Evaluation of antigen-specific responses using

in vitro enriched T cells

N. Jonesa, D. Agrawalb, M. Elrefaeia, A. Hansona,V. Novitskyc, J.T. Wongd, Huyen Caoa,*

aCalifornia Department of Health Services, 850 Marina Bay Parkway, VRDL Richmond, CA 94804, USAbPartners AIDS Research Center, Charlestown, MA, USA

cHarvard School of Public Health, Boston, MA, USAdDepartment of Medicine, Massachusetts General Hospital, Boston, MA, USA

Received 25 July 2002; received in revised form 10 October 2002; accepted 4 November 2002

Abstract

Antigen-specific lymphocytes are important in the immune response to viral infection. Peripheral blood mononuclear cells

(PBMC) are traditionally used as a source of effector cells in most immunological studies. We described here the use of the

bispecific monoclonal antibodies (BSMAB) anti CD3:CD8 (CD3,8) and anti CD3:CD4 (CD3,4B) to expand and selectively

enrich CD4+ and CD8+ T cell populations, respectively. The expanded cells demonstrated >90% CD3+CD4+ or CD3 +CD8+

by 14 days. We measured HIV- and CMV-specific responses of these subset-enriched T cell and found that sensitivity and

specificity is similar or higher when compared to PBMC in various cellular immunology assays (CMI). Vbeta analysis of

BSMAB-enriched cells demonstrated comparable repertoire to the parent PBMC. Although both CD45RAhi and CD45ROhi cell

populations were expanded with the BSMAB, selective subset depletion demonstrated that the antigen-specific T cell responses

were restricted to the initial CD45ROhi memory effector subgroup. In conclusion, BSMAB in vitro enrichment of T cells allows

significant expansion of the cell population without loss of specificity. This technique of cell expansion permits studies of T cell

subset function in situations where the initial cell source is scarce, and presents an alternative for viable and functional T cells in

immunological assays.

D 2002 Elsevier Science B.V. All rights reserved.

Keywords: T cell; CMI; Bispecific monoclonal antibody

1. Introduction

T cells play an important role against viral

infection. The recent development of highly sensi-

tive assays to measure T cell responses such as the

ELISpot (Larsson et al., 1999; Altfeld et al., 2000)

and intracellular cytokine flow cytometry (Appay et

al., 2000; Goulder et al., 2000) allows rapid quan-

titation of specific T cell responses directly from

PBMC. However, the detailed examination of virus-

specific T cell response, including epitope mapping

and HLA restriction, still relies heavily on a large

viable source of functional T cells. Cellular immu-

nology assays (CMI) have traditionally utilized

0022-1759/02/$ - see front matter D 2002 Elsevier Science B.V. All rights reserved.

PII: S0022 -1759 (02 )00510 -0

* Corresponding author.

E-mail address: Hcao@dhs.ca.gov (H. Cao).

www.elsevier.com/locate/jim

Journal of Immunological Methods 274 (2003) 139–147

peripheral blood mononuclear cells (PBMC) in

evaluating antigen-specific responses and the quan-

tity and quality of these cells are often limiting.

Furthermore, cryopreservation of PBMC invariably

leads to some degree of decreased cell viability with

the best recovery < 100% (Weingberg et al., 2000),

excluding other factors such as storage and shipping

conditions, and often limits the scope of immuno-

logical studies.

Bispecific monoclonal antibodies (BSMAB) have

two different antigen binding sites, but share a

common Fc region. The CD3,4B BSMAB has both

an anti-CD3 and an anti-CD4 Ag recognition sites,

whereas the CD3,8 BSMAB has both an anti-CD3

and an anti-CD8 Ag recognition sites. When added

to PBMC in the presence of IL2, CD3,4B BSMAB

leads to selective elimination of CD4+ cells, pri-

marily by bridging CD4 molecules on target cells

to the CD3 molecular complexes of CD8+ cells.

Concurrently, CD3,4B activates residual CD3+ (pre-

dominantly CD8+) T cells and results in their rapid

expansion. Similarly, CD3,8 leads to selective cy-

tolysis of CD8+ T cells, and results in rapid pro-

liferation of the residual CD3 + CD4+ T cells

(Wong and Colvin, 1987; Wilson et al., 1995). In

this study, we evaluated cellular immune responses

of BSMAB-expanded T cell population in standard

immunological assays. Our data support the

hypothesis that selective expansion of functional

T cells by BSMAB maintains the antigen-specific

responses and provides a reliable source of viable

cells for in depth analysis of virus-specific T cell

responses.

2. Materials and methods

2.1. Bispecific monoclonal antibody

BSMA CD3,4B and CD3,8 recognize CD3 and

CD4 or CD3 and CD8 antigens, respectively. The

BSMAB are produced by hybrid-hybridomas. The

production of the hybrid-hybridomas and the

BSMAB have been previously described (Wong

and Colvin, 1987; Wong et al., 1989). In brief, the

CD3,8 hybrid-hybridoma was produced by the

fusion of the 12F6 (anti-CD3) and OKT8 (anti-

CD8) hybridomas. It produces the CD3,8 BSMAB

along with a mixture of the parental 12F6, OKT8,

and other hybrid mAb. CD3,8 BSMAB has one CD3

Ag recognition site and one CD8 Ag recognition site

that share a common Fc. The CD3,4B hybrid-

hybridoma was similarly produced by the fusion of

12F6 and Jet4B (anti-CD4) hybridomas. This hybrid-

hybridoma produces the CD3,4B BSMAB along

with a mixture of the parental 12F6 and Jet4B and

other recombinants. The CD3,8 BSMAB and

CD3,4B BSMAB were purified by preparative iso-

electric focusing as previously described. Fractions

that were pure as judged by analytical IEF were

pooled, and their binding properties confirmed by

the staining patterns on peripheral blood lympho-

cytes.

2.2. Cell lines and culture conditions

PBMC from HIV seropositive volunteer donors

from Botswana (SP1) (Novitsky et al., 2001) and

Uganda (SP2, SP3) (Cao et al., 2000) with known

Gag- and Nef-specific responses as well as from CMV

positive healthy HIV seronegative individual (SN1)

were evaluated. Cryopreserved cells were resus-

pended in RPMI 1640 medium containing 10% FCS

at a concentration of 1–2 million per ml with 5 Ag/ml

of CD3,4B or CD3,8 BSMAB. Next day, and sub-

sequently twice a week, the cells were fed with RPMI

1640 media containing 10% FCS and IL-2. The cells

were kept in culture for 10–14 days before being

tested for activity. Cells were sensitized with CMV

lysate (Advanced Biotechnologies, Columbia, MD)

and HIV synthetic peptides (NIH AIDS Research and

Reference Reagent Program, Bethesda, MD) at 1 and

2 Ag/ml, respectively.

2.3. Flow cytometry

2.3.1. Quantifying Vbeta repertoire

The IOTest Beta Mark (Immunotech, Marseille,

France) kit was used for quantitative analysis of the

TCR Vbeta repertoire of T lymphocytes by flow

cytometric analysis. The kit is composed of eight

vials containing mixtures of conjugated TCR Vbeta

antibodies corresponding to 24 different specificities

(about 70% coverage of normal human TCR Vbeta

repertoire). PBMCs or expanded CD8+ cells were

used in the analysis. Tubes were labeled A through

N. Jones et al. / Journal of Immunological Methods 274 (2003) 139–147140

H with extra tubes used for compensation and as

controls. The cells were incubated at 4 jC for 30

min. They were then washed twice and fixed by

adding 100 Al of 1% paraformaldehyde. Analysis

was done on Becton Dickinson FACScalibur flow

cytometer using the Cell Quest software.

2.3.2. T cell subset study

Depletion of CD45RA+ or CD45RO+ cells was

performed according to manufacturer’s procedures

(Stem Cell Technologies, Canada). Briefly, freshly

isolated PBMC were washed twice with PBS plus

5% FCS (wash buffer). PBMC were resuspended

at 2–5� 107 cells/ml in wash buffer. For depletion

of CD45RA+ cells, 1 ml aliquot of the PBMC

was stained with anti-human CD45A tetramer (3

Ag/ml) for 15 min at room temperature (RT). For

depletion of CD45RO+ cells, 1 ml aliquot of the

PBMC was first stained with biotinylated anti-

human CD45RO (clone UCHL1) for 15 min at

RT. Cells were washed once with wash buffer then

stained with anti-biotin tetrameric antibody (100 Al/

1 ml of cells) for 15 min at RT. Cells were washed

once and resuspended in 1 ml of wash buffer.

Magnetic colloid (60 Al/ml of cells) was added to

stained PBMC from both groups and incubated for

15 min at RT. Depletion was accomplished by bind-

ing of the labeled cells to a 3Wmagnetic column

(Stem Cell Technologies). The percentage of

CD45RA+ or CD45RO+ cells in the depleted ali-

quots was reduced to V 3% as determined by flow

cytometry.

2.3.3. ELISpot

ELISpot assay was performed on cryopreserved

PBMCs and BSMAB-enriched cells generated from

PBMC. Ninety-six-well nitrocellulose plates (Milli-

titer, Millipore, Bedford, MA) were pre-coated over-

night at 4 jC with 0.5 Ag/ml of monoclonal

antibody 1-D1K (Mabtech, Stockholm, Sweden).

The plates were then washed 6X with phosphate-

buffered saline (PBS), and PBMCs were added at

100,000 cells/well in duplicate wells. The CD8 cell

lines were washed twice in RPMI 1640 media and

Fig. 1. (A) PBMC and CD8-enriched cells from an HIV seropositive volunteer was tested for HIV-specific response using overlapping Nef

synthetic pooled peptides (20 amino acids) in the ELISpot assay. (B) Cells from the same individual was tested for activity against CMV

antigens. (C) PBMC and CD8-enriched cells from an HIV seropositive Ugandan volunteer tested against pools of synthetic peptides. All assays

were done with duplicate wells of 100 K/well. Results are reported as spot forming cells (SFC) per million.

N. Jones et al. / Journal of Immunological Methods 274 (2003) 139–147 141

added at 50,000 cells/well. The plates were incu-

bated overnight at 37 jC in 5% CO2, and biotiny-

lated monoclonal antibody anti-IFN-Mab (Mabtech)

was then added at 0.5 Ag/ml for 100 min followed

by streptavidin-ALP (Mabtech) for 1 h at room

temperature. The plates were washed three times

with PBS and 5-bromo-4-chloro-3-indolyl phos-

phate, and nitro blue (Sigma) was added to develop

the reaction. Tap water was added to stop the

reaction after 15 min. Individual cytokine-producing

cells were detected as dark spots, and subsequently

counted on the ELISpot reader (CTL, Cleveland,

OH). CTL frequency was calculated from the num-

ber of spots (SFC, spot forming cells) subtracted

from the control wells and averaged from the

replicate wells. The final frequency of specific T

cell response was reported as the average frequency

per 106 cells (SFC/million). Responses were consid-

ered positive if the SFCs were at least twice that of

the control and >10 spots per well. Background SFC

was on average less than 10/well.

2.3.4. Chromium release assay

Autologous BLCL infected with recombinant vac-

cinia virus and subsequently inactivated by UV/Psor-

alen were co-cultured with CD8+ enriched cells in the

presence of recombinant IL-2 and tested for activity

within 14 days (Lubaki et al., 1994; Cao et al., 1997).

Target cells were autologous B-LCL infected with

recombinant vaccina virus expressing Gag antigen

and subsequently labeled with radioactive chromium

(51Cr). Effector/target cell ratio (E:T) was 20:1 in a

final volume of 200 Al with all assays performed in

duplicate wells. Supernatant fluid was harvested after

4 h and the percent specific lysis was determined

from the formula: 100� [(experimental release�spontaneous release)/(maximum release�spontane-

ous release)]. HIV-specific CTL activity was defined

Fig. 2. BSMAB enriched CD8+ cells are functional. CD8 cells from a HIV seropositive Ugandan volunteer (SP3) were stimulated with

inactivated autologous BLCL expressing Gag and irradiated feeders and tested for activity after 14 days. (A) Cytolytic activity against Gag was

demonstrated in the standard chromium release assay. (B) ELISpot assay using CD8-enriched cells from the same individual against overlapping

synthetic Gag peptides demonstrated Gag-specific activity against peptide 22. (C) HLA restriction using partially matched BLCL pulsed with

peptide 22 and CD8-expanded T cells demonstrated that the epitope is HLA B14-restricted.

N. Jones et al. / Journal of Immunological Methods 274 (2003) 139–147142

as 10% above background/control. Spontaneous

release was < 30% of maximal release for all as-

says.

3. Results

3.1. BSMAB-enriched CD8+ cells maintain specific-

ity and functionality

CD8-enriched T cells and PBMCs of HIV sero-

positive individuals (SP1, SP2) were tested for HIV-

specific activity. Expanded CD8+ T cells demon-

strated similar or increased sensitivity and specificity

to HIV antigens when compared to PBMC in the

ELISpot assay (Figs. 1 and 2). In addition, function-

ality as measured by cytolytic activity in the stand-

ard chromium release assay was demonstrated from

CD8-enriched cells from an HIV seropositive indi-

vidual (SP3). CD8+ T cells were stimulated with

autologous BLCL infected with recombinant vacci-

nia viruses expressing HIV antigens that were inac-

tivated with UV/Psoralen as previously described.

Cytolytic activity to Gag was demonstrated for and

corresponded to Gag activity detected by the ELI-

Fig. 3. PBMC from a healthy volunteer with known CD4-mediated

CMV responses as demonstrated by CD4 depletion. PBMC was

depleted for CD45RO and CD45RA and subsequently BSMAB

enriched for CD4+Tcells. Activity against CMV lysate was tested in

the ELISpot assay. CMV-specific response in CD4-enriched cells was

detected in mock depletion and CD45RA-depleted cells but

abrogated with CD45RO depletion. CMV-specific response is

mediated by CD4 +CD45RO T cells. CD4-enriched cells that were

expanded from undepleted, CD45RA or CD45RO depleted PBMC

were tested for CMV-specific activity in the ELISpot assay. Depletion

of CD45RO prior to BSMAB CD4 enrichment resulted in loss of

antigen specific activity, indicating that specific memory phenotype T

cells are expanded. Specific spot forming cells (SFC) were subtracted

over background. Background SFC/million were < 50.

Fig. 4. Quantitative analysis of Vbeta repertoire of PBMC and CD8 enriched T cells were compared using the IOTest Beta Mark kit. No marked

shift in Vbeta expansion was seen in CD8-enriched cells compared to the Vbeta regions from PBMC of a seronegative volunteer.

N. Jones et al. / Journal of Immunological Methods 274 (2003) 139–147 143

Fig. 5. CD3,8 selective expansion of PBMC results in CD45ROhi phenotype. (A) CD45RA and RO phenotype in PBMC and CD45RA depleted

PBMC. (B) Following addition of BSMAB, the CD4 enriched cells acquired a CD45ROhi phenotype. (C) Depletion of CD45RA followed by

BSMAB expansion demonstrated that the initial CD45ROhi phenotype is maintained with concomitant CCR7-phenotype.

N. Jones et al. / Journal of Immunological Methods 274 (2003) 139–147144

SPot assay (Fig. 2). The increased yield in expanded

cells allowed for HLA restriction and finer epitope

mapping (Fig. 2).

Similarly, CD4-mediated responses were evaluated

using BSMAB CD3,8 stimulation. CD4-enriched

cells from a CMV-positive healthy volunteer (SN1)

were tested for activity in the ELISpot assay (Fig. 3).

Abrogation of the response after CD4 depletion dem-

onstrated that CMV-specific response is CD4-medi-

ated.

3.2. BSMAB expansion of T cells is polyclonal

Virus-specific activity of CD8-enriched cells was

tested for activity against HIVand CMV in an individ-

ual with known dual responses. Both HIV- and CMV-

specific responses are maintained in the same individ-

ual suggesting that the BSMAB expansion of T cells is

polyclonal (Fig. 1). We also examined the Vbeta

repertoire of PBMC and CD8-expanded cells from

seronegative volunteer SN1. We observed no major

shift in the T cell receptor (TCR) population of the

CD8-enriched T cell and PBMC of seronegative vol-

unteer SN1 (Fig. 4) with similar data obtained from

HIV seropositive individuals (data not shown). To

evaluate whether specific T cell populations were

expanded, we performed CD45RA and CD45RO

depletion of PBMC before BSMAB expansion (Fig.

5).Although the expandedTcells acquired aCD45ROhi

phenotype, only the CD45RA-depleted population

Fig. 6. BSMAB 3,8 addition to PBMC leads to selective enrichment of >90% CD4+ at 14 days by flow cytometry with plots gated on

lymphocyte population. (A) CMV specific response as measured by ELISpot over same period of expansion. Media alone was used as negative

control with responses reported as spot forming cells (SFC)/million cells. CD4+ T cells selected by magnetic bead depletion and expanded with

anti-CD3 was tested for CMV-specific activity in parallel to BSMAB-enriched cells (B).

N. Jones et al. / Journal of Immunological Methods 274 (2003) 139–147 145

maintained CMV-specificity (Fig. 3), indicating that

antigen-specific T cells are derived from the initial

memory population phenotype of CD45ROhi.

3.3. Kinetics and phenotype bispecific monoclonal

antibody expansion

Addition of BSMSB typically leads to selective

expansion of T cells to >90% after 14 days (range 75–

97%, data not shown). We also evaluated the magnetic

bead depletion method (which required>2M PBMC)

followed by anti CD3 expansion and did not observe

similar expansion kinetic (range 40–65%, data not

shown) with nonspecific activity more significant

(Fig. 6). High level of nonspecific responses is seen

with BSMAB prior to 2 weeks and activity typically

wanes after 4 weeks without restimulation. Pheno-

typic purity of expanded cell population decreases

after 21 days, without restimulation, with concurrent

loss of specific responses (Fig. 6). We found that the

time window for testing BSMAB-enriched T cells is

optimal between 14 and 21 days.

4. Discussion

Detailed analysis of T cell responses is critical in

the understanding of the immunopathogenesis of viral

and other intracellular infections. Recent development

of assays such as the ELISpot and intracellular cyto-

kine flow cytometry, which detect the secretion of

cytokines following antigenic stimulation, has al-

lowed greater sensitivity in the evaluation of anti-

gen-specific T cell response. However, the need for a

viable and functional source of T cells continues to

limit detailed analysis of cellular immune responses.

The addition of bispecific monoclonal antibody to

PBMC results in significant in vitro enrichment and

expansion of the CD8+ and CD4+ T cells. The unique

feature of the BSMAB is that it has only one CD3 Ag

recognition site and one CD4 Ag recognition site that

share a common Fc. Alternative methods that cross-

link anti-CD3 and anti-CD4 or CD8 with anti-CD4

would produce antibody aggregates with two or more

anti-CD3 and anti-CD4 or CD8 binding sites that may

lead to redirected cytolysis of multiple subsets, instead

of predominantly the subset that is targeted by the

specific BSMAB (Wong et al., 1990). BSMAB expan-

sion also offers several major advantages over current

expansion methods such as selective depletion fol-

lowed by anti-CD3 expansion. The technique is

relatively simple (and hence the lower contamination

probability) and does not require extensive purifica-

tion steps with significantly higher yield of cell

subsets. In addition, the method allows investigators

to optimize limited sources of viable PBMC.

We evaluated the functionality of CD4- and CD8-

enriched T cell populations to CMV and HIV in

various immunological assays. We demonstrated that

the sensitivity and specificity of BSMAB expanded T

cells are similar when compared to PBMC using the

standard chromium release assay or the ELISpot and

intracellular cytokine flow cytometry assays. The

polyclonal expansion of T cells, without loss of

specific responses, offers an important tool in the

evaluation of cellular immune response in generating

a source of functional T cells. The expansion of

antigen-specific T cells permits fine mapping of a

response to a protein antigen as well as HLA restric-

tion of a specific response, which typically requires

large quantities of viable PBMC. This technique also

offers the added advantage of known phenotype of

antigen-specific responses (CD8- or CD4-mediated).

The requirement of IL-2 has both advantages and

limitations. It permits the expansion of purified T cell

subsets in the presence of the BSMAB, and enhances

the overall immune reactivity. We noted evidence for

early nonspecific activity that subsided and permitted

assessment of antigen-specific response beyond the

first 10 days of expansion. In this particular study, the

PBMC were selectively stimulated with the CD3,4 or

CD3,8 BSMAB only once and the cell population

gradually loses the immunoactivity after prolonged in

vitro culture condition. BSMAB-expanded T cell

populations were viable for extensive period of time

in artificial capillary cartridges or with restimulation

in previous studies (Liu and Wong, 1999). Without

further assay by subcloning or limiting dilution, the

bulk polyclonal expansion of T cell population pro-

vides qualitative rather than quantitative detection and

analysis of specific responses. Although we found that

both breadth and specificity of responses of these T

cells are comparable to those of PBMC, it is possible

that marginal low frequency responses may not be

favored during the expansion of T cell populations.

We have previously used this technique to assess

N. Jones et al. / Journal of Immunological Methods 274 (2003) 139–147146

extensive T cell subset function from the limited

source of cryopreserved PBMC in AIDS vaccine trial

(Cao and Al, in press).

In conclusion, we found that bispecific monoclonal

antibody selective enrichment of T cells represents a

simple method of obtaining a large number of T cells

that remain functional after a short period of TCR

stimulation. We demonstrated that this approach has

useful potential and application in the studies of

antigen-specific T cell immunology.

References

Altfeld, M., Trocha, A., Elridge, R., Rosengerg, E.S., Phillips, M.,

Addo, M., Sekaly, R., Kalams, S.A., Burchett, S., McIntosh, K.,

Walker, B.D., Goulder, P.J., 2000. Identification of dominant

optimal HLA B-60 and HLA-B61-restricted cytotoxic T lym-

phocyte epitopes: rapid characterization of CTL responses by

enzyme-linked immunospot assay. J. Virol. 74, 8541–8549.

Appay, V., Nixon, D.F., Donahoe, S.M., Gillespie, M., Dong, T.,

King, A., Ogg, G.S., Speigel, H.M., Conlon, C., Spina, C.A.,

Havlir, D.V., Richmand, D.D., Waters, A., Easterbrook, P.,

McMichael, A.J., Rowland-Jones, S., 2000. HIV-specific

CD8+ T cells produce antiviral cytokines but are impaired in

cytolytic function. J. Exp. Med. 192, 63–75.

Cao, H., Al, E., in press. Immunogenicity to a canarypox HIV

vaccine in seronegative Ugandan volunteers: result of the HIV-

NET 007 vaccine study. J. Infec. Dis.

Cao, H., Kanki, P., Sankale, J.-L., Dieng-Sarr, A., Mazzara, G.P.,

Kalams, S.A., Korber, B., MBoup, S., Walker, B.D., 1997. Cy-

totoxic T-lymphocyte cross-reactivity among different human

immunodeficiency virus type 1 clades: implication for vaccine

development. J. Virol. 71, 8615–8623.

Cao, H., Mani, I., Vincent, R., Mugerwa, R., Mugyenyi, P., Kanki,

P., Ellner, J., Walker, B.D., 2000. Cellular immunity to HIV-1

Clades: relevance to HIV-1 vaccine trials in Uganda. J. Infect.

Dis. 182, 1350–1356.

Goulder, P.J., Tang, Y., Brander, C., Betts, M.R., Altfeld, M., An-

namalai, K., Trocha, A., He, S., Rosengerg, E.S., Ogg, G.,

O’Callaghan, C.A., Kalams, S.A., McKinney, R.E.J., Mayer,

K., Koup, R., Pelton, S.I., MBurchett, S.K., McIntosh, K.,

Walker, B.D., 2000. Functionally inert HIV-specific cytotoxic

T lymphocytes do not play a major role in chronically infected

adults and children. J. Exp. Med. 192, 1819–1832.

Larsson, M., Jin, Z., Ramratnam, B., Ogg, G.S., Engelmayer, J.,

Demoitie, M.-A., McMichael, A.J., Cox, W.I., Steinman, R.M.,

Nixon, D., Bhardwaj, N., 1999. A recombinant vaccinia virus

based ELISPOT assay detects high frequencies of Pol-specific

CD8 T cells in HIV-1-positive individuals. AIDS 13, 767–777.

Liu, Z., Wong, J.T., 1999. Proliferative and regenerative capacities

of CD4+ T cells upton TCR stimulation. Clin. Immunol. 93,

16–23.

Lubaki, M.N., Egan, M.A., Siliciano, R.F., Weinhold, K.J., Bollin-

ger, R.C., 1994. A novel technique for the detection and ex vivo

expansion of HIV type 1-specific cytotoxic T lymphocytes.

AIDS Res. Hum. Retrovir. 10, 1427–1431.

Novitsky, V., Rybak, N., McLane, M.F., Gilbert, P., Chigwedere, P.,

Klein, I., Gaolekwe, S., Chang, S.Y., Peter, T., Thior, I., NDungu,

T., Vannberg, F., Foley, B.T., Marlink, R., Lee, T.H., Essex, M.,

2001. Identification of HIV-1 subtype C Gag-, Tat-, Rev-, and

Nef-specific Elispot-based cytotoxic T-lymphocyte response for

AIDS vaccine design. J. Virol. 75, 9210–9228.

Weingberg, A., Zhang, L., Brown, D., Erice, A., Polsky, B., Hirsch,

M.S., Owens, S., Lamb, K., 2000. Viability and functional ac-

tivity of cryopreserved mononuclear cells. Clin. Diagn. Lab.

Immunol. 7, 714–716.

Wilson, C.C., Wong, J.T., Girard, D., Merril, D.P., Dynan, M., An,

D., Kalams, S.A., Johnson, R.P., Hirsch, M.S., D’Aquila, R.T.,

Walker, B.D., 1995. Ex vivo expansion of CD4+ lymphocytes

from HIV-1 infected persons in the presence of combination

antiretroviral therapy. JID 172, 88–96.

Wong, J.T., Colvin, R.B., 1987. Bi-specific monoclonal antibodies:

selective binding and complement fixation to cells that express

two different surface antigens. J. Immunol. 139, 1369–1374.

Wong, J.T., Pinto, C.E., Gifford, J.D., Kurnick, J.T., Kradin, R.L.,

1989. Characterization of the CD4+ and CD8+ tumor infiltrat-

ing lymphocytes propagated with bispecific monoclonal anti-

bodies. J. Immunol. 143, 3404–3411.

Wong, J.T., Schott, E., Sabga, E.M., Kielpinski, G.G., Colvin, R.B.,

1990. Immunogenic epitopes of the p55 chain of the interleukin

2 (IL2) receptor: relationship to high afficiny IL2 binding and

modulation of the p55 chain. Transplantation 49, 587–596.

N. Jones et al. / Journal of Immunological Methods 274 (2003) 139–147 147