.

14 Ahmad R, Sindhu ST, Toma E, Morisset R, Vincelette J, Menezes J et al. Evidence for acorrelation between antibody-dependent cellular cytotoxicity-mediating anti-HIV-1antibodies and prognostic predictors of HIV infection. J Clin Immunol 2001; 21:227–233.

15 Johansson SE, Rollman E, Chung AW, Center RJ, Hejdeman B, Stratov I et al. NK cellfunction and antibodies mediating ADCC in HIV-1-infected viremic and controllerpatients. Viral Immunol 2011; 24: 359–368.

16 Wren LH, Chung AW, Isitman G, Kelleher AD, Parsons MS, Amin J et al. Specificantibody-dependent cellular cytotoxicity responses associated with slow progression ofHIV infection. Immunology 2013; 138: 116–123.

17 Chung AW, Navis M, Isitman G, Wren L, Silvers J, Amin J et al. Activation of NK cellsby ADCC antibodies and HIV disease progression. J Acquir Immune Defic Syndr 2011;58: 127–131.

18 Bonsignori M, Pollara J, Moody MA, Alpert MD. ADCC-mediating antibodies from anHIV-1 vaccine efficacy trial target multiple epitopes and preferentially use the VH1gene family. J Virol 2012; 86: 11521–11532.

19 Ackerman ME, Dugast A-S, McAndrew EG, Tsoukas S, Licht AF, Irvine DJ et al.Enhanced phagocytic activity of HIV-specific antibodies correlates with naturalproduction of immunoglobulins with skewed affinity for FcgR2a and FcgR2b. J Virol2013; 87: 5468–5476.

20 Barouch DH, Stephenson KE, Borducchi EN, Smith K, Stanley K, McNally AG et al.Protective efficacy of a global HIV-1 mosaic vaccine against heterologous SHIVchallenges in rhesus monkeys. Cell 2013; 155: 531–539.

21 Rerks-Ngarm S, Pitisuttithum P, Nitayaphan S, Kaewkungwal J, Chiu J, Paris R et al.Vaccination with ALVAC and AIDSVAX to prevent HIV-1 infection in Thailand. New EnglJ Med 2009; 361: 2209–2220.

22 Haynes BFB, Gilbert PBP, McElrath MJM, Zolla-Pazner SS, Tomaras GDG, Alam SMSet al. Immune-correlates analysis of an HIV-1 vaccine efficacy trial. New Engl J Med2012; 366: 1275–1286.

23 Tomaras GD, Ferrari G, Shen X, Alam SM, Liao H-X, Pollara J et al. Vaccine-inducedplasma IgA specific for the C1 region of the HIV-1 envelope blocks binding and effectorfunction of IgG. Proc Natl Acad Sci USA 2013; 110: 9019–9024.

24 Ackerman ME, Moldt B, Wyatt RT, Dugast A-S, McAndrew E, Tsoukas S et al. A robust,high-throughput assay to determine the phagocytic activity of clinical antibodysamples. J Immunol Methods 2011; 366: 8–19.

25 McAndrew EG, Dugast A-S, Licht AF, Eusebio JR, Alter G, Ackerman ME. Determiningthe phagocytic activity of clinical antibody samples. J Vis Exp 2011; (57), e3588–e3588.

26 Liu H, Johnston A. A programmable sensor to probe the internalization ofproteins and nanoparticles in live cells. Angew Chem Int Ed Engl 2013; 52:5744–5783.

27 Forthal DN, Gabriel EE, Wang A. Association of Fcg receptor IIIa (FcgRIIIa) genotypewith the rate of HIV infection following gp120 vaccination. Blood 2012; 120:2836–2842.

28 French MA, Tanaskovic S, Law MG, Lim A, Fernandez S, Ward LD et al. Vaccine-induced IgG2 anti-HIV p24 is associated with control of HIV in patients with a‘high-affinity’ FcgammaRIIa genotype. AIDS 2010; 24: 1983–1990.

29 Ngo-Giang-Huong N, Candotti D, Goubar A, Autran B, Maynart M, Sicard D et al. HIVtype 1-specific IgG2 antibodies: markers of helper T cell type 1 response andprognostic marker of long-term nonprogression. AIDS Res Hum Retroviruses 2001;17: 1435–1446.

30 Dugast A-S, Tonelli A, Berger CT, Ackerman ME, Sciaranghella G, Liu Q et al.Decreased Fc receptor expression on innate immune cells is associated with impairedantibody-mediated cellular phagocytic activity in chronically HIV-1 infected indivi-duals. Virology 2011; 415: 160–167.

31 Chung AW, Ghebremichael M, Robinson H, Brown E, Choi I, Lane S et al.Polyfunctional Fc-effector profiles mediated by IgG subclass selection distinguishRV144 and VAX003 vaccines. Sci Transl Med 2014; 6: 228–238.

32 Center RJ, Wheatley AK, Campbell SM, Gaeguta AJ, Peut V, Alcantara S et al.Induction of HIV-1 subtype B and AE-specific neutralizing antibodies in mice andmacaques with DNA prime and recombinant gp140 protein boost regimens. Vaccine2009; 27: 6605–6612.

The Supplementary Information that accompanies this paper is available on the Immunology and Cell Biology website (http://www.nature.com/icb)

ADP in HIVFA-S Batiz et al

9

Immunology and Cell Biology