Human virus-specific T cells in peripheral blood and lymph nodes: Phenotype, function and clonal relationships - Chapter 6: Clonal evolution of CD8+ T cell responses against latent viruses: Relationship between phenotype, localization

(1)

UvA-DARE is a service provided by the library of the University of Amsterdam (https://dare.uva.nl)

UvA-DARE (Digital Academic Repository)

Human virus-specific T cells in peripheral blood and lymph nodes: Phenotype,

function and clonal relationships

Remmerswaal, E.B.M.

Publication date

2014

Document Version

Final published version

Link to publication

Citation for published version (APA):

Remmerswaal, E. B. M. (2014). Human virus-specific T cells in peripheral blood and lymph

nodes: Phenotype, function and clonal relationships.

General rights

It is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s)

and/or copyright holder(s), other than for strictly personal, individual use, unless the work is under an open

content license (like Creative Commons).

Disclaimer/Complaints regulations

If you believe that digital publication of certain material infringes any of your rights or (privacy) interests, please

let the Library know, stating your reasons. In case of a legitimate complaint, the Library will make the material

inaccessible and/or remove it from the website. Please Ask the Library: https://uba.uva.nl/en/contact, or a letter

to: Library of the University of Amsterdam, Secretariat, Singel 425, 1012 WP Amsterdam, The Netherlands. You

will be contacted as soon as possible.

(2)

(3)

CLONAL EVOLUTION OF CD8

+

_{T CELL}

RESPONSES AGAINST LATENT VIRUSES:

RELATIONSHIP BETwEEN PHENOTYPE,

LOCALIZATION AND FUNCTION

Ester B.M. Remmerswaal

1,2,

_{*, Paul L. Klarenbeek}

1,3,4,

_{*, Nuno L. Alves}

1

_,

Marieke E. Doorenspleet

1,3,4

_{, Barbera D.C. van Schaik}

5

_,

Rebecca E.E. Esveldt

1,3,4

_{, Mirza M. Idu}

6

_{, Ester M.M. van Leeuwen}

1

_,

Nelly van der Bom-Baylon

1,2

_{, Antoine H.C. van Kampen}

5

_{, Sven D. Koch*,}

Hanspeter Pircher

7

_{, Frederike J. Bemelman}

2

_{, Anja ten Brinke}

8

_{, Frank Baas}

4

_,

Ineke J.M. ten Berge

2,

_{*, Rene A.W. van Lier}

1,8,

_{* and Niek de Vries}

3,

_*

1_{Department of Experimental Immunology,}

Academic Medical Center, Amsterdam, the Netherlands

2_{Renal Transplant Unit, Department of Internal Medicine,}

3_{Department of Clinical Immunology and Rheumatology,}

4_{Department of Genome Analysis, Academic Medical Center,}

Amsterdam, the Netherlands

5_{Department of Clinical Epidemiology, Biostatistics and Bioinformatics,}

6_{Department of Surgery, Academic Medical Center, Amsterdam, the Netherlands}

7_{Institute of Immunology, University of Freiburg, Freiburg, Germany}

8_{Sanquin Research at CLB, Amsterdam, the Netherlands}

*authors contributed equally to this manuscript

(4)

CLONAL EVOLUTION OF VIRAL CD8

+

T CELL RESPONSES

6 ABSTRACT

Human cytomegalovirus (hCMV) infection is characterized by a vast expansion of resting effector-type virus-specific T cells in the circulation. In mice, IL-7R

α

-expressing cells contain the precursors for long-lived antigen-experienced CD8+_{T cells, but it is}

unclear if similar mechanisms operate to maintain these pools in humans. Here we studied whether IL-7R

α

-expressing cells obtained from peripheral blood (PB) or lymph nodes (LNs) sustain the circulating effector-type hCMV-specific pool.

Using flowcytometry and functional assays we found that the IL-7Rα+_{hCMV-specific}

T cell population comprise cells that have a memory-phenotype and lack effector-features. We used next-generation sequencing of the T-cell receptor to compare the clonal repertoires of IL-7Rα+_{and IL-7Rαˉ subsets. We observed limited overlap of clones}

between these subsets during acute infection and after one year. When comparing the hCMV-specific repertoire between PB and paired LN we found that LNs often contain unique clones. These clones only rarely appear in the PB during reactivation of hCMV.

Thus, although PB IL-7Rα-expressing and LN hCMV-specific CD8+_{T cells show}

typical traits of memory-type cells, these populations do not seem to contain the precursors for the hCMV-specific CD8+_{T-cell pool during latency or upon antigen}

recall. IL-7Rα+_{PB and LN hCMV-specific memory cells form separate virus-specific}

compartments, and precursors for the peripheral blood hCMV-specific CD8+

effector-type T cells are possibly located in other secondary lymphoid tissues or are being recruited from the naïve CD8+_{T cell pool.}

(5)

+

T CELL RESPONSES

6 INTRODUCTION

Adaptive immune responses against transient viral infections typically consist of three phases. First, viral antigens are recognized by naive CD8+_{T cells in LNs, where}

after activated T cells expand vigorously to form effector clones that eliminate virus-infected cells. Second – after clearance of the virus – the majority of the activated CD8+_{T cells undergo apoptosis. Third, a proportion of virus-specific T cells survives}

to provide long-lasting immunological memory (1-3). Although this response is well established for cleared infections, responses against persistent viruses are more complex. The immune-surveillance required to control these infections, triggers regular activation of virus-specific CD8+_{T cells. Persistent infections can therefore}

challenge the immune system for decades and may be associated with lympho-proliferative disorders and opportunistic infections in immunocompromised patients. Understanding how viral latency is maintained is important in designing strategies that may prevent complications from these infections. HCMV is an attractive virus to study persistent infections in humans as the primary infection can be studied longitudinally in recipients of solid organ transplants, such as kidneys. Here we used this approach to study the clonal and phenotypic relations between peripheral blood (PB) and LN memory and PB effector-type subsets in primary and latent phases of hCMV infection.

The majority of the latent phase circulating hCMV-specific CD8+_{T cells is CD28⁻CD27⁻}

CD45RA+_{granzyme B}+_perforin+ _quiescent_{effector-type cells. These CD8}+_{T cell populations}

consist of large clonal expansions that are maintained for many years (4). As such, these cells were thought to be long-lived (5). Recent findings in a murine CMV (MCMV) model on the other hand showed that MCMV-specific effector CD8+_{T-cell pool was maintained}

by constant recruitment of CD27-expressing memory T cells and, to a limited extent, naïve T cells (6, 7). A “buffered memory” concept was suggested (6), proposing that a memory-like T-cell pool, shielded from high antigenic loads by compartmentalization, would be supplementing the effector-type pool at times of rechallenge.

Such a concept has not been investigated in hCMV. It has been shown that, besides the prevalent resting effector-type cells, at least two hCMV-specific CD8+_{T-cell pools}

with memory-like features can be distinguished. In PB, IL-7Rα (CD127) - a marker that distinguishes memory-precursor cells during transient viral infections - is expressed on a minority of the PB hCMV-specific CD8+_{T cells (8). Further in LNs, the chemokine}

receptor CCR7, which can be used to identify central memory T cells, is found on a substantial percentage of the hCMV-specific CD8+_{T cells (9). We hypothesized that}

either one or both of these hCMV-specific CD8+_{T cell memory pools might fuel the}

PB effector-type pool. To address this issue, we combined flowcytometry with next-generation sequencing to investigate phenotype and longitudinal clonal relationships between PB and LN derived subsets of hCMV-specific CD8+_{T cells.}

(6)

+

T CELL RESPONSES

6 MATERIALS AND METHODS

Subjects

The phenotype and cytokine producing capacities of IL-7R

α

ˉ and IL-7R

α

+

hCMV-specific CD8+_{T cells in latency was studied in healthy individuals (n=7). In order to}

study the contribution of acute phase IL-7R

α

-expressing hCMV-specific CD8+_{T cells}

to those present during the latency phase, hCMV-specific CD8+_{T cells from 2}

hCMV-seronegative recipients of a hCMV-seropositive kidney transplant were analyzed longitudinally (table 1, pt 1-2). The relationship between paired PB and LN derived virus-specific CD8+_{T cells was studied in 7 hCMV-seropositive renal transplant recipients}

prior to transplantation (pt 3-9). The contribution of LN-derived hCMV-specific CD8+

T cells upon hCMV reactivation was longitudinally analyzed in four of these hCMV-seropositive renal transplant recipients (pt 3-6); three received a hCMV-hCMV-seropositive and one a hCMV-seronegative kidney. Immunosuppressivedrug treatment of these patients is summarized in table 1. Previously, we did not find an important effect of induction treatment with CD25mAb on the immune response against hCMV. These patients were transplanted in the time-period 2000-2004, during which no antiviral prophylaxis was administered. Because of clinical symptoms, patients 2 and 4 were treated with valgancyclovir until the hCMV-PCR was negative at 2 subsequent time points. None of the patients experienced an acute rejection episode. The study was performed according to the Declaration of Helsinki and approved by the local medical ethics committee. All patients gave written informed consent.

Isolation of mononuclear cells from PB and LN

Heparinized PB samples were obtained from healthy individuals and from patients before transplantation (pre-TX) and at regular intervals thereafter up to one year after transplantation. Peripheral blood mononuclear cells (PBMCs) were isolated using standard density gradient centrifugation. Para-iliac LNs were collected from recipients during living donor kidney transplantation as described before (9). Directly after extirpation, the gathered LNs were chopped into small pieces. A cell suspension was obtained by grinding the material through a flow-through chamber. PBMCs and LNMCs were subsequently cryopreserved until further analysis.

Virological analysis

HCMV-PCR, EBV-PCR quantitative polymerase chain reaction (PCR) for hCMV and EBV was performed in EDTA (ethylenediaminetetraacetic acid) whole blood samples, as described (10). To determine hCMV serostatus, anti-hCMV IgG was measured in the serum using the AxSYM microparticle enzyme immunoassay (Abbott Laboratories). Measurements were calibrated relative to a standard serum. EBV serostatus was determined by qualitative measurement of specific IgG against the viral capsid Ag

(7)

+

T CELL RESPONSES

6

and against the nuclear Ag of EBV using, respectively, the anti-EBV viral capsid Ag IgG ELISA and the anti-EBV nuclear Ag of EBV IgG ELISA (Biotest). All tests were performed following the instructions of the manufacturers.

Immunofluorescence staining, flowcytometry

PBMCs were washed in phosphate-buffered saline containing0.01% (wt/vol) NaN₃ and 0.5% (wt/vol) bovine serum albumin.Two million PBMCs were incubated with APC-labeled tetrameric-complexes for either hCMV-pp65, hCMV-IE-1, BZLF1, BMLF-1, EBV-EBNA-1, EBV-EBNA-3a and Influenza A virus-matrix protein 1 (table 2), followed by incubation with a combination of the following antibodies: IL-7R

α

PE-Cy7, IL-7R

α

APC-eFluor780 and CD27 APC-APC-eFluor780, CX₃CR1 PE (eBioscience Inc, San Diego, CA, USA), CXCR3 PE (Caltag, Buckingham, UK), CCR5 PE, CD45RA BrilliantViolet650 (BioLegend, San Diego, CA, USA), CD8 BrilliantViolet421, CD3 V500, CXCR3 AlexaFluor488 (BD Pharmingen, San Diego, CA, USA), CD28 PE, CD45RA PE-Cy7, CCR7 PE-Cy7 (BD Biosciences, San Jose, CA, USA), KLRG1 AlexaFluor488 (11). The FOX-p3 staining kit (eBioscience) was used for intracellular stainings with the following antibodies: granzyme B AlexaFluor700, granzyme A PE (BD Pharmingen), perforin PerCP-eFluor710, Eomesodermin PerCP-eFluor710 (eBioscience), granzyme K PE (Immunotools, Friesoythe, Germany) and T-Bet BrilliantViolet421 (Biolegend). Measurements were done on a LSR Fortessa flow cytometer (BD) and analysis was performed with FlowJo software (FlowJo, Ashland, OR, USA). The gating strategy as well as a representative example of the immunofluorescent staining can be found in supplemental figure 1.

Intracellular cytokine staining

Cytokine release after cognate peptide or phorbol 12-myristate 13-acetate (PMA)/ ionomycin stimulation was performed as described by Lamoreaux (12). Freshly isolated PBMC from hCMVseropositive healthy individuals were stained with anti-CD3 PE-Cy7 and - IL-7R

α

APC-Alexa Fluor 750 and sorted into CD3ˉ, IL-7R

α

+_CD3+_{and IL-7R}

α

_ˉCD3+

populations on a FACSAria. Purity of the obtained sorted cells were checked by flowcytometry and had to be at least 90%. Sorted IL-7R

α

+_CD3+_{and IL-7R}

α

_ˉCD3+_cells

were rested overnight in suspension flasks (Greiner) in RPMI supplemented with 10% FCS, penicillin, and streptomycin. Next, IL-7R

α

+_CD3+_{cells and IL-7R}

α

_ˉCD3+_{cells were}

supplemented with the CD3ˉ sorted cells which were used as APC (1 CD3ˉ cell to 2.5 IL-7R

α

+_CD3+_{or IL-7R}

α

_ˉCD3+_{cells). Subsequently, the cells were stimulated with PMA/}

ionomycin or with the viral peptides in culture medium supplemented with of

α

CD28 (15E8; 2 μg/mL),

α

CD29 (TS 2/16; 1 μg/mL), brefeldin A (Invitrogen; 10 μg/mL) and GolgiStop (BD Biosciences) in a final volume of 200 μL for 4 hours (PMA at 10 ng/ mL; ionomycin at 1 μg/mL) or 6 hours (1 μg of peptide) in untreated, round-bottom, 96-well plates (Corning). Thereafter cells were incubated with the appropriate tetramers, followed by incubation with CD8 PE-AlexaFluor610 (Invitrogen). The cells were then

(8)

+

T CELL RESPONSES

6

washed, fixed, and permeabilized (Cytofix/Cytoperm reagent; BD Biosciences) and subsequently incubated with the following intracellular mAbs: TNF

α

FITC, IL-2 PE, IFN-

γ

PerCP-Cy5.5 (BD Biosciences). Acquisition was done on a FACSCanto flow cytometer (BD) and analysis was done with FlowJo. An representative example of a sort and of the immunofluorescent staining after stimulation can be found in supplemental figure 2

Isolation of IL-7R

α

+

_{and IL-7R}

α

_{ˉ hCMV-specific CD8}

+

_{T cells}

For isolation ofIL-7R

α

+_{and IL-7R}

α

_{ˉ hCMV-specific CD8}+_{T cells, at least 10*10e6 PBMC}

were first stained with the appropriate tetramer followed by anti-CD8 PerCP-Cy5.5 and - IL-7R

α

PE (Beckman Coulter, Indianapolis, USA). PBMC were than first sorted into hCMV-specific CD8+_{T cells on a FACsARIA (BD) and subsequently subjected to a second sort}

to obtain high purity IL-7R

α

ˉ and IL-7R

α

+ _{hCMV-specific CD8}+ _{T cells. For isolation of}

hCMV-specific CD8+_{T cells from LN, paired PB samples and subsequent longitudinal}

PB samples, at least 10*10e6 PBMC were first stained with the appropriate tetramer followed by incubation with anti-CD3 V500 and -CD8 BrilliantViolet421. CD3+_CD8+

tetramer+_{cells were subsequently sorted, followed by a second sort if the purity did not}

exceed 95% after the first sort (example of sort can be found in supplemental figure 3). The numbers of sorted cells are given in table 3. Purity of the obtained sorted hCMV-specific cells was checked by flowcytometry and was at least 95 %.

RNA isolation and cDNA synthesis

RNA was isolated from PB IL-7R

α

ˉ and IL-7R

α

+_{and from total LN and PB hCMV-specific}

sorted cells by the Nucleospin RNA xs kit (Machery Nagel, Düren, Germany) and subsequently subjected to template switch-anchored RT-PCR by using the Smarter pico cDNA PCR synthesis kit and the Advantage 2 PCR kit (both: Clontech, Mountain View, CA, USA) according to manufacturer’s instructions.

Linear amplification, next-generation sequencing

and bioinformatics

10 uL of SMARTer pico treated DNA was amplified using a primerset to cover all functional Vß-gene segments (13). Here, all Vß-gene-segments are denoted according to the HUGO-nomenclature (14). The linear amplification, next-generation sequencing protocol (4, 15) and the bioinformatics pipeline (16) have previously been described. Clones were identified by their unique TCR-Vß sequence. The degree of expansion of each clone was expressed as percentage of all obtained TCR-sequences in a sample. To exclude clonal signal from sort impurities, only clones that exceeded 1% of total were included. Clones smaller than 1% were only included when the same clone exceeded 1% in the other subsets analyzed.

(9)

+

T CELL RESPONSES

6 Statistical analysis

The two-tailed paired T-test was used for analysis of differences between groups. A p-value of < 0.05 was considered statistically significant.

RESULTS

Circulating IL-7R

α

+

_{hCMV specific CD8}

+

_{T cells in healthy}

individuals have phenotypical traits of memory cells

Murine IL-7R

α

+_{virus-specific CD8}+_{T cells have been shown to be antigen-independent,}

long lived memory cells (17). We wanted to test whether PB hCMV-specific IL-7R

α

+_CD8+

T cells had a true memory phenotype by first measuring their surface expression of several differentiation-associated molecules (figure 1A). We found a substantially higher percentage of cells expressing the co-stimulatory receptors CD27 and CD28 in the IL-7R

α

+

hCMV-specific CD8+_{T cells. PD1 expression was not different between the IL7R}

α

+_{and IL-7R}

α⁻

subsets (data not shown). Furthermore, neither population appeared to be activated, since the expression of HLA-DR, CD38 and Ki-67 was uniformly low (data not shown). We also tested killer cell lectin-like receptor G1 (KLRG1) expression as the lack of this receptor has been shown to be another feature of memory precursor cells generated during LCMV infection in mice (18, 19). Although both IL-7R

α

ˉ and IL-7R

α

+_{subsets expressed KLRG1,}

significantly less KLRG1-expressing cells were found in the IL-7R

α

+_subset.

As memory-fate and cytotoxic functions have been shown to be linked to the transcription factors Eomesodermin and T-Bet (18, 20, 21), we analyzed the subsets for expression of these proteins (figure 1B). Both transcription factors were expressed on the majority of cells within both subsets and no significant differences were found.

Finally, we evaluated the expression of homing receptors CCR7 (homing to LN), CCR5, CXCR3 (homing to a wide array of infected tissues) (22-28) and CX3CR1 (homing

to stressed endothelium) (22, 29-31). CCR7 was not expressed on either population of PB hCMV-specific CD8+_{T cells (figure 1C). CCR5 and CXCR3 were found on the majority of}

cells in both subsets, but the IL-7Rα+_{subset contained slightly more CXCR3 expressing}

cells. In contrast more CX₃CR1 expressing cells were found in the IL-7Rαˉ subset,

In summary, the PB IL-7Rα+_{hCMV-specific CD8}+_{T cells contained substantially}

more cells that expressed the costimulatory molecules CD27 and CD28, a typical feature of ‘classical’ memory cells.

Many circulating IL-7R

α

+

_{hCMV-specific CD8}

+

_{T cells lack}

typical effector-type properties

A feature of hCMV-specific T cells is the abundant expression of components of the cytolytic granule exocytosis machinery (32, 33). We found that within the PB hCMV-specific CD8+_{T cells the vast majority of the IL-7R}

α

_{ˉ cells contained both granzyme B}

(10)

+

T CELL RESPONSES

6

and perforin (figure 1D), compared to only half of the IL-7R

α

+_{hCMV-specific CD8}+_T

cells. The latter subset however did harbor significantly more granzyme K containing cells. Granzyme A was equally present in both subsets.

We have previously shown that IL-7Rα-expressing h-CMV-specific CD8+_{T cells have}

superior proliferation capacity (8). We next wanted to evaluate the ability of both subsets to produce cytokines after in vitro restimulation by either PMA-ionomycin or their cognate peptide (figure 1E). No difference was detected in IFNγ and TNFα synthesis, and although no difference in the production of IL-2 could be detected after PMA-ionomycin stimulation, a significantly higher number of IL-7Rα+_{hCMV-specific CD8}+_T

cells were able to produce IL-2 after stimulation with cognate peptide. Consequently, more polyfunctional hCMV-specific CD8+_{T cells (producing IFNγ, TNFα and IL-2}

simultaneously) were found in the IL-7Rα+_{pool compared to the IL-7Rα}-_{pool (figure 1E).}

In conclusion, with respect to both phenotypic and functional attributes, circulating hCMV-specific CD8+_{T cells that express IL-7Rα were found to}_{contain cells with}

classical memory traits.

Circulating IL-7R

α

+

_{hCMV-specific CD8}

+

_{T cells}

do not seem to feed the IL-7R

α

ˉ effector-type

hCMV-specific CD8

+

_{T cell pool found in latency}

Since the above analyses showed that IL-7R

α

+_{hCMV-specific CD8}+_{T-cell subset}

contained cells with all typical properties of memory-cells, we set out to analyze if they would contribute to the establishment and/or maintenance of the IL-7R

α

ˉ specific effector-type pool. To this end, we studied two patients who were hCMV-seronegative prior to kidney transplantation. Each of them received a kidney from a hCMV-seropositive donor, resulting in a primary hCMV infection (figure 2 and table 1). From peripheral blood, IL-7R

α

+_{and IL-7R}

α

_{ˉ subsets of hCMV-pp65 tetramer-binding}

CD8+_{T cells were sorted to high purity and their clonal composition was determined by}

next-generation sequencing. Clones were identified by their unique TCRß sequences. In agreement with our earlier observations (8) the frequencies of IL-7Rα+_{were very}

low at the early time point, but just after the peak of viral replication enough cells could be isolated to reliably compare the clonal compositions of the IL-7Rα+_{and IL-7Rαˉ}

hCMV-specific cell subsets (figure 2A and C). In both patients multiple hCMV-pp65-specific clones were identified that carried diverse Vß- and Jß-genes (supplemental table 1 and 2). In patient 1, the IL-7Rαˉ subset at the peak of the primary infection consisted of a single clone that had a frequency of only 6.2% in the IL-7Rα+_population

(figure 2B), whereas the majority of the clones in the IL-7Rα+_{subset were unique. In}

patient 2 (figure 2D) there were 3 clones that were identical between the IL-7Rα+_and

IL-7Rαˉ hCMV-specific CD8+_{T cells, however both subsets also contained many}

non-overlapping clones. These results showed that during the acute phase, IL-7Rα expression characterized overlapping but not identical hCMV-specific CD8+_{T-cell pools.}

(11)

+

T CELL RESPONSES

6

We reasoned that if, after the primary effector phase, the IL-7Rα+_{clones would}

indeed feed the ‘latent-phase’ hCMV-specific CD8+_{T-cell population, the IL-7Rα}+_and

IL-7Rαˉ hCMV-specific CD8+_{T-cell pools at latency would be enriched for clones found}

in the acute phase IL-7Rα-expressing CD8+_{T cells. We therefore studied the IL-7Rα}+

and IL-7Rαˉ pool 1 year after primary infection (figure 2B and 2D). In patient 1, only 53% and 26% of the sequences of IL-7Rα+_{and IL-7Rαˉ}_{subset respectively at the 1 year}

time point were attributable to clones found in the IL-7Rα+ _{subset at the acute phase.}

In patient 2, this was 7% and 80%, respectively, however the 80% overlap in the IL-7Rα+

Table 1: patient characteristics

patient age HLA-I typing

donor hCMV serostatus

immunosuppressive regimen*

1 23 A1A3 B7B35 Primary hCMV positive CsA/P/MMF

2 18 A2A3 B7B58 Primary hCMV positive CsA/P/MMF/CD25 mAb

3 58 A1A24 B35B44 Reactivation negative CsA/P/MMF/CD25 mAb

4 65 A2A24 B44B39 Reactivation or

superinfection

positive CsA/P/MMF/CD25 mAb

superinfection

positive CsA/P/MMF

superinfection

positive CsA/P/MMF/CD25 mAb

7 41 A2A26 B35 CsA/P/MMF/CD25 mAb

8 49 A3A24 B7B35 CsA/P/MMF/CD25 mAb

9 45 A2A32 B27B35 CsA/P/MMF/CD25 mAb

*: CsA=cyclosporine A; P=Prednisolon; MMF=mycophenolate mofetil, CD25mAb=Basilixumab induction therapy

Table 2: tetrameric-complexes used

Name HLA Virus Protein Peptide

hCMV-pp65-YSE HLA-A*0101 hCMV pp65 YSEHPTFTSQY

hCMV-pp65-NLV HLA-A*0201 hCMV pp66 NLVPMVATV

hCMV-pp65-TPR HLA-B*0702 hCMV pp65 TPRVTGGGAM

hCMV-IE-QIK HLA-B*0801 hCMV IE-1 QIKVRVKMV

hCMV-IE-VLE HLA-A*0201 hCMV IE-1 VLEETSVML

EBV-BZLF-EPL HLA-B*3501 EBV BZLF-1 EPLPQGQLTAY

EBV-BMLF-GLC HLA-A*0201 EBV BMLF-1 GLCTLVAML

EBV-EBNA-HPV HLA-B*3501 EBV EBNA-1 HPVGEADYFEY

EBV-EBNA-RPP HLA-B*0702 EBV EBNA-3a RPPIFIRRL

(12)

+

T CELL RESPONSES

6

subset was caused by 3 clones that were also found in the IL-7Rαˉsubset from the acute phase. It should be noted that the diversity of the total hCMV-specific CD8+_T-cell

pool remained comparable to that found during the acute response but that both the composition and hierarchy of the clones in the IL-7Rαˉ pools had changed. This latter

PMA / Ionomycin cognate peptide (1 µg/ml )

E * * IFNγ IL-2 TNFα IL-2 IFNγ TNFα

IL-7Rα+ _{hCMV-specific CD8 T cells}

IL-7Rα¯ hCMV-specific CD8 T cells Total hCMV-specific CD8 T cells

B % E om es % T -bet C * * % CCR7 % C X3 CR1 % CCR5 % CX CR3 D ** * ** % gr anz y m e A % gr anz y m e K % gr anz y m e B % per for in % A * ** * % CD2 7 % CD2 8 % K L R G 1 % I L-7R α + IFNγ IL-2 TNFα IL-2 IFNγ TNFα %

FIGURE 1. Flowcytometric analysis of total (black), IL-7Rα-expressing (green) and IL-7Rαˉ (blue)

hCMV-specific CD8+_{T cells. Percentage of (A) CD27, CD28 and KLRG1; (B) intracellular Eomes}

and T-Bet; (C) CCR7, CX3CR1, CCR5 and CXCR3; (D) intracellular granzyme A, granzyme K,

granzyme B and perforin; (E) percentage of cytokine-producing hCMV-specific CD8+_{T cells after}

PMA/Ionomycin stimulation for 4 hours (left) or cognate peptide for 6 hours (1ug/ml) (right). Statistical analysis with the paired student T test: *p=<0.05; **p=<0.01.

(13)

CLONAL EVOLUTION OF VIRAL CD8 + T CELL RESPONSES

6

A C B D

hCMV load (copies/ml) IL-7Rα¯ IL-7Rα+

Overlapping clone at peak Unique IL-7Rα+_clones

New clones, overlapping at 1 year New clones, no overlap at 1 year

Overlapping clones at peak Unique IL-7Rα+_clones

Unique IL-7Rα¯ clones New clones, overlapping at 1 year New clones, no overlap at 1 year

0 20 40 60 0 50000 100000 150000 0 20000 40000 60000hC M V load (c opi es /m l) A bs ol ut e nr of hC M V -pp65 CD8 + T c e lls

Weeks after transplantation

1:TPR 0 20 40 60 0 10000 20000 30000 40000 0 10000 20000 30000 40000 hC M V load (c opi es /m l) A bs ol ut e nr of hC M V -pp65 CD8 + T c e lls

Weeks after transplantation

2:TPR 0 20 40 60 80 100 0 20 40 60 80 100 0 20 40 60 80 100 0 20 40 60 80 100 peak 1 year % of t ot al c lones % of t ot al c lones 1:TPR IL-7Rα¯ IL-7Rα+ 0 20 40 60 80 100 0 20 40 60 80 100 0 20 40 60 80 100 0 20 40 60 80 100 peak 1 year % of t ot al c lones % of t ot al c lones 2:TPR IL-7Rα¯ IL-7Rα+

FIGURE 2. Longitudinal analysis of IL-7Rα-expressing and IL-7Rαˉ hCMV-specific CD8+_{T cells}

following primary hCMV (left: patient 1, right patient 2). (A and C) Absolute number of

IL-7Rα-expressing (green) and IL-7Rαˉ (blue) hCMV-pp65-TPR-specific CD8+_{T cells in two patients}

experiencing a primary hCMV infection (hCMV load (copies/ml) in grey)(B and D) TCRβ repertoire

analysis of IL-7Rα-expressing (upper panel) and IL-7Rαˉ (lower panel) hCMV-pp65-TPR-specific

CD8+_{T cells just after the peak viral load and at 1 year after transplantation of same two patients.}

Orange tones represent clones that are found in both IL-7Rα+ and IL-7Rαˉ hCMV-specific pools,

green tones represent clones that are only found in IL-7Rα+ and blue tones represent clones

that are only found in IL-7Rαˉ hCMV-specific pools. Red tones represent overlapping IL-7Rα+_and

IL-7Rαˉ clones only found at one year and black tones represent clones only found at one year

without any overlap. Different colors in either tone represent a different clone. Identical tones within one patient represent an identical clone. No overlapping clones between patients were found. (Sequences can be found in supplemental tables 1 and 2).

(14)

+

T CELL RESPONSES

6

observation appears to be in contrast with our earlier findings that showed stabile clonal composition from the acute response until five years of follow-up. First, for this study we used different patients and we already noted a large variability in immune responses between individuals (4). Perhaps more importantly, in the current study we used samples that were fairly close to the peak of viral replication, which could have resulted in monitoring of immune repertoires in an early stage of clonal competition.

Irrespective of these considerations, these data suggested that acute phase circulating IL-7Rα+_{hCMV-specific CD8}+_{T cells did not form the precursor pool for the}

IL-7Rαˉ hCMV-specific CD8+_{T cells.}

Lymph nodes contain unique virus-specific CD8

+

_{T-cell clones}

LN hCMV-specific CD8+_{T cells have a (central-)memory phenotype (9). As we could}

not detect the precursor pool of effector-type hCMV-specific CD8+_{T cells in PB,}

we next investigated the possibility that those precursors reside in the lymphoid compartment (34). To this end, we sorted and analyzed total IE-1- and hCMV-pp65-specific CD8+_{T cells obtained from paired LN and PB from 7 hCMV-seropositive}

renal-transplant recipients before transplantation (table 3).

A large variance in the clonal breadth of both LN and PB hCMV-specific CD8+_{T cells}

was observed and both mono- and oligoclonal responses were found within one individual (figure 3A and B; pt5:YSE and pt5:QIK) or against one epitope (pt3:YSE and pt5:YSE).

When analyzing the relationship between the hCMV-specific CD8+_{T cell clones of}

the paired samples, again a large variability was observed. In some cases a monoclonal response was found that was identical in both compartments (pt6:QIK and pt7:VLE), while in other cases highly expanded and unique clones were found in the PB (pt3:YSE and pt5:QIK) and surprisingly also in the LN (pt3:YSE, pt4:NLV, pt5:QIK).

These patterns were not unique for hCMV-specific responses as the same variation in breadth and overlap was found for EBV- and influenza-specific CD8+_{T cells (figure 3C-E).}

Thus, long after primary infection LNs often contained unique virus-specific clones.

Upon viral reactivation, the majority of the unique

LN hCMV-specific CD8

+

_{T cell clones are not recruited}

to the circulating pool

We next studied the clonal relationships between hCMV-specific CD8+_{T cells from}

pre-transplantation LN samples to those from PB samples taken at the moment of viral replication in 4 hCMV-seropositive patients who experienced a reactivation (or superinfection) after kidney transplantation (patients 3-6) (figure 4-5 and table 1). In these patients, 5 hCMV-epitope-specific CD8+_{T-cell pools could be studied longitudinally.}

Although the pre transplant overlapping clones (in orange) most likely have a common ancestor, the fact that they are identical does not allow for the analysis of recruitment

(15)

+

T CELL RESPONSES

6

of LN clones towards the PB upon viral reactivation. Therefore we focused on the discrepant clones between LN and PB as well as any new clone detected during or after reactivation. In only 1 out of 5 hCMV-specific CD8+_{T cell pools studied, 2 of the}

4 unique clones detected in the LN compartment before transplantation were found at the time of viral replication (figure 5: pt3:YSE, blue clones). Interestingly, we observed a dichotomy between the responses: if the initial hCMV-specific CD8+_{T-cell pool was}

monoclonal, it largely remained so during the antigenic rechallenge (figure 5 pt4:NLV, pt6:QIK and pt5:YSE), while responses that were initially broad led to even broader

Table 2: Materials and sorts

patient tetramer time-point sort

cells

IL-7Rαˉ IL-7Rα+

1 hCMV-pp65-TPR just after peak viral replication 59400 2590

1 year after transplantation 225000 14140

2 hCMV-pp65-TPR just after peak viral replication 6000 770

1 year after transplantation 11100 3400

PB LN

3 hCMV-pp65-YSE pre-transplantation 7400 2700

peak CD8-expansion after reactivation 22000

1 year after transplantation 11000

EBV-BZLF-EPL pre-transplantation 9600 7500

4 hCMV-pp65-NLV pre-transplantation 5700 600

peak CD8-expansion after reactivation 1 12400

peak CD8-expansion after reactivation 2 46900

EBV-BMLF-GLC pre-transplantation 2350 550

FLU-MP-GIL pre-transplantation 500 490

5 hCMV-IE-QIK pre-transplantation 61600 1400

hCMV-pp65-YSE pre-transplantation 14600 500

6 hCMV-IE-QIK pre-transplantation 20500 530

7 hCMV-IE-VLE pre-transplantation 7500 2150

8 EBV-EBNA-RPP pre-transplantation 1400 2000

EBV-EBNA-HPV pre-transplantation 1600 1600

(16)

6

hCMV IE % of t ot al c lones 0 20 40 60 80 100 PB LN 5: QIK % of t ot al c lones 0 20 40 60 80 100 PB LN 6: QIK % of t ot al c lones 0 20 40 60 80 100 PB LN 7: VLE B hCMV pp65 0 20 40 60 80 100 4: NLV PB LN % of t ot al c lones % of t ot al c lones 0 20 40 60 80 100 PB LN 5: YSE 0 20 40 60 80 100 3: YSE PB LN % of t ot al c lones A EBV lytic % of t ot al c lones 0 20 40 60 80 100 PB LN 4: GLC % of t ot al c lones 0 20 40 60 80 100 PB LN 3: EPL C EBV latent % of t ot al c lones 0 20 40 60 80 100 PB LN 8: RPP % of t ot al c lones 0 20 40 60 80 100 PB LN 8: HPV D Influenza A Virus % of t ot al c lones 0 20 40 60 80 100 PB LN 4: GIL % of t ot al c lones 0 20 40 60 80 100 PB LN 9: GIL E

overlapping PB and LN clones unique PB clones unique LN clones

responses during reactivation, accompanied by the appearance of many new clones (pt5:QIK and pt3:YSE) (red clones). Both type of reponses can occur within one patient (pt5:YSE and pt5:QIK), and they also do not seem to depend on the epitope analysed (pt3:YSE and pt5:YSE).

Overall these data suggested that neither the PB IL-7Rα-expressing hCMV-specific CD8+_{T cells with ‘classical’ memory-like features, nor the LN-derived hCMV-specific}

CD8+_{T cells with (central-) memory-like features represented a major precursor-pool}

for the circulating, effector-type hCMV-specific CD8+_{T cell population.}

FIGURE 3. TCRβ repertoire analysis of paired PB and LN samples. Orange tones represent clones

that are found in both PB and LN, green tones represent clones that are only found in PB and blue tones represent clones that are only found in LN. Different colors in either tone represent a different clone. Identical tones within one epitope/patient represent an identical clone. No overlapping clones between different epitopes or patients were found. (A) Analysis of hCMV-pp65-specific

CD8+_{T cells: patient 3 specific (3:YSE); patient 4 NLV-specific (4:NLV) and patient 5}

YSE-specific (5:YSE), (B) Analysis of hCMV-IE-YSE-specific CD8+_{T cells: patient 5 QIK-specific (5:QIK); patient}

6 QIK-specific (6:QIK) and patient 7 VLE-specific (7:VLE), (C) Analysis of lytic EBV-specific CD8+_T

cells: patient 3 BZLF-1-EPL-specific (3:EPL) and patient 4 BMLF-1-GLC-specific (4:GLC), (D) Analysis

of latent EBV-specific CD8+_{T cells: patient 3 EBNA-3a-EPL-specific (3:EPL) and patient 8}

ebna-1-HPV-specific (8:HPV), (E) Analysis of Influenza A-specific CD8+_{T cells: patient 9 GIL-specific (9:GIL)}

(17)

6

0 ₂₀ ₄₀ ₆₀ 0 10000 20000 30000 40000 0 10000 20000 30000 40000 50000 0 20 40 60 80 100 LN 0 ₂₀ ₄₀ ₆₀0 10000 20000 30000 40000 50000 5: QIK hCMV load (copies/ml) HLA-DR+_CD38+_(activated) CCR7+ hCMV load (copies/ml)

Absolute nr of hCMV-specific CD8 T cells

Phenotype Absolute nr 0 ₂₀ ₄₀ ₆₀ 0 5000 10000 15000 20000 25000 0 5000 10000 15000 20000 0 20 40 60 0 5000 10000 15000 20000 0 20 40 60 80 100 LN 3: YSE A 0 ₂₀ ₄₀ ₆₀ 0 5000 10000 15000 20000 25000 0 100000 200000 300000 0 20 40 60 80 100 LN 0 ₂₀ ₄₀ ₆₀0 100000 200000 300000 4: NLV B 0 ₂₀ ₄₀ ₆₀ 0 5000 10000 15000 20000 25000 0 10000 20000 30000 40000 50000 0 20 40 60 80 100 LN 0 ₂₀ ₄₀ ₆₀0 10000 20000 30000 40000 50000 5: YSE C D 0 ₂₀ ₄₀ ₆₀ 0 5000 10000 15000 20000 25000 0 1000 2000 3000 4000 0 20 40 60 80 100 L N 0 20 40 60 0 1000 2000 3000 4000 6: QIK E A bs nr of hC M V -I E -s pec if ic C D 8 T c el ls A bs nr of hC M V -pp65-s pec if ic C D 8 T c el ls hC M V load (c opi es /m l) hC M V load (c opi es /m l) % ex pr es s ion

Weeks after transplantation Weeks after transplantation CX3CR1

CD27¯

FIGURE 4. Flowcytometric longitudinal analysis of PB hCMV-specific CD8+_{T cells. Left panels:}

absolute number of hCMV-specific CD8+_{T cells (blue, left y-axis) and hCMV load (gray shades, right}

y-axis). Right panels: hCMV load (gray shades, right y-axis) and percentage of CCR7-expressing

(green), activated (CD38+_HLA-DR+_{; red), CD27-negative (orange) and CX}

3CR1-expressing cells

(black)(all on the left y-axis) within the total hCMV-specific CD8+_{T cell population analyzed. (A)}

Analysis of CMV-pp65-YSE-specific CD8+_{T cells in patient 3, (B) hCMV-pp65-NLV-specific CD8}+_T

cells in patient 4, (C) hCMV-pp65-YSE-specific CD8+_{T cells in patient 5, (D) hCMV-IE-QIK-specific}

(18)

6

5: YSE 0 20 40 60 80 100 LN Pre TX % o f t o ta l c lo n e s 0 20 40 60 80 100 PB % of total clones reactivation PB % of total clones C 6: QIK 0 20 40 60 80 100 LN Pre TX % of total clones PB reactivation % of total clones 0 20 40 60 80 100 PB % of total clones 1 year PB % of total clones B 0 20 40 60 80 100 PB 3: YSE LN Pre TX % o f t o ta l c lo n e s % o f t o ta l c lo n e s 0 20 40 60 80 100 reactivation PB % of total clones E 5: QIK LN Pre TX % o f t o ta l c lo n e s 0 20 40 60 80 100 PB % of total clones reactivation PB % of total clones D 4: NLV 0 20 40 60 80 100 LN Pre TX % o f t o ta l c lo n e s 0 20 40 60 80 100 PB % of total clones reactivation 1 PB % of total clones 0 20 40 60 80 100reactivation 2 PB % of total clones 0 20 40 60 80 100 1 year PB % of total clones A new clones PB / LN derived clones LN derived clones PB derived clones 0 20 40 60 80 100 1 year % of total clones PB

FIGURE 5. Longitudinal TCRβ repertoire analysis of five patients experiencing hCMV reactivation

after transplantation. Orange tones represent clones that were found in both pre-transplantation PB and LN, green tones represent clones unique pre-TX PB clones and blue tones represent clones unique pre-TX LN clones. Red tones represent clones that could only be detected at the peak of CD8 expansion after hCMV reactivation and 1 year after. Different colors in either tone represent a different clone. Identical tones within one epitope/patient represent an identical clone. No overlapping clones between different epitopes or patients were found. (A) Analysis of

hCMV-pp65-NLV-specific CD8+_{T cells of patient 4 (4:NLV), (B) hCMV-IE-QIK-specific CD8}+_{T cells}

of patient 6 (6:QIK), (C) hCMV-pp65-YSE-specific CD8+_{T cells of patient 5 (5:YSE), (D)}

hCMV-IE-QIK-specific CD8+_{T cells of patient 5 (5:QIK) and (E) hCMV-pp65-YSE-specific CD8}+_{T cells of}

(19)

+

T CELL RESPONSES

6 DISCUSSION

Although IL-7R

α

+_{hCMV-specific CD8}+_{T cells display several typical memory cell}

features we did not find strong evidence that there is recruitment from the acute phase IL-7R

α

+_{to the latency phase IL-7R}

α

_{ˉ hCMV-specific CD8}+_{T-cell pool. There are}

several putative explanations for the difference between our findings on the human virus-specific T-cell pools and data obtained in mice. First, the initial experiments in mice were performed during and following acute viral infections (e.g. LCMV) and not during persistent infection as hCMV. Second, the IL-7R

α

+_{cells during and after acute}

viral infection in mice bear some features that hCMV-specific CD8+_{memory T cells do}

not have. For example, it has been suggested that only IL-7R

α

+_{cells lacking KLRG1}

expression are proper memory precursor cells (18, 19, 35). Although we showed that the IL-7R

α

+_{subset contained cells lacking KLRG1 expression the percentage of KLRG1}

negative IL-7R

α

+_{hCMV-specific CD8}+_{T cells was too low obtain the required amount}

of cells to perform TCR

β

repertoire analysis. Third, the IL-7R

α

+_{populations that were}

used for the murine studies (17) were isolated from the spleen and could thus be quite different from the circulating IL-7R

α

+_{cells that we analyzed. Indeed, several studies}

including those from our own group have shown that phenotype and function of (virus-specific) CD8+_{T cells can differ greatly depending on the tissue studied (9, 28, 36-39).}

We also found that during hCMV reactivation, recruitment of LN derived hCMV-specific CD8+_{T cells towards the PB compartment appears to be a rare event, being}

detectable in only one of the five analyzed hCMV-specific CD8+_{T-cell populations.}

This analysis however has a number of potential caveats. First, based on the fact that HCMV is a systemic infection, we assume that by sampling a number of para-iliac LN we sample the full LN compartment in terms of clonal representation. Indeed, at least part of these cells expresses CCR7 (9), and they therefore have the capacity to circulate from LN to LN in search of their cognate antigen. Further, in the context of secondary lymphoid organs it is worthwhile to analyze if spleen and bone marrow hCMV-specific CD8+_{T cells may serve as a source of circulating effector-type cells (28, 36, 40).}

Second, three out of four patients received a graft from a hCMV seropositive donor making it impossible to conclude whether the T-cell response was triggered by superinfection or by reactivation of the endogenous virus. Although the epitopes studied are immunodominant and have not yet been reported to vary between different strains of hCMV, it is feasible that the route of reactivation versus superinfection by the transplanted kidney plays a role too. Only in one patient reactivation of the latent endogenous virus did occur for certain, and it is in this patient that we detected unique LN clones in the PB upon reactivation.

Third, we cannot distinguish between LN-derived and PB-derived Il-7Rαˉ and IL-7Rα+

hCMV-specific CD8+_{T-cell clones where it concerns overlapping clones. Thus we are}

unable to exclude that overlapping clones are indeed maintained by recruitment from the LN or from the IL-7Rα-expressing hCMV-specific CD8+_{T-cell pool. Furthermore,}

(20)

+

T CELL RESPONSES

6

earlier studies have shown that effector-type human CD8+_{T cells are long lived (5) and}

it cannot be excluded that the effector-type pool is, at least partially, maintained by homeostatic proliferation. In this respect, it is of interest that Tesselaar et al have shown that, different from mice, for the maintenance of the human naïve pool, homeostatic proliferation is far more important than influx of new cells from the thymus (41).

Finally, one can never analyze all virus-specific CD8+_{T cells clones in a certain}

compartment due to limitations in available materials and the techniques used. Although at least 10*10e_{6 MNC were used for each analysis and when analyzing overlap, all}

clones, including those that comprised less than 1% of the total reads were considered, it is possible that extremely rare virus-specific clones were missed in the analysis.

Whether a cell will become a memory cell or an effector cell has been shown to depend on many factors including the amount of antigen, cytokine environment, costimulation and TCR affinity (2, 42-45). However it has been shown that one clone can give rise to both acute phase short-lived effector cells and memory precursor cells (46) and that this balance is altered by the kind of infection, route of infection, tissue specific events and other factors (46-48). Price et al (44) showed that during latency, EBV- and hCMV-specific CD8+_{T cells with low-affinity TCRs have a}

memory-type phenomemory-type whereas the high-affinity TCR-expressing cells have an effector-memory-type phenotype. However, Griffiths et al. recently reported that especially in the elderly, the effector-type cells actually have a lower affinity for the MHC-peptide complex (49). Our preliminary analyses yielded no apparent evidence for different affinities between the total IL-7Rα+_{and IL-7Rαˉ hCMV-specific CD8}+_{T-cell pools since tetramer binding was}

indistinguishable between both subsets (supplemental figure 4). However, it cannot formally be excluded that unique IL-7Rα+_{clones as well as unique LN clones indeed}

have a different affinity when compared to unique IL-7Rαˉ PB clones.

Thus, although we show that the circulating IL-7Rα+_{hCMV-specific CD8}+_T-cell

pool contains memory-phenotype cells that lack typical effector features, no prove could be found that either this population or the LN-residing (central-) memory hCMV-specific CD8+_{T cells harbor the precursors for the terminally differentiated effector PB}

hCMV-specific CD8+_{T-cell pool.}

ACkNOwLEGDEMENTS

We thank B. Hooibrink for sorting hCMV-specific cells and drs K.A.M.I. van der Pant and N. van der Weerd for providing clinical data.

(21)

+

T CELL RESPONSES

6

1. Gamadia LE, ten Berge IJ, Picker LJ, and van Lier RA. Skewed maturation of virus-specific CTLs? NatImmunol. 2002;3(3):203.

2. Kaech SM, Wherry EJ, and Ahmed R. Effector and memory T-cell differentiation: implications for vaccine development.

NatRevImmunol. 2002;2(4):251-62.

3. Zimmerman C, Brduscha-Riem K, Blaser C, Zinkernagel RM, and Pircher H. Visualization, characterization, and turnover of CD8+ memory T cells in virus-infected hosts. JExpMed. 1996;183(4):1367-75. 4. Klarenbeek PL, Remmerswaal EB, ten

Berge IJ, Doorenspleet ME, van Schaik BD, Esveldt RE, Koch SD, ten Brinke A, van Kampen AH, Bemelman FJ, et al. Deep sequencing of antiviral T-cell responses to HCMV and EBV in humans reveals a stable repertoire that is maintained for many years. PLoS pathogens. 2012;8(9):e1002889.

5. Beverley PC. Kinetics and clonality of immunological memory in humans. Seminars

in immunology. 2004;16(5):315-21.

6. Snyder CM. Buffered memory: a hypothesis for the maintenance of functional, virus-specific CD8(+) T cells during cytomegalovirus infection.

ImmunolRes. 2011;51(2-3):195-204.

7. Snyder CM, Cho KS, Bonnett EL, van Dommelen S, Shellam GR, and Hill AB. Memory inflation during chronic viral infection is maintained by continuous production of short-lived, functional T cells. Immunity. 2008;29(4):650-9.

8. van Leeuwen EM, de Bree GJ, Remmerswaal EB, Yong SL, Tesselaar K, ten Berge IJ, and van Lier RA. IL-7 receptor alpha chain expression distinguishes functional subsets of virus-specific human CD8+ T cells. Blood. 2005;106(6):2091-8. 9. Remmerswaal EB, Havenith SH, Idu MM,

van Leeuwen EM, van Donselaar KA, ten Brinke A, Bom-Baylon N, Bemelman FJ, van Lier RA, and ten Berge IJ. Human virus-specific effector-type T cells accumulate in blood but not in lymph nodes. Blood. 2012;119(7):1702-12.

10. Boom R, Sol C, Weel J, Gerrits Y, de Boer M, and Wertheim-van Dillen P. A highly sensitive assay for detection and quantitation of human cytomegalovirus

DNA in serum and plasma by PCR and electrochemiluminescence. Journal of

clinical microbiology. 1999;37(5):1489-97.

11. Voehringer D, Koschella M, and Pircher H. Lack of proliferative capacity of human effector and memory T cells expressing killer cell lectinlike receptor G1 (KLRG1).

Blood. 2002;100(10):3698-702.

12. Lamoreaux L, Roederer M, and Koup R. Intracellular cytokine optimization and standard operating procedure. NatProtoc. 2006;1(3):1507-16.

13. van Dongen JJ, Langerak AW, Bruggemann M, Evans PA, Hummel M, Lavender FL, Delabesse E, Davi F, Schuuring E, Garcia-Sanz R, et al. Design and standardization of PCR primers and protocols for detection of clonal immunoglobulin and T-cell receptor gene recombinations in suspect lymphoproliferations: report of the BIOMED-2 Concerted Action BMH4-CT98-3936. Leukemia. 2003;17(12):2257-317.

14. Folch G, and Lefranc MP. The human T cell receptor beta variable (TRBV) genes.

ExpClinImmunogenet. 2000;17(1):42-54.

15. Klarenbeek PL, de Hair MJ, Doorenspleet ME, van Schaik BD, Esveldt RE, van de Sande MG, Cantaert T, Gerlag DM, Baeten D, van Kampen AH, et al. Inflamed target tissue provides a specific niche for highly expanded T-cell clones in early human autoimmune disease. Annals of the

rheumatic diseases. 2012;71(6):1088-93.

16. Klarenbeek PL, Tak PP, van Schaik BD, Zwinderman AH, Jakobs ME, Zhang Z, van Kampen AH, van Lier RA, Baas F, and de Vries N. Human T-cell memory consists mainly of unexpanded clones.

ImmunolLett. 2010;133(1):42-8.

17. Kaech SM, Tan JT, Wherry EJ, Konieczny BT, Surh CD, and Ahmed R. Selective expression of the interleukin 7 receptor identifies effector CD8 T cells that give rise to long-lived memory cells. NatImmunol. 2003;4(12):1191-8.

18. Joshi NS, Cui W, Chandele A, Lee HK, Urso DR, Hagman J, Gapin L, and Kaech SM. Inflammation directs memory precursor and short-lived effector CD8(+) T cell fates via the graded expression of T-bet transcription factor. Immunity. 2007;27(2):281-95.

(22)

+

T CELL RESPONSES

6

19. Sarkar S, Kalia V, Haining WN, Konieczny BT, Subramaniam S, and Ahmed R. Functional and genomic profiling of effector CD8 T cell subsets with distinct memory fates.

JExpMed. 2008;205(3):625-40.

20. Banerjee A, Gordon SM, Intlekofer AM, Paley MA, Mooney EC, Lindsten T, Wherry EJ, and Reiner SL. Cutting edge: The transcription factor eomesodermin enables CD8+ T cells to compete for the memory cell niche. JImmunol. 2010;185(9):4988-92.

21. Intlekofer AM, Takemoto N, Wherry EJ, Longworth SA, Northrup JT, Palanivel VR, Mullen AC, Gasink CR, Kaech SM, Miller JD, et al. Effector and memory CD8+ T cell fate coupled by T-bet and eomesodermin.

NatImmunol. 2005;6(12):1236-44.

22. Bolovan-Fritts CA, Trout RN, and Spector SA. High T-cell response to human cytomegalovirus induces chemokine-mediated endothelial cell damage. Blood. 2007;110(6):1857-63.

23. Galkina E, Thatte J, Dabak V, Williams MB, Ley K, and Braciale TJ. Preferential migration of effector CD8+ T cells into the interstitium of the normal lung.

JClinInvest. 2005;115(12):3473-83.

24. Kohlmeier JE, Cookenham T, Miller SC, Roberts AD, Christensen JP, Thomsen AR, and Woodland DL. CXCR3 directs antigen-specific effector CD4+ T cell migration to the lung during parainfluenza virus infection. JImmunol. 2009;183(7):4378-84. 25. Kohlmeier JE, Miller SC, Smith J, Lu B,

Gerard C, Cookenham T, Roberts AD, and Woodland DL. The chemokine receptor CCR5 plays a key role in the early memory CD8+ T cell response to respiratory virus infections. Immunity. 2008;29(1):101-13. 26. Kunkel EJ, Boisvert J, Murphy K, Vierra

MA, Genovese MC, Wardlaw AJ, Greenberg HB, Hodge MR, Wu L, Butcher EC, et al. Expression of the chemokine receptors CCR4, CCR5, and CXCR3 by human tissue-infiltrating lymphocytes.

AmJPathol. 2002;160(1):347-55.

27. Liu L, Fuhlbrigge RC, Karibian K, Tian T, and Kupper TS. Dynamic programming of CD8+ T cell trafficking after live viral immunization.

Immunity. 2006;25(3):511-20.

28. Palendira U, Chinn R, Raza W, Piper K, Pratt G, Machado L, Bell A, Khan N, Hislop

AD, Steyn R, et al. Selective accumulation of virus-specific CD8+ T cells with unique homing phenotype within the human bone marrow. Blood. 2008;112(8):3293-302. 29. Chi C, Sun Q, Wang S, Zhang Z, Li X,

Cardona CJ, Jin Y, and Xing Z. Robust antiviral responses to enterovirus 71 infection in human intestinal epithelial cells. Virus Res. 2013.

30. Sauty A, Dziejman M, Taha RA, Iarossi AS, Neote K, Garcia-Zepeda EA, Hamid Q, and Luster AD. The T cell-specific CXC chemokines IP-10, Mig, and I-TAC are expressed by activated human bronchial epithelial cells. JImmunol. 1999;162(6):3549-58.

31. van de Berg PJ, Yong SL, Remmerswaal EB, van Lier RA, and ten Berge IJ. Cytomegalovirus-induced effector T cells cause endothelial cell damage.

ClinVaccine Immunol. 2012;19(5):772-9.

32. van Aalderen MC, Remmerswaal EB, ten Berge IJ, and van Lier RA. Blood and beyond: properties of circulating and tissue-resident human virus-specific alphabeta CD8(+) T cells. European journal

of immunology. 2014;44(4):934-44.

33. van Lier RA, ten Berge IJ, and Gamadia LE. Human CD8(+) T-cell differentiation in response to viruses. NatRevImmunol. 2003;3(12):931-9.

34. Torti N, Walton SM, Brocker T, Rulicke T, and Oxenius A. Non-hematopoietic cells in lymph nodes drive memory CD8 T cell inflation during murine cytomegalovirus infection.

PLoS pathogens. 2011;7(10):e1002313.

35. Rubinstein MP, Lind NA, Purton JF, Filippou P, Best JA, McGhee PA, Surh CD, and Goldrath AW. IL-7 and IL-15 differentially regulate CD8+ T-cell subsets during contraction of the immune response.

Blood. 2008;112(9):3704-12.

36. Letsch A, Knoedler M, Na IK, Kern F, Asemissen AM, Keilholz U, Loesch M, Thiel E, Volk HD, and Scheibenbogen C. CMV-specific central memory T cells reside in bone marrow. European journal

of immunology. 2007;37(11):3063-8.

37. Piet B, de Bree GJ, Smids-Dierdorp BS, van der Loos CM, Remmerswaal EB, von der Thusen JH, van Haarst JM, Eerenberg JP, ten Brinke A, van der Bij W, et al. CD8(+) T cells with an intraepithelial

(23)

+

T CELL RESPONSES

6

phenotype upregulate cytotoxic function upon influenza infection in human lung.

JClinInvest. 2011;121(6):2254-63.

38. Smolders J, Remmerswaal EB, Schuurman KG, Melief J, van Eden CG, van Lier RA, Huitinga I, and Hamann J. Characteristics of differentiated CD8(+) and CD4 (+) T cells present in the human brain. Acta

Neuropathol. 2013;126(4):525-35.

39. van Aalderen MC, Remmerswaal EB, Heutinck KM, ten Brinke A, Pircher H, van Lier RA, and ten Berge IJ. Phenotypic and functional characterization of circulating polyomavirus BK VP1-specific CD8+ T cells in healthy adults. Journal of virology. 2013;87(18):10263-72.

40. Langeveld M, Gamadia LE, and ten Berge IJ. T-lymphocyte subset distribution in human spleen. EurJClinInvest. 2006;36(4):250-6.

41. den Braber I, Mugwagwa T, Vrisekoop N, Westera L, Mogling R, de Boer AB, Willems N, Schrijver EH, Spierenburg G, Gaiser K, et al. Maintenance of peripheral naive T cells is sustained by thymus output in mice but not humans. Immunity. 2012;36(2):288-97.

42. Badovinac VP, Messingham KA, Jabbari A, Haring JS, and Harty JT. Accelerated CD8+ T-cell memory and prime-boost response after dendritic-cell vaccination.

NatMed. 2005;11(7):748-56.

43. Badovinac VP, Porter BB, and Harty JT. CD8+ T cell contraction is controlled by early inflammation. NatImmunol. 2004;5(8):809-17.

44. Price DA, Brenchley JM, Ruff LE, Betts MR, Hill BJ, Roederer M, Koup RA, Migueles SA, Gostick E, Wooldridge L, et al. Avidity for antigen shapes clonal dominance in CD8+ T cell populations specific for persistent DNA viruses. JExpMed. 2005;202(10):1349-61. 45. Sarkar S, Teichgraber V, Kalia V, Polley A,

Masopust D, Harrington LE, Ahmed R, and Wherry EJ. Strength of stimulus and clonal competition impact the rate of memory CD8 T cell differentiation. JImmunol. 2007;179(10):6704-14.

46. Plumlee CR, Sheridan BS, Cicek BB, and Lefrancois L. Environmental cues dictate the fate of individual CD8+ T cells responding to infection. Immunity. 2013;39(2):347-56.

47. Buchholz VR, Flossdorf M, Hensel I, Kretschmer L, Weissbrich B, Graf P, Verschoor A, Schiemann M, Hofer T, and Busch DH. Disparate individual fates compose robust CD8+ T cell immunity.

Science. 2013;340(6132):630-5.

48. Gerlach C, van Heijst JW, Swart E, Sie D, Armstrong N, Kerkhoven RM, Zehn D, Bevan MJ, Schepers K, and Schumacher TN. One naive T cell, multiple fates in CD8+ T cell differentiation. The Journal of experimental

medicine. 2010;207(6):1235-46.

49. Griffiths SJ, Riddell NE, Masters J, Libri V, Henson SM, Wertheimer A, Wallace D, Sims S, Rivino L, Larbi A, et al. Age-associated increase of low-avidity cytomegalovirus-specific CD8+ T cells that re-express CD45RA. JImmunol. 2013;190(11):5363-72.

(24)

6

IL7-Rα⁻ IL7-Rα+ Granzyme B Pe rf ori n Granzyme B Gr an zy me K Granzyme A CD27 T-bet Eo mes CCR7 CX3 CR 1 CD27 CD45RA CXCR3 CD95 CCR5 KL RG1 hCMV-specific CD8 T cells CD28 CCR5 CD27 CCR7 CX3CR1 CD8 tetramer IL-7Rα IL-7Rα KLRG1 CD28 KL RG1 CD95 CXCR3 Eomes T-bet Granzyme B Perforin CD27 Granzyme A Granzyme K Granzyme B CD45RA KLRG1 IL-7Rα¯ IL-7Rα+ FSC-A SSC -A FSC-H FSC -W SSC-H SSC -W CD3 CD8 tet ra mer CD8 total A B

FIGURE S1: (A) gating strategy for IL-7Rα+

and IL-7Rα⁻ hCMV-specific CD8+

T cells. (B) example of staining of one healthy donor. From top to bottom: CD27 vs CD45RA, CD28 vs KLRG1, CXCR3 vs CD95, CCR7 vs CX

3CR1, CCR5 vs KLRG1, T-bet vs eomes, granzyme B vs perforin, granzyme

A vs CD27 and granzyme B vs granzyme K. Right column total CD8+

T cells, second column total

hCMV-specific CD8+

T cells, third column IL-7Rα+

hCMV-specific CD8+

T cells, fourth column

IL-7Rα⁻ hCMV-specific CD8+

T cells, two most right columns overlay of IL-7Rα+

(green) and

IL-7Rα⁻ (blue) hCMV-specific CD8+

(25)

6

IL-7R α + IL-7R α⁻ medium PMA/Iono peptide IFNγ T NF α Tetramer+ TNFα+ IFNγ IFNγ IL -2 IL -2 medium PMA/Iono peptide CD3+ CD8 tet ram er Tetramer+ Tetramer+ A CD3 IL-7R α B IL-7R α CD3 CD3¯ CD3 IL-7R+ _α¯ _CD3+_IL-7Rα+ Before sort After sort C

FIGURE S2: A-B Sorting strategy of IL-7Rα+_CD3+_{, IL-7R}α_ˉCD3+_{and CD3ˉ cells. (A) Sample before}

sort. PBMC were stained with anti-CD3 PE-Cy7 and CD127 APC-Alexa Fluor 750. (B) Sample after sort. Left dotplot shows collected cells sorted from CD3ˉ gate, middle dotplot shows collected

cells from IL-7RαˉCD3+_{gate and right dotplot shows collected cells from the IL-7R}α+_CD3+_gate.

Representative data for 4 sorts. (C) example of staining after medium (top), PMA/Ionomycin

(middle) and cognate peptide stimulation (bottom) of IL-7Rα+_{(top half) and IL-7R}α⁻_{(lower half)}

hCMV-specific CD8+ _{T cells. Left column tetramer gating within total CD3+ cells, second column}

IFNγ vs TNFα, third column IFNγ vs IL-2 both within hCMV-specific CD8+_{T cells, last column}

(26)

+

T CELL RESPONSES

6

FIGURE S4. IL-7Rα+_{and IL-7R}α_{ˉ hCMV-specific CD8}+_{T cells have similar tetramer binding}

affinities. Dotplots of CD8+_{T cells are shown. The amount of tetramer compared to manufacturers}

instructions used is indicated above each dotplot. The last dotplot represents the CD8+_{T cell dot}

plot if no tetramer is added.

B

IL-7Rα CM V B 7 T PR

C

CM V B 7 T PR IL-7Rα FSC SSC SSC CD8 CM V B 7 T PR

A

IL-7Rα Before sort

After 1rst_sort After 2ndsort

IL-7R_α+ C M V-B7 T PR 2x 1x 0.5x 0.25x 0.1x 0.05x 0.025x

-FIGURE S3. Sorting strategy for IL-7Rα+_{and IL-7R}α_{ˉ hCMV-specific CD8}+_{T cells. (A) Gating of}

lymphocytes (left panel), CD8+_{cells (middle panel) and subsequent hCMV-specific CD8}+_{T cells}

(right panel). (B) Result of hCMV-specific CD8+_{T cell sort as shown in a; sorting gate for IL-7R}α+

and IL-7Rαˉ hCMV-specific CD8+_{T cells is shown. (C) Result of IL-7R}α+_{(left panel) and IL-7R}α_ˉ

(27)

+

T CELL RESPONSES

6

Supplemental table 1. Patient 1: hCMV-pp65 B7 TPR-specific CD8+_{T cell clones (primary hCMV)}

Vβ jβ CDR3 amino acid sequence

peak 1 year + - + -7.9 1.6 CASSTQDLQGARSPLHFG ND ND 1.4% 58.0% 10.3 1.2 CAISATGGDYGYTFG 2.5% ND ND ND 18 2.5 CASSPGKLDSEETQYFG 59.8% ND ND ND 21.1 1.2 CASSKDPPGTGGASGYTFG 2.1% ND ND ND 27 1.2 CASSLGPNNYYGYTFG 6.2% 96.6% 38.5% 21.6% 27 1.3 CASSVGPGNTIYFG ND ND 1.0% ND 27 1.4 CASSLGAAGRELFFG ND ND 5.1% 1.8% 27 1.5 CASSSSSWDRPQHFG 4.5% ND ND ND 27 1.6 CASSIGTANNSPLHFG ND ND 4.8% 1.1% 27 2.1 CASRIGGAPNNEQFFG 1.0% ND ND ND 27 2.1 CASGLGSSPMFNEQFFG 3.4% ND 13.5% 3.1% 27 2.1 CASRVGLAGNNEQFFG 4.2% ND ND ND 27 2.7 CASRTGTGSYRQYFG ND ND 1.1% ND 27 2.7 CASRTGTGNLYEQYFG 4.1% ND ND ND 27 2.7 CASSLGPASYEQYFG ND ND 21.7% 6.0% 27 2.7 CASSLGPATYEQYFG 2.6% ND ND ND 27 2.7 CASSLSPGQKYEQYFG 1.0% ND ND ND

Each row represents a different clone. Vβ = TCR-Vβ gene; Jβ = TCR-Jβ gene; - = IL-7Rαˉ;

(28)

+

T CELL RESPONSES

6

Supplemental table 2. Patient 2: hCMV-pp65 B7 TPR/specific CD8+_{T cell clones (primary hCMV)}

Vβ jβ CDR3 aminoacid sequence peak 1 year + - + -5.1 1.6 CASSLEGTGANSPLHFG ND ND 2.0% ND 5.6 1.2 CASSFQGADYGYTFG 2.7% ND ND ND 6.6 2.7 CASDSGVFSYEQYFG ND 1.1% ND ND 11.2 2.1 CASTETSGSSYNEQFFG ND ND 1.3% ND 18 1.2 CASSPRDSQPYGYTFG 9.3% ND ND ND 19 1.1 KNPTAFG 24.7% 16.6% ND 8.7% 19 2.2 CATAEGAGELFFG ND ND 17.0% ND 21.1 2.5 CASSKEDREKETQYFG 42.2% 40.5% 4.6% 15.8% 27 1.1 CASSLSVGVNTEAFFG 3.1% ND ND ND 27 1.1 CASSSGQEAFFG ND ND ND 1.6% 27 1.1 CASSLGPAGNTEAFFG ND 3.9% ND ND 27 1.2 CASSLSHSTGNYGYTFG ND ND 11.8% ND 27 2.1 CASRLGLAGGINEQFFG ND ND 13.4% 1.3% 27 2.1 CASTLGSAGPQFNEQFFG ND ND 14.0% 1.0% 27 2.3 CASALGASGITQYFG ND 2.5% ND ND 27 2.3 CASSLGSANTDTQYFG 7.4% ND ND ND 27 2.6 CVSSSGANVLTFG ND 1.8% ND 1.7% 27 2.7 CASSLSPSTGRVEQYFG ND ND 14.1% ND 27 2.7 CASSLGSTPYEQYFG ND ND 14.2% ND 27 2.7 CASSLSYSTLSYEQYFG ND 21.7% ND 5.4% 29.1 1.2 CSGGALGQDFGGYTFG 3.3% 3.2% ND 48.4% 29.1 1.2 CSVEDLRNYGYTFG ND ND ND 4.5%

Each row represents a different clone. Vβ = TCR-Vβ gene; Jβ = TCR-Jβ gene; - = IL-7Rαˉ;

(29)

+

T CELL RESPONSES

6

Supplemental table 3. Patient 3: hCMV-pp65 A1 YSE-specific CD8+_{T cells}

Vβ jβ CDR3 aminoacid sequence pre Tx _peak PB 1 yr PB PB LN 6.5 2.7 CASSYSPPIPGQGIDEQYFG 30.4% 0.06% 5.4% 10.0% 7.2 2.3 CASSSQRESSDTQYFG 21.2% 42.1% 0.4% 0.6% 4.3 2.3 CASSQWERDPQYFG 10.3% 13.9% 1.2% 2.4% 6.5 2.2 CASSSPRLAGGPSGTGELFFG 2.5% 0.3% 0.1% 2.8% 4.3 2.1 CASSQVLRVLSYNEQFFG 1.6% 12.5% 0.2% 0.4% 12.3 2.1 CASSSTLAGGPYEQFFG 0.9% 1.9% 0.2% 0.2% 15 2.1 CATSSRDTSSDEQFFG 5.1% ND 0.2% 0.02% 4.3 2.7 CASSQVLGVLSSYEQYFG 4.6% ND 0.02% 0.05% 15 2.1 CATSLLVRDNEQFFG 3.4% ND 0.02% ND 29.1 1.1 CSAPDRDTEAFFG 3.2% ND 0.04% 0.9% 6.2 2.3 CASSLLRDSPTDTQYFG 1.4% ND 11.3% 5.3% 6.2 2.2 CASTSWGLQGPTGELFFG 1.0% ND 0.3% 0.02% 15 2.7 CATSSPQGASDEQYFG 1.0% ND 1.1% 0.06% 18 2.1 CASSPDRGLNNEQFFG 0.3% ND 4.2% 1.1% 25.1 2.1 CASSEPSTANEQFFG 1.4% ND 0.01% ND 5.6 1.2 CASSLGQLYGYTFG 1.4% ND ND ND 18 2.2 CASTRADTGELFFG ND 15.5% 39.4% 9.0% 15 2.3 CATSSPMGSTDTQYFG ND 0.05% 2.6% 5.6% 15 2.5 CATAQNRLQETQYFG ND 7.4% ND ND 7.2 1.4 CASSPVGPWDEKLFFG ND 1.2% ND ND 29.1 2.2 CSVEGLAGTGELFFG ND ND 5.4% 0.8% 27 2.3 CASSPGSWGSTDTQYFG ND ND 4.4% 0.9% 15 2.1 CATSRPTSGSDEQFFG ND ND 2.5% 33.5% 29.1 2.5 CSADIQGAIGETQYFG ND ND 2.0% ND 29.1 1.1 CSAPGWDTEAFFG ND ND 2.0% 0.02% 29.1 1.1 CSVPTVNTEAFFG ND ND 1.4% 0.01% 10.3 2.3 CAISEPTSGRDTQYFG ND ND 1.3% 2.5% 27 2.1 CASMGLPGYEQFFG ND ND 1.3% ND 27 2.7 CASSVFG ND ND 1.2% 0.03% 29.1 1.1 CSVEPLWSPEAFFG ND ND ND 3.8% 27 2.7 CASSLEGDMDSEQYFG ND ND 0.5% 3.2% 27 2.3 CASSGGSSDTQYFG ND ND 0.2% 3.1% 29.1 2.3 CSAEDTSGITDTQYFG ND ND ND 2.2%

(30)

+

T CELL RESPONSES

6

Supplemental table 4: Patient 4; hCMV-pp65 A2 NLV-specific CD8+_{T cells}

Vβ jβ CDR3 aminoacid sequence pre Tx _{peak 1} PB peak 2 PB 1 year PB PB LN 10.3 1.3 CAISEWDIHSSGNTIYFG 91.8% 20.1% 99.0% 99.0% 99.0% 28 2.5 CASSPLFGREPDGETQYFG ND 75.8% ND ND ND

Each row represents a different clone. Vβ = TCR-Vβ gene; Jβ = TCR-Jβ gene; ND = not detected.

Supplemental table 5. Patient 5: hCMV-pp65 A1 YSE-specific CD8+_{T cells}

Vβ jβ CDR3 aminoacid sequence pre Tx peak PB LN PB 15 2.3 CATSRPLDSTDTQYFG 95.7% 85.1% 60.5% 27 2.3 CASSSPRESTDTQYFG 0.1% 1.1% 0.6% 29.1 1.2 CSVGEGANYGYTFG ND 3.1% ND 29.1 2.7 CSVGTGEDGEQYFG ND 2.1% ND 27 2.7 CASSLFLAGGYEQYFG ND 1.2% ND 29.1 2.1 CSGVVGRGAYNEQFFG ND 1.1% ND 10.3 2.3 CAISEPGTSTDTQYFG 0.1% ND 2.7% 7.8 1.1 CASSPGLGAEAFFG ND ND 12.0% 29.1 2.5 CSVWGETQYFG ND ND 6.6% 10.3 2.7 CAISESGSPEQYFG ND ND 5.4% 10.3 2.2 CAISGPGGPTGELFFG ND ND 2.0% 29.1 2.7 CSVTGGTYEQYFG ND ND 1.1%

(31)

+

T CELL RESPONSES

6

Supplemental table 6. Patient 5: hCMV-IE B8 QIK-specific CD8+_{T cells}

Vβ jβ CDR3 aminoacid sequence pre Tx _peak PB PB LN 7.3 2.2 CASSRLAGGTGELFFG 15.5% 0.3% 4.3% 7.3 2.3 CASSRLAGSTDTQYFG 1.0% 0.1% 0.5% 6.5 2.7 CASSLSRGAYEQYFG ND 86.4% ND 4.2 1.5 CASSPGGQGGQPQHFG ND 7.5% ND 29.1 2.2 CSVEESGLAGNTGELFFG 13.9% ND 4.0% 5.6 2.7 CASSLWGALYEQYFG 4.3% ND 0.5% 7.3 2.2 CASSRLGGGHRGAVFG 2.2% ND 0.2% 12.3 2.3 CASSLEGFTDTQYFG 1.3% ND 0.3% 18 1.5 CASSLGVSEPQHFG 28.3% ND ND 5.7 2.7 CASSSGTSGYEQYFG 15.9% ND ND 6.1 2.2 CASSESEQGYTGELFFG 2.4% ND ND 29.1 1.2 CSVGRDRDHGYTFG ND ND 40.7% 29.1 2.1 CSGVVGRGAYNEQFFG ND ND 9.6% 6.6 1.1 CASSYGPVETEAFFG ND ND 6.0% 29.1 2.1 CSVDRHSYNEQFFG ND ND 4.8% 27 1.2 CARNATGNYGYTFG ND ND 1.8% 27 2.7 CASRRRESHAYEQYFG ND ND 1.5% 5.5 2.1 CASSLGWEYNEQFFG ND ND 1.1% 29.1 1.2 CSGWGGTRDHGYTFG ND ND 1.0%

Supplemental table 7: Patient 6; hCMV-IE B8 QIK-specific CD8+_{T cells}

Vβ jβ CDR3 aminoacid sequence pre Tx peak 1 yr PB LN PB PB 5.1 2.2 CASSSREVTGELFFG 95.9% 97.2% 78.3% 81.7% 2 2.2 CASKWTSGGPNTGELFFG 0.4% 0.1% 9.9% 5.9% 7.8 2.7 CASSLTQGLAGVEQYFG 0.7% ND 0.7% 2.7% 2 1.1 CASSGGQGRTMNTEAFFG 0.1% ND 0.01% 1.8% 29.1 1.2 CSVRFSANYGYTFG ND ND 4.3% 1.3%

(32)

+

T CELL RESPONSES

6

Supplemental table 8: Patient 7; hCMV-IE A2 VLE-specific CD8+_{T cells}

Vβ jβ CDR3 aminoacid sequence

pre Tx

PB LN

29.1 2.7 CSVGQGATEQYFG 85.7% 84.3%

7.8 2.1 CASSVRQEPYNEQFFG 9.7% 10.6%

Supplemental table 9: Patient 3; EBV-bzlf-1 B35 EPL-specific CD8+_{T cells}

Vβ jβ CDR3 aminoacid sequence pre Tx PB LN 6.4 1.2 CASSDTPQPFSSGYTFG 57.4% 84.1% 6.4 1.2 CASSDVAPAPFYGYTFG 12.2% 4.6% 19 2.4 CASSIDLRGYQNIQYFG 13.5% 2.8% 6.6 2.7 CASQDTNSYEQYFG 4.2% ND 6.6 2.1 CASSPASGAFGYNEQFFG 3.6% ND

Supplemental table 10: Patient 4; EBV-bmlf-1 A2 GLC-specific CD8+_{T cells}

Vβ jβ CDR3 aminoacid sequence pre Tx PB LN 4.2 2.1 CASSQDGAGGLGEQFFG 20.6% 51.7% 20.1 1.2 CSARDKTGNGYTFG ND 41.0% 4.2 2.1 CASSQDGLAGGAFNEQFFG ND 2.3% 20.1 1.3 CSARSRVGNTIYFG 54.7% ND 20.1 1.2 CSARDSPGNGYTFG 8.5% ND 12.3 2.6 CASSPSDPGANVLTFG 2.3% ND 27 2.1 CASSSGTSGYYNEQFFG 1.1% ND 20.1 1.5 CSARDRPGGPSNQPQHFG 1.1% ND

Supplemental table 11: Patient 8; EBV-ebna-3a B7 RPP-specific CD8+_{T cells}

Vβ jβ CDR3 aminoacid sequence

pre Tx

PB LN

9 2.1 CASSLPSGGTNNEQFFG 93.2% 9.2%

20.1 2.5 CSARDQGEWGIKETQYFG ND 85.5%