Targeted lysis of HIV-infected cells by natural killer cells armed and triggered by a recombinant immunoglobulin fusion protein: implications for immunotherapy.

Abstract

Natural killer (NK) cells play an important role in both innate and adaptive antiviral immune responses. The adaptive response typically requires that virus-specific antibodies decorate infected cells which then direct NK cell lysis through a CD16 mediated process termed antibody-dependent cellular cytotoxicity (ADCC). In this report, we employ a highly polymerized chimeric IgG1/IgA immunoglobulin (Ig) fusion protein that, by virtue of its capacity to extensively crosslink CD16, activates NK cells while directing the lysis of infected target cells. We employ HIV as a model system, and demonstrate that freshly isolated NK cells preloaded with an HIV gp120-specific chimeric IgG1/IgA fusion protein efficiently lyse HIV-infected target cells at picomolar concentrations. NK cells pre-armed in this manner retain the capacity to kill targets over an extended period of time. This strategy may have application to other disease states including various viral infections and cancers.

Full text

Page 1

Targeted lysis of HIV-infected cells by natural killer cells armed and
triggered by a recombinant immunoglobulin fusion protein:
implications for immunotherapy
Neil Guptaa,1, James Arthosa,*,1, Prateeti Khazaniea, Tavis D. Steenbekea, Nina M. Censoplanoa,
Eva A. Chunga, Catherine C. Cruza, Margery A. Chaikina, Marybeth Dauchera, Shyam Kottilila,
Domenico Mavilioa, Peter Schuckb, Peter D. Sunc, Ronald L. Rabind, Sergei Radaevc,
Donald Van Ryka, Claudia Cicalaa, Anthony S. Faucia
aLaboratory of Immunoregulation, NIAID, NIH Bldg. #10 6A08, 9000 Rockville Pike, Bethesda MD 20892, USA
bProtein Biophysics Resource, DBPS, ORS, NIH, Bethesda, MD 20892, USA
cLaboratory of Immunogenetics, NIAID, NIH, Bethesda, MD 20892, USA
dDivision of Bacterial, Parasitic, and Allergenic Products, CBER, FDA, Bethesda, MD 20892, USA
Received 18 June 2004; returned to author for revision 15 September 2004; accepted 13 December 2004
Available online 13 January 2005
Abstract
Natural killer (NK) cells play an important role in both innate and adaptive antiviral immune responses. The adaptive response typically
requires that virus-specific antibodies decorate infected cells which then direct NK cell lysis through a CD16 mediated process termed
antibody-dependent cellular cytotoxicity (ADCC). In this report, we employ a highly polymerized chimeric IgG1/IgA immunoglobulin (Ig)
fusion protein that, by virtue of its capacity to extensively crosslink CD16, activates NK cells while directing the lysis of infected target cells.
We employ HIVas a model system, and demonstrate that freshly isolated NK cells preloaded with an HIV gp120-specific chimeric IgG1/IgA
fusion protein efficiently lyse HIV-infected target cells at picomolar concentrations. NK cells pre-armed in this manner retain the capacity to
kill targets over an extended period of time. This strategy may have application to other disease states including various viral infections and
cancers.
Published by Elsevier Inc.
Keywords: Natural killer cell; Antibody dependent cellular cytotoxicity; Immunotherapy; HIV; CD16; Recombinant antibody
Introduction
Natural killer cells in conjunction with IgG antibodies
kill virus-infected cells through a process termed antibody
dependent cell-mediated cytotoxicity (ADCC). Antibodies
initially decorate virus-infected cells, and subsequently bind
to, and crosslink CD16, the low affinity Fc receptor
expressed on Natural killer cells. Natural killer cells then
deliver a perforin-mediated lethal hit to the target. Because
of their extraordinary potential to kill targets, the use of
natural killer cells as immunotherapeutic agents has been
extensively investigated (Ruggeri et al., 2002). In this
regard, it is appropriate to develop additional therapeutic
strategies to fully harness the potential anti tumor/anti-viral
activity of natural killer cells (Farag et al., 2002). HIV-
infected CD4+ T-cells are subject to ADCC, although the
degree to which this process plays a role in controlling viral
replication is unclear (Riviere et al., 1989; Weinhold et al.,
1988). The viral envelope expressed on the surface of HIV-
infected cells is the principal target of this process (Tanneau
et al., 1990; Tyler et al., 1989). In this report, using HIV
0042-6822/$ - see front matter. Published by Elsevier Inc.
doi:10.1016/j.virol.2004.12.018
* Corresponding author. Fax: +1 301 442 7184.
E-mail address: jarthos@niaid.nih.gov (J. Arthos).
1Contributed equally to this study.
Virology 332 (2005) 491–497
www.elsevier.com/locate/yviro

Page 2

infected cells as a model system, we describe a novel
approach in which freshly isolated natural killer cells are
preloaded with a highly polymerized recombinant antibody
fusion protein (Arthos et al., 2002). This fusion protein is
distinct in its capacity to stimulate NK cells through CD16.
Freshly isolated cells armed and activated with this protein
exhibit the capacity to efficiently lyse HIV-infected cells in a
specific manner over an extended period of time. In
principle, this strategy may be adapted to other viral or
tumor-associated antigens.
Results
In previous studies, we have described a highly
polymerized CD4-Ig fusion protein that exhibits potent
neutralizing activity against patient derived isolates of HIV
(Arthos et al., 2002; Kwong et al., 2002). This recombinant
protein fuses the two amino terminal domains of CD4
(D1D2), the CH2CH3 domains of a human IgG1 heavy
chain, and the eighteen amino acid carboxy-terminal
tailpiece of IgA. The tailpiece of this protein drives its
polymerization into a cystine stabilized dodecamer termed
D1D2-Igatp, comprised of 12 D1D2-Ig chains that can
simultaneously bind up to twelve gp120s, the surface unit of
the HIV envelope. Because the Fc receptor recognition
epitope of antibodies resides within the CH2CH3 domains,
we considered the possibility that this highly polymerized Ig
fusion protein could hyper-crosslink CD16 and as a
consequence deliver an exceptionally strong signal to
CD16 bearing natural killer (NK) cells, thus promoting
potent ADCC. To address this possibility, we characterized
the physical and functional interaction between D1D2-Igatp
and CD16.
CD16 is the low-affinity Fc receptor, and is recognized
by antibodies only after they have aggregated on target cells
or pathogens (Radaev and Sun, 2002). We first determined
the effect of extensive polymerization of D1D2-Igatp on its
interaction with CD16 using an optical biosensor (Biacore).
Increasing concentrations of D1D2-Igatp were passed over
a sensor surface to which a recombinant soluble CD16
protein (Radaev et al., 2001) was immobilized (Fig. 1a).
Although difficult to measure precisely due to the highly
avid nature of the binding interaction, we estimate the
dissociation rate constant (kd) to be less than 10?6s?1,
resulting in an affinity (KD) of less than 10?12M. (Fig. 1a).
Under the same experimental conditions a two chain D1D2-
Ig, lacking the alpha tailpiece, failed to accumulate on the
CD16 surface (Fig. 1a insert). Only after its concentration
was increased 1000-fold were we able to estimate its affinity
for CD16 to be on the order of 3.6e?6M (data not shown),
underscoring the pronounced avidity effect associated with
D1D2-Igatp-CD16 interactions. In order to better under-
stand this apparent increase in affinity, we determined how
many CD16 molecules could bind to a single D1D2-Igatp
by performing analytical ultracentrifugation experiments
in which the sedimentation velocity of the D1D2-Igatp
was measured in the presence and absence of soluble
CD16. Since CD16 redistributes much more slowly in the
centrifugal field than does D1D2-Igatp (Fig. 1b inset),
maintaining excess CD16 in solution promotes a high
occupancy of potential ligand sites throughout the sed-
Fig. 1. Stoichiometry and kinetics of D1D2-Igatp/CD16 binding. (a) SPR sensorgram overlay of increasing concentrations of D1D2-Igatp binding to
immobilized recombinant CD16. CD16 was covalently bound to the biosensor chip through direct amine coupling. Increasing concentrations of D1D2-Igatp
were injected over a period of 2 min ((a) association phase), at which point running buffer in the absence of D1D2-Igatp was passed over the surface ((b)
dissociation phase), to allow dissociation of bound proteins. For comparison, increasing concentrations of a standard human IgG1 mAb were passed over the
CD16 surface (insert). (b) Distribution of apparent sedimentation coefficients of D1D2-Igatp in the presence (solid line) and absence (dotted line) of a molar
excess of CD16. The inset shows a superposition of concentration distributions measured for the D1D2-Igatp/CD16 mixture at time intervals of 10 min at a
rotor speed of 30,000 rpm.
N. Gupta et al. / Virology 332 (2005) 491–497
492

Page 3

imentation process. In the presence of CD16, we observed a
~3.5 S shift in the peak s values (Fig. 1b). This corresponds
to a mass increase of ~27%, indicating that a single D1D2-
Igatp binds at least ten CD16 monomers (see Materials and
methods). We conclude that the functionally irreversible
association between CD16 and D1D2-Igatp occurs as a
consequence of avidity effects associated with the simulta-
neous binding of a single D1D2-Igatp to multiple CD16
molecules.
We then employed flow-cytometry to evaluate the
capacity of D1D2-Igatp to bind to CD16 on freshly isolated
NK cells. Because NK cells bear virtually no CD4, we
employed a fluoresceine labeled anti-CD4 monoclonal
antibody (mAb) as a secondary antibody (Ab) in this assay.
We could not detect anti-CD4 binding to freshly isolated
NK cells in the absence of D1D2-Igatp (Fig. 2a). The
majority of NK cells reacted positively to anti-CD4 when
D1D2-Igatp was added. The anti-CD16 mAb 3G8 abo-
lished anti-CD4 reactivity, demonstrating that D1D2-Igatp
bound to NK cells through CD16.
Crosslinking CD16 on NK cells initiates a signal
transduction cascade that ultimately promotes perforin-
mediated ADCC (Windebank et al., 1988). Early in this
process, phospholipase C-g is cleaved into inositol triphos-
phate and diacylglycerol, followed by a rapid increase in the
concentration of cytosolic calcium. We determined whether
crosslinking of CD16 on freshly isolated NK cells by
D1D2-Igatp could induce mobilization of calcium. We
employed a flow cytometry-based assay in which NK cells
are loaded with the calcium chelating fluorophore indo-1
(Weissman et al., 1997). Less than 100 s after treatment with
D1D2-Igatp levels of cytosolic-free calcium increased
markedly in these cells (Fig. 2b panel 2). This responses
was prolonged, lasting at least 2 min. In order to de-
monstrate that this response was mediated through CD16,
we expressed and employed a variant of D1D2-Igatp
termed FD1D2-Igatp. In FD1D2-Igatp we substituted
several residues in human IgG1 that interact with CD16
with the corresponding residues encoded in human IgG2.
Since IgG2 exhibits an affinity for CD16 at least 100-fold
lower than that of IgG1, we reasoned that this molecule
would demonstrate a reduced capacity to crosslink CD16
and mobilize calcium. As predicted, FD1D2-Igatp induced
little or no calcium mobilization in NK cells (Fig. 2b panel
3). Treatment with a standard human IgG1 also failed to
mobilize calcium (data not shown). We conclude that
Fig. 2. Binding to, and signaling through CD16 on NK cells by D1D2-Igatp. (a) Flow cytometric analysis of freshly isolated NK cells incubated with FITC
anti-CD4 in the absence (panel 1) or presence (panel 2) of D1D2-Igatp. An anti CD16 Fab (3g8) was included to demonstrate specificity (panel 3). (b) Calcium
mobilization in NK cells treated with D1D2-Igatp and FD1D2-Igatp. NK cells loaded with indo-1, were stimulated with D1D2-Igatp or FD1D2-Igatp.
Fluorescence intensity was derived from the ratio of bound (390/20 nm) over free (530/20 nm) emission. Baseline fluorescence was established prior to
injection, and fluorescence was measured over time. A sham (PBS) injection was included as a control.
N. Gupta et al. / Virology 332 (2005) 491–497
493

Page 4

D1D2-Igatp crosslinks CD16 and delivers a potent stimulus
to freshly isolated NK cells.
We next determined whether NK cells preloaded with
D1D2-Igatp could mediate ADCC of HIV-infected cells by
employing a flow-cytometry based lytic cell assay that
measures the integrity of the target cell membrane as
indicated by the uptake of propidium-iodide (Derby et al.,
2001; Sheehy et al., 2001). HIV-infected CEM NKR cells
were employed as targets, and the parental CEM NKR cell
line was included as a control (Ahmad et al., 2001). Freshly
isolated NK cells were pre-incubated with D1D2-Igatp for
15 min and then rinsed to remove unbound protein. The
cells were then co-cultured with target cells for 90 min. We
observed significant lytic activity. The degree of lysis varied
from donor to donor; however, typically, close to 90% of the
target cells took up PI (Fig. 3a panel 2). Minimal
background lysis was observed in the absence of D1D2-
Igatp (Fig. 3a panel 1). Uninfected CEM NKR cells to
which D1D2-Igatp was added as well as infected CEM
NKR cells in the absence of D1D2Igatp also yielded
minimal background lysis (Fig. 3a panels 3 and 4). We
conclude that D1D2-Igatp directs the specific lysis of HIV-
infected target cells. We then titrated D1D2-Igatp and
compared its ADCC activity to that of FD1D2-Igatp.
D1D2-Igatp repeatedly displayed peak lytic activity at
concentrations below 1 nM (Fig. 3b). At higher concen-
trations, lytic activity was reduced, possibly reflecting a
decrease in the degree of CD16 crosslinking. When
Fig. 3. D1D2-Igatp mediates NK cell killing of HIV-infected target cells. (a) Flow cytometric analysis of IIIBCEMNKR cells incubated with freshly isolated
NK cells loaded with a green fluorescent dye at a 5:1 ratio in the absence (panel 1) or presence (panel 2) of 0.625 nM D1D2-Igatp. Membrane uptake was
measured by the uptake of PI. Unlabeled NK cells were gated out of each dot plot. Uninfected CEM NKR cells were similarly treated (panels 3 and 4) as
specificity controls. (b) D1D2-Igatp and FD1D2-Igatp were compared for the capacity to induce killing. Increasing concentrations of D1D2-Igatp (n) or
FD1D2-Igatp (.) were added to IIIBCEMNKR cells, and the percentage of PI positive IIIBCEMNKR cells are reported for each concentration. The percentage
of PI positive cells in the absence of D1D2-Igatp is identified by a dashed line denoted as bbackground killingQ. (c) A time course of D1D2-Igatp mediated
killing was established by pre-incubating D1D2-Igatp with NK cells for a specified time denoted on the x axis followed by incubation with IIIBCEMNKR
cells. The percentage of PI-positive cells is reported on the left y axis, and the background level of killing in the absence of D1D2-Igatp is denoted by the lower
dashed line. The percent PI-positive IIIBCEMNKR cells relative to NK cells plus D1D2-Igatp added simultaneously to IIIBCEMNKR cells without any pre-
incubation (t = 0) are reported on the right y axis. The level of killing at t = 0 is denoted by the upper dashed line. The percentage of PI positive cells in the
absence of D1D2-Igatp at each of the four time points falls at or below the dashed line denoted as bbackground killingQ. (d) A comparison of D1D2-Igatp
mediated killing relative to mAbs 2G12 and 2F5, and HIVIG. Equal mass amounts of each protein were employed, and the fold increase in killing over
background killing is reported on the y axis.
N. Gupta et al. / Virology 332 (2005) 491–497
494

Page 5

employing FD1D2-Igatp, we did not rinse NK cells after
preincubation to avoid removing any loosely bound protein.
At concentrations below 1 nM, FD1D2-Igatp exhibited little
measurable lytic activity over background (Fig. 3b). We
observed peak activity at concentrations ~100X higher than
that required for D1D2-Igatp mediated lysis.
We next determined the duration of time that NK cells
armed with D1D2-Igatp could retain the capacity to kill
targets. NK cells were pre-incubated with D1D2-Igatp for
15 min, rinsed and then placed into culture at 37 8C. At 1-h
increments, target cells were added to loaded NK cells and
incubation proceeded for an additional 90 min. NK cells
retained significant lytic activity for more than 4 h post-
treatment (Fig. 3c). Over this time period, specific lytic
activity actually increased. Consistent with this observation,
flow cytometric binding experiments revealed the stable
display of D1D2-Igatp on the surface of NK cells for up to
5 h (data not shown). After 5 h, activity decreased
substantially. We conclude that NK cells armed with
D1D2-Igatp can mediate the specific lysis of target cells
for an extended period of time.
In order to better understand the efficiency with which
D1D2-Igatp mediates ADCC, we compared it to a highly
enriched pool of human hyper immune anti-HIV Ig (HIVIG)
with potent neutralizing activity (Cummins et al., 1991). We
also evaluated mAbs 2G12 and 2F5, two of the most potent
HIV envelope-specific neutralizing human mAbs characte-
rized to date (Purtscher et al., 1996; Trkola et al., 1996). On
an equimolar basis, HIVIG, 2G12, and 2F5 mediated no
specific lytic activity at the concentration that D1D2-Igatp
generated maximum lysis (data not shown). Employing
equal mass concentrations, HIVIG exhibited less than 24%
of the activity of D1D2-Igatp, while 2G12, and 2F5
exhibited less than 22% and 8%, respectively (Fig. 3d).
Discussion
In summary, we have demonstrated that a highly
polymerized Ig fusion protein can be used to target and
induce freshly isolated NK cell to lyse HIV-infected cells
over an extended period of time. Previous efforts to employ
NK cells as immunotherapeutic agents have relied primarily
upon ex vivo activation with cytokines (Farag et al., 2002),
and these attempts have met with limited success. The
strategy that we have outlined here results in ADCC activity
that is highly specific, rapid and efficient, requiring
picomolar concentrations of the recombinant protein.
Polymerization of antibodies is known to enhance effector
functions (Smith et al., 1995). Thus, we conclude that the
marked efficiency with which D1D2-Igatp mediates ADCC
reflects the demonstrated potential of each recombinant
protein to crosslink at least 10 CD16 receptors, delivering a
strong stimulus to NK cells, as evidenced by the rapid
mobilization of calcium. Although we reversed the bnormalQ
order of antibody binding that occurs in the context of
ADCC, we did not observe non-specific killing of unin-
fected cells. Of note, a subset of naturally occurring
cytophilic antibodies, generated in response to HIV infec-
tion, has been described (Tyler et al., 1989). Similar to
D1D2-Igatp, these Abs bind first to CD16. Thus, with
respect to the order of binding, we may be recapitulating a
naturally occurring phenomenon.
The potency with which D1D2-Igatp mediates ADCC is
underscored by comparisons with neutralizing Abs. D1D2-
Igatp was significantly more effective than either of the two
envelope neutralizing mAbs tested and HIVIG. Since NK
cells exhibit reduced lytic activity in individuals chronically
infected with HIV, the stimulus provided by D1D2-Igatp
may provide a useful boost to NK cells derived from such
individuals (Brenner et al., 1989, 1993; Weinhold et al.,
1990). NK cells retained 80% of their maximal killing
activity after 4 h, after which the capacity to mediate ADCC
was lost. It remains to be determined whether providing
addition stimuli (e.g., cytokines) could extend this time
period. Finally, the strategy outlined here may have utility in
eliminating productively infected cells that persist in the
face of antiretroviral therapy. Moreover, this approach may
have broader applications against targets associated with
tumors and other infectious agents.
Materials and methods
Cells and reagents
Expression and purification of D1D2-Igatp and CD16
were described elsewhere (Arthos et al., 2002; Radaev et al.,
2001). FD1D2-Igatp was created using the ExSite PCR
based site-directed mutagenesis kit (Stratagene La Jolla,
CA). Amino acid residues 233–238 (Kabat numbering
system, www.kabatdatabase.com) in the hinge/CH2 inter-
face of human IgG1 (ELLGGP) were substituted with the
corresponding residues encoded in human IgG2 (PV-AGP).
FD1D2-Igatp was expressed in CHO cells and grown in a
hollow-fiber bioreactor (Fibercell Systems, Frederick Mary-
land). CEM NKR and IIIB CEM NKR (Tyler et al., 1990)
were provided by Dr. A. Pinter, Public Health Research
Institute, New York). NK cells were isolated and purified
from leukopheresis products using a negative selection
antibody-magnetic bead cocktail (StemCell Technologies
Vancouver Canada). Cells were employed on the day of
isolation and maintained in RPMI plus 10% fetal bovine
serum (FBS) without the addition of cytokines.
Analytical ultracentrifugation
Sedimentation velocity experiments were conducted in a
Beckman Optima XL-I/A (Beckman Coulter, Fullerton,
CA). 300 Al of a mixture of D1D2-Igatp at 0.2 mg/ml and
CD16 at 0.085 mg/ml (a ~15-fold molar excess) dissolved
in PBS were sedimented at a rotor speed of 30,000 rpm at
N. Gupta et al. / Virology 332 (2005) 491–497
495

Page 6

20 8C. The evolution of the concentration distributions
were measured with Rayleigh interference optics. In
parallel control experiments, the proteins were also studied
separately. The sedimentation coefficient distribution of
D1D2-Igatp and its complexes were calculated with the
software SEDFIT, www.analyticalultracentrifugation.com,
using the model ls-g*(s) for non-diffusing species (Schuck
and Rossmanith, 2000), superimposed by a single-compo-
nent Lamm equation solution (Schuck, 1998), describing
the slower redistribution of excess CD16. We employed
the scale relationship s–M2/3to estimate that the increase
of mass causing the observed increase in sedimentation
rate (Lebowitz et al., 2002).
Optical biosensor analysis
All binding assays were performed using a BIA3000
optical biosensor (Biacore, Inc., Uppsala, Sweden). CD16
was immobilized onto the surface of a CM5 sensor chip
using a standard amine coupling procedure described by
Biacore, Inc. Briefly, the carboxyl groups on the sensor
surface were activated by injecting 35 Al of 0.2 M N-ethyl-
N’-(3-diethylamino-propyl) carbodiimide (EDC), 0.05 M
N-hydroxy-succinimide (NHS) at a flow rate of 5 Al/min.
The ligand, suspended in 10 mM acetate buffer, pH 4.0–
5.5 (depending on the ligand used) was passed over the
activated surface until the desired surface density was
reached. Unreacted carboxyl groups were capped by
injecting 35 Al of 1 M ethanolamine (pH 8.0). Ovalbumin
(Ova) was immobilized on the surface of one flow cell as a
reference surface to control for non-specific binding of
analyte. The running buffer used was 10 mM HEPES, pH
7.4, 150 mM NaCl, 3 mM EDTA, 0.01% surfactant P-20,
0.5% soluble carboxymethyl dextran (Fluka BioChemika,
Inc., Buchs, Switzerland). All binding experiments were
performed in duplicate and at 25 8C.
Flow cytometric binding assay
1 ? 105freshly isolated NK cells were incubated with
10 Ag of D1D2-Igatp in 200 Al phosphate buffered saline
(PBS) plus 2% FBS and 0.2% sodium azide at 4 8C for
30 min. Cells were rinsed three times and then stained
with Leu 3A, a FITC-conjugated anti CD4 mAb (BD
Pharmingen, San Diego, CA) under the same conditions.
After rinsing, cells were analyzed on a FACSCAN flow
cytometer (Becton Dickinson Immunocytometry Systems,
San Jose, CA).
Calcium mobilization
Calcium mobilization assays were carried out as
previously described (Weissman et al., 1997). Briefly,
freshly isolated NK cells were resuspended in Hanks
balanced salt solution (HBSS) with calcium and magne-
sium with 10 mM HEPES. The fluorescent probe indo-1/
acetoxymethylester (final concentration, 10 mM) (Mole-
cular Probes, Eugene, OR) and pleuronic acid (final
concentration, 300 mg/ml) (Molecular Probes, Eugene,
OR) were added, and the cell suspension was incubated at
30 8C for 45 min with occasional gentle mixing. Cells
were washed twice in HBSS/2% FBS. Aliquots of 106
cells in 1 ml were analyzed on a LSR flow cytometer
(Becton Dickinson Immunocytometry Systems, San Jose,
CA) at 25 8C. Indo-1 fluorescence was measured at
wavelengths of 390/20 nm (bound) and 530/20 nm (free).
Fluorescence data were collected for 30 s, and then cells
were injected with either a buffer sham or a recombinant
protein. Data analysis was carried out using FLOWJO
software (Treestar, Stanford, CA).
ADCC
CEM NKR or IIIB CEM NKR target cells were loaded
with PKH 67 (Sigma, Milwaukee, WI) for 5 min at 25 8C
followed by the addition of 2 ml FBS for 1 min at 25 8C,
after which cells were washed extensively in RPMI/10%
FBS and resuspended at a final concentration of 0.5 ? 106
cells/ml. Where specified, freshly isolated NK cells were
preloaded with D1D2-Igatp for 15 min at 25 8C, and then
rinsed to remove unbound protein. When other mAbs and
recombinant proteins were employed, the rinse step was
eliminated. 2.5 ? 104target cells were combined with NK
cells at a ratio of 1:5 in a final volume of 200 Al and
incubated for 90 min at 37 8C. In the final 15 min of this
incubation, propidium-iodide (50 Ag/ml) was added to the
culture. Cells were then analyzed on a FACSCAN flow
cytometer (Becton Dickinson Immunocytometry Systems,
San Jose, CA). 5 ? 104target cells from the live cell gate,
determined by light scatter, were acquired for subsequent
analysis.
Acknowledgments
We wish to thank Dr. Abraham Pinter for providing
CECM NKR and IIIBCEM NKR cell lines. Neil Gupta was
supported through the Howard Hughes NIH Research
Scholars Program.
References
Ahmad, R., Sindhu, S.T., Toma, E., Morisset, R., Vincelette, J., Menezes, J.,
Ahmad, A., 2001. Evidence for a correlation between antibody-depen-
dent cellular cytotoxicity-mediating anti-HIV-1 antibodies and prognos-
tic predictors of HIV infection. J. Clin. Immunol. 21 (3), 227–233.
Arthos, J., Cicala, C., Steenbeke, T.D., Chun, T.W., Dela Cruz, C.,
Hanback, D.B., Khazanie, P., Nam, D., Schuck, P., Selig, S.M., Van
Ryk, D., Chaikin, M.A., Fauci, A.S., 2002. Biochemical and
biological characterization of a dodecameric CD4-Ig fusion protein:
implications for therapeutic and vaccine strategies. J. Biol. Chem. 277
(13), 11456–11464.
Brenner, B.G., Dascal, A., Margolese, R.G., Wainberg, M.A., 1989. Natural
N. Gupta et al. / Virology 332 (2005) 491–497
496

Page 7

killer cell function in patients with acquired immunodeficiency
syndrome and related diseases. J. Leukocyte Biol. 46 (1), 75–83.
Brenner, B.G., Gryllis, C., Gornitsky, M., Wainberg, M.A., 1993. Changes
in natural immunity during the course of HIV-1 infection. Clin. Exp.
Immunol. 93 (2), 142–148.
Cummins, L.M., Weinhold, K.J., Matthews, T.J., Langlois, A.J., Perno,
C.F., Condie, R.M., Allain, J.P., 1991. Preparation and characterization
of an intravenous solution of IgG from human immunodeficiency virus-
seropositive donors. Blood 77 (5), 1111–1117.
Derby, E., Reddy, V., Kopp, W., Nelson, E., Baseler, M., Sayers, T.,
Malyguine, A., 2001. Three-color flow cytometric assay for the study of
the mechanisms of cell-mediated cytotoxicity. Immunol. Lett. 78 (1),
35–39.
Farag, S.S., Fehniger, T.A., Ruggeri, L., Velardi, A., Caligiuri, M.A., 2002.
Natural killer cell receptors: new biology and insights into the graft-
versus-leukemia effect. Blood 100 (6), 1935–1947.
Kwong, P.D., Doyle, M.L., Casper, D.J., Cicala, C., Leavitt, S.A., Majeed,
S., Steenbeke, T.D., Venturi, M., Chaiken, I., Fung, M., Katinger, H.,
Parren, P.W., Robinson, J., Van Ryk, D., Wang, L., Burton, D.R.,
Freire, E., Wyatt, R., Sodroski, J., Hendrickson, W.A., Arthos, J.,
2002. HIV-1 evades antibody-mediated neutralization through con-
formational masking of receptor-binding sites. Nature 420 (6916),
678–682.
Lebowitz, J., Lewis, M.S., Schuck, P., 2002. Modern analytical ultra-
centrifugation in protein science: a tutorial review. Protein Sci. 11 (9),
2067–2079.
Purtscher, M., Trkola, A., Grassauer, A., Schulz, P.M., Klima, A., Dopper,
S., Gruber, G., Buchacher, A., Muster, T., Katinger, H., 1996. Restricted
antigenic variability of the epitope recognized by the neutralizing gp41
antibody 2F5. Aids 10 (6), 587–593.
Radaev, S., Sun, P., 2002. Recognition of immunoglobulins by Fc gamma
receptors. Mol. Immunol. 38 (14), 1073–1083.
Radaev, S., Motyka, S., Fridman, W.H., Sautes-Fridman, C., Sun, P.D.,
2001. The structure of a human type III Fc gamma receptor in complex
with Fc. J. Biol. Chem. 276 (19), 16469–16477.
Riviere, Y., Tanneau-Salvadori, F., Regnault, A., Lopez, O., Sansonetti, P.,
Guy, B., Kieny, M.P., Fournel, J.J., Montagnier, L., 1989. Human
immunodeficiency virus-specific cytotoxic responses of seropositive
individuals: distinct types of effector cells mediate killing of targets
expressing gag and env proteins. J. Virol. 63 (5), 2270–2277.
Ruggeri, L., Capanni, M., Urbani, E., Perruccio, K., Shlomchik, W.D.,
Tosti, A., Posati, S., Rogaia, D., Frassoni, F., Aversa, F., Martelli,
M.F., Velardi, A., 2002. Effectiveness of donor natural killer cell
alloreactivity in mismatched hematopoietic transplants. Science 295
(5562), 2097–2100.
Schuck, P., 1998. Sedimentation analysis of noninteracting and self-
associating solutes using numerical solutions to the Lamm equation.
Biophys. J. 75 (3), 1503–1512.
Schuck, P., Rossmanith, P., 2000. Determination of the sedimentation
coefficient distribution by least-squares boundary modeling. Biopol-
ymers 54 (5), 328–341.
Sheehy, M.E., McDermott, A.B., Furlan, S.N., Klenerman, P., Nixon, D.F.,
2001. A novel technique for the fluorometric assessment of T lympho-
cyte antigen specific lysis. J. Immunol. Methods 249 (1–2), 99–110.
Smith, R.I., Coloma, M.J., Morrison, S.L., 1995. Addition of a mu-tailpiece
to IgG results in polymeric antibodies with enhanced effector functions
including complement-mediated cytolysis by IgG4. J. Immunol. 154
(5), 2226–2236.
Tanneau, F., McChesney, M., Lopez, O., Sansonetti, P., Montagnier, L.,
Riviere,Y.,1990.Primarycytotoxicityagainsttheenvelopeglycoprotein
of human immunodeficiency virus-1: evidence for antibody-dependent
cellular cytotoxicity in vivo. J. Infect. Dis. 162 (4), 837–843.
Trkola, A., Purtscher, M., Muster, T., Ballaun, C., Buchacher, A., Sullivan,
N., Srinivasan, K., Sodroski, J., Moore, J.P., Katinger, H., 1996. Human
monoclonal antibody 2G12 defines a distinctive neutralization epitope
on the gp120 glycoprotein of human immunodeficiency virus type 1.
J. Virol. 70 (2), 1100–1108.
Tyler, D.S., Nastala, C.L., Stanley, S.D., Matthews, T.J., Lyerly, H.K.,
Bolognesi, D.P., Weinhold, K.J., 1989. GP120 specific cellular
cytotoxicity in HIV-1 seropositive individuals. Evidence for circulating
CD16+ effector cells armed in vivo with cytophilic antibody.
J. Immunol. 142 (4), 1177–1182.
Tyler, D.S., Stanley, S.D., Zolla-Pazner, S., Gorny, M.K., Shadduck, P.P.,
Langlois, A.J., Matthews, T.J., Bolognesi, D.P., Palker, T.J., Weinhold,
K.J., 1990. Identification of sites within gp41 that serve as targets for
antibody-dependent cellular cytotoxicity by using human monoclonal
antibodies. J. Immunol. 145 (10), 3276–3282.
Weinhold, K.J., Lyerly, H.K., Matthews, T.J., Tyler, D.S., Ahearne, P.M.,
Stine, K.C., Langlois, A.J., Durack, D.T., Bolognesi, D.P., 1988.
Cellular anti-GP120 cytolytic reactivities in HIV-1 seropositive
individuals. Lancet 1 (8591), 902–905.
Weinhold, K.J., Tyler, D.S., Lyerly, H.K., 1990. Measurement of direct and
indirect forms of anti-HIV-1 ADCC: implications for other retroviral
disease. Dev. Biol. Stand. 72, 343–348.
Weissman, D., Rabin, R.L., Arthos, J., Rubbert, A., Dybul, M., Swofford,
R., Venkatesan, S., Farber, J.M., Fauci, A.S., 1997. Macrophage-tropic
HIV and SIV envelope proteins induce a signal through the CCR5
chemokine receptor. Nature 389 (6654), 981–985.
Windebank, K.P., Abraham, R.T., Powis, G., Olsen, R.A., Barna, T.J.,
Leibson, P.J., 1988. Signal transduction during human natural killer cell
activation: inositol phosphate generation and regulation by cyclic AMP.
J. Immunol. 141 (11), 3951–3957.
N. Gupta et al. / Virology 332 (2005) 491–497
497