The JI
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     
 

The Journal of Immunology, 2007, 179, 2013 -2022
Copyright © 2007 by The American Association of Immunologists, Inc.
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by O’Keefe, C. L.
Right arrow Articles by Maciejewski, J. P.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by O’Keefe, C. L.
Right arrow Articles by Maciejewski, J. P.

Molecular Analysis of Alloreactive CTL Post-Hemopoietic Stem Cell Transplantation1

Christine L. O’Keefe*, Lukasz Gondek*, Randall Davis{dagger}, Elizabeth Kuczkowski{dagger}, Ronald M. Sobecks{dagger}, Alexander Rodriguez*, Yadira Narvaez*, Zachariah McIver*, Ralph Tuthill{ddagger}, Mary Laughlin§, Brian Bolwell{dagger} and Jaroslaw P. Maciejewski2,*

* Experimental Hematology and Hematopoiesis Section, {dagger} Bone Marrow Transplantation Program, {ddagger} Division of Pathology and Laboratory Medicine, Cleveland Clinic Foundation, Cleveland, OH 44195; and § Case Comprehensive Cancer Center, Cleveland OH 41106


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 Disclosures
 References
 
The development of laboratory tests for the diagnosis and monitoring of graft-vs-host disease (GVHD) is hampered by a lack of knowledge of minor histocompatibility Ags triggering alloresponses. We hypothesized that the unique molecular structure of the TCR could be used as a marker for the unidentified Ags and exploited for molecular monitoring of GVHD posttransplant. To identify alloreactive T cell clones, we performed in vitro allostimulation cultures for a cohort of patients undergoing hemopoietic stem cell transplantation and determined the sequence of the CDR3 of immunodominant alloreactive clones; 10 corresponding clonotypes restricted to activated T cells were identified. As an alternative method for the identification of alloreactive clones, molecular TCR analysis was applied to biopsies of GVHD-affected tissues. Culture- and biopsy-derived clonotypes were used to design sequence-specific quantitative PCR assays to monitor the levels of putative allospecific clonotypes in posttransplant blood samples and subsequent biopsies. Because of the rational design of the methods used to identify immunodominant clonotypes, we were able to follow the behavior of potentially GVHD-specific T cells during the transplant course. Based on our results, we conclude that molecular T cell diagnostics can be a powerful tool for monitoring immune responses posttransplantation.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 Disclosures
 References
 
The diagnosis of graft-vs-host disease (GVHD)3 in hemopoietic stem cell transplantation (HSCT) can be a significant clinical problem and currently there are only a few objective laboratory tests that can aid in diagnosis. In many circumstances the available diagnostic tools may not be able to distinguish GVHD lesions from other pathologies, such as drug toxicity or infection, or predict impending GVHD.

In a fully molecular HLA-matched setting, GVHD is likely induced by a mismatch in minor histocompatibility Ags (mHA). Ideally, knowledge of the Ags involved in GVHD could be used to design more accurate clinical tests and better match the donor to the recipient. However, the nature of these Ags is in most instances unknown and it is possible that for each donor/recipient pair different Ags will be involved. In addition, through tissue destruction additional Ags can be recruited and further excite the immune response and add to its complexity.

As an alternative, molecular markers of disease activity and pathology that do not rely on knowledge of the precipitating Ags could be used. Clonotypic molecular analysis of the TCR repertoire has been proposed as one of such methods (1, 2). The principle of this approach relies on the theory that, although in general GVHD is mediated by polyclonal responses, some T cell clones may be more abundant due to their affinity for or the amount and distribution of the Ags in tissue. Consequently, such immunodominant clones may be indicative of the activity of the polyclonal alloresponses. The clonotype (or sequence) of the TCR variable beta-chain (VB) CDR3 is unique and may be used as a marker of the unknown Ags (3). Additionally, clonotypes can be exploited to track the behavior of these clones during the posttransplant course (3).

Clonotypic diagnostics have been applied to studying the GVHD-related T cell repertoire posttransplant in both blood (4, 5) and tissues affected by GVHD (6, 7). Previously, we characterized the VB utilization spectrum and immunodominant clonotypes among CD8+ T cells in 29 HSCT patients and in biopsies from nine patients posttransplant (8, 9). The frequency of the CD8 and biopsy-derived clonotypes in blood posttransplant and the extent of clonotype sharing between patients were determined by sequencing and clonotype-specific PCR. Others isolated allospecific and leukemia-specific CD4+ T cell clones from mixed leukocyte reactions of nine patient/donor pairs; five immunodominant clonotypes from one patient were followed in the posttransplant course using quantitative PCR (10). A later report from the same group correlated the presence of alloreactive CTL as measured by quantitative PCR with clinical events such as biopsies and immunosuppression in another patient (2). These pilot studies provide a proof of principle for clonotypic diagnostics and underline the applicability for studying allospecific clonal T cell response in GVHD.

Despite the rational approach and precision that clonotypic diagnostics provide, its future utility will depend upon the resolution of several problems inherent to clonotypic specificity and the methods of clonotype identification. The mHA responsible for GVHD and, therefore, the clonotypes of the Ag-specific CTL clones are most likely individualized. Although CTL clonotypes can be used as surrogates for the unknown Ags, the association between the presence of the clonotype and the immune response is indirect. Those blood-derived clonotypes that are most likely alloreactive must be correlated with GVHD activity or, more suggestively, in tissues affected by GVHD.

We have used a rationally designed experimental plan to identify a putative alloprecursor CTL in hemopoietic stem cell grafts by performing in vitro allostimulation cultures for patient/donor pairs; additionally, immunodominant clonotypes were isolated from tissue biopsies for some of the patients. The clonotypes were exploited in sequence-specific molecular assays, including a highly sensitive quantitative PCR, to follow the behavior of the clones posttransplant.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 Disclosures
 References
 
Patients and controls

Blood samples pretransplant and posttransplant were obtained during clinically indicated testing from a total of 17 patients who underwent allogeneic HSCT. Peripheral blood (PB), donor bone marrow (BM), and archival (formalin-fixed, paraffin-embedded) biopsies were obtained according to protocols approved by the Institutional Review Board of the Cleveland Clinic Foundation (Cleveland, OH). Patient characteristics are listed in Table I.


View this table:
[in this window]
[in a new window]

  		Table I. Clinical features of patients receiving allogeneic HSCT

 
In vitro allostimulation

Mononuclear cells were separated from total PB and BM (BM-derived mononuclear cells) by density gradient sedimentation (Mediatech). Recipient PBMC were labeled with PKH26 (Sigma-Aldrich) as per the manufacturer’s instructions and cultured with or without donor BM mononuclear cells at a 1:10 ratio in RPMI 1640 medium (Invitrogen Life Technologies) supplemented with 10% heat-inactivated FCS, penicillin/streptomycin (1x), and IL-2 (50 U/ml) for 72 h. Positive controls consisted of cultures of donor cells stimulated with PHA (nos.1–8) or PMA (nos.9–17).

Isolation of activated and unactivated donor CD8 cells by flow cytometry

Cultures were stained with a PC5 CD8 mAb (Beckman Coulter) and a CD69 mAb labeled with FITC (BD Pharmingen). PKH–CD8+ cells were sorted to obtain donor CD8 cells expressing CD69 on an Epics Altra high-speed flow cytometer (Beckman Coulter) (Fig. 1).


Figure 1
View larger version (36K):
[in this window]
[in a new window]

  		FIGURE 1. Flow cytometric analysis of in vitro allostimulation cultures. A–C, Demonstration of donor/recipient cell distinction using the membrane-bound dye PKH26 and flow cytometry. A, Unlabeled donor cells. B, Labeled recipient cells. C, Result of coculture. Following coculture, donor cells were gated based on a lack of PKH26 fluorescence. D–G, scattergrams of cells within the PKH26– lymphocyte gate. D, No labeling. E, Positive control after stimulation with PHA in the absence of recipient cells. F, Negative control (no recipient cells added). G, Coculture of donor and recipient cells. Representative gates used for sorting are indicated by the boxes in G.

 
RNA isolation and cDNA synthesis

Total RNA was extracted from sorted donor CD8 cell fractions with TRIzol reagent (Invitrogen Life Technologies). cDNA was generated by first strand cDNA synthesis using the SuperScript II RT kit (Invitrogen Life Technologies).

CDR3 region amplification, cloning, and sequencing

cDNA was amplified from PB and BM mononuclear cells and cultured cells using PCR with 22 different VB family-specific primers and a constant region (CB) primer (11, 12). VB10 and VB19 are pseudogenes and therefore were excluded. PCR amplification was performed as described previously (13).

PCR products were gel purified using the QIAquick gel extraction kit (Qiagen) following the manufacturer’s instructions. Four microliters of the purified product was ligated into a TA vector (Invitrogen Life Technologies) and transformed into bacteria. Sequences were generated as described (9). The sequences were run on a 3100-Avant genetic analyzer (Applied Biosystems) and analyzed using the DNA Sequencing Analysis software version 3.7 (ABI Prism). A pathologically expanded immunodominant clonotype was defined based on its frequency among cloned sequences within an expanded VB family.

Isolation of genomic DNA from tissue

Ten sections (10 µm) of a paraffin-embedded biopsy were placed in lysis buffer (50 mM Tris-HCl (pH 8.6), 1 mM EDTA (pH 8.6), and 0.5% Tween 20) and incubated at 110°C for 10 min. Five microliters of proteinase K (20 mg/ml) was added to the lysate and incubated overnight at 65°C. The enzyme was inactivated at 110°C for 10 min. The sample was spun down and DNA was extracted from the aqueous phase using an Eppendorf Phase Lock Gel. Glycogen (Gentra) was used as a carrier to precipitate the DNA.

Multiplex VB CDR3 amplification

VB TCR gene rearrangements in genomic DNA isolated from biopsies were performed as described previously (9, 14). PCR products were gel purified, cloned, and sequenced as described above.

Clonotype-specific PCR

Clonotype-specific PCR primers were designed from immunodominant sequences identified in allostimulated mononuclear cells and biopsies to span 3–5 nucleotides of the D/J junction while covering as much of the NDN region as possible.

Clonotype-specific TaqMan PCR on CD8+ cells from PB

Sequences derived from expanded clones detected in the allostimulation cultures and in archival biopsies were used to design a clonotypic assay using clonotype-specific forward primers, JB-specific TaqMan probes, and a CB reverse primer (15). In brief: 15 µl of 2 x TaqMan Universal Master Mix (Applied Biosystems) were used with 3.6 µl (5 µM) of both forward clonotypic and reverse CB primers, 0.66 µl (10 µm) of JB TaqMan probe, 5 µl cDNA (diluted 1/5 with water), and 1.14 µl of water in a total volume of 30 µl. All samples were run in duplicate on a 7500 real-time PCR system (Applied Biosystems). An initial 10 min at 95°C was followed by 48 cycles of 95°C for 15 s and 60° for 60 s. For each sample an endogenous GAPDH control was run in duplicate under the same conditions as the quantitative clonotypic PCR. Relative frequencies of the clonotypes were determined with the 7500 SDS software (Applied Biosystems) using the initial graft as the calibrator sample.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 Disclosures
 References
 
Clinical features of allogeneic HSCT recipients

We hypothesized that clonotypic sequences identified in activated CTL after allostimulation could be used for the monitoring of GVHD activity in blood and tissues affected by GVHD following BM transplantation. We investigated the in vitro allostimulation-derived clonal TCR repertoire in a cohort of 17 patients who received allogeneic hemopoietic stem cell transplants (Table I). Nine patients received a graft from an HLA-matched sibling donor, five from an HLA-matched unrelated donor, and three from HLA-mismatched unrelated donors; the mismatches were at one allele (HLA-A for patients no.5 and 10, HLA-C for no.6). Eight patients developed GVHD of grade 2 or higher: five with skin GVHD, two with gastrointestinal GVHD, and one with liver, gastrointestinal, and skin GVHD. In addition, four patients developed chronic GVHD of the liver, gastrointestinal tract, or skin; the diagnosis of GVHD was confirmed with biopsy. Six patients developed CMV viremia during the posttransplant course while two others had no evidence of any infection. There was no temporal relationship between the sampling for GVHD and the detection of CMV viremia.

In vitro allostimulation of donor BM mononuclear cells

We used an in vitro allostimulation protocol to prospectively identify T cell clones (and their clonotypes) potentially involved in GVHD before transplantation in the BM lymphocyte fractions derived from donors. Activation, as measured by the expression of CD69, was used to identify allospecific CTL present in the grafts. These preprimed allospecific precursors were used as a source for the cloning of CDR3 regions and the identification of immunodominant clonotypes. For the coculture experiments, recipient "stimulator" cells were labeled with PKH dye to facilitate the distinction of target cells from effector cells by flow cytometric analysis (Fig. 1, A–C). The stimulator cells were not irradiated to preserve cell viability. Patient and donor mononuclear cells were cultured at a ratio of 1:10 for 72 h in the presence of low-dose IL-2 to generate a population of alloreactive cells large enough to be sorted and analyzed downstream. Additionally, due to the differences in cell count pretransplant between the patients, this ratio was achievable for all patient/donor pairs unlike the higher target:effector cell ratios. Donor cells cultured under the same conditions but without patient cells or exogenous polyclonal stimulation showed only baseline activation (Fig. 1D). As expected, donor cells stimulated with PHA or PMA exhibited strong activation (Fig. 1E) but little activation was seen after exposure to IL-2 alone (Fig. 1F). Distinct populations of activated (CD69 expressing) and quiescent CD8 cells were generated by allostimulation with target BM cells (Fig. 1G). However, when cultured with allogeneic target cells the recipient BM-derived mononuclear cells demonstrated varying levels of activation above those seen with IL-2 (Table II). Overall, the activation difference between allostimulated and control cultures reflects the T cell response to allogeneic targets.


View this table:
[in this window]
[in a new window]

  		Table II. In vitro allostimulation of donor CD8+ T cellsa

 
To explore the correlation between our in vitro system and alloresponses in vivo, we compared the level of alloreactive precursors present in the graft with several parameters posttransplantation. No correlation was found between the highest grade of GVHD or the number of VB expansions as measured by flow cytometry and alloprecursor frequencies.

Molecular characterization of VB usage in allostimulated donor CTL

To identify immunodominant clonotypes in the activated CD8+ donor lymphocyte population, the experimental cultures were gated on PKH26 staining; PKH26– cells were sorted for the CD8+CD69+ (or activated) and CD8+CD69– (or unactivated) fractions. The gating for the FACS sorting was based on both the control and IL2-stimulated cultures. Only CD8+ cells were sorted; CD8dim and CD4+ cells were excluded. The technique for sorting activated and unactivated CD8+ cells is illustrated in Fig. 1G. We performed PCR using a panel of VB-specific primers and cDNA derived from the activated, sorted cell fraction. Amplification of at least one VB family was observed for each in vitro allostimulation culture (range 2–12 VB families; median 9; data not shown). Positive PCR products were cloned into a TA vector and at least 10 individual colonies were sequenced per family (Fig. 2A). Sequence analysis was performed for six of the original eight patient/donor pairs. In general, the activated fraction demonstrated a lower diversity (number of clonotypes identified per total number of clones sequenced) as compared with the unactivated fraction (Fig. 2B). Thirteen immunodominant clonotypes were identified, 10 were with restricted to activated CTL (Table III). The immunodominant clonotypes generated by in vitro allostimulation ranged in percentage of redundancy (number of clones of a specific clonotype per the total number of clones sequenced) from 20 to 100% (median 43%; SD 27%). Five patients had immunodominant clones in two or more VB families, whereas one had an immunodominant clone in a single family. At the same time, amplification of the unactivated fraction and cells from the original graft yielded a broad spectrum of VB amplicons (Fig. 2B).


Figure 2
View larger version (64K):
[in this window]
[in a new window]

  		FIGURE 2. Identification of immunodominant clonotypes from in vitro allostimulated cultures for use in molecular assays. A, The research plan for identifying alloreactive immunodominant clonotypes limited to the activated (CD69+CD8+) T cells generated by in vitro allostimulation. B, Representative sequencing results for the activated (CD8+CD69+) and unactivated (CD8+CD69–) cells after allostimulation as well as the original graft (donor stem cells (SC)) are shown for three patients. The sequence of the CDR3 region includes the last invariant amino acid of the V region and the first invariant amino acid of the J region. The contribution of each clonotype to the total pool of sequences within the VB family (%VB) was determined by sequencing at least 10 clones for each family.

 

View this table:
[in this window]
[in a new window]

  		Table III. Clonotypes of CD8+CD69+ T cells expanded by in vitro allostimulationa

 
If the behavior of immunodominant CTL clones identified by allostimulation correlates with clinical events, their allospecificity can be indirectly inferred. To determine the presence and frequency of the clonotypes in the individual patients through the posttransplant course, serial samples were amplified for the families harboring immunodominant clones in three exemplary patients (nos.1, 6 and 7) and cloned, and a large number of clones were sequenced. In patient no.1, the VB 9 immunodominant clone was identified in the initial graft (Fig. 2A) and in blood derived from the recipient posttransplantation; the VB13 and 27 clonotypes were not detected. Clonotypes identified by coculture for patients no.6 (VB 6 and 24) and no.7 (VB 8, 20, and 13) were not amplified from posttransplant blood samples, although the VB 24 clonotype was found in the original graft (Fig. 3A). Additionally, the frequency of expanded clonotypes for VB 28 and 6 were analyzed in 10 healthy control individuals (Fig. 3B).


Figure 3
View larger version (14K):
[in this window]
[in a new window]

  		FIGURE 3. Frequency of immunodominant clonotypes in allostimulation cultures by direct sequence analysis. PCR amplification of CDR3 regions was performed using a VB family-specific sense primer and a CB-specific antisense primer. The frequencies were obtained by sequencing at least 10 clones for each family and represent the contribution of each clone to their specific VB families. Frequencies of immunodominant clonotypes in allostimulated precursor T cells (left for patient) and the initial graft (right) are shown. A, For patient no.1, three immunodominant clonotypes were identified in the in vitro allostimulation cultures: VB 9 JB 1.2 (filled bars), VB 27 JB 1.2 (gray bars), and VB 13 JB 2.7 (open bars). VB 6 JB 2.1 (filled bars) and VB 24 JB 2.1 (gray bars) expansions were detected in the allostimulation culture for patient no.6. VB 8 JB 1.2 (black bars), VB 20 JB 2.2 (gray bars), and BV 13 JB 2.2 (open bars) were identified for patient no.7. B, The frequency of expanded clonotypes in VB families 28 and 8 were determined for 10 healthy individuals.

 
Direct and quantitative clonotypic PCR analysis of patient samples

Direct sequencing of VB amplicons as a means of identifying and quantitating immunodominant clonotypes is very labor intensive and clinically not feasible. Therefore, we developed highly specific PCR assays that exploited the immunodominant CDR3 sequences derived from our in vitro allostimulation cultures to determine the presence and relative levels of alloresponsive T cell clones. Clonotype-specific primers were designed for the immunodominant clones of four patients: no.1 (VB13 JB 2.7 and VB 27 JB 1.2), no.5 (VB 25 JB 2.7), no.6 (VB 6 JB 2.1 and VB 24 JB 2.1), and no.7 (VB 20 JB 2.2). The specificity of the primers was determined by sequencing of amplification products generated from the activated samples in which the clonotypes were first identified. In both patients the allostimulation clonotypes were detected in samples up to 8 mo posttransplant by traditional PCR methods (data not shown).

Next, we exploited the increased level of sensitivity of quantitative PCR as compared with traditional PCR methods to design an assay to further investigate the frequency of alloreactive CTL after HSCT for four patients (no.1, 5, 6 and 7). This assay used the previously designed clonotype-specific primers, a CB primer, and probes for the JB used by the expanded clone. For analysis of relative clonotype frequencies, the total graft was used as the calibrator sample.

Patient no.1 had chronic phase CML and underwent a matched sibling donor allogeneic BM transplant. Of three posttransplant samples tested for this patient, the first two were obtained while the patient still had molecular evidence of disease but was on immunosuppressive therapy and without evidence of GVHD (Fig. 4). The third posttransplant specimen was obtained after a donor lymphocyte infusion from which the patient achieved a molecular remission but also developed skin and gastrointestinal GVHD requiring the resumption of immunosuppressive therapy. However, at the time the specimen was obtained the GVHD had resolved. The frequency of the VB 13 JB 2.7 clonotype identified in coculture for patient no.1 was elevated in three posttransplant samples but showed a general decrease over time (Fig. 4). In contrast, the VB 27 JB 1.2 clonotype was present at a frequency less than that in the graft for the first and last posttransplant sample but was elevated in the second sample; these findings are in accordance with the direct sequencing analysis (data not shown).


Figure 4
View larger version (28K):
[in this window]
[in a new window]

  		FIGURE 4. Correlation of clinical parameters and allospecific clonotypes posttransplantation. Clinical parameters (A), absolute lymphocyte count (B), and relative frequency of in vitro allostimulation-derived clonotypes (C) are shown for patient no.1. DLI, Donor lymphocyte infusion.

 
Patient no.6 underwent an unrelated donor allogeneic PB stem cell transplant but developed graft failure and had evidence of residual myelodysplastic syndrome on day 39 posttransplant at which time immunosuppressive therapy was discontinued. The frequency of the VB 6 JB 2.1 clonotype in a pretransplant sample from this patient was higher than that of the total graft, and the frequency increased posttransplant (Fig. 5A). The VB 24 JB 2.1 clonotype also increased posttransplant as compared with the pretransplant sample.


Figure 5
View larger version (28K):
[in this window]
[in a new window]

  		FIGURE 5. Quantitative PCR analysis of allospecific clonotypes in blood posttransplantation and correlation with GVHD and treatment. The initial graft was used as the calibrator sample to determine relative frequencies of the clonotypes. Relative frequencies were correlated with immunosuppression, antifungal and antiviral treatment, and the occurrence of GVHD and infection. A, For patient no.6, the relative frequency of two immunodominant clonotypes, VB 6 JB 2.1 and VB 24 JB 2.1, was determined in the patient pretransplant (Pre-Tx) and in two posttransplant samples. MMF, Mycophenolate mofetil. B and C, The relative frequencies of a VB 20 JB 2. 2 clonotype in patient no.7 (B) and a VB 25 JB 2.7 clonotype in patient no.5 (C) were determined at two time points posttransplantation.

 
Patient no.5 had refractory chronic lymphocytic leukemia and had received an unrelated allogeneic HSCT. The patient had developed GVHD of the colon by day 42 posttransplant that was successfully managed with further immunosuppression. A VB 25 JB 2.7 clonotype was subsequently identified at a decreased frequency as compared with the initial graft up to 3 mo posttransplant (Fig. 5C). Patient no.7 had chronic myelogenous leukemia and also received an unrelated donor allogeneic HSCT. This was complicated by the development of skin and gastrointestinal acute GVHD requiring additional immunosuppressive treatment. Thereafter, the levels of a VB 20 JB 2.2 clonotype identified by an allostimulation culture increased steadily posttransplant (Fig. 5B).

Molecular characterization of immunodominant clonotypes in archival biopsies

Biopsies of tissues affected by GVHD can also serve as a source of putatively alloreactive CTL clones. To identify expanded immunodominant clonotypes associated with GVHD, we obtained a total of 15 paraffin-embedded biopsies for five of the patients for whom an in vitro allostimulation coculture was performed (Table IV). Ten biopsies were from the gastrointestinal tract and five were from skin. Five biopsies were diagnostic of GVHD grades 1 or 2 while 10 were negative for GVHD or represented drug eruptions. As a positive control for quality the extracted DNA was amplified with GAPDH-specific primers; products were obtained for all but one biopsy sample (Bx3 for patient no. 6). A multiplex VB PCR assay performed on GAPDH+ samples amplified sequences for all but patient no.6 Bx2 and no.7 Bx2. Sequencing of a large number of clones identified immunodominant clonotypes in 11 biopsies (frequency 19–95%, median 30%; Fig. 6). No sharing of clonotypes was found between patients; however, an identical VB 24 JB 1.5 clonotype was identified in skin and gut biopsies from patient no.7. When the sequences derived from allostimulation cocultures and posttransplant PB samples were compared with the biopsy-derived immunodominant clonotypes, no overlap was identified. Interestingly, for patient no.1 the expanded CTL clones in both the coculture and two biopsies (stomach and skin) belonged to the VB 27 family, even though the CDR3 sequences were not similar.


View this table:
[in this window]
[in a new window]

  		Table IV. Immunodominant sequences identified in biopsies by VB multiplex PCRa

 

Figure 6
View larger version (55K):
[in this window]
[in a new window]

  		FIGURE 6. Immunodominant clonotypes and their frequency by sequencing in biopsies. A, Skin punch biopsy from patient no.1 of a lesion suspected to be a manifestation of skin GVHD and demonstrating interface inflammation consistent with GVHD grade 2. The sequence of a VB 27 JB 2.3 clone with a frequency of 25% is given. B, Skin punch biopsy from the right neck of patient no.7, demonstrating interface inflammation consistent with GVHD grade 2. A VB 24 JB 1.5 clone (sequence given) was found at a frequency of 92%. C, Histologically unremarkable biopsy of the duodenum of patient no.6 with no evidence of GVHD. Two immunodominant clones, VB 6 JB 1.5 (60% frequency) and VB 6 JB 2.3 (37% frequency), were identified.

 
Biopsy-derived, clonotype-specific primers for patients no. 1 (no.1biop4 VB 27A and B) and 6 (no. 6biop1 VB6) were designed and used in a quantitative PCR assay on posttransplant samples. No expression of either clonotype from patient no.1 was detected posttransplantation. However, sequences derived from a gastrointestinal tract biopsy for patient no.6 were detected in peripheral blood samples posttransplant, as well as pretransplant and in the graft.


   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
Disclosures
 References
 
Technologies based on the analysis of the molecular structure of the rearranged TCR offer many advantages for the investigation of cellular immune responses, notably in situations in which the precise nature of the target Ags is not known (as in the case of a multitude of possible mHa). However, while TCR-based assays overcome some of the limitations of functional assays, including low reproducibility and a large time expense, molecular methods have other sets of inherent obstacles. This study is a continuation of our earlier work in which we identified immunodominant T cell clones in blood (1) and tissues (9) of patients up to 7 mo following allogeneic HSCT. These clones were hypothesized to correspond to most permissive alloreactive CTL specificities.

Using a rationally based molecular approach, we devised methods that would allow for selective recognition of CTL specificities directed against alloantigens. We chose to use two independent strategies: an in vitro allostimulation culture system for the generation of alloreactive T cells and an analysis of T cell clones infiltrating those tissues affected by GVHD. Rationally, graft-derived clonotypes can be used for clonotypic assays performed in blood and tissue biopsies after HSCT. Conversely, tissue-derived clonotypes can either be used retrospectively to determine the frequency of allospecific lymphocyte precursors in the grafts or prospectively during the subsequent posttransplant course. Our experimental strategies were augmented by the application of new laboratory techniques, including a VB multiplex PCR that has many advantages over traditional PCR methods (9). Once the sequence and frequency of putative alloreactive clonotypes are identified, their levels in posttransplant samples can be measured using a clonotype-specific quantitative PCR assay. Use of this method in the diagnostic setting provides proof of concept for the feasibility of molecular TCR-based diagnostics not only in GVHD but also in other immune-mediated conditions such as infections or autoimmune diseases (14).

In the current study, putatively alloreactive precursor T cells or their TCR clonotypes present in the HSC grafts were identified by coculture with recipient cells obtained before the onset of pretransplant conditioning and their activation was measured by positivity for an activation marker. Immune recovery post-HSCT is a complex procedure and is affected by a large variety of factors, including preparative regimens, transplant types, immunosuppressive treatments, infections, and the underlying diagnosis (Figs. 4 and 5). Our study was designed to test the feasibility of using molecular methods to identify alloresponsive T cell clones before transplantation and therefore was limited to allotransplants. To assay the widest sample pool, patients undergoing either matched sibling or matched unrelated transplants were selected; we limited the number of donor/recipient pairs studied to achieve the maximal comparability. Increased levels of activation upon challenge with recipient cells were observed in a majority of the donor/recipient pairs. The numbers of activated cells induced in response to allostimulation did not correlate with the degree of mismatch, suggesting that it is mostly driven by responses to mHa. Moreover, the lack of correlation of preprimed alloreactive precursor frequency and the grade of GVHD suggests that the degree of mismatch in mHA, rather than priming, is of importance in GVHD.

The conditions chosen for the coculture experiments may have affected both the quantity and quality of activation. In addition, the flow cytometric method used in these studies does not distinguish the specificity of activation. The preprimed T cells identified in the graft are not likely to be exclusively alloreactive but could also represent those cross-reactive for recipient antigens (16). If the alloreactive Ag is similar to a common Ag, such as one for the CMV or influenza, the frequency of preprimed CTL identified by coculture might be higher than expected from purely naive specificities. Conversely, if the alloantigen resembles an Ag for which the donor has had little or no exposure, the level of activation in the allostimulation culture may be low or nonexistent (only naive T cell precursors present).

The T cell repertoire of the alloreactive CTL, as measured by a VB family-specific PCR assay, was more restricted than that of total lymphocytes. Perhaps not surprisingly this finding may reflect artificial oligoclonality, most likely due to the relatively small numbers of cells cultured and/or recovered. A more probable explanation is that immunodominant, allospecific, preprimed CTL clones exist within the tissues and are more likely to be detected there than in a polyclonal background such as that found in the peripheral circulation. However, we were able to identify immunodominant T cell clones for most of the coculture experiments molecularly analyzed. Several of the clonotypes were present at a very high frequency, approaching levels more reminiscent of clonal expansions seen in other immune conditions (i.e., CMV and EBV) (17, 18, 19) or in large granular lymphocytic leukemia (13, 14) and similar to those previously reported in blood posttransplantation (5, 6, 8, 20). No VB family was over-represented among the clonotypes and no sharing was identified between the patients. In comparison, the clonal size of VB family expansions detected in healthy individuals is smaller; the largest expansion contributed to 21.4% of the VB family.

It is true that not all of the potential tissue-specific alloantigens are present in the cells used as targets for the allostimulation cultures; however, it is possible that some are. In addition, it is also possible that the mononuclear cells used as stimulators present peptides derived from tissue-specific Ags, as Ag-derived peptides are the target for the T cell response. Organ presentation is influenced not by Ag distribution alone but perhaps also by tissue-specific accessibility, organ injury or the organ-specific cytokine milieu. The organ specificity of GVHD is a subject of much speculation; clearly, the presence of tissue-specific minor alloantigens is one possibility.

Although we demonstrated that in vitro allostimulation could potentially be a powerful tool for identifying alloreactive CTL, we hypothesized that true GVHD-specific clones would be easier to identify by the molecular analysis of tissues affected by GVHD (7, 9). Additionally, studying the repertoire of infiltrating T cells would allow us to recognize clonotypes specific for tissue-restricted Ags that would not have been activated in our in vitro assay by stimulation with PB. Although it has not been definitively shown that T cells infiltrating GVHD-affected tissues are causative, it has been demonstrated that these clones are alloreactive (2, 21). In addition, the number of infiltrating CTL is a minute fraction of that found in peripheral circulation, making identification of expanded T cell clones simpler than in a large polyclonal background. However, T cell infiltration may not be consistent throughout the affected tissues and the frequency of any given T cell clone may be artificially elevated due to a low number of T cells actually infiltrating the tissue, particularly if biopsies are obtained after the administration of further immunosuppressive therapy.

We identified immunodominant clonotypes in archival biopsies from a subset of our cohort, including ones scored as GVHD, normal, or drug eruption. The presence of immunodominant clonotypes in lesions without a pathologic diagnosis of GVHD may suggest that they represent a very early stage of GVHD not recognized by standard morphologic analysis. As in the allostimulation cultures, we identified multiple coimmunodominant clones in the biopsies. Each was unique to a single patient, implying that the Ags the clones were reacting against were "private" or individual specific. However, in patient no. 7 we found identical clonotypes in biopsies of two distinct sites, the skin and gastrointestinal tract. Such a finding suggests that this clone may have a role in the pathogenesis of GVHD in both tissues.

Once identified, clonotypes can be exploited to monitor the putative alloreactive clone during the posttransplant course. As an initial analysis, we used a VB family -specific PCR and sequencing assay. The samples were obtained during times of active GVHD as well as during times with no clinical evidence of GVHD (Figs. 4 and 5). Allostimulation-derived immunodominant clonotypes were detected in posttransplant blood samples for one exemplary patient only. Possibly, at the time of sampling the clones may have already left the circulatory system and infiltrated the tissue sites affected by GVHD. Most likely, this finding reflects the level of sensitivity of sequence analysis; if a clonotype is extremely rare, a prohibitively large number of clones may need to be sequenced. It still remains unclear whether the clones responsible for GVHD can be detected in blood, although we have detected tissue-derived clonotypes posttransplant (9).

Sequencing of a large number of CDR3 regions is a feasible albeit labor-intensive method of molecular monitoring. Traditional PCR using clonotypic primers is a rapid and sensitive method for detecting the presence of immunodominant clonotypes; however, the absolute frequency of the alloreactive clone may play an important role in the onset of GVHD. Quantitative PCR with sequence-specific primers has been shown to be a powerful tool for studying alloreactive CTL posttransplantation (2, 10). The presence of identical immunodominant clonotypes from allostimulation cultures and tissue biopsies in the same patient or in tissues and posttransplant blood samples would help to support the functional role the clonotypes play in the pathogenesis of GVHD.

We were able to detect the presence of sequences derived from coculture in posttransplant samples for four patients using clonotype-specific quantitative PCR. For example, in one of our patients the VB 27 JB 1.2 clonotype peaked at the second posttransplant time point, whereas the VB 13 JB 2.7 clonotype frequency remained relatively stable (Fig. 3A). This latter clonotype may represent a T cell clone resistant to conventional immunosuppressant agents such as cyclosporine or steroids. Alternatively, this clonotype may have potentially been important in achieving a graft-vs-leukemia response from which the patient achieved molecular remission. It is important to note that although this patient did develop GVHD, all samples tested were collected at time points where there was no clinical evidence of GVHD.

Biopsy-derived clonotypes were more difficult to identify in blood even when using clonotype-specific quantitative PCR. Two skin-derived clonotypes for patient no.1 were undetectable posttransplant. It is possible that at the time the blood samples were obtained the alloreactive CTL had already left the circulation and entered the affected tissues. It is interesting to note that an immunodominant clonotype from the same VB family (VB 27) was identified by coculture and detected posttransplant. This may reflect an initial polyclonal response within the VB family of which only one clone was capable of infiltrating the skin, leaving other immunodominant clones in the periphery. The gut-derived clonotype for patient no. 6 was identified in both pretransplant and posttransplant samples as well as the graft. This particular clone may be specific for a common intestinal pathogen given its ubiquity among the different samples.

In summary, we have demonstrated that in vitro allostimulation may be a powerful tool for the identification of preprimed alloreactive CTL in HSCT. Additionally, analysis of biopsy material can be used to identify potentially GVHD-related immunodominant T cell clones. As tissue analysis can only be performed after a patient has developed GVHD, only coculture experiments can be truly predictive. Although the detection of allostimulation- and tissue-derived clonotypes in the posttransplant course and their correlation with clinical events are suggestive of a role for these CTL in GVHD, ultimately this will need to be determined by a functional analysis of these putative alloreactive T cell clones. However, the studies presented here are proof of a principle for molecular T cell diagnostics and underline its potential utility in the clinical setting. Future studies will include an analysis of mHa-specific CTL using tetramers and further refinement of coculture conditions.


   Disclosures
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 Disclosures
References
 
The authors have no financial conflict of interest.


   Footnotes
 
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1 C.L.O. designed the research, performed the research, analyzed data, and wrote the paper. L.G. performed the research and analyzed data. R.M.S., Z.M., and R.D. analyzed clinical data. R.M.S. and B.B. provided patient samples for the study. Y.N. performed research. R.T. analyzed biopsies. J.P.M. designed the research and analyzed data. Back

2 Address correspondence and reprint requests to Dr. Jaroslaw P. Maciejewski, R40 Taussig Cancer Center, The Cleveland Clinic Foundation, 9600 Euclid Avenue, Cleveland, OH 44195. Back

3 Abbreviations used in this paper: GVHD, graft -vs-host disease; BM, bone marrow; Bx, biopsy sample; CB, beta-chain constant region; HSC, hemopoietic stem cell transplant; mHA, minor histocompatibility Ag; PB, peripheral blood; VB, variable beta-chain. Back

Received for publication June 27, 2006. Accepted for publication April 22, 2007.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 Disclosures
 References
 

  1. O’Keefe, C. L., R. Beck, A. Rodriguez, E. Howe, K. Bell, B. Bolwell, J. P. Maciejewski. 2003. Molecular T cell receptor diagnostics can be used to identify shared clonotypes in blood and tissue during GVHD. Blood 102: 714a
  2. Michalek, J., R. H. Collins, B. J. Hill, J. M. Brenchley, D. C. Douek. 2003. Identification and monitoring of graft-versus-host specific T-cell clone in stem cell transplantation. Lancet 361: 1183-1185. [Medline]
  3. O’Keefe, C. L., A. M. Risitano, J. P. Maciejewski. 2004. Clinical implications of T cell receptor repertoire analysis after allogeneic stem cell transplantation. Hematology 9: 189-198. [Medline]
  4. Godthelp, B. C., M. J. van Tol, J. M. Vossen, P. J. van den Elsen. 1999. T-cell immune reconstitution in pediatric leukemia patients after allogeneic bone marrow transplantation with T-cell-depleted or unmanipulated grafts: evaluation of overall and antigen-specific T-cell repertoires. Blood 94: 4358-4369. [Abstract/Free Full Text]
  5. Matsutani, T., T. Yoshioka, Y. Tsuruta, S. Iwagami, T. Toyosaki-Maeda, T. Horiuchi, A. B. Miura, A. Watanabe, G. Takada, R. Suzuki, M. Hirokawa. 2000. Restricted usage of T-cell receptor {alpha}-chain variable region (TCRAV) and T-cell receptor beta-chain variable region (TCRBV) repertoires after human allogeneic haematopoietic transplantation. Br. J. Haematol. 109: 759-769. [Medline]
  6. Hirokawa, M., T. Matsutani, H. Saitoh, Y. Ichikawa, Y. Kawabata, T. Horiuchi, A. Kitabayashi, T. Yoshioka, Y. Tsuruta, R. Suzuki, et al 2002. Distinct TCRAV and TCRBV repertoire and CDR3 sequence of T lymphocytes clonally expanded in blood and GVHD lesions after human allogeneic bone marrow transplantation. Bone Marrow Transplant. 30: 915-923. [Medline]
  7. Margolis, D. A., J. T. Casper, A. D. Segura, T. Janczak, L. McOlash, B. Fisher, K. Miller, J. Gorski. 2000. Infiltrating T cells during liver graft-versus-host disease show a restricted T-cell repertoire. Biol. Blood Marrow Transplant. 6: 408-415. [Medline]
  8. O’Keefe, C. L., A. Rodriguez, R. M. Sobecks, B. Bolwell, J. P. Maciejewski. 2004. Molecular TCR diagnostics can be used to identify shared clonotypes after allogeneic hematopoietic stem cell transplantation. Exp. Hematol. 32: 1010-1022. [Medline]
  9. Beck, R. C., M. Wlodarski, L. Gondek, K. S. Theil, R. J. Tuthill, R. Sobeck, B. Bolwell, J. P. Maciejewski. 2005. Efficient identification of T-cell clones associated with graft-versus-host disease in target tissue allows for subsequent detection in peripheral blood. Br. J. Haematol. 129: 411-419. [Medline]
  10. Michalek, J., R. H. Collins, H. P. Durrani, P. Vaclavkova, L. E. Ruff, D. C. Douek, E. S. Vitetta. 2003. Definitive separation of graft-versus-leukemia- and graft-versus-host-specific CD4+ T cells by virtue of their receptor beta loci sequences. Proc. Natl. Acad. Sci. USA 100: 1180-1184. [Abstract/Free Full Text]
  11. Gorski, J., M. Yassai, X. Zhu, B. Kissela, B. Kissella, C. Keever, N. Flomenberg. 1994. Circulating T cell repertoire complexity in normal individuals and bone marrow recipients analyzed by CDR3 size spectratyping: correlation with immune status. J. Immunol. 152: 5109-5119. [Abstract]
  12. Pannetier, C., J. Even, P. Kourilsky. 1995. T-cell repertoire diversity and clonal expansions in normal and clinical samples. Immunol. Today 16: 176-181. [Medline]
  13. O’Keefe, C. L., M. Plasilova, M. Wlodarski, A. M. Risitano, A. R. Rodriguez, E. Howe, N. S. Young, E. Hsi, J. P. Maciejewski. 2004. Molecular analysis of TCR clonotypes in LGL: a clonal model for polyclonal responses. J. Immunol. 172: 1960-1969. [Abstract/Free Full Text]
  14. Wlodarski, M., C. L. O’Keefe, E. C. Howe, A. Risitano, A. Rodriguez, I. Warshawsky, T. P. Loughran, Jr, J. P. Maciejewski. 2005. Pathologic clonal cytotoxic T cell responses: non random nature of the T cell receptor in large granular lymphocytic leukemia. Blood 106: 2769-2780. [Abstract/Free Full Text]
  15. Wlodarski, M. W., L. P. Gondek, Z. P. Nearman, M. Plasilova, M. Kalaycio, J. P. Maciejewski. 2006. Molecular strategies for detection and quantitation of clonal cytotoxic T cell responses in aplastic anemia and myelodysplastic syndrome. Blood. 108: 2632-2641. [Abstract/Free Full Text]
  16. Heeger, P. S., N. S. Greenspan, S. Kuhlenschmidt, C. Dejelo, D. E. Hricik, J. A. Schulak, M. Tary-Lehmann. 1999. Pretransplant frequency of donor-specific, IFN-{gamma}-producing lymphocytes is a manifestation of immunologic memory and correlates with the risk of posttransplant rejection episodes. J. Immunol. 163: 2267-2275. [Abstract/Free Full Text]
  17. Gillespie, G. M., M. R. Wills, V. Appay, C. O’Callaghan, M. Murphy, N. Smith, P. Sissons, S. Rowland-Jones, J. I. Bell, P. A. Moss. 2000. Functional heterogeneity and high frequencies of cytomegalovirus-specific CD8+ T lymphocytes in healthy seropositive donors. J. Virol. 74: 8140-8150. [Abstract/Free Full Text]
  18. Kuzushima, K., Y. Hoshino, K. Fujii, N. Yokoyama, M. Fujita, T. Kiyono, H. Kimura, T. Morishima, Y. Morishima, T. Tsurumi. 1999. Rapid determination of Epstein-Barr virus-specific CD8+ T-cell frequencies by flow cytometry. Blood 94: 3094-3100. [Abstract/Free Full Text]
  19. Peggs, K., S. Verfuerth, A. Pizzey, J. Ainsworth, P. Moss, S. Mackinnon. 2002. Characterization of human cytomegalovirus peptide-specific CD8+ T-cell repertoire diversity following in vitro restimulation by antigen-pulsed dendritic cells. Blood 99: 213-223. [Abstract/Free Full Text]
  20. Horiuchi, T., M. Hirokawa, Y. Kawabata, A. Kitabayashi, T. Matsutani, T. Yoshioka, Y. Tsuruta, R. Suzuki, A. B. Miura. 2001. Identification of the T cell clones expanding within both CD8+CD28+ and CD8+CD28– T cell subsets in recipients of allogeneic hematopoietic cell grafts and its implication in post-transplant skewing of T cell receptor repertoire. Bone Marrow Transplant. 27: 731-739. [Medline]
  21. Gaschet, J., M. A. Trevino, M. Cherel, R. Vivien, A. Garcia-Sahuquillo, M. M. Hallet, M. Bonneville, J. L. Harrousseau, R. Bragado, N. Milpied, H. Vie. 1996. HLA-target antigens and T-cell receptor diversity of activated T cells invading the skin during acute graft-versus-host disease. Blood 87: 2345-2353. [Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 J. Immunol.Home page
K. Sugimoto, M. Murata, S. Terakura, and T. Naoe
CTL Clones Isolated from an HLA-Cw-Mismatched Bone Marrow Transplant Recipient with Acute Graft-Versus-Host Disease
J. Immunol., November 1, 2009; 183(9): 5991 - 5998.
[Abstract] [Full Text] [PDF]

Home page
 BloodHome page
W. J. Murphy
Raising the spectra of T-cell profiling
Blood, October 15, 2008; 112(8): 3008 - 3009.
[Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by O’Keefe, C. L.
Right arrow Articles by Maciejewski, J. P.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by O’Keefe, C. L.
Right arrow Articles by Maciejewski, J. P.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS