www.elsevier.com/locate/trim

Transplant Immunology 14

EBV-specific memory CD8+ T cell phenotype and function in stable solid

organ transplant patients

Camila Macedoa,b, Albert Donnenbergb,c, Iulia Popescua,b, Jorge Reyesa,b,

Kareem Abu-Elmagda,b, Ron Shapiroa,b, Adriana Zeevia,d, John J. Funga,b,

Walter J. Storkusc,e,f, Diana Metesa,b,c,f,*

aThomas E. Starzl Transplantation Institute, University of Pittsburgh, Pittsburgh, PA, United StatesbDepartment of Surgery, University of Pittsburgh, Pittsburgh, PA, United States

cUniversity of Pittsburgh Cancer Institute, University of Pittsburgh, Pittsburgh, PA, United StatesdDepartment of Pathology, University of Pittsburgh, Pittsburgh, PA, United States

eDepartment of Dermatology, University of Pittsburgh, Pittsburgh, PA, United StatesfDepartment of Immunology, University of Pittsburgh, Pittsburgh, PA, United States

Received 14 January 2005; received in revised form 9 February 2005; accepted 10 February 2005

Abstract

Immune responses to EBV in immunosuppressed (IS) solid organ transplant (SOTx) recipients have not been well characterized. Here we

evaluate the phenotype and function of EBV-specific CD8+ T cells in peripheral blood isolated from ‘‘stable’’ IS SOTx recipients. The EBV-

specific CD8+ T cell memory subset distribution in the peripheral blood of patients was examined by flow cytometric analysis using HLA-A2

tetramers incorporating BMLF1 (lytic), and LMP2 and EBNA3A (latent)-derived peptides, in conjunction with mAbs against the CD45RO,

CD45RA, and CD62L markers. The ability of CD8+ T cells to produce IFN-g in response to the same EBV-derived peptides was measured by

ELISPOT assay. Patients and healthy normal donors exhibited similar anti-EBV CD8+ T cell frequencies and specificities against the EBV

epitopes evaluated. When compared to healthy normal donors, an overall significant expansion of the CD8+ T cell ‘‘effector memory’’

(CD45RO+/CD62L�) pool, including that of EBV ‘‘latent’’ (LMP2 and EBNA3A)-specific CD8+ T cells was detected in IS SOTx patients.

However, the patients’ EBV-specific CD8+ T cells showed decreased IFN-g production to the EBV-peptide stimulation. These results indicate

that the impairment of EBV-specific CD8+ T cell activity is not due to clonal depletion, but is mainly due to impaired functional activation.

D 2005 Elsevier B.V. All rights reserved.

Keywords: Solid organ transplantation; EBV; Memory CD8+ T cells; IFN-g

1. Introduction

Epstein Barr Virus (EBV) is a ubiquitous human g-

herpes virus with tropism for B lymphocytes [1,2]. More

than 90% of the adult world’s population is infected with

EBV, and carry the virus life-long, as an asymptomatic

0966-3274/$ - see front matter D 2005 Elsevier B.V. All rights reserved.

doi:10.1016/j.trim.2005.02.001

* Corresponding author. Section of Cellular Transplantation, Thomas E.

Starzl Transplantation Institute, E1551 Biomedical Science Tower, 200

Lothrop Street, Pittsburgh, PA 15261, United States. Tel.: +1 412 648 3291;

fax: +1 412 624 6666.

E-mail address: metesdm@msx.upmc.edu (D. Metes).

infection. EBV replication and persistence are maintained

through a fine balance of productive (‘‘lytic’’) and non-

productive (‘‘latent’’) infections [3,4]. In healthy individ-

uals, EBV-specific CD8+ T cells play a pivotal role in

controlling viral replication and latency establishment

during the primary infection, and throughout the life-long

carrier state, preventing EBV-associated diseases [5,6].

Since chronic administration of immunosuppressive (IS)

drugs to prevent graft rejection impairs cellular immune

surveillance, EBV is an important source of morbi-

mortality in solid organ transplant (SOTx) recipients,

rendering them susceptible to post-transplant lymphoproli-

ferative disorders (PTLD) [7]. These are typically B cell

(2005) 109 – 116

Table 1

Patient information

Patientsa,b (n =15) Regimenc

FK506 (mg/day) Prednisone (mg/day)

Liver (n =4) 2–6 5

Small Bowel (n =5) 2–12 5–15

Multivisceral (n =6) 4–8 5

Controlsb (n =9) N/A N/A

a Stable SOTx patients=patients with no recent rejection episodes or

EBV re-activations.b The age/gender distribution of subjects was as follows: Controls – 4F/

2M, median age 45; Patients – 4F/6M, median age 49.c All patients were on two-drug maintenance therapy with the exception

of two multivisceral recipients who were on FK506 only.

C. Macedo et al. / Transplant Immunology 14 (2005) 109–116110

malignancies, ranging from polyclonal proliferations to

aggressive monomorphic lymphomas. They display a

broad spectrum of latency genes, also referred to as

latency III phenotype [8]. Recent reports also imply that

EBV lytic replication might actively participate in estab-

lishment and pathogenesis of latency.

Memory CD8+ T cells can be characterized by distinct

phenotypic, migratory, and effector functions. ‘‘Central

memory’’ T cells are CD45RO+ cells that constitutively

express high levels of the L-selectin CD62L marker, and

lack immediate effector function. They home to secondary

lymphoid organs, where they proliferate and differentiate

into effectors upon proper antigenic re-challenge [9,10]. In

contrast, ‘‘effector memory’’ cells exhibit a down-regulation

in CD62L expression, are found mostly in the peripheral

inflamed tissues, and can readily secrete IFN-g, IL-4, and/or

perforin in response to antigen stimulation [9,10]. Some

CD8+ T cells in this functional compartment are CD45RA+,

and are known as ‘‘stable effector memory’’ cells, whereas

others are CD45RO+, and are known as ‘‘effector memory’’

cells. Thus, in humans, the ‘‘effector’’ memory pool

contains the bonafide functional Th1, Th2, and CTL cell

populations [9,10].

The repertoire and activation status of EBV-specific

CD8+ T cells have been well documented in EBV+ healthy

individuals [11–13]. Interestingly, recent studies on

healthy EBV chronic carriers have revealed profound

differences in the composition of EBV-specific CD8+ T

cell memory populations and their functionality, suggesting

that immune response to ‘‘lytic’’ vs. ‘‘latent’’-associated

antigens is independent [14]. EBV ‘‘lytic’’-epitope specific

CD8+ T cells are CD45RO+ or CD45RA+ and CD62L+,

whereas EBV ‘‘latent’’-epitope specific CD8+ T cells

express mainly CD45RO+ and CD62L+ in normal subjects

[11,14].

To date, most studies regarding anti-EBV cellular

immunity in IS SOTx patients have focused on pediatric

transplant recipients with PTLD [15–17]. Little is known

about EBV-specific CD8+ T cell responses in the context of

impaired immunity seen with quiescent SOTx patients. Here

we have characterized the antigen specificities, phenotype,

and function of EBV-specific CD8+T cells in adult stable

transplant patients receiving two-drug maintenance therapy

(Prednisone and Tacrolimus). Our results provide a better

understanding of the anti-viral immune responses influenced

by this unique iatrogenic-biased virus–host relationship.

2. Materials and methods

2.1. Subjects

Fifteen stable IS SOTx patients and nine healthy

individuals (Table 1) were consented under IRB-approved

protocols. All subjects were EBV+ as determined by

serology. Ten stable patients and six healthy individuals

were HLA-A2+ as determined by molecular HLA-typing,

performed at the University of Pittsburgh Medical Center

(Table 1). At the time of the blood donation, all patients

were quiescent, without evidence of rejection or EBV

(re)activation. All patients had been receiving chronic

daily administration of Prednisone (5–15 mg/day) and

FK506 (2–12 mg/day) for more than 3 years. These

patients had never received Tymoglobulin or Campath1H

as preconditioning regimen or for rejection– rescue

intervention.

2.2. Cells

Ten milliliters of heparinized whole blood was collected

from each subject every 6 months over a period of 1 year.

Whole blood was used in flow cytometry analyses, while

peripheral blood mononuclear cells (PBMC) were isolated

by density gradient centrifugation for functional assays [18].

The TAP-deficient T2 cell line was pulsed with the relevant

peptides, and used as an antigen-presenting cell to stimulate

autologous PBMC in ELISPOT assays [19].

2.3. Peptides

The following EBV-derived peptides presented by HLA-

A2 were synthesized at the Peptide Synthesis Facility at

University of Pittsburgh: GLCTLVAML (BMLF1),

CLGGLLTMV (LMP2), and SVRDRLARL (EBNA3A)

[4,5]. These peptides were used in ELISPOT assays, as

well as for generation of streptavidin-PE conjugated HLA-

A2 tetramers at the NIAID MHC Tetramer Core Facility

(Emory University, Atlanta, GA).

2.4. Flow cytometry

Whole blood (200 Al) was incubated with a given

tetramer and a mixture of 3 mAbs for 30 min in the dark

at room temperature [19]. Surface staining markers used

included: anti-CD8 Cy-Chrome and anti-CD45RO FITC

(BD Bioscience Pharmingen, San Jose, CA); anti-CD62L

FITC, and anti-CD45RA ECD (Immunotech, Miami, FL).

C. Macedo et al. / Transplant Immunology 14 (2005) 109–116 111

Red blood cells were lysed with FACS Lysing Solution (BD

Biosciences); remaining cells were washed and re-sus-

pended in 1% paraformaldehyde. The samples were

analyzed on a Coulter EPICSElite flow cytometer (Beck-

man-Coulter, Miami FL). At least two independent analyses

were performed for each subject, with representative data

reported.

2.5. IFN-c ELISPOT assay

ELISPOT assays were performed as previously described

[19]. Briefly, 96 well plates (Millipore, Bedford, MA) were

pre-coated with anti-IFN-g mAb1-D1K (10 Ag/ml, Mab-

tech, Sweden). PBMC (105) were stimulated with T2 cells

(2�104) pre-pulsed with each EBV-peptide (10 Ag/ml) at

37 -C and 5%CO2 for 24 h. PMA (5 ng/ml) plus Ionomycin

(100 ng/ml) (Sigma, St. Louis, MO) stimulation served as a

positive control. After washing and addition of biotinylated

anti-IFNgmAb 7-B6-1 (2 Ag/ml, Mabtech), the reaction was

developed as described [19]. The spots were counted using

an ELISPOT reader (ImmunoSpotTM, Cellular Technology,

Cleveland, OH).

2.6. Statistical analysis

The tetramer and ELISPOT values were log normally

distributed and were log-transformed prior to analysis [20].

Student’s t test (2-tailed) was used for comparison of

groups. Linear regression was used to evaluate the

correlation of parameters. p�0.05 was considered statisti-

cally significant.

0

20

40

60

80

100

0

20

40

60

80

100 1

1

% o

f pos

itive

cel

ls

CD8

CD45 RO

controlspatients

Fig. 1. CD8+ T cell phenotyes in peripheral blood of stable IS SOTx patients. Fou

was performed by gating on lymphocytes and subsequent gating on CD8+ T cel

circulating CD8+ T cell population in peripheral blood of control subjects and pa

3. Results

3.1. CD8+ T cell phenotypes in peripheral blood of solid

organ transplant recipients

The patients showed a trend of higher frequencies of

CD8+ T cells in their peripheral blood as compared to

healthy individuals (Fig. 1). The proportion of CD8+ T

cells expressing the CD45RA marker was lower in patients

than controls, with a complementary increase in the

frequency of cells bearing a CD45RO phenotype (Fig.

1). In addition, we found that patients CD8+ T cells

exhibited significantly lower expression levels of CD62L

marker ( p <0.03) as compared to controls, suggesting their

shift from ‘‘central memory’’ into ‘‘effector memory’’

compartment (Fig. 1).

3.2. Analysis of the EBV-specific CD8+ T cell profiles in

peripheral blood of solid organ transplant recipients

The EBV-specific CD8+ T cell profiles were analyzed

by using EBV-specific HLA-A2 tetramers and flow

cytometric analysis. Overall, CD8+ T cells specific for

BMLF1 (lytic) and LMP2 and EBNA3A (latent) EBV

antigens were easily detected in the peripheral blood of

both normal controls and patients (Fig. 2A). In healthy

EBV+ subjects, the BMLF1-specific CD8+ T cells were

most frequently detected in peripheral blood (in 5 out of

6 subjects), with the mean values of 0.49%T0.4 of total

CD8+ T cells, while the LMP2 and EBNA3A specific

CD8+ T cells were comparatively less frequent (3 out of 6

0

20

40

60

80

00

0

20

40

60

80

00

CD45 RA

CD62L

*p< 0.03

r-color flow cytometry was used for whole blood staining. Data acquisition

ls to detect the percentages of CD45RO+, CD45RA+, and CD62L+ within

tients *p�0.03.

0

0.5

1

1.5

2

2.5

.5 1 1.5 2 2.50

0.5

1

1.5

2

2.5

0 0.5 1 1.5LMP2 EBNA3A BMLF1 LMP2 EBNA3A BMLF1

Controls Patients

% E

BV

-spe

cific

CD

8+T

cel

ls

A

-1 0 1 2

LR_EBNA3A

-1

0

1

2

LR_L

MP

2B

p=0.001

Fig. 2. Analysis of the repertoire of EBV-specific CD8+ T cells in peripheral blood of stable IS SOTx patients. Panel A compares the EBV-specific repertoire

and the percentages of EBV tetramer+ cells among total CD8+ T cells in 6 normal controls and 10 patients. Panel B depicts the correlation of the frequencies of

LMP2 and EBNA3A tetramer+ CD8+ T cells in peripheral blood of control subjects and patients (n =16, R2=0.56, p =0.001).

C. Macedo et al. / Transplant Immunology 14 (2005) 109–116112

subjects), with mean values of 0.13%T0.1 and 0.12%T0.1, respectively (Fig. 2A). Among the study patients, the

BMLF1-specific CD8+ T cell responses were also

frequent (in 7 out of 10 subjects), with the mean

frequencies of 0.50%T0.7. Interestingly, more patients

displayed ‘‘latent’’ anti-LMP2 and anti-EBNA3A specific

CD8+ T cells responses (in 6 out of 10 donors evaluated

for LMP2 and in 8 out of 10 subjects assessed for

EBNA3A), with mean values of 0.24% T0.5 and

0.36%T0.7, respectively, when compared to normal

control subjects (Fig. 2A). All normal controls and five

out of 10 patients were analyzed at least twice over a 12

months period to determine the stability of these sub-

population frequencies. These staining patterns remained

consistent for each EBV-epitope over the study period

(data not shown).

We also analyzed the data as logarithmic reciprocal

frequencies using linear regression. Our results showed

that LMP2 and EBNA3A ‘‘latent’’-specific CD8+ T cell

frequencies were well correlated in both patients and

normal controls (Fig. 2B), whereas the BMLF1 ‘‘lytic’’-

specific CD8+ T cell frequencies did not correlate with

either of the latent epitopes (data not shown). Thus, the

frequency of cells responsive to LMP2 correlated in

magnitude to those responsive to EBNA3A, despite the

fact that these represent discrete epitopes.

3.3. EBV-specific CD8+ T cells memory subset distribution

in the peripheral blood of stable transplant recipients

Phenotypic analysis of EBV-specific CD8+ T cells

provided us with information on their maturational status.

Our results show that for healthy EBV+ controls, the

CD8+ T cells specific for LMP2 and EBNA3A epitopes

express primarily CD45RO+ phenotype, with 60–70% of

them accumulating in the CD45RO+CD62L+ ‘‘central

memory’’ pool (Fig. 3A,B). In contrast, CD8+ T cells

specific for BMLF1 epitope appear equally distributed

between the CD45RA+/CD62L� (stable effector memory)

and the CD45RO+/CD62L� (effector memory) pools,

which emphasizes the important relationship between

EBV-epitope specificity and CD8+ T cell memory

phenotype [21]. In SOTx patients, the circulating

LMP2- and EBNA3A-specific CD8+ T cells have

significantly decreased the CD62L expression, as com-

pared to control subjects ( p <0.003 and p <0.03, respec-

tively). These results suggest that in transplant recipients,

the LMP2 and EBNA3A (latent)-specific CD8+ T cells

shift from the ‘‘central memory’’ (CD45RO+/CD62L+)

into the ‘‘effector memory’’ (CD45RO+/CD62L�) pool,

similar to the phenotype seen with BMLF1 (lytic)-

specific CD8+ T cells memory subset distribution

(Fig 3A,B).

A

0.01 0.47

60.5

0.34 0.14

19.9

0.01 0.43

57.2

0.39 0.05

18.3

0.91 1.26

58.5

1.68 0.07

18.3

0.11 0.4

43.3

0.16 0.24

39.8

0.15 0.4

41.2

0.2 0.32

39.3

0.3 0.15

34.8

0.28 0.15

50.5

LM

P2a

PE

BM

LF

1 P

EE

BN

A3A

PE

CD45RO FITC CD62L FITC CD45RO FITC CD62L FITC

PatientsControls

Controls Patients

0

20

40

60

80

100

CD45 RA

CD45 RO

CD62L

0

20

40

60

80

100

CD45 RACD45 ROCD62L

% o

f pos

itive

cel

ls

LMP2 (n=3)

EBNA3A (n=3)

BMLF1 (n=5)

LMP2 (n=6)

EBNA3A (n=8)

BMLF1 (n=7)

p≤0.03p≤0.003

**

B

103

102

101

100

103102101100

103

102

101

100

103102101100

103

102

101

100

103102101100

103

102

101

100

103102101100

103

102

101

100

103102101100

103

102

101

100

103102101100

103

102

101

100

103102101100

103

102

101

100

103102101100

103

102

101

100

103102101100

103

102

101

100

103102101100

103

102

101

100

103102101100

103

102

101

100

103102101100

Fig. 3. The memory subset distribution of EBV-specific CD8+ T cells in the peripheral blood of stable IS SOTx patients. Panel A shows the memory subset

distribution (CD45RO+ and CD62L+) of EBV-specific HLA-A2 tetramer+ CD8+ T cells in one representative experiment performed with one normal control

vs. one IS SOTx patient. Data acquisition was performed by gating on lymphocytes and subsequent gating on CD8+ T cells. Panel B shows the mean TS.D.

percentages of CD45RO+, CD45RA+, and CD62L+ co-stained with EBV-specific HLA-A2 tetramers+ in IS SOTx patients and controls. *p�0.003,

**p�0.03.

C. Macedo et al. / Transplant Immunology 14 (2005) 109–116 113

3.4. IFN-c production by EBV-specific peripheral blood

CD8+ T cells in SOTx patients

The IFN-g production by CD8+ T cells in response to

PMA/Ionomycin (non-specific) stimulation was comparable

in both SOTx patients and control subjects, suggesting that

the patients were not over-immunosuppressed (Fig. 4A).

When patients’ CD8+ T cells were challenged in vitro

against all three EBV-specific peptides, we detected lower

frequencies of IFN-g production (27T12 spots/105 for

BMLF1, 17T16 spots/105 for LMP2, and 23T22 spots/

105 for EBNA3A), as compared to normal controls (58T70spots/105 for BMLF1, 29T27 spots/105 for LMP2, and

27T15 spots/105 for EBNA3A) (Fig. 4B).

In addition, the IFN-g ELISPOT frequencies to the

‘‘latent’’-specific epitopes correlated well, whereas the IFN-

g ELISPOT frequencies in response to ‘‘latent’’ vs. ‘‘lytic’’-

specific peptides were uncorrelated (Fig. 4C and data not

shown).

3.5. Correlation of EBV-specific tetramer and ELISPOT

frequencies

We compared the frequency of EBV-specific CD8+ T

cells detected by tetramer staining with the frequency of

EBV-specific CD8+ T cells to produce IFN-g as detected by

ELISPOT. The ratio of tetramer+ to ELISPOT+ cells was

higher in patients than in controls for the ‘‘latency’’-

associated peptides LMP2 (11.1 vs. 5.3) and EBNA3A

(12.6 vs. 3.9) (Fig. 5). In contrast, the tetramer+ to

ELISPOT+ ratio for ‘‘lytic’’-associated peptide BMLF1

was comparable in patients and controls (11.5 vs. 9.5)

6

8

10

12

14

er+

/ELI

SP

OT

+ controls

patients

0

200

400

600

800

1000

1 2

A

B

020406080

100120140160180200

1 2 3 4LMP2 EBNA3A BMLF1 LMP2 EBNA3A BMLF1

PMA+Ca2+ PMA+Ca2+

# of

IFN

- s

ecre

ting

cells

/105

cells

controls

patients

controls

patients

1000 100001000

10000

LR (EBNA3A)

C

p=0.009

LR (

LMP

2)

Fig. 4. IFN-g response to EBV epitopes by CD8+T cells isolated from the

peripheral blood of stable IS SOTx patients. Panel A indicates the

frequencies of IFN-g secreting effector cells after PMA/Ionomycin

stimulation in control subjects (n =6) and patients (n =10). Panel B depicts

the meanTS.D. of the frequency of IFN-g producing cells in response to

EBV-derived peptide stimulation in control subjects and patients. Panel C

illustrates the correlation of the frequencies of IFN-g secreting cells in

response to EBNA3A and LMP2 epitope-specific stimulation. Control data

are shown as Os and patient data as Xs. The least-squares line of best fit

with 95% confidence intervals is shown (n =16, R2=0.42, p =0.009).

C. Macedo et al. / Transplant Immunology 14 (2005) 109–116114

(Fig. 5). These results demonstrate the independence of

immune responses to lytic and latent antigens, and show a

preferential tendency towards a decrease in IFN-g secreting

CD8+ T cell frequency associated with latent epitopes in

patients as compared to normal controls.

0

2

4

LMP2 EBNA3A BMLF1

Tet

ram

Fig. 5. Correlation between tetramer and ELISPOT CD8+ T cell

frequencies. The results represent the ratios between absolute numbers of

‘‘lytic’’ and ‘‘latent’’ tetramer specific CD8+ T cells and the absolute

numbers of ‘‘lytic’’ and ‘‘latent’’-specific IFN-g producing CD8+ T cells

obtained from both IS SOTx patients (n =10) and healthy controls (n =6).

4. Discussions

In healthy individuals, it is well established that Type-1

CD8+ cytotoxic T cells are critical for mediating immune

surveillance against EBV [5,6]. Chronic administration of

immunosuppressive drugs in SOTx patients non-specifically

impairs T cell functions, leading to persistent EBV infection

that may result in fatal virus-driven lymphoproliferation

[7,8]. The few studies evaluating pediatric immunosup-

pressed SOTx patients longitudinally to the time of

diagnosis of PTLD describe a profound deficiency in anti-

EBV cellular immunity, paralleled by very high EBV loads

[16,17]. If modulation of immunosuppression and treatment

is followed by recruitment and expansion of effective Type-

1 CD8+ T cells, this should be linked to a favorable outcome

[16,17]. In this study, we have analyzed the impact of

chronic administration of immunosuppressive drugs on the

maturation, function, and specificity of peripheral blood

EBV-specific CD8+ T cells from EBV+ quiescent SOTx

patients. It is well documented that calcineurin inhibitors

down-modulate T cell function by blocking transcription of

the IL-2 gene, while corticosteroids have a broader effect on

T cells, by blocking pro-inflammatory cytokines synthesis

and promoting apoptosis [22,23]. Overall, the study patients

displayed higher levels of CD8+ T cells, with significant

expansion of the effector memory (CD45RO+/CD62�) cell

subset, as compared to normal controls (Fig. 1). These

phenotypic changes may be a direct consequence of chronic

administration of immunosuppressive drugs, or may be

related to a prolonged (higher) antigenic challenge (due to

impaired T cell immune-surveillance) that may occur in

SOTx patients.

By using MHC class I tetramers specific for BMLF1

(lytic) and LMP2 and EBNA3A (latent)-derived peptides,

we determined that EBV-specific CD8+ T cells are readily

detectable in the peripheral blood of stable transplant

recipients, and that their specificities and frequencies are

comparable to that of healthy EBV chronic carriers (Fig.

2A). The MHC class I tetramer technique was also used to

detect the EBV-specific CD8+ T cells in the peripheral blood

of pediatric transplant recipients [15].

Here we have detected a significant shift of EBNA3A

and LMP2 (latent)-specific CD8+ T cells from ‘‘central

memory’’ (CD45RO+/CD62L+) to ‘‘effector memory’’

(CD45RO+/CD62L�) compartment in study patients,

whereas in healthy chronic EBV carriers, the EBV latent-

specific CD8+ T cells displayed mostly a ‘‘central memory

C. Macedo et al. / Transplant Immunology 14 (2005) 109–116 115

phenotype’’. The BMLF1 (lytic)-specific CD8+ T cells (Fig.

3A,B) in both patients and controls are mostly part of the

‘‘effector memory’’ subset [21]. When taken together, these

results support the idea that stable transplant recipients may

experience chronic higher endogenous EBV ‘‘latent’’-anti-

gen challenge, leading to ‘‘latent’’-specific clonal prolifer-

ation of potential ‘‘promptly’’ functional T cells (Fig. 3A,B).

Unfortunately, we were not able to quantify the viral load in

PBMC of the patients at the time of phenotypic analysis.

Previous reports suggest that transplant patients have an

increased number of circulating EBV-infected B cells in

their peripheral blood, and that these cells carry an elevated

EBV copy number per genome [24,25]. Babcock et al.

speculate that the increased EBV genome burden per B cell

is caused by a more frequent viral replication seen in

patients [25]. In stable SOTx patients, although circulating

EBV-infected B cells are resting memory B cells, they may

frequently convert into proliferating B cell blasts due to

impaired T cell immune-surveillance caused by persistent

immunosuppression. These B cell blasts that express latency

III phenotype, may therefore represent a constant source of

EBV ‘‘latent’’ antigens [25].

We next examined the frequency of IFN-g producing,

EBV-specific CD8+ T cells in the peripheral blood of stable

IS SOTx patients using the ELISPOT assay. Previous

reports, including our own published results, have deter-

mined that in healthy EBV+ controls approximately 10–

20% of circulating EBV-specific CD8+ T cells readily

respond to in vitro antigen-specific stimulation by produc-

ing IFN-g [19]. The frequency of IFN-g producing BMLF1-

specific CD8+ T cells was (on average) higher than the

frequencies of IFN-g producing EBNA3A and LMP2

specific CD8+ T cells (Fig. 4B). This is most likely due to

the higher frequency of BMLF1-specific CD8+ T cells,

which display an ‘‘effector memory’’ phenotype (CD45RO+/

CD62L� or CD45RA+/CD62L�) in peripheral blood

[11,19].

Our current results show that the frequency of CD8+ T

cells producing IFN-g in patients and controls was similar

when assessed by PMA and ionomycin stimulation,

supporting the notion that stable SOTx recipients were not

over-immunosuppressed. In contrast, the response to EBV-

specific stimulation induced lower frequencies of IFN-g

producing CD8+ T cells in patients, as compared with

healthy controls (Fig. 4B). This is intriguing, in view of an

increased proportion of ‘‘effector memory’’ cells among the

patients’ LMP2 and EBNA3A tetramer-specific CD8+ T

cells, as compared with that of healthy control donors.

Based on this phenotype, patients’ cells would be expected

to respond readily (i.e. IFN-g production) upon EBV-

specific stimulation [9,10]. We therefore speculate that the

antigen-specific functional (IFN-g) impairment of CD8+ T

cells in study patients may be related to the chronic

administration of IS drugs, and may be facilitated by the

development of antigen-specific suppressor/regulatory cells

[26]. The relationship between tetramer-specific and IFN-

gproducing antigen-specific CD8+ T cells in SOTx patients

has also been analyzed in the setting of alternate viral

infections [27]. Engstrand et al. showed that CMV-specific

CD8+ T cells are maintained in the peripheral blood of

SOTx patients, at levels comparable to those detected in

healthy subjects [27]. However, even though they express

an activated effector phenotype, they are functionally

impaired at the level of IFN-g production [27]. Of note, in

our study, the ratio of tetramer+ to ELISPOT+ cells further

revealed a significant decrease in IFN-g secretion specific

for the ‘‘latency’’ associated epitopes in stable patients (Fig.

5). These subtle alterations in the phenotype and function of

LMP2- and EBNA3A-specific memory CD8+ T cells in

patients may therefore, at least in part, explain the tendency

towards increased susceptibility of these individuals to

EBV-latency associated complications. Further studies are

needed to identify additional immune defects that may

render certain patients to be at greater risk than others to

develop EBV-related malignancies.

Acknowledgments

We thank Allison Logar for her outstanding assistance

with flow cytometric data analysis; Dr. Russell Salter

(University of Pittsburgh) for his gift of the T2 cell line.

This work was supported by grants from American Cancer

Society and American Heart Association. [DM]

References

[1] Rickinson A. Epstein-Barr virus. Virus Res 2002;82:109–13.

[2] Kuppers R. B cells under the influence: transformation of B cells by

Epstein-Barr Virus. Nat Rev Immunol 2003;3:801–12.

[3] Murray PG, Young LS. Epstein-Barr virus infection: basis of

malignancy and potential for therapy. Expert Rev Mol Med

2001:1–20.

[4] Thorley-Lawson DA. Epstein-Barr virus: exploiting the immune

system. Nat Rev Immunol 2001;1:75–82.

[5] Rickinson AB, Moss DJ. Human cytotoxic T lymphocyte responses to

Epstein Barr virus infection. Annu Rev Immunol 1997;15:405–13.

[6] Khanna R, Moss DJ, Burrows SR. Vaccine strategies against

Epstein-Barr virus-associated diseases: lessons from studies on

cytotoxic T-cell-mediated immune regulation. Immunol Rev 1999;

170:49–64.

[7] Nalesnik MA. Clinicopathologic characteristics of post-transplant

lymphoproliferative disorders. Recent Results Cancer Res

2002;159:9–18.

[8] Damania B. Oncogenic gamma-herpes viruses: comparison of viral

proteins involved in tumorigenesis. Nat Rev Microbiol 2004;8:

656–68.

[9] Sallusto F, Lenig D, Forster R, Lipp M, Lanzavecchia A. Two subsets

of memory T lymphocytes with distinct homing potentials and effector

functions. Nature 1999;401:708–12.

[10] Sallusto F, Lanzavecchia A. Exploring pathways of memory T cell

generation. J Clin Invest 2001;108:805–6.

[11] Hislop AD, Annels NE, Gudgeon NH, Leese AM, Rickinson AB.

Epitope-specific evolution of human CD8(+) T cell responses from

primary to persistent phases of Epstein-Barr virus infection. J Exp

Med 2002;195:893–905.

C. Macedo et al. / Transplant Immunology 14 (2005) 109–116116

[12] Catalina MD, Sullivan JL, Brody RM, Luzuriaga K. Phenotypic and

functional heterogeneity of EBV epitope-specific CD8+T cells. J

Immunol 2002;168:4184–91.

[13] Tussey L, Speller S, Gallimore A, Vessey R. Functionally distinct

CD8+ memory T cell subsets in persistent EBV infection are

differentiated by migratory receptor expression. Eur J Immunol

2000;30:1823–9.

[14] Faint JM, Annels NE, Curnow SJ, Shields P, Pilling D, Hislop AD,

et al. Memory T cells constitute a subset of the human

CD8+CD45RA+ pool with distinct phenotypic and migratory

characteristics. J Immunol 2001;167:212–20.

[15] Falco DA, Nepomuceno RR, Krams SM, Lee PP, Davis MM,

Salvatierra O, et al. Identification of Epstein-Barr virus-specific

CD8+ T lymphocytes in the circulation of pediatric transplant

recipients. Transplantation 2002;74:501–10.

[16] Smets F, Latinne D, Bazin H, Reding R, Otte JB, Buts JP, et al. Ratio

between Epstein-Barr viral load and anti-Epstein-Barr virus specific T-

cell response as a predictive marker of posttransplant lymphoprolifer-

ative disease. Transplantation 2002;73(10):1603–10.

[17] Baudouin V, Dehee A, Pedron-Grossetete B, Ansart-Pirenne H,

Haddad E, Maisin A, et al. Relationship between CD8+ T-cell

phenotype and function, Epstein-Barr virus load, and clinical outcome

in pediatric renal transplant recipients: a prospective study. Trans-

plantation 2004;77(11):1706–13.

[18] Boyum A, Lovhaug D, Tresland L, Nordlie EM. Separation of

leucocytes: improved cell purity by fine adjustments of gradient

medium density and osmolarity. Scand J Immunol 1991;34:697–712.

[19] Popescu I, Macedo C, Zeevi A, Nellis J, Patterson KR, Logar A, et al.

Ex vivo priming of naı̈ve T cells into EBV-specific Th1/Tc1 effector

cells by mature autologous DC loaded with apoptotic/necrotic LCL.

Am J Transplant 2003;3:1369–77.

[20] Hoffmann TK, Donnenberg VS, Friebe-Hoffmann U, Meyer EM,

Rinaldo CR, DeLeo AB, et al. Competition of peptide-MHC class I

tetrameric complexes with anti-CD3 provides evidence for specific-

ity of peptide binding to the TCR complex. Cytometry 2000;41:

321–8.

[21] Hislop AD, Gudgeon NH, Callan MF, Fazou C, Hasegawa H, Salmon

M, et al. EBV-specific CD8+ T cell memory: relationship between

specificity, cell phenotype, and immediate effector function. J Immunol

2001;167:2019–29.

[22] Siekierka JJ, Sigal NH. FK-506 and Cyclosporin A: immunosuppres-

sive mechanism of action and beyond. Curr Opin Immunol 1992;4:

548–52.

[23] Goulding NJ, Guyre PM. Glucocorticoids, lipocortins and the immune

response. Curr Opin Immunol 1993;5:108–13.

[24] Rose C, Green M, Webber S, Kingsley L, Day R, Watkins S, et al.

Detection of Epstein-Barr virus genomes in peripheral blood B cells

from solid-organ transplant recipients by fluorescence in situ hybrid-

ization. J Clin Microbiol 2002;40:2533–44.

[25] Babcock GJ, Decker LL, Freeman RB, Thorley-Lawson DA. Epstein-

Barr virus-infected resting memory B cells, not proliferating lympho-

blasts, accumulate in the peripheral blood of immunosuppressed

patients. J Exp Med 1999;190(4):567–76.

[26] Li Y, Koshiba T, Yoshizawa A, Yonekawa Y, Masuda K, Ito A, et al.

Analyses of peripheral blood mononuclear cells in operational

tolerance after pediatric living donor liver transplantation. Am J

Transplant 2004;4(12):2118–25.

[27] Engstrand M, Lidehall AK, Totterman TH, Herrman B, Eriksson BM,

Korsgren O. Cellular responses to cytomegalovirus in immunosup-

pressed patients: circulating CD8+ T cells recognizing CMVpp65 are

present but display functional impairment. Clin Exp Immunol 2003;

132(1):96–104.