Scholarly article on topic 'Long-Term Immune Reconstitution of Naive and Memory T Cell Pools after Haploidentical Hematopoietic Stem Cell Transplantation'

Long-Term Immune Reconstitution of Naive and Memory T Cell Pools after Haploidentical Hematopoietic Stem Cell Transplantation Academic research paper on "Biological sciences"

CC BY-NC-ND
0
0
Share paper
OECD Field of science
Keywords
{"Hematopoietic stem cell transplantation" / "Allogenic haploidentical donors" / "Immune reconstitution" / "T cell homeostasis"}

Abstract of research paper on Biological sciences, author of scientific article — Rita I. Azevedo, Maria V.D. Soares, Adriana S. Albuquerque, Rita Tendeiro, Rui S. Soares, et al.

Abstract Haploidentical hematopoietic stem cell transplantation (HSCT) constitutes an important alternative for patients lacking a human leukocyte antigen (HLA)-matched donor. Although the use of haploidentical donors is increasingly common, the long-term impact of generating a donor-derived immune system in the context of an HLA-mismatched thymic environment remains poorly characterized. We performed an in-depth assessment of immune reconstitution in a group of haploidentical HSCT recipients 4 to 6 years posttransplantation, in parallel with the respective parental donors and age-matched healthy control subjects. Our data show that the proportion of naive and memory subsets in the recipients, both within CD8+ and CD4+ T cells, more closely resembled that observed in age-matched control subjects than in the donors. HSCT recipients displayed relatively high signal-joint T cell–receptor excision circle levels and a high frequency of the recent thymic emigrant–enriched CD31+ subset within naive CD4+ and naive regulatory T cells. Moreover, CD8+, CD4+, and regulatory T cells from HSCT recipients displayed a diverse T cell repertoire. These results support a key role for thymic output in T cell reconstitution. Nevertheless, HSCT recipients had significantly shorter telomeres within a naive-enriched CD4+ T cell population than age-matched control subjects, despite the similar telomere length observed within the most differentiated CD8+ and CD4+ T cell subsets. Overall, our data suggest that long-term immune reconstitution was successfully achieved after haploidentical HSCT, a process that appears to have largely relied on de novo T cell production.

Academic research paper on topic "Long-Term Immune Reconstitution of Naive and Memory T Cell Pools after Haploidentical Hematopoietic Stem Cell Transplantation"

Long-Term Immune Reconstitution of Naive and Memory T Cell Pools after Haploidentical Hematopoietic Stem Cell Transplantation

Rita I. Azevedo1, Maria V.D. Soares1, *, Adriana S. Albuquerque1, Rita Tendeiro1, Rui S. Soares1, Miguel Martins 2, Dário Ligeiro 2, Rui M.M. Victorino1, Joao F. Lacerda13, Ana E. Sousa1

1 Instituto de Medicina Molecular, Faculdade de Medicina, Universidade de Lisboa, Lisbon, Portugal

2 Immunogenetics Laboratory, Centro de Histocompatibilidade do Sul—CHSul, Lisboa, Portugal

3 Hospital de Santa Maria-CHLN, Lisboa, Portugal

American Society for Blood and Marrow Transplantation

Article history: Received 12 October 2012 Accepted 19 January 2013

Key Words:

Hematopoietic stem cell transplantation

Allogenic haploidentical donors Immune reconstitution T cell homeostasis

ABSTRACT

Haploidentical hematopoietic stem cell transplantation (HSCT) constitutes an important alternative for patients lacking a human leukocyte antigen (HLA)-matched donor. Although the use of haploidentical donors is increasingly common, the long-term impact of generating a donor-derived immune system in the context of an HLA-mismatched thymic environment remains poorly characterized. We performed an in-depth assessment of immune reconstitution in a group of haploidentical HSCT recipients 4 to 6 years posttransplantation, in parallel with the respective parental donors and age-matched healthy control subjects. Our data show that the proportion of naive and memory subsets in the recipients, both within CD8+ and CD4+ T cells, more closely resembled that observed in age-matched control subjects than in the donors. HSCT recipients displayed relatively high signal-joint T cell—receptor excision circle levels and a high frequency of the recent thymic emigrant—enriched CD31+ subset within naive CD4+ and naive regulatory T cells. Moreover, CD8+, CD4+, and regulatory T cells from HSCT recipients displayed a diverse T cell repertoire. These results support a key role for thymic output in T cell reconstitution. Nevertheless, HSCT recipients had significantly shorter telomeres within a naive-enriched CD4+ T cell population than age-matched control subjects, despite the similar telomere length observed within the most differentiated CD8+ and CD4+ T cell subsets. Overall, our data suggest that long-term immune reconstitution was successfully achieved after haploidentical HSCT, a process that appears to have largely relied on de novo T cell production.

© 2013 American Society for Blood and Marrow Transplantation.

INTRODUCTION

Allogeneic hematopoietic stem cell transplantation (HSCT) has been increasingly performed in patients with severe hematologic and immunologic diseases [1]. Because most patients lack a human leukocyte antigen (HLA)-matched sibling, the use of volunteer unrelated donors is a common alternative [2]. The search for an unrelated HLA-matched donor is, however, a lengthy process, with variable success (10% to 70%) depending on patient ethnicity [3]. Haploidentical related donors (ie, first-degree relatives sharing 1 haplotype with the patient while being variably mismatched for the other HLA alleles) represent a good alternative for patients lacking an HLA-matched donor [2]. The advantages of haploidentical donors are their prompt availability for most patients and a potentially enhanced graft-versus-tumor effect [2,4]. Moreover, when several haploidentical donors are available, the most suitable donor can be chosen according to age, cytomegalovirus status, and donor-versus-recipient natural killer cell alloreactivity [5,6]. On the other hand, haploidentical HSCT has attendant risks associated with crossing the HLA barrier, namely graft failure, severe graft-versus-host disease (GVHD), and delayed immune reconstitution.

Financial disclosure: See Acknowledgments on page 711.

* Correspondence and reprint requests: Maria V.D. Soares, Unidade de Citometria de Fluxo, Instituto de Medicina Molecular, Av. Prof. Egas Moniz, 1649-028 Lisbon, Portugal.

E-mail address: msoares@fm.ul.pt (M.V.D. Soares).

1083-8791/$ — see front matter © 2013 American Society for Blood and Marrow http://dx.doi.org/10.1016/j.bbmt.2013.01.017

The outcome of haploidentical HSCT has improved significantly over the past 2 decades due to developments in conditioning regimens, graft engineering, and posttransplantation immune suppression. Reduced-intensity conditioning regimens have helped circumvent myeloa-blative regimen—related toxicity and mortality [4,7]. Graft failure and severe GVHD have been partly surmounted through the use of extensively T cell—depleted grafts [8-10], which leads to the generation of a donor-derived immune system wherein thymic T cell selection takes place in an HLA-mismatched environment [11]. We and others have shown that the infusion of purified CD34+ cells provides sufficient T and B cell depletion for haploidentical HSCT from related donors [10,12]. Unfortunately, this approach is still associated with delayed immune reconstitution and consequently also with substantial rates of infection-related mortality and leukemia relapse [13].

Although haploidentical HSCT is increasingly performed in patients who lack an HLA-matched donor, few studies have characterized immune reconstitution after this type of transplantation [7], particularly the long-term impact on the composition of naive and memory T cell pools. In the present study, we performed a detailed evaluation of the long-term immune reconstitution achieved by patients with acute myeloid leukemia or aplastic anemia who underwent hap-loidentical related HSCT after a chemotherapy-alone conditioning regimen [12]. These patients were, at the time of the study, 4 to 6 years posttransplantation and were analyzed in parallel with their respective donors, who were always 1 of Transplantation.

the parents, and age-matched healthy control subjects. We were particularly interested in investigating the relative contribution of thymic output and peripheral expansion of mature T cells to the process of T cell reconstitution in these patients.

METHODS

Patients and Sample Collection

A cross-sectional study was performed involving 5 haploidentical related HSCT recipients, 4 to 6 years posttransplantation. During a 62-month period, 15 patients (11 with acute myeloid leukemia, 3 with chronic myeloid leukemia, and 1 with aplastic anemia) were treated with an allogeneic stem cell transplantation from a fully haploidentical donor, because a related or unrelated HLA-matched donor was unavailable [12,14].

The conditioning regimen consisted of thiotepa 5 mg/kg/d on days -9 and -8, fludarabine 40 mg/m2/d on days -9 to -5, rabbit antithymocyte globulin 5 mg/kg/d on days -7to -3 as a 12- to 24-hour infusion, melphalan 60 mg/m2/d on days -4 and -3, cyclosporine 3 mg/kg/d on days -10 to -2, and prednisolone 2 mg/kg/d on days -7 to -3 [12,14]. The 14 high-risk leukemia patients received a median of 5.4 x 106 (range, 2.9 to 13.8 x 106) CD34+ cells/kg, 1.62 x 104 (range, .33 to 5.96 x 104) CD3+ cells/ kg, and 9.32 x 104 (range, 5.5 to 12.55 x 104) CD19+ cells/kg [12]. The patient with aplastic anemia received a total of 8.66 x 106 CD34+ cells/kg, 1.17 x 104 CD3+ cells/kg, and 11.3 x 104 CD19+ cells/kg [14]. T cell depletion was the only GVHD prophylaxis [12,14].

All patients received at least 1 infusion of donor whole blood containing 5, 7,10, 25, or 50 x 103 CD3+ cells/kg between days 25 and 95 after transplantation [12,14]. Four to 6 years posttransplantation, we studied 5 of the 6 surviving members of this patient group. These patients, who comprised 3 men and 2 women with a median age of 30 years (range, 24 to 34), had poor-prognosis acute myeloid leukemia (n = 4) or aplastic anemia (n = 1). The donors were always 1 of the parents, comprising 4 men and 1 woman with a median age of 55 years (range, 50 to 62). All donor—patient pairs shared 3/ 6 HLA antigens. The age-matched control group comprised 3 men and 2 women with a median age of 31 years (range, 24 to 35).

Peripheral blood mononuclear cells were isolated by Ficoll-Hypaque density gradient (Amersham Pharmacia Biotech, Uppsala, Sweden) from heparinized peripheral blood, and serum samples were collected in parallel. This study was approved by the Ethical Board of the Faculty of Medicine, University of Lisbon, and all patients gave written informed consent.

Flow Cytometric Analysis

Freshly isolated peripheral blood mononuclear cells (PBMCs) were resuspended in phosphate-buffered saline (PBS) containing 1% bovine serum albumin (Sigma-Aldrich) St. Louis, MO and .1% sodium azide (Sigma-Aldrich) and stained for 20 minutes at room temperature with the following anti-human monoclonal antibodies: CD4 allophycocyanin—cyanin 7 (Cy7) or fluorescein isothiocyanate (FITC; clone RPA-T4, BD Biosciences); CD8 peri-dinin chlorophyll protein (PerCP-Cy5.5), eFluor 605 (clone RPA-T8, eBioscience, San Diego, CA) or biotin (clone OKT8, eBioscience); CD45RA PerCP-Cy5.5, allophycocyanin, or biotin (clone H1100, eBioscience); CD27 FITC, phycoerythrin, or phycoerythrin -Cy7 (clone O323, eBioscience); CD31 allophycocyanin (clone WM59, eBioscience); and Streptavidin Cy3 (Cedar-lane Laboratories, Ontario, Canada). Intracellular staining for Ki67 FITC (clone B56, BD Biosciences) and Foxp3 Pacific Blue (clone PCH101, eBio-science) was performed using fixation and permeabilization reagents from eBioscience. Samples were acquired on a BD FACSAria flow cytometer (BD Biosciences) after fixation with 1% formaldehyde (Sigma-Aldrich). Data were analyzed using FlowJo software version 8.1.1 (TreeStar, Ashland, OR).

Telomere Length Measurement by Flow Cytometry—Fluorescent In Situ Hybridization

Telomere length was measured using a modified version of the fluorescent in situ hybridization coupled with the flow cytometry protocol previously described [15,16]. In brief, PBMCs were surface stained for CD4, CD8, CD45RA, and CD27. After washing in PBS, cells were fixed in 1 mM BS3 (Perbio Science, Northumberland, UK). The reaction was quenched with 50 mM Tris (pH 7.2) in PBS. After washing in PBS followed by hybridization buffer, cells were incubated in .75 mg/mL protein nucleic acid telomeric probe (C3TA2)3 conjugated to Cy5 (Panagene, Daejeon, Korea). After being heated for 10 minutes at 82°C, samples were left to hybridize. Samples were washed in posthybridization buffer followed by PBS and analyzed immediately by flow cytometry. All samples were run in triplicate alongside cryopreserved PBMCs with known telomere fluorescence to ensure consistency of results.

Signal-Joint T Cell—Receptor Excision Circle Quantification by Real-Time Polymerase Chain Reaction

DNA was purified from 106 PBMCs using DNAzol reagent (Gibco Life Technologies Carlsbad, California). Signal-joint T cell—receptor excision circles (sjTRECs) were quantified by nested real-time polymerase chain reaction (PCR) using Power SYBR Green PCR Master Mix (Applied Biosystems, Life Technologies, Carlsbad, California) and AB1 PR1SM 7000 Sequence Detection System (Applied Biosystems). Specific primers and probes were used for sjTREC, and CD3g was used as a housekeeping gene for absolute quantification of sjTREC levels: sj-out5,5'-CTCTCCTATCTCTGCT CTGAA-3'; sj-out3, 5'-ACTCACTTTTCCGAGGCTGA-3'; sj-in5, 5'-CCTCTGT CAACAAAGGTGAT-3'; sj-in3, 5'-GTGCTGGCATCAGAGTGTGT-3'; CD3-out5, 5'-ACTGACATGGAACAGGGG AAG-3'; CD3-out3, 5'-CCAGCTCTGAAGTAGGG AACATAT-3'; CD3-in5, 5'-GGCTATCATTCTTCTTCAAGGT-3'; CD3-in3, 5'-CCTCT CTTCAGCCATTTAA GTA-3'; sj-Probe1, 5'-AATAAGTTCAGCCCTCCATGTCACAC Tf-3'; sj-Probe2, 5'-XTGTTTTCCATCCTGGGGAGTGTTTCAp-3'; CD3-Probe1, 5'-GGCTGAAGGTTAG GGATACCAATATTCCTGTCTCf-3'; CD3-Probe2,5'-XCTAGT GATGGGCTCTTCC CTTGAGCCCTTCp-3'.

T Cell Receptor Chain Complementarity-Determining Region 3 Spectratyping

TCRVB complementarity-determining region 3 (CDR3) spectratyping was performed as previously described [17]. Briefly, total RNA was extracted from 105 to 106 cells with RNeasy kit (Qiagen, Dusseldorf, Germany) and first strand cDNA synthesized from 1 to 2 mg of RNA with the Superscript 111 kit (Invitrogen Life Technologies, Carlsbad, California) using an equivolume mixture of random hexamers and oligo (dT). Amplification of the TCRVB CDR3 was performed using primers specific for each Vb family [18], except for Vb6 and VP21 [19], and a common CB reverse primer [18]; followed by a run-off reaction extending each different PCR product with a constant CB FAM labeled primer [18]; and a third step, in which each different Vb PCR-labeled fragment was separated on a capillary electrophoresis—based DNA automated sequencer. Data were collected and analyzed with GeneMapper v4.0 (Applied Biosystems) for size and fluorescence intensity determination.

Determination of Vb Family Usage by Flow Cytometry

TCR Vb family frequency was quantified in whole blood using 1OTest Beta Mark (Beckman Coulter, Brea, California), according to the manufacturer's instructions. Samples were labeled with T cell receptor variable beta chain (TCR Vb)-specific antibodies conjugated to FITC, phycoerythrin, or both, together with antibodies against CD4, CD8, and CD27. Fixed/per-meabilized cells were then stained with anti-human Foxp3 and analyzed by flow cytometry. The frequency of the individual TCR Vb families was assessed within T cell populations of interest.

Statistical Analysis

Statistical analysis was performed using GraphPad Prism version 4.00 (GraphPad Software, San Diego, CA). Data are presented as mean with standard error of mean. Gaussian distribution was confirmed with the Kolmogorov-Smirnov normality test, and statistical analysis was performed using a paired or unpaired t-test, as appropriate. P values <.05 were considered significant.

RESULTS

Proportion of Naive and Memory CD4+ and CD8+ T Cell Subsets

We first evaluated the degree of immune reconstitution achieved 4 to 6 years posttransplantation in recipients of fully haploidentical HSCT by assessing the absolute numbers of the main lymphocyte populations in peripheral blood (Table 1). A trend was observed for HSCT recipients who had higher numbers of total lymphocytes, total CD3+ T cells, CD8+ T cells, and natural killer cells, which were identified as CD3- cells expressing CD16 and/or CD56, than both donors and age-matched control subjects, although these differences did not reach statistical significance (Table 1). Similarly, the absolute numbers of CD19+ B cells tended to be higher in HSCT recipients than the other 2 groups, reaching statistical significance in comparison with age-matched control subjects (Table 1). The observed similarities in the absolute numbers of the major lymphocyte subsets in haploidentical HSCT recipients and age-matched control subjects were indicative of successful recovery of the size of these peripheral lymphocyte pools. Additionally,

Table 1

Absolute Counts of the Main Lymphocyte Populations

Donors Recipients Control Subjects

Absolute counts per mL

Lymphocytes 3979 ± 470 5187 ± 896 3809 ± 460

Total T cells 2686 ± 358 2872 ± 480 2401 ± 341

CD8+ T cells 486 ± 97 851 ± 266 615 ± 127

CD4+ T cells 1976 ± 247 1635 ± 180 1616 ± 275

B cells 224 ± 91 417 ± 126* 241 ± 80

NK cells 52 ± 26 118 ± 62 85 ± 42

CD4/CD8 4.45 ± .59 2.32 ± .34 2.96 ± .72

P = .0399, recipients versus control subjects.

we documented in vitro proliferative responses to mitogenic as well as antigen-specific stimuli (data not shown), attesting to the degree of functional competence achieved by HSCT recipients.

To investigate whether T cell reconstitution was associated with imbalances in naive and memory T cell subset distribution, we determined the frequency of naive and memory subsets, defined by the expression of CD45RA and CD27, within CD8+ and CD4+ T lymphocytes: CD45RA+ CD27+ naive, CD45RA~CD27+ central memory, CD45RA~

CD27~ effector memory, and CD45RA+CD27~ effector memory re-expressing CD45RA (EMRA) cells [20] (Figure 1). HSCT recipients had a significantly higher frequency of CD45RA+CD27+ naive cells and a significantly lower percentage of CD45RA+CD27~ EMRA cells within CD8+ T lymphocytes as compared with donors (paired t-test: P = .0249 and P = .0116, respectively) (Figure 1A). These differences were less striking within the CD4+ T cell population, with the 3 groups featuring similar frequencies of CD45RA+CD27+ naive cells (Figure 1B). Nevertheless, we also observed a tendency for the recipients to have a lower proportion of highly differentiated CD45RA~CD27~ effector memory and CD45RA+CD27~ EMRA subsets within the CD4+ T cell population than the donors (Figure 1B). The levels of naive T cells have been shown to decrease, whereas highly differentiated memory T cells increase during aging [21-23]. Thus, the distinct differentiation profiles observed in recipients and donors may be due to the age difference between these 2 groups, given that the donors were always 1 of the parents. Nonetheless, recipients tended to have a less differentiated profile than age-matched control subjects, particularly within CD8+ T cells (Figure 1A), suggesting that age is not the sole factor contributing to these

Figure 1. Frequency of naive and memory subsets within the CD8+ and CD4+ T cell pools after haploidentical HSCT. The frequency of naive (CD45RA+CD27+), central memory (CM; CD45RA~CD27+), effector memory (EM; CD45RA~CD27~), and CD45RA+ EM (EMRA; CD45RA+CD27~) cell subsets was determined within CD8+ (A) and CD4+ (B) T lymphocytes in HSCT recipients, donors, and age-matched control subjects (*P < .05 between recipients and donors or control subjects and donors). Flow cytometry images from a representative recipient, the respective donor, and age-matched control subject are shown.

Figure 2. Assessment of CD31 expression, sjTREC content, and telomere length after haploidentical HSCT. (A) CD31 expression was used to identify a population enriched in RTEs within naive (CD45RA+CD27+) CD4+ T lymphocytes. Flow cytometry images from a representative recipient, the respective donor, and age-matched control subject are shown. (B) Levels of sjTRECs within PBMCs. Results are expressed as the copy number of sjTRECs per 106 PBMCs. (C) Telomere length measured within CD8+CD27+ and CD27-, as well as within CD4+CD45RA+ and CD45RA- T lymphocytes. Each symbol represents an individual. Mean values are shown as horizontal lines. *P < .05.

differences. Overall, HSCT recipients showed effective immune reconstitution 4 to 6 years posttransplantation, with evidence of successful recovery and maintenance of the naive T cell pool.

Mechanisms Underlying the Reconstitution of the Naive T Cell Pool

We next sought to assess the relative contribution of thymic output and peripheral expansion to T cell reconstitution in these patients. Several studies used CD31 expression within naive CD4+ T cells to indirectly assess thymic output after HSCT [24,25]. Although CD31 expression cannot be considered an absolute marker of recent thymic emigrants (RTEs), because CD31+ naive CD4+ T cells are able to undergo interleukin (lL)-7-driven homeostatic proliferation without losing CD31 expression [26], it identifies the population that is most enriched in newly generated T cells as estimated by the quantification of sjTREC [27,28].

To evaluate the relative contribution of thymic output and peripheral expansion to T cell reconstitution, we first assessed the expression of CD31 within naive CD4+ T cells (Figure 2A). We observed that the proportion of CD31-expressing cells within naive CD45RA+CD27+ CD4+ T cells was significantly higher in recipients than in donors (paired t-test: P = .0135) and similar to that observed in age-

matched control subjects (Figure 2A). We further assessed the contribution of thymic output to immune reconstitution in recipients via the quantification of sjTREC levels in PBMCs (Figure 2B). At a cell population level, sjTREC content reflects the overall outcome of the TREC-enriching contribution of thymic output and the TREC-diluting effect of peripheral expansion [29]. 1n agreement with the high proportion of CD31 + naive CD4+ T cells, sjTREC levels in recipients tended to be higher than donors, which were similar to those observed in age-matched control subjects (Figure 2B). The observation that the recipients had relatively high sjTREC content suggested that long-term T cell reconstitution and maintenance did not rely heavily on peripheral expansion but rather on a substantial contribution of thymic output.

The replicative history of a cell population may be estimated by the assessment of telomere length. Thus, to further assess the impact of peripheral expansion on T cell reconstitution, we evaluated telomere length in CD8+ and CD4+ T cell subsets (Figure 2C). Naive CD4+ T cells are known to have longer telomeres than their memory counterparts [30,31]. In agreement, we observed higher telomere-specific fluorescence intensity in the naive T cell subset than in CD4+ T cells lacking CD45RA expression in all 3 groups (paired t-test: donors, P = .0125; recipients, P = .0364; control subjects, P = .0004). As shown in

Figure 2C, when telomere length within CD45RA+ or CD45RA- CD4+ T cells was compared among the 3 groups, recipients tended to have longer telomeres than the respective donors but had significantly shorter telomeres within CD4+CD45RA+ cells as compared with age-matched control subjects (paired t-test: P = .0316). The latter had significantly longer telomeres within both CD4+CD45RA+ and CD4+CD45RA- subsets than donors (unpaired t-test: P = .0220 and P = .0386, respectively). The observation that telomere length within the naive-enriched CD45RA+ CD4+ T cell subset was significantly shorter in recipients than in age-matched control subjects may suggest that peripheral expansion may also have contributed to T cell recovery in these individuals. Additionally, it may reflect the inherently shorter telomeres of CD34+ hematopoietic stem cells from the parental donors.

For the measurement of telomere length within CD8+ T cells, CD8 was costained with CD27, as we could only use 2 surface markers. CD27 was selected because it allows the discrimination between a CD27+ population comprising CD45RA+CD27+ naive and CD45RA-CD27+ central memory CD8+ T cells and a CD27- population containing highly differentiated CD45RA-CD27- effector memory and CD45RA+CD27- EMRA CD8+ T cells (Figure 2C). As expected, CD27+ CD8+ T cells had significantly longer telomeres than their CD27- counterparts in all 3 groups (paired t-test: donors, P = .002; recipients, P = .0173; control subjects, P = .0018). Similar to the results obtained for CD4+ T cell subsets, recipients tended to have longer telomeres than the respective donors (Figure 2C).

Overall, our data support a major contribution of the thymus to the replenishment of the T cell compartment. However, peripheral expansion of naive T cells was also likely to play a role.

T Cell Receptor Repertoire of the CD4+ and CD8+ T Cell Pools

Next, we assessed whether the restoration of T cell numbers was accompanied by the establishment of a diverse T cell receptor repertoire. For this purpose, we performed a spectratyping analysis in purified CD4+ and CD8+ T cells (Figure 3A, B). This analysis provides a measure of diversity at the level of CDR3 length, reflecting the overall sequence heterogeneity. T cell pools with a diverse poly-clonal T cell receptor repertoire present a Gaussian distribution of CDR3 length, whereas skewed T cell receptor repertoires have a CDR3 length distribution with a reduced number of peaks or may even comprise a single peak, in the case of clonal dominance [32,33]. Figure 3A illustrates representative spectratypes of CD8+ T cells from a transplantation recipient together with the respective donor and age-matched control subject. The CD8 TCRVB repertoire was largely polyclonal in all recipients, although CD8+ T cells from 1 individual displayed a monoclonal distribution within the Vb9 and Vb11 families, whereas another had a monoclonal distribution within the Vb22 family (data not shown).

As for the CD4+ T cell pool, all recipients displayed a polyclonal distribution of CDR3 length, except for 1 individual who had an oligoclonal distribution of the Vb13 and Vb22 families (data not shown). This markedly polyclonal spectratype profile closely resembled that observed in CD4+ T cells from age-matched control subjects, as illustrated by the representative individuals shown in Figure 3B. As previously described [34,35], we observed more

perturbations in TCRVB repertoire diversity in CD8+ T cells (Figure 3A) than in CD4+ T cells (Figure 3B). This may be due to the more robust and prolonged proliferative response upon antigen encounter observed in CD8+ compared with CD4+ T cells [35].

We also performed a spectratyping analysis within gS T cells (data not shown). All patients showed a Gaussian polyclonal distribution of these 2 major VS subsets, whereas the donor and age-matched control groups displayed a more restricted clonal profile of the VS1 and VS2 families (data not shown). As for the main Vg families, all 3 groups showed a predominantly polyclonal profile (data not shown). When we compared the distribution of the different VS and Vg among the 3 groups, the only statistically significant difference observed was in the frequency of VS4 expression between recipients and donors (recipients, 18.39 ± 5.33; donors, 2.09 ± 1.06; paired t-test: P = .0474). The frequency of VS4 expression in the age-matched control subjects was intermediate compared with the other 2 groups (10.95 ± 4.61).

We further analyzed the TCR Vb family usage within naive CD8+ (Figure 3C) and CD4+ (Figure 3D) T cells. We used the lOTest b Mark kit (Beckman Coulter), a flow cytometry—based assay for quantitative analysis of the TCR Vb repertoire, which assesses the frequency of 24 Vb families covering z70% of the T cell receptor repertoire of human T lymphocytes. The results of the TCR Vb distribution analysis for the CD45RA+CD27+ naive CD8+ subset are shown in Figure 3C and those for the RTE-enriched CD31+ naive CD4+ T cell subset in Figure 3D, illustrating the diverse set of TCR Vb family usage. lnterestingly, the Vb2, Vb5.1, Vb5.3, and Vb17 families were overrepresented in CD31+ naive CD4+ T cells from donors as compared with recipients (Figure 3D), with the latter displaying an even broader distribution of Vb family usage. Thus, TCR Vb analysis revealed a largely poly-clonal repertoire both within the CD4+ and CD8+ T cell pools, indicating that the immune reconstitution accomplished in the recipients was associated with a diverse T cell repertoire.

Reconstitution of the Regulatory T Cell Pool

A major complication post-HSCT is GVHD, which is associated with poor regulatory T cell (Treg) reconstitution in the first year posttransplantation [36]. All patients received a small dose of donor leukocytes posttransplantation (between 6 x 103/kg and 5 x 104/kg CD3+ cells) to improve engraftment and/or to boost immune recovery [12,14]. Four of 5 patients received these cells until day 72 posttransplantation and developed acute GVHD grades I (n = 1), ll (n = 1), Ill (n = 1), and IV (n = 1), which was successfully treated with immunosuppression. None of them developed chronic GVHD. We were thus very interested in assessing the long-term reconstitution of Tregs in this particular group of patients submitted to haploidentical HSCT.

The frequency of Tregs within the CD4+ T cell pool was similar in the 3 groups (Figure 4A). Recipients had Treg levels comparable with those observed in age-matched control subjects, both in terms of frequency and of absolute numbers (recipients, 49 ± 11 counts/pL; control subjects, 44 ± 6 counts/pL). As for the differentiation state of these Tregs, the proportion of naive cells within the Treg population was also comparable in all groups (Figure 4B). lnterestingly, when we assessed the frequency of CD31+ cells within the naive Treg pool, transplantation recipients displayed significantly higher levels of this RTE-enriched population in comparison

Figure 3. TCR VP repertoire of CD8+ and CD4+ T lymphocytes after haploidentical HSCT. Spectratyping analysis of the CDR3 VB regions of the CD8+ (A) and CD4+ (B) T cell pools. Spectratypes for each VB family from a representative recipient and the respective donor and age-matched control subject are shown. Flow cytometry analysis of TCR VP family usage within naive CD8+ (C) and CD31+ naive CD4+ T cells (D). Bar graphs show mean with standard error of the mean. *P < .05, **P < .01.

with donors (paired t-test: P = .0133) (Figure 4C). The maintenance of the naive Treg compartment in HSCT recipients did not seem to mainly rely on enhanced peripheral expansion, given that the frequency of cycling cells, as assessed by Ki67 expression, within CD31+ naive Tregs (Figure 4D) and the overall naive CD45RA+CD27+ Treg pool (data not shown) was very low in these individuals. There

were no statistically significant differences in the frequency of Ki67+ cells within total CD4+Foxp3+ and memory CD45RA~ Tregs among the 3 groups (data not shown).

We further assessed the TCR Vb repertoire within the Treg pool (Figure 4E). The recipient group displayed a balanced TCR Vb family usage, whereas the frequency of the Vb5.2 and Vb14 families was significantly higher in donors as compared

Figure 4. Analysis of the Treg pool after haploidentical HSCT. (A) Frequency of Tregs (CD4+Foxp3+) within the CD4+ Tcell population. (B) Proportion of naive cells (CD45RA+CD27+) within the Treg pool. (C) Proportion of CD31+ cells within naive Tregs. (D) Frequency of cycling CD31+ naive Tregs, identified by Ki67 expression. Mean values are shown as horizontal lines. Each symbol represents an individual. (E) Flow cytometry analysis of TCR Vß family usage within Tregs. Bar graph shows mean with standard error of the mean. *P < .05, **P < .01.

with recipients (Figure 4E). These significant differences were only observed within CD45RA~ but not CD45RA+ Tregs (data not shown). Overall, HSCT recipients have successfully reconstituted a diverse Treg pool through a mechanism that involves de novo thymic production.

DISCUSSION

Haploidentical HSCT constitutes a particularly challenging clinical setting because the number of mature T cells in the graft needs to be minimal to prevent GVHD, leading to delayed immune reconstitution. Moreover, the conditioning regimen may damage thymic and/or peripheral lymphoid tissues, further hindering naive T cell output and memory T cell homeostasis. To assess whether immune reconstitution was successfully achieved after haploidentical HSCT, as well as the potential mechanisms underlying this recovery, we performed a cross-sectional study of long-term immune reconstitution in a group of 5 patients who had received a CD34+ purified stem cell graft from full-haplotype mismatched related donors. The main limitation of our study

was the small number of HSCT recipients who were available for evaluation 4 to 6 years posttransplantation. Nevertheless, our study group comprised a substantial proportion of the original cohort of haploidentical HSCT recipients (5 of 15, 33.3%) and of the surviving patient group (5 of 6, 83,33%). We evaluated T cell reconstitution in these patients by combining naive/memory T cell phenotyping, sjTREC quantification, telomere length measurement, and T cell repertoire analysis.

We observed a successful recovery of absolute CD8+ and CD4+ Tcell counts, as well as of B and natural killer cells, in the transplantation recipients. We further observed that transplantation recipients tended to have increased frequencies of naive CD45RA+CD27+ cells and lower frequencies of EMRA CD45RA+CD27~ cells than their respective donors, with these differences reaching statistical significance within CD8+ T cells. The CD45RA+CD27~ subset has been described in human CD8+ Tcells as the most differentiated type of memory cells, which is supported by their low proliferative capacity, high susceptibility to apoptosis, and loss of CD28 and CCR7

expression [37,38]. This highly differentiated subset has been shown to accumulate during aging [39] and chronic viral infections [40,41], comprising large clonal expansions of virus-specific cells [41]. As for CD4+ T cells, we previously demonstrated that CD45RA+CD27~ CD4+ T cells are prone to telomere-independent senescence through a process partly driven by the p38 MAPK pathway [20] and that these cells are accumulated during chronic cytomegalovirus infection [42], a setting associated with persistent antigen stimulation. Hence, the absence of inflated CD45RA+CD27~ T cell populations in these transplantation recipients suggests that restoration of T cell numbers was not mainly driven by clonal expansion.

We sought to investigate the relative contribution of thymic-dependent and -independent mechanisms to the replenishment of the T cell pool in these patients. Thymic-independent mechanisms include antigen- and/or cytokine-driven expansion of mature donor T cells infused with the graft, resulting in limited T cell-receptor diversity. Conversely, thymic-dependent mechanisms require de novo production of naive T cells from donor-derived precursor cells, generating a more long-lasting and diverse T cell reconstitution [43]. We found that the frequency of CD31+ cells, a population enriched in RTEs, was significantly higher in recipients than in donors and similar to that observed in age-matched control subjects. The absolute numbers, as well as the frequency, of CD31 + naive CD4+ T cells in human peripheral blood have been shown to decrease with aging, in parallel with the decline in TREC content [27,44].

The significantly higher frequencies of CD31+ naive CD4+ T cells observed in the recipients are likely to reflect not only the age gap between the recipients and their parental donors but also the contribution of de novo naive T cell production to immune reconstitution. We previously reported that the CD31 + naive CD4+ T cell subset can be maintained by an 1L-7-driven peripheral expansion [26]. 1L-7 levels have been described as being increased during lymphopenic states, such as those taking place immediately after stem cell transplantation [45]. Once normal CD4+ T cell numbers are restored, particularly within the naïve CD4+ T cell subset, 1L-7 has been shown to return to steady state levels [46-48]. At the time of the study, transplantation recipients did not have higher 1L-7 serum levels than the other 2 groups (data not shown), which is in agreement with the observation of comparable CD4+ T cell counts (Table 1 ), as well as of CD31+ naive CD4+ T cell frequencies, between recipients and age-matched control subjects. Moreover, the expression levels (both in terms of frequency and median fluorescence intensity) of 1L-7Ra within naive and memory CD4+ and CD8+ T cells, particularly within CD31+naive CD4+ Tcells, were similar in all 3 groups (data not shown). Although we observed low levels of proliferation within naive CD4+ Tcells at the time of our study ( < 1% Ki67+ cells both within total naive CD45RA+CD27+ CD4+ Tcells and the CD31+ naive subset), 1L-7-driven homeostatic proliferation is likely to have contributed to the replenishment of the CD31 + naive CD4+ T cell subset in the initial stages of immune reconstitution, during which the cells would have been exposed to elevated 1L-7 levels.

To further ascertain the contribution of thymic-dependent and -independent mechanisms to immune reconstitution, we compared sjTREC levels and telomere length in our 3 groups. The relatively high sjTREC content within PBMCs from recipients as compared with age-matched control subjects supports a contribution of thymic output to the immune reconstitution. On the other hand,

although recipients and age-matched control subjects displayed similarly long telomeres within the more differentiated CD4+ and CD8+ T cell subsets, telomere length within the naive-enriched CD4+CD45RA+ subset in the recipients was significantly shorter than in the age-matched control subjects, more closely resembling that observed in the donors. Telomere length in HSCT recipients has been found to correlate with donor age [49], indicating that hemato-poietic stem cells harvested from older donors have shorter telomeres. Although the low levels of telomerase activity displayed by hematopoietic stem cells are sufficient to ensure long-term survival, they do not appear to prevent age-associated telomere erosion in these cells [50]. Hence, the observation that HSCT recipients tended to have shorter telomeres than their age-matched control subjects, particularly within naive-enriched subsets, may be a consequence of infusing CD34+ hematopoietic stem cells from parental donors. Additionally, as mentioned previously, this may also be partly due to increased levels of 1L-7-driven homeostatic proliferation of the CD31 + naive CD4+ T cell subset in the recipients during the initial stages of immune reconstitution.

The presence of a broad and largely polyclonal T cell repertoire in transplantation recipients, comparable with that observed in age-matched control subjects, supports the view that immune reconstitution likely relied on thymic output together with peripheral T cell proliferation driven by homeostatic triggers.

The induction and maintenance of immune tolerance after allogeneic HSCT is of the utmost importance, given that its loss may lead to major clinical complications, particularly chronic GVHD, which remains a significant cause of posttransplantation morbidity and mortality [51]. Furthermore, acute and chronic GVHD have been shown to target thymic function and to be associated with poor immune reconstitution [52-54]. On the other hand, a sustained contribution of thymic output to immune reconstitution might decrease the risk of GVHD through negative selection of self-reactive Tcells during de novo Tcell generation [55]. Tregs have been shown to play a critical role in the establishment of post-HSCT tolerance in murine studies [56,57]. 1n humans, lower levels of Tregs have been associated with a higher risk of chronic GVHD incidence and severity [36]. A study showed for the first time in humans that early infusion of donor Tregs, in the absence of pharmacological GVHD prophylaxis, allows for the administration of significant numbers of conventional CD4+ T cells at the time of haploidentical HSCT, leading to a faster recovery of posttransplantation immunity without GVHD [58]. Hence, a timely reconstitution of the Treg pool appears to be vital for overall immune reconstitution and tolerance. Although all but 1 of our HSCT recipients developed acute GVHD after the infusion of a limited dose of donor leukocytes, none developed chronic GVHD. The observation that all patients successfully replenished the Treg compartment is in line with this subset's ability to control and prevent chronic GVHD. All recipients had Treg levels comparable with those observed in age-matched control subjects, both in terms of frequency and of absolute numbers, which may have crucially contributed to immune reconstitution and tolerance establishment post-haploidentical HSCT. 1nterestingly, the frequency of CD31 + cells within the naive Treg population was significantly higher in the recipients compared with donors, despite the frequency of cycling cells within the CD31 + naive Treg subset being lower in the former, further suggesting a substantial contribution of thymic output to the replenishment of the T cell pool.

Overall, our data show that once the initial posttransplantation period is successfully overcome, full immune reconstitution can be achieved after haploidentical HSCT. Our results further suggest that T cell homeostasis was effectively restored in these patients, a process that appears to have mostly relied on thymus-dependent mechanisms, despite the high degree of HLA mismatch between donor and recipient in this setting.

ACKNOWLEDGMENTS

The authors are grateful to Russell B. Foxall (1nstituto de Medicina Molecular, Lisboa) for scientific discussion and manuscript revision. We acknowledge Remy Cheynier (1nstitut Cochin, Paris) for kindly providing the pCD3-TREC plasmid.

Financial disclosure: Supported by grant PTDC/SAU-M11/ 67662/2006 from Funda^ao para a Ciencia e Tecnologia (FCT) and by Programa Operacional Ciencia e Inova^ao 2010 (POC12010) to M.V.D.S. 1n addition, R.1.A., A.S.A., R.T., and R.S.S. received scholarships from FCT cofinanced by POC1 2010 and FSE.

Conflict of interest statement: There are no conflicts of interest to report.

Authorship statement: R.I.A. and M.V.D.S. contributed equally to this work.

REFERENCES

1. Huang XJ, Chang YJ. Unmanipulated HLA-mismatched/haploidentical blood and marrow hematopoietic stem cell transplantation. Biol Blood Marrow Transplant. 2011;17:197-204.

2. Huang XJ. Current status of haploidentical stem cell transplantation for leukemia. J Hematol Oncol. 2008;1:27.

3. Hansen JA, Petersdorf E, Martin PJ, Anasetti C. Hematopoietic stem cell transplants from unrelated donors. Immunol Rev. 1997;157:141-151.

4. Spitzer TR. Haploidentical stem cell transplantation: the always present but overlooked donor. Hematology Am Soc Hematol Educ Program. 2005; 390-395.

5. Zhao XY, Huang XJ, Liu KY, et al. Prognosis after unmanipulated HLA-haploidentical blood and marrow transplantation is correlated to the numbers of KIR ligands in recipients. Eur J Haematol. 2007;78:338-346.

6. Ruggeri L, Mancusi A, Burchielli E, et al. NK cell alloreactivity and allogeneic hematopoietic stem cell transplantation. Blood Cells Mol Dis. 2008;40:84-90.

7. Dodero A, Carniti C, Raganato A, et al. Haploidentical stem cell transplantation after a reduced-intensity conditioning regimen for the treatment of advanced hematologic malignancies: posttransplantation CD8-depleted donor lymphocyte infusions contribute to improve T-cell recovery. Blood. 2009;113:4771-4779.

8. Chen X, Hale GA, Barfield R, et al. Rapid immune reconstitution after a reduced-intensity conditioning regimen and a CD3-depleted hap-loidentical stem cell graft for paediatric refractory haematological malignancies. Br J Haematol. 2006;135:524-532.

9. Bethge WA, Haegele M, Faul C, et al. Haploidentical allogeneic hema-topoietic cell transplantation in adults with reduced-intensity conditioning and CD3/CD19 depletion: fast engraftment and low toxicity. Exp Hematol. 2006;34:1746-1752.

10. Aversa F, Terenzi A, Tabilio A, et al. Full haplotype-mismatched hema-topoietic stem-cell transplantation: a phase 11 study in patients with acute leukemia at high risk of relapse. J Clin Oncol. 2005;23:3447-3454.

11. Huang XJ, Liu DH, Liu KY, et al. Treatment of acute leukemia with unma-nipulated HLA-mismatched/haploidentical blood and bone marrow transplantation. Biol Blood Marrow Transplant. 2009;15:257-265.

12. Lacerda JF, Martins C, Carmo JA, et al. Haploidentical stem cell transplantation with purified CD34 cells after a chemotherapy-alone conditioning regimen. Biol Blood Marrow Transplant. 2003;9:633-642.

13. Rizzieri DA, Koh LP, Long GD, et al. Partially matched, non-myeloablative allogeneic transplantation: clinical outcomes and immune reconstitution. J Clin Oncol. 2007;25:690-697.

14. Lacerda JF, Martins C, Carmo JA, et al. Haploidentical stem cell transplantation with purified CD34+ cells after a chemotherapy-alone conditioning regimen in heavily transfused severe aplastic anemia. Biol Blood Marrow Transplant. 2005;11:399-400.

15. Plunkett FJ, Franzese O, Finney HM, et al. The loss of telomerase activity in highly differentiated CD8+CD28-CD27- T cells is associated with decreased Akt (Ser473) phosphorylation. J Immunol. 2007;178: 7710-7719.

16. Henson SM, Franzese O, Macaulay R, et al. KLRG1 signaling induces defective Akt (ser473) phosphorylation and proliferative dysfunction of highly differentiated CD8+ T cells. Blood. 2009;113:6619-6628.

17. Albuquerque AS, Marques JG, Silva SL, et al. Human FOXN1-deficiency is associated with aß double-negative and FoxP3+ T-cell expansions that are distinctly modulated upon thymic transplantation. PLoS One. 2012;7:e37042.

18. Puisieux I, Even J, Pannetier C, et al. Oligoclonality of tumor-infiltrating lymphocytes from human melanomas. J Immunol. 1994;153: 2807-2818.

19. Garderet L, Dulphy N, Douay C, et al. The umbilical cord blood alphabeta T-cell repertoire: characteristics of a polyclonal and naive but completely formed repertoire. Blood. 1998;91:340-346.

20. Di Mitri D, Azevedo RI, Henson SM, et al. Reversible senescence in human CD4+CD45RA+CD27- memory T cells. J Immunol. 2011;187: 2093-2100.

21. Mackall CL, Gress RE. Thymic aging and T-cell regeneration. Immunol Rev. 1997;160:91-102.

22. Lazuardi L, Jenewein B, Wolf AM, et al. Age-related loss of naive T cells and dysregulation of T-cell/B-cell interactions in human lymph nodes. Immunology. 2005;114:37-43.

23. Weyand CM, Brandes JC, Schmidt D, et al. Functional properties of CD4+ CD28- T cells in the aging immune system. Mech Ageing Dev. 1998;102:131-147.

24. Muraro PA, Douek DC, Packer A, et al. Thymic output generates a new and diverse TCR repertoire after autologous stem cell transplantation in multiple sclerosis patients. J Exp Med. 2005;201:805-816.

25. Thiel A, Alexander T, Schmidt CA, et al. Direct assessment of thymic reactivation after autologous stem cell transplantation. Acta Haematol. 2008;119:22-27.

26. Azevedo RI, Soares MV, Barata JT, et al. 1L-7 sustains CD31 expression in human naive CD4+ T cells and preferentially expands the CD31+ subset in a P13K-dependent manner. Blood. 2009;113: 2999-3007.

27. Kimmig S, Przybylski GK, Schmidt CA, et al. Two subsets of naive T helper cells with distinct T cell receptor excision circle content in human adult peripheral blood. J Exp Med. 2002;195:789-794.

28. Junge S, Kloeckener-Gruissem B, Zufferey R, et al. Correlation between recent thymic emigrants and CD31+ (PECAM-1) CD4+ T cells in normal individuals during aging and in lymphopenic children. Eur J Immunol. 2007;37:3270-3280.

29. Douek DC, McFarland RD, Keiser PH, et al. Changes in thymic function with age and during the treatment of HIV infection. Nature. 1998;396: 690-695.

30. Weng NP, Levine BL, June CH, Hodes RJ. Human naive and memory T lymphocytes differ in telomeric length and replicative potential. Proc Natl Acad Sci USA. 1995;92:11091-11094.

31. Fletcher JM, Vukmanovic-Stejic M, Dunne PJ, et al. Cytomegalovirus-specific CD4+ T cells in healthy carriers are continuously driven to replicative exhaustion. J Immunol. 2005;175:8218-8225.

32. Pannetier C, Even J, Kourilsky P. T-cell repertoire diversity and clonal expansions in normal and clinical samples. Immunol Today. 1995;16: 176-181.

33. Gorski J, Yassai M, Zhu X, et al. Circulating T cell repertoire complexity in normal individuals and bone marrow recipients analyzed by CDR3 size spectratyping. Correlation with immune status. J Immunol. 1994; 152:5109-5119.

34. Kharbanda M, McCloskey TW, Pahwa R, et al. Alterations in T-cell receptor Vbeta repertoire of CD4 and CD8 T lymphocytes in human immunodeficiency virus-infected children. Clin Diagn Lab Immunol. 2003;10:53-58.

35. Maini MK, Casorati G, Dellabona P, et al. T-cell clonality in immune responses. Immunol Today. 1999;20:262-266.

36. Matsuoka K, Kim HT, McDonough S, et al. Altered regulatory T cell homeostasis in patients with CD4+ lymphopenia following allogeneic hematopoietic stem cell transplantation. J Clin Invest. 2010;120: 1479-1493.

37. Geginat J, Lanzavecchia A, Sallusto F. Proliferation and differentiation potentialof human CD8+ memory T-cell subsets in response to antigen or homeostatic cytokines. Blood. 2003;101:4260-4266.

38. Hamann D, Baars PA, Rep MH, et al. Phenotypic and functional separation of memory and effector human CD8+ T cells. J Exp Med. 1997; 186:1407-1418.

39. Plunkett FJ, Franzese O, Belaramani LL, et al. The impact of telomere erosion on memory CD8+ T cells in patients with X-linked lympho-proliferative syndrome. Mech Ageing Dev. 2005;126:855-865.

40. Dunne PJ, Faint JM, Gudgeon NH, et al. Epstein-Barr virus-specific CD8(+) T cells that re-express CD45RA are apoptosis-resistant memory cells that retain replicative potential. Blood. 2002;100: 933-940.

41. Wills MR, Okecha G, Weekes MP, et al. 1dentification of naive or antigen-experienced human CD8(+) T cells by expression of costimulation and chemokine receptors: analysis of the human cytomegalovirus-specific CD8(+) T cell response. J Immunol. 2002;168: 5455-5464.

42. Libri V, Azevedo RI, Jackson SE, et al. Cytomegalovirus infection induces the accumulation of short-lived, multifunctional CD4+CD45RA+ CD27+ T cells: the potential involvement of interleukin-7 in this process. Immunology. 2011;132:326-339.

43. Mackall CL, Fry TJ, Bare C, et al. IL-7 increases both thymic-dependent and thymic-independent T-cell regeneration after bone marrow transplantation. Blood. 2001;97:1491-1497.

44. Kilpatrick RD, Rickabaugh T, Hultin LE, et al. Homeostasis of the naive CD4+ T cell compartment during aging. J Immunol. 2008;180: 1499-1507.

45. Bolotin E, Annett G, Parkman R, Weinberg K. Serum levels of IL-7 in bone marrow transplant recipients: relationship to clinical characteristics and lymphocyte count. Bone Marrow Transplant. 1999;23: 783-788.

46. Fry TJ, Connick E, Falloon J, et al. A potential role for interleukin-7 in T-cell homeostasis. Blood. 2001;97:2983-2990.

47. Napolitano LA, Grant RM, Deeks SG, et al. Increased production of IL-7 accompanies HIV-1-mediated T-cell depletion: implications for T-cell homeostasis. Nat Med. 2001;7:73-79.

48. Smith DK, Neal JJ, Holmberg SD. Unexplained opportunistic infections and CD4+ T-lymphocytopenia without HIV infection. An investigation of cases in the United States. The Centers for Disease Control Idiopathic CD4+ T-lymphocytopenia Task Force. N Engl J Med. 1993;328:373-379.

49. Akiyama M, Asai O, Kuraishi Y, et al. Shortening of telomeres in recipients of both autologous and allogeneic hematopoietic stem cell transplantation. Bone Marrow Transplant. 2000;25:441-447.

50. Wang J, Geiger H, Rudolph KL. Immunoaging induced by hematopoietic stem cell aging. Curr Opin Immunol. 2011;23:532-536.

51. Broers AE, van der Holt B, Haze S, et al. A comparison of postengraft-ment infectious morbidity and mortality after allogeneic partially T cell-depleted peripheral blood progenitor cell transplantation versus T cell-depleted bone marrow transplantation. Exp Hematol. 2005;33: 912-919.

52. Poulin JF, Sylvestre M, Champagne P, et al. Evidence for adequate thymic function but impaired naive T-cell survival following allogeneic hematopoietic stem cell transplantation in the absence of chronic graft-versus-host disease. Blood. 2003;102:4600-4607.

53. Hazenberg MD, Otto SA, de Pauw ES, et al. T-cell receptor excision circle and T-cell dynamics after allogeneic stem cell transplantation are related to clinical events. Blood. 2002;99:3449-3453.

54. Clave E, Busson M, Douay C, et al. Acute graft-versus-host disease transiently impairs thymic output in young patients after allogeneic hematopoietic stem cell transplantation. Blood. 2009;113:6477-6484.

55. Snyder KM, Mackall CL, Fry TJ. IL-7 in allogeneic transplant: clinical promise and potential pitfalls. Leuk Lymph. 2006;47:1222-1228.

56. Trenado A, Charlotte F, Fisson S, et al. Recipient-type specific CD4+ CD25+ regulatory T cells favor immune reconstitution and control graft-versus-host disease while maintaining graft-versus-leukemia. J Clin Invest. 2003;112:1688-1696.

57. Edinger M, Hoffmann P, Ermann J, et al. CD4+CD25+ regulatory T cells preserve graft-versus-tumor activity while inhibiting graft-versus-host disease after bone marrow transplantation. Nat Med. 2003;9: 1144-1150.

58. Di Ianni M, Falzetti F, Carotti A, et al. Tregs prevent GVHD and promote immune reconstitution in HLA-haploidentical transplantation. Blood. 2011;117:3921-3928.