doi:10.1182/blood-2004-10-4045Prepublished online May 24, 2005;2005 106: 2769-2780   

 Ilka Warshawsky, Thomas P. Loughran, Jr and Jaroslaw P. MaciejewskiMarcin W. Wlodarski, Christine O'Keefe, Evan C. Howe, Antonio M. Risitano, Alexander Rodriguez, 

receptor restriction in large granular lymphocyte leukemia−T-cellPathologic clonal cytotoxic T-cell responses: nonrandom nature of the

http://bloodjournal.hematologylibrary.org/content/106/8/2769.full.htmlUpdated information and services can be found at:

(4212 articles)Neoplasia   � (5124 articles)Immunobiology   �

Articles on similar topics can be found in the following Blood collections

http://bloodjournal.hematologylibrary.org/site/misc/rights.xhtml#repub_requestsInformation about reproducing this article in parts or in its entirety may be found online at:

http://bloodjournal.hematologylibrary.org/site/misc/rights.xhtml#reprintsInformation about ordering reprints may be found online at:

http://bloodjournal.hematologylibrary.org/site/subscriptions/index.xhtmlInformation about subscriptions and ASH membership may be found online at:

Copyright 2011 by The American Society of Hematology; all rights reserved.Washington DC 20036.by the American Society of Hematology, 2021 L St, NW, Suite 900, Blood (print ISSN 0006-4971, online ISSN 1528-0020), is published weekly    

For personal use only. by guest on December 18, 2013. bloodjournal.hematologylibrary.orgFrom For personal use only. by guest on December 18, 2013. bloodjournal.hematologylibrary.orgFrom

IMMUNOBIOLOGY

Pathologic clonal cytotoxic T-cell responses: nonrandom nature of the T-cell–receptor restriction in large granular lymphocyte leukemiaMarcin W. Wlodarski, Christine O’Keefe, Evan C. Howe, Antonio M. Risitano, Alexander Rodriguez, Ilka Warshawsky,Thomas P. Loughran Jr, and Jaroslaw P. Maciejewski

T-cell large granular lymphocyte (T-LGL)leukemia is a clonal lymphoproliferationof cytotoxic T cells (CTLs) associatedwith cytopenias. T-LGL proliferationseems to be triggered/sustained by anti-genic drive; it is likely that hematopoieticprogenitors are the targets in this pro-cess. The antigen-specific portion of theT-cell receptor (TCR), the variable beta(VB)–chain complementarity-determiningregion 3 (CDR3), can serve as a molecularsignature (clonotype) of a T-cell clone.We hypothesized that clonal CTL prolifera-

tion develops not randomly but in thecontext of an autoimmune response. Weidentified the clonotypic sequence ofT-LGL clones in 60 patients, including 56with known T-LGL and 4 with unspecifiedneutropenia. Our method also allowed forthe measurement of clonal frequencies; adecrease in or loss of the pathogenicclonotype and restoration of the TCRrepertoire was found after hematologicremission. We identified 2 patients withidentical immunodominant CDR3 se-quence. Moreover, we found similarity

between multiple immunodominant clono-types and codominant as well as a nonex-panded, “supporting” clonotypes. Thedata suggest a nonrandom clonal selec-tion in T-LGL, possibly driven by a com-mon antigen. In contrast, the physiologicclonal CTL repertoire is highly diverseand we were not able to detect any signifi-cant clonal sharing in 26 healthy controls.(Blood. 2005;106:2769-2780)

© 2005 by The American Society of Hematology

Introduction

Physiologic and pathologic cellular immune responses to antigenicproteins can be polyclonal but can also be highly polarized forstrong antigenic stimuli.1-7 Due to the diversity of peptide process-ing and HLA backgrounds8-10 responsible for antigen presentation,the patterns of clonal utilization are of complex nature.11-13

Proliferation of an individual T-cell clone may be difficult todiscriminate in the context of the simultaneous polyclonal expan-sions but in some disorders, such as viral infections, highlypolarized responses were described.14-16 Consequently, under cer-tain circumstances an immunodominant clonal expansion may beso significant that it is detectable above the polyclonal background.It can result in the overrepresentation of the corresponding T-cellreceptor (TCR) and consequently the entire � (A)– or � (B)–chainfamilies.14,16-18 Complementarity-determining region 3 (CDR3) ofthe TCR VB chain is highly unique and can be used as a marker ofindividual T-cell clones and as a measure of the T-cell repertoirediversity.11,12,19 TCR � chain is less suitable for detection of��–T-cell clonality due to the paradigm of dual receptor T cell: upto 30% of rearranged human peripheral blood T cells express 2distinct functional � chains as opposed to only one � chain.20 Thestructure of the VB CDR3 region results from somatic VDJ generecombination and junctional diversification.11,12,19 Due to theNDN modification of � chain, VB CDR3 region is the most diverseregion of the TCR.21 While TCR CDR3 region appears to mostdirectly ligate to the antigenic peptide presented in the context of

HLA, CDR1 and CDR2 loops have a major function in stabilizingthe ligated CDR3 regions.22 In addition, VB CDR3 sequencecovers over half of the peptide surface23 and appears to be highlyimportant in determining ligand specificity.24 Finally, sequencesubstitution of VB CDR3 region can abolish the response, asconfirmed by site-directed mutagenesis studies.25-28

Analysis of antigen-specific cells sorted using, eg, tetramers,demonstrates that a specific antigenic epitope can be recognized bya variety of unique TCR clonotypes.7,29,30 However, in the presenceof identical HLA restriction element, TCRs with identical VBCDR3 regions strongly suggest recognition of the same antigenicpeptides, assuming that VB clonality is associated with the usage ofthe same � chain.31-34

Highly significant clonal expansions occur in large granularlymphocyte (LGL) leukemia, a disease resulting from the prolifera-tion of individual cytotoxic T-cell (CTL) clones.1,10,35,36 Tradition-ally, this condition is diagnosed through detection of an abnormalCD3�CD8�CD57� cell population and the presence of a clonalTCR gene rearrangement.36-38 Moreover, the presence of LGLclones can often be uncovered using V� antibodies.39,40 Molecularidentification of the clonal CTL expansions can establish the exactdiagnosis of LGL leukemia but also possibly other diseases withunderlying T-cell lymphoproliferations.10,19,41,42 The mechanism oftransformation of the LGL clone is not well understood and thedisease often does not behave as a typical malignancy; exuberant

From the Experimental Hematology and Hematopoiesis Section, TaussigCancer Center, and the Department of Clinical Pathology, Cleveland ClinicFoundation, Cleveland, OH; Institute of Medical Immunology, Charite MedicalSchool, Berlin, Germany; Penn State Cancer Institute, Hershey, PA; andHematology Branch, National Heart, Lung, and Blood Institute (NHLBI),National Institutes of Health (NIH), Bethesda, MD.

Submitted October 20, 2004; accepted April 12, 2005. Prepublished online asBlood First Edition Paper, May 24, 2005; DOI 10.1182/blood-2004-10-4045.

Supported in part by grants RO1-HL043429 (J.P.M.), U54-RR019397 (J.P.M.),

and RO1-CA113792 (J.P.M.); a grant from the Aplastic Anemia and MDSInternational Foundation (J.P.M.); and a generous gift from the Trotter family.

Reprints: Jaroslaw P. Maciejewski, Taussig Cancer Center/R40, 9500 EuclidAvenue, Cleveland, OH 44195; e-mail: maciejj@ccf.org.

The publication costs of this article were defrayed in part by page chargepayment. Therefore, and solely to indicate this fact, this article is herebymarked ‘‘advertisement’’ in accordance with 18 U.S.C. section 1734.

© 2005 by The American Society of Hematology

2769BLOOD, 15 OCTOBER 2005 � VOLUME 106, NUMBER 8

For personal use only. by guest on December 18, 2013. bloodjournal.hematologylibrary.orgFrom

accumulation of malignant cells is often absent and the acquisitionof a more malignant phenotype is rare. Instead, clonal CTLs inLGL leukemia while apparently dysregulated, appear to retainsome physiologic properties of normal CTLs and resemble anexaggerated response to an immunodominant antigen. One of thetheories postulates that the initial events in the genesis of LGLleukemia may involve an underlying immune response to exog-enous and cross-reactive or intrinsic antigens. It is possible that,prior to the outgrowth of the immunodominant LGL clone, such aresponse may be polyclonal in nature. Clinical evidence supportsthis notion. LGL leukemia often occurs in the context of a specificautoimmune pathology.1,10 LGL-leukemia–associated paraneoplas-tic syndrome can involve isolated cytopenias.35,43-47 Some authorshypothesize that the mechanisms of cytopenia include secretion ofFas-L/soluble Fas receptors that are constitutively expressed byLGL cells and also found elevated in patient’s serum.48,49 However,lineage-specific cytopenias cannot be easily explained by displace-ment of hematopoiesis or a global cytokine effect. Consequently, itis possible that the TCR specificity of the LGL clone determines theclinical presentation spectrum by highly specific recognition andkilling of individual hematopoietic cell lineages. Thus, the transfor-mation event may not be random and may occur in the context of apreexisting immune response (see for example Starkebaum et al47).The nature of this event remains unknown but may includeblockade of apoptotic pathways or persisting upregulation ofcell-cycle activating or antiapoptotic mechanisms. For example,constitutive activation of signal transducers and activators oftranscription 3 (STAT3) has been described in LGL cells.50 It wasalso shown that patient’s LGL cells, although resistant to Fas-mediated apoptosis, express high levels of Fas, similar to phyto-hemagglutinin (PHA)–activated normal T cells, suggesting an invivo antigen activation.51 Of note is that the transforming eventsmay involve the memory cell, which feeds into the mature effectorCTL compartment.52

The extreme expansion of LGL clones can be easily character-ized and there is no interference from other codominant clonescontributing to a polyclonal T-cell response. Therefore, LGLleukemia can serve as a simplified model of natural and lessexaggerated polyclonal CTL responses. Such processes may in-clude, for example, typical aplastic anemia53 and classic autoim-mune diseases such as rheumatoid arthritis and multiple sclero-sis.54-57 Of note is that in all of these conditions, evolution of LGLleukemia has been described,45,47,58-60 and the LGL clone may haveevolved from an initially polyclonal process. The TCR in LGLleukemia had been investigated at DNA sequence level in only alimited number of cases,54,59,61-63 and perhaps not surprisingly (dueto diversity of the HLA background of patients) identical specifici-ties were not found.

Here, we systematically studied the TCR repertoire in a largecohort of patients with LGL leukemia. For that purpose we used therearranged VB CDR3 region as a clonal marker and focused ourwork on TCR � rather than � chain. We hypothesize that theevolution of the LGL clone may not be a random event butrepresents an overreactive response of an individual CTL clone inthe context of an otherwise polyclonal reaction. Consequently,shared TCR specificities may be found not only between theexpanded clonotypes within the TCR repertoire of an individualpatient but also between immunodominant clones and thosecontained within polyclonal background. Moreover, if LGL leuke-mia reactivity is directed against an antigen also present in healthyindividuals, identical clonotypes will be found in these individualsalbeit at a very low level. We have studied disease-associated

immunodominant clonotypes in a large cohort of LGL leukemiapatients and compared their CDR3 sequences against a clonotypicdatabase containing CDR3 regions from healthy individuals se-quenced in our laboratory and CDR3 sequences described in theliterature. Our results demonstrate that shared clonotypic se-quences can be identified and may be used to track the presence ofthe corresponding CTL clones for diagnostic purposes.

Patients, materials, and methods

Patients and controls

We obtained peripheral blood specimens from 60 patients with known LGLleukemia, 13 patients with single-lineage cytopenias, and 26 healthycontrols. Informed consent for sample collection was given by theindividuals according to protocols approved by the Institutional ReviewBoard of the Cleveland Clinic Foundation (Cleveland, OH), the NationalHeart, Lung, and Blood Institute, National Institutes of Health (Bethesda,MD), and the Penn State Cancer Institute (Hershey, PA). Previously, flowcytometry and CDR3 results were reported in part for 29 of these patients.63

The original diagnosis of LGL leukemia was established by clinical andlaboratory parameters as suggested by Berliner et al,38 Semenzato et al,64

and Herling et al65 but we also included patients with otherwise unexplainedcytopenias and those with an expansion in a specific CTL population whodid not fulfill the current criteria for LGL leukemia.66 A diagnosis ofmyelodysplastic syndrome (MDS), when applicable, was established bybone marrow biopsy and peripheral blood counts and classified according toFrench American British (FAB) classification.67

Detection strategy

Our strategy for the detection of expanded (immunodominant) LGL clonesinitially included flow cytometric VB typing (Figure 1). When a signifi-cantly expanded VB family within the CD8� CTL population was found,the respective CDR3 amplification product was subcloned and transformedinto Escherichia coli. A pathologic VB expansion was defined as greaterthan mean � 2 standard deviations (SD) of the distribution found in healthycontrols, but, in most of the cases, the expansion of VB families weregreater than 20% of all CD8� cells. On average, 25 colonies weresequenced in order to determine the nucleotide sequence of the immunodom-inant clones. The frequency at which a particular clone occurred was

Figure 1. Rational approach for detecting immunodominant clonotypes inpatients with T-LGL leukemia. For explanations, see “Detection strategy.”

2770 WLODARSKI et al BLOOD, 15 OCTOBER 2005 � VOLUME 106, NUMBER 8

For personal use only. by guest on December 18, 2013. bloodjournal.hematologylibrary.orgFrom

calculated as the % of identical sequences divided by the total number ofclones analyzed.

In several patients the panel of VB antibodies used failed to identify aspecific VB family expansion. In such cases, the VB families not coveredby the antibody panel (VB6, 15, and 24) were subjected to reversetranscriptase–polymerase chain reaction (RT-PCR) and subcloned/sequenced if oligoclonal or monoclonal genotyping results were obtained.Samples negative by both flow cytometric VB typing and by RT-PCR forVB families 6, 15, and 24 were subjected to multiplex VB PCR using aBIOMED2 primer set,68 which amplifies all CDR3 regions. Samples with amonoclonal/oligoclonal pattern were subcloned and sequenced (Figure 1).Redundant clonotypes (� 2) detected through multiplex PCR were used toidentify the VB family harboring the immunodominant clone; cloning andsequencing of specific VB CDR3 was used to determine the clonotypicfrequency within a given VB class.

Flow cytometric VB typing

Fresh peripheral blood was stained for VB flow cytometry analysis toquantitate the contribution of each VB family to the CD4� and CD8�

lymphocyte populations. The manufacturer’s instructions (IOTest BetaMark kit; Beckman-Coulter, Fullerton, CA) were modified as follows: 5 �Lof phycoerythrin cyanin 5 (PC5)-conjugated CD4, 5 �L energy-coupleddye (ECD)–conjugated CD8 monoclonal antibodies (mAbs), and 20 �L ofanti-VB antibody (VB1-5, VB7-9, VB11-14, VB16-18, VB20-23) wereadded. A 4-color protocol was used. The lymphocyte gate was set accordingto the size and forward scatter properties. VB family usage was determinedwithin the CD4 and CD8 lymphocyte fractions using the appropriate gates.Initially, 9 normal samples were characterized to define the average size andstandard deviations for the VB repertoire as detected by the antibodies. Asignificant expansion was defined as one that was mean plus 2 SD over theaverage VB family size.39 Because the antibody set contained in the IOBetaMark kit does not cover the whole VB spectrum, VB PCR was performedfor VB families 6, 15, and 24 on samples that were negative for expansionby flow cytometry. Samples negative for VB families 6, 15, and 24 and forVB families covered by the flow cytometry panel were subjected tomultiplex PCR using BIOMED2 (see “RNA isolation and RT-PCR” ).

Lymphocyte separation

Mononuclear cells were separated from peripheral blood by densitygradient sedimentation (Mediatech, Herndon, VA). CD4� and CD8� T cellswere isolated by flow cytometric sorting. Samples were stained with a CD8mAb conjugated with fluorescein isothiocyanate (FITC; Pharmingen, SanDiego, CA) and sorted on an Epics Altra high-speed flow cytometer(Beckman Coulter, Miami, FL).

RNA isolation and RT-PCR

Total RNA was extracted from CD8� T cells with TRIZOL reagent(Invitrogen, Carlsbad, CA) and dissolved in a final volume of 20 �L ofdiethyl pyrocarbonate (DEPC) water. cDNA was generated from 6 �L ofRNA by first-strand cDNA synthesis using either SuperScript II RT Kit orSuperScript III RT Kit (Invitrogen). cDNA was amplified using a nonla-beled or FAM-labeled constant region antisense primer (CB) and 22different VB family–specific sense primers (including VB6, 15, and 24;pseudogenes VB10 and VB19 were not covered).11,12 PCR master mix wasset up as follows: 2.5 �L of 10 � PCR buffer containing 15 mMmagnesium acetate (Eppendorf, Hamburg, Germany), 2 �L of 2.5 mMdeoxynucleoside triphosphates (dNTPs), 2 �L of 3 �M antisense unlabeledor FAM-labeled CB primer, 2 �L of 3 �M sense VB primer, 0.25 �L of Taq(5 U/�L; Eppendorf), and 1.5 to 2 �L of cDNA were mixed with dH2O to afinal volume of 25 �L. Thermocycling was performed either on an MJthermocycler (MJ Research, Waltham, MA) or on a Mastercycler (Eppen-dorf). After an initial denaturation step at 95°C for 5 minutes and 10touch-down cycles (denaturation at 95°C for 30 seconds; annealing at 63°C,� 1°C per step, for 40 seconds and extension at 72°C for 50 seconds), thefinal 20 PCR cycles were performed (30 seconds at 95°C, 40 seconds at54°C, 50 seconds at 72°C).

Alternatively we applied a 2-tube multiplex PCR that covers all VBTCR gene rearrangements (developed by the BIOMED-2 Concerted ActionGroup) as previously described.68

CDR3 region genotyping

One microliter of amplification products generated either with theBIOMED-2 VB multiplex primer set or the 22 VB/FAM-CB primers wasmixed with 0.5 �L Genescan 400HD ROX size standard (AppliedBiosystems, Foster City, CA) and 12 �L Hi-Di formamide (AppliedBiosystems). The samples were run on an ABI Prism 310 GeneticAnalyzer (Applied Biosystems) and analyzed using GeneScan AnalysisSoftware v3.7 (Applied Biosystems). A set of numeric standards wasadopted to differentiate normal from abnormal profiles. Abnormal(skewed) profiles were defined as those that deviated from a Gaussiandistribution of intensity peaks (� 5 peaks, or contained one or severaldominant peaks). Accuracy of an abnormal classification was verified bysequencing, as described by us previously.63

CDR3 region cloning and sequencing

PCR products were gel purified using the QIAquick Gel Extraction Kit(Qiagen, Valencia, CA) or Eppendorf Gel Extraction Kit (Eppendorf)following the manufacturer’s instructions. Four microliters of the purifiedPCR product was ligated into the TA cloning vector pCR2.1 (Invitrogen)overnight at 14°C and heat-shock transformed into TOP10F E coli. ColonyPCR was performed on an MJ thermocycler as follows: on average 25bacterial colonies were picked and the inserts were amplified using M13Fand M13R primers provided in the TA cloning kit (2 �L of 3-�M solutioneach) and 3.0 �L of 10 � PCR buffer (Invitrogen), 1 �L of 50 mM MgCl2,2 �L dNTP (2.5 mM), 0.1 �L Taq polymerase (Invitrogen), in a totalvolume of 30 �L. After 5 minutes of denaturation at 94°C, 19 PCR cycles(30 seconds at 94°C, 30 seconds at 55°C, and 40 seconds at 72°C) wereperformed. Positive PCR products containing vector and insert wereconfirmed on a 1.2% agarose gel and purified using the Montage PCR96

Cleanup Kit (Millipore, Billerica, MA). Purified PCR products (5 �L) weresequenced using 1 �L of Big Dye Terminator v3.0 or v3.1 (AppliedBiosystems), 1 �L of 3 �M M13F primer, and 3 �L dH2O. Cyclesequencing was performed on an MJ thermocycler (MJ Research) asfollows: an initial denaturation at 96°C for 40 seconds was followed by 31cycles (10 seconds at 96°C, 15 seconds at 55°C, 4 minutes at 60°C).Sequencing reactions were purified using the Montage SEQ96 SequencingReaction Cleanup Kit (Millipore) run on a 3100-Avant Genetic Analyzer(Applied Biosystems). A pathologically expanded, immunodominant clono-type was defined based on its frequency among cloned sequences within anexpanded VB family. In this study, clonal diversity was defined as % ofunique clonotypes out of a total number of clonotypes. We have defined theclonal redundancy as the frequency of the identical clonotypes within allclones sequenced (at least 12 clonotypes were sequenced per VB andpatient). Based on the average size of the expansions in healthy individuals,pathologically expanded “major” clonotypes were defined as those occur-ring at the frequency greater than 12.5% of all sequenced clones (see“Identification strategy for immunodominant clones in LGL leukemia”).Minor, nonexpanded clonotypes were defined as CDR3 sequences occur-ring at the frequency of 1/N or greater but lesser than the size of theimmunodominant (major) clonotypes.

Sequence analysis

Annotation nomenclature for VB, joining beta (JB), and CDR3 regionswas adapted from the T cell receptor Factsbook69 and CDR3 regionsequences were analyzed using the ImMunoGeneTics informationsystem TCR alignment tool.70 For the purpose of this study we comparedCDR3 regions rather than NDN regions. A CDR3 region includesinvariant portions of VB and JB chains according to the nomenclatureproposed on the ImMunoGeneTics website.71 The clonotypic databaseused for sequence comparisons contains 4893 sequences, 2273 gener-ated in our laboratory and 2620 from other sources.

CLONAL CTL RESPONSES IN LGL LEUKEMIA 2771BLOOD, 15 OCTOBER 2005 � VOLUME 106, NUMBER 8

For personal use only. by guest on December 18, 2013. bloodjournal.hematologylibrary.orgFrom

A 7– and 8–amino acid (AA) substitution code was applied forcomparison of CDR3 sequences. Each amino acid residue was substitutedwith a category label corresponding to physicochemical properties of theindividual amino acids. In the first system, the groups were as follows:GAVLI, aliphatic amino acids (B); FYW, aromatic amino acids (J); CM,sulfhydryl-group–containing amino acids (O); S, T, aliphatic, hydroxyl-group–containing amino acids (U); KRH, basic amino acids (X); DENQ,acidic amino acids (Z); P, proline (P).72 The second substitution code hadthe following amino acid groupings: WFY, aromatic (A); LIVPA, aliphatic(B); MC, hydrophobic (C); G, �/� hydrophilic (D); STNQ, hydrophilic(E); H, �/� charged (F); DE, acidic (G); KR, basic (H).

Results

Identification strategy for immunodominant clones inLGL leukemia

We studied 60 patients with LGL leukemia and 13 patients withneutropenia for whom drug-induced hematopoietic toxicity andother hematologic diseases were excluded. In 56 of the LGLpatients we characterized a pathologically expanded CTL clonethat was dominant within the respective TCR VB family and theentire CD8� cell population. In addition, we were able to detect thepresence of previously unknown immunodominant CTL clones in 4patients with neutropenia (Tables 1-2).

Based on the sequencing of a large number of CDR3 ampliconsin controls (57 25 [mean standard deviation; SD] colonies perVB family and sample; n 26), we defined a clonal frequency ofgreater than 12.5% (mean�2 � SD of the largest clones found inhealthy controls) as pathologic. On average, expansions in patientswere 64% 30% of the CDR3 repertoire within a given VB family(Figure 2). A total of 86 immunodominant LGL clonotypes wereidentified (Figure 3). Their frequency was 18% to 100%(65% 30%) of a given VB family or 1.8% to 95% (35.4% 31%)of the total CD8 population (Figure 3). By comparison, the mostexpanded VB13 clones in healthy controls contribute to only 0.7%of the CD8 repertoire (Figure 2). Remarkably, in 21 patients wefound more than one immunodominant clone, suggesting that someLGL lymphoproliferations are biclonal (Tables 1-2; Figure 3).

Structural analysis of immunodominant LGL clones

Given the immense variability of physiologic TCR repertoire,detection of identical clonotypes shared by patients with LGLleukemia could suggest a nonrandom nature of clonal transforma-tion. Translation of nucleotide sequences into AA code resulted in86 LGL CDR3 clonotypes (Figure 3). These sequences were alsocompared with LGL-specific immunodominant CDR3 sequences(n 5) described in literature.59,61,62 We found 2 patients (nos. 50and 54 in Table 3) with identical immunodominant clonotype.Interestingly, this clonotype was also shared with a third patient(no. 27) in whom it was not significantly expanded. Analysis of amuch larger number of CDR3 sequences in healthy controlsdemonstrates that such a finding is not likely to be a totally randomevent. The presence of this sequence was confirmed throughmultiple independent amplifications and sequencing. As docu-mented by HLA typing, these 3 patients (nos. 27, 50, 54) share theclass I antigen B7. In addition, we found sequence identity betweenan immunodominant and a nonimmunodominant (or minor) clono-type and between 2 minor clonotypes in 2 LGL patient pairs (nos.27 and 54; nos. 27 and 50 in Table 3). In 2 other patients, a majorclonotype showed a striking homology to a minor clonotype alsopresent within the TCR repertoire (nos. 27 and 49 in Table 3). Not

only were identical CDR3 sequences shared between the patientsbut a high degree of similarity was also found between a largenumber of minor (nonexpanded) clonotypes (Table 3). Codominantclonotypic sequences within the TCR repertoire of individualpatients were found that differ only in one or a very fewnucleotides. In some instances, these nucleotide mismatches didnot result in a new AA sequence (Figure 4), consistent withselection of high-affinity T-cell receptors during the initial poly-clonal response. By comparison, when up to 172 clones weresequenced per VB family in 26 healthy donors, only one sharedclonotype was found in 2 donors and only a total of 11 sequenceswere detected to show a similar level of homology between thehealthy controls as described in LGLs (Table 4).

We have also employed more redundant AA codes (8-AA and7-AA instead of 20-AA code), allowing for the comparison ofsequences based on the physical and chemical properties of theAAs (Tables 5-6). Although the rules governing the recognition ofthe antigenic peptide by the TCR are not fully known, the VBCDR3 region appears to be most intricately involved in the contactof the target peptide and B chain. Thus similarity of the structure ofthe VB CDR3 region if occurring in the context of the commonHLA restriction element may indicate shared specificity. However,coincidental similarity cannot be excluded. The comparison revealsan even higher degree of sequence homology than found using20-AA translation. In contrast, analogous analysis performed onCD8� and on CD8�CD57� cell populations derived from healthydonors resulted in identification of a significantly lesser CDR3sequence homology in 20-, 8-, and 7-AA coding (data not shown).

Clonotypic background

In addition to the major clone, other clonotypes occurring at lowfrequencies may be present. These clonotypes are likely randomlyselected due to, for example, polyclonal expansion. If LGL clonesarise in the context of initially polyclonal CTL responses, it ispossible that structurally similar clonotypes will be present.Consequently, we compared AA sequences of the major (immuno-dominant) and minor clonotypes identified in each patient. Wefound a high degree of homology between major and minorclonotypes within the repertoire of individual patients (Figure 4).Single base differences were also detected but they were notincluded in this homology analysis, as such clonotypes may or maynot be a result of polymerase reading errors. However, all theresults were based on sequencing of a large numbers of clones, andclonotypes that varied in one or only a few AAs were found atmultiple occasions.

Clonotypic dynamics and disease progress

Immunodominant clonotypes can be used as tumor markers and wehave investigated the changes in the contribution of the LGL clonein correlation with disease activity (Figure 5). The size of thepathologic clone was calculated by multiplication of the contribu-tion of the VB family to the total CD8 spectrum and the relativesize of the clone within a given VB family. For most patients, amarked decrease in the frequency of the immunodominant clonewas observed with successful immunosuppressive therapy. In somepatients the drop appeared to be less significant while in others thepathogenic clone was not detectable following the therapy (Figure5A). Decrease in the contribution of the immunodominant clone tothe CD8 repertoire was associated with the appearance of minorclones that were previously “diluted out” by the expanded cloneand not detectable. This process resulted in increased clonal

2772 WLODARSKI et al BLOOD, 15 OCTOBER 2005 � VOLUME 106, NUMBER 8

For personal use only. by guest on December 18, 2013. bloodjournal.hematologylibrary.orgFrom

Table 1. Patient characteristics, clinical presentation, and laboratory data

Patientno.

Age,y

Main clinicalpresentation Diagnosis

Additionaldiagnosis

TCR gammarearrangement

by PCR HLA-A HLA-B HLA-C DR

1 25 Pan LGL — Y 26, 32 27, 44 ND 1, 4

2 72 Pan LGL RA/PNH Y 2, 3 13, 35 4, 8 11, 13

3 58 RCA/Neu LGL — Y 2, 2 58, 55 ND 1, 4

4 76 RCA LGL — Y 2, 68 27, 51 ND 1, 13

5 62 Neu LGL RA Y 1, 2 14, 51 2, 8 1, 13

6 66 Neu LGL — Y NA NA NA NA

7 61 Neu LGL — Y 3, 24 37, 40 3, 6 4, 10

8 36 Neu LGL — Y 2, 11 18, 44 7, 16 11, 13

9 58 Neu LGL MS Y 1, 32 8, 44 5, 7 4, 15

10 65 Pan LGL — Y 1, 68 14, 37 6, 8 15, 13

11 42 RCA LGL Hepatitis Y 1, 2 7, 37 6, 7 N.D.

12 61 Neu LGL RhAr Y 1, 2 8, 15 3, 7 3, 4

13 63 Neu LGL — Y ND ND ND ND

14 77 RCA LGL — Y ND ND ND ND

15 37 Neu LGL SLE N ND ND ND ND

16 50 RCA LGL — Y ND ND ND ND

17 64 Pan LGL RARS Y 29, 08 40, 45 ND 11, 15

18 63 RCA LGL UC Y 2, 2 8, 44 7, 7 3, 11

19 65 Neu LGL — Y ND ND ND ND

20 78 RCA LGL RhAr Y ND ND ND ND

21 51 Neu LGL — Y 2, 24 7, 73 7, 15 1, 4

22 76 Pan LGL RA Y 24, 24 7, 35 4, 7 1, NA

23 80 RCA LGL RA Y 1, 29 13, 44 6, 16 4, 7

24 32 Pan LGL Hepatitis Y 2, 24 40, 51 3, 3 4, 12

25 63 RCA LGL RA Y 11, 33 51, 65 7, 8 1, 13

26 73 Neu LGL RhAr Y 2, 29 44, 44 12, 16 4, 4

27 62 Neu LGL Hepatitis Y 2, 2 7, 51 1, 7 11, 4

28 46 Neu LGL RhAr Y 3, 24 37, 40 3, 6 4, 10

29 62 Neu LGL — Y 23, 32 18, 35 4, 12 1, 13

30 78 Neu LGL — NA ND ND ND ND

31 71 RCA LGL — Y ND ND ND ND

32 57 — LGL — Y ND ND ND ND

33 72 RCA LGL — Y 1, 31 8, 39 7, 7 1, 3

34 71 Neu LGL — Y 1, 2 8, 44 7, 7 16, 3

35 66 Neu LGL — Y ND ND ND ND

36 80 Neu LGL — Y ND ND ND ND

37 62 Neu LGL — Y ND ND ND ND

38 NA Neu LGL — Y ND ND ND ND

39 NA Neu LGL — Y ND ND ND ND

40 NA Neu LGL — Y ND ND ND ND

41 NA Neu LGL — Y ND ND ND ND

42 64 RCA LGL AIHA Y ND ND ND ND

43 NA Neu LGL — Y ND ND ND ND

44 77 Neu LGL RA Y ND ND ND ND

45 76 Neu LGL — Y ND ND ND ND

46 NA Neu LGL — Y ND ND ND ND

47 59 Neu LGL — Y ND ND ND ND

48 NA Neu LGL — N 3, 31 18, 51 12, 14 10, 15

49 70 Neu LGL — NA ND ND ND ND

50 65 Neu AIN — N 2, 11 7, 15 3, 7 15, 4

51 72 Pan LGL RA N 1, 68 35, 44 4, 4 7, 8

52 51 Neu LGL — N 11, 26 45, 51 3, 6 4, 9

53 29 Neu AIN Hepatitis N ND ND ND ND

54 43 Neu AIN Hepatitis N 3, 3 7, 35 4, 7 1, 15

55 64 Neu LGL RhAr Y 24, 02 44, 51 2, 4 11, 13

56 39 Pan AIN — Y 2, NA 15, 44 3, 5 4, 13

57 56 Neu LGL RhAr Y 1, 25 8, 18 7, 12 4, 11

58 60 Pan LGL PNH NA 68, 68 14, 14 8, 8 1, 1

59 53 Neu LGL AIH Y 1, 3 7, 37 6, 7 13, 15

60 51 Pan LGL — N 11, 24 27, 22 1, 1 4, 15

Pan indicates pancytopenia; LGL, large granular lymphocyte leukemia; —, not applicable; Y, yes; ND, not done; RA, refractory anemia; PNH, paroxysmal nocturnalhemoglobinuria; RCA, red cell aplasia; Neu, neutropenia; NA, not available; MS, multiple sclerosis; RhAr, rheumatoid arthritis; SLE, systemic lupus erythematodus; N, no;RARS, RA with ringed sideroblasts; UC, ulcerative colitis; AIHA, autoimmune hemolytic anemia; and AIN, autoimmune neutropenia.

CLONAL CTL RESPONSES IN LGL LEUKEMIA 2773BLOOD, 15 OCTOBER 2005 � VOLUME 106, NUMBER 8

For personal use only. by guest on December 18, 2013. bloodjournal.hematologylibrary.orgFrom

diversity (defined as percentage of different clones out of a totalnumber of clones) and decreased redundancy (defined as percent-age expansion of the largest clone; Figure 5B). Clinical correlationbetween the size of immunodominant clones and hematologicparameters is shown for 2 patients with LGL leukemia associatedwith pure red cell aplasia (Figure 5C-D). When we analyzed thestructure of clonotypes that appeared after therapy we found that insome instances they showed similarity to the original immunodom-inant clones (data not shown).

Clonotype sharing with other diseases

We also investigated minor and immunodominant clonotype shar-ing between LGL patients and patients with other diseases (n 156).We found 3 expanded LGL-associated clonotypes to also bepresent in patients studied after allogeneic bone marrow transplan-tation (Table 4). In addition, sharing of minor clonotypes was alsofound between LGL and bone marrow transplantation patients.Similarly, a minor LGL clonotype was detected in a patient withHIV. Two of the immunodominant LGL CDR3 sequences werefound in healthy individuals tested and furthermore 2 minorLGL-associated clonotypes were shared with healthy donors.

Discussion

Our study addresses several important aspects of the physiologicand aberrant cellular immune response. LGL leukemia, albeitrare, can act as a model system to study CTL expansions, as theclonal CTLs show a behavior that is highly reminiscent of thatof their normal counterparts. While flow cytometric40 andmolecular75 VB typing has previously been applied to relativelylarge cohorts of LGL patients, LGL-like clonotypic expansionswere described only in very small series including 10 patientswith Felty syndrome (9 clonotypes),59 3 patients with MDS anderythroid hypoplasia (3 clonotypes),62 32 patients with rheuma-toid arthritis (9 sequences),54 and in a donor/recipient pair afterBMT (5 sequences).61

We have applied a rational strategy not only to identifyindividual LGL-leukemia–associated immunodominant clonotypesbut also to quantitate their contribution to the total CTL repertoire

Table 2. VB-JB restriction of immunodominant clones instudied patients

Patient no. VB% VB familyexpansion JB

1 14 90 JB2.1

2 3 88 JB2.4

3 22 89 JB2.7

4 17 83 JB2.3

5 9 64 JB1.1

6 8 26 JB1.5

7 6 NA JB2.1

8 23 15 JB1.1, JB2.7*

8 14 16 JB2.7

9 3 63 JB1.1

10 7 76 JB2.7

11 20 23 JB2.1, JB2.7*

12 14 65 JB2.7

13 12 24 JB2.6

14 2 40 JB2.7

15 18 8 JB1.1, JB2.3*

16 2 90 JB2.7

17 2 20 JB2.1

17 21 31 JB2.1

18 17 45 JB2.3

18 3 25 JB2.1

19 2 38 JB2.3

20 14 22 JB1.2

21 13 7 JB1.6

22 3 26 JB1.5

23 3 33 JB2.7

24 2 80 JB2.7

25 15 NA JB2.1, JB2.3*

26 9 25 JB2.5

27 13 43 JB1.1

27 18 25 JB2.7

27 22 NA JB2.7

28 13 65 JB2.7

29 5.3 25 JB2.7

30 15 NA JB1.2, JB2.3*

31 5 NA JB1.2

32 15 NA JB2.5, JB2.1, JB2.5*

33 18 27 JB2.7

34 3 10 JB2.7

34 22 60 JB1.2

35 7 NA JB1.6

36 9 78 JB2.4

37 13 80 JB2.5

38 12 93 JB2.1

39 7 46 JB2.2

40 24 NA JB2.2

40 13 NA JB1.2

41 15 NA JB2.7

41 24 NA JB2.5

42 6 NA JB2.5

42 15 NA JB2.4

43 6 NA JB2.3

43 24 NA JB2.1

43 12 NA JB2.3

44 9 81 JB2.3

44 21 11 JB1.1

45 24 NA JB1.1

46 6 NA JB2.1, JB2.7*

46 17 8 JB2.1

47 6 NA JB2.7

47 24 NA JB2.3

48 13 93 JB1.1

49 1 70 JB1.5

50 13 6 JB2.7

Table 2. (Continued)

Patient no. VB% VB familyexpansion JB

51 12 8 JB2.1

51 6 NA JB1.1

52 1 NA JB1.1

53 18 42 JB2.7

54 13 6 JB2.7, JB2.5*

55 13 7 JB2.3

55 18 78 JB2.5

56 4 22 JB2.1

57 13 89 JB2.5

58 3 NA JB2.7

58 13 NA JB1.5

59 1 95 JB1.6

60 14 28 JB2.1

VB indicates variable beta chain (VB family) of the TCR; JB, joining beta chain ofthe expanded clonotypes; NA, not available.

*Cases in which 2 (3 in patient 32) immunodominant clonotypes with identical VBbut different JB families were detected.

2774 WLODARSKI et al BLOOD, 15 OCTOBER 2005 � VOLUME 106, NUMBER 8

For personal use only. by guest on December 18, 2013. bloodjournal.hematologylibrary.orgFrom

during therapy. We demonstrate that either flow cytometric detec-tion of phenotypically abnormal CTL populations or VB CDR3genotyping may leave a significant number of CTL proliferationsunidentified. By combining these methods and sequencing ex-panded clones, we found several patients with significant expan-sions of clonal CTLs that were unsuspected based on traditionalclinical tests. It is likely that clinically manifested cases of LGLleukemia represent the extreme clonal/oligoclonal lymphoprolifera-tions that, to a lesser extent, may also be present in certain patientswith otherwise unexplained cytopenias. In addition, some of theexpansions, when asymptomatic, can represent benign CTL reac-tions also referred to as T-cell clonopathy of undeterminedsignificance.76

Our molecular clonotypic analysis was based on the B chain.The VB CDR3 region appears to interact with the antigenicpeptide but A chain may also play an important role, especially

in the initial phase of clonal high-affinity TCR selection.77 Asexpected, VB clonality was accompanied by VA clonality inanalyzed patients. For 3 of our patients we have obtainedanalogous results showing that their clones express TCRphenotypes: VB1/JB1.6 and VA9-2/JA41*01 (patient no. 59),VB9/JB2.5 and VA26-1/JA9*01 (patient no. 26), VB13/JB2.5

Figure 3. CDR3 motifs and expansions of immunodominant T-cell clones withinVB family and CD8� population. Pat indicates patient number; CDR3, sequence ofthe complementarity determining region 3; inv, first 3 invariant AAs of the JB region;VB-JB, restriction of the variable and joining � region of the TCR; imm clone/VBfamily, expansion of immunodominant, disease-associated T-cell clone within a givenTCR VB family; imm clone/CD8�, expansion of immunodominant clone within theentire CD8� population (clonal expansion within given VB family multiplied byexpansion of this VB family within CD8� cell population); and *, VB family expansionnot available.

Figure 2. Clonotypic expansions in patients and healthy controls.The size ofpathogenic LGL clones was compared with the most expanded clones found forcorresponding VB families in healthy individuals. The figure depicts selectiveexamples for individual patients/healthy controls and VB families. u indicatespercent clonotypic expansion within a given VB family (clone/VB). f indicatespercent expansion within the entire CD8� cell population (% clonotypic expansionwithin a VB family multiplied by VB family contribution to CD8� population;clone/CD8�). VB3-restricted clonotypic expansions were found in 6 patients andcompared with the average size of the most redundant (expanded) clonotypeswithin VB3 in 9 healthy controls (a total of 356 CDR3 sequences). VB9-restrictedexpansions were found in 4 patients and compared with the average size of themost redundant clones for VB9 family studied in 2 healthy donors (a total of 45CDR3 sequences). VB13-restricted expansions were found in a total of 10patients and compared with the average size of the most redundant clones forVB13 family detected in 8 healthy controls (a total of 608 CDR3 sequences). Atotal of 6 patients were found to have VB18-specific immunodominant clonotypesand were compared with the average size of the most redundant clones for VB18family derived from 8 healthy individuals (a total of 447 CDR3 sequences). Inaddition, repertoires of mature effector CTL cells (CD8�CD57�) derived fromhealthy individuals (n 2) were studied. Pat no. indicates patient number.

CLONAL CTL RESPONSES IN LGL LEUKEMIA 2775BLOOD, 15 OCTOBER 2005 � VOLUME 106, NUMBER 8

For personal use only. by guest on December 18, 2013. bloodjournal.hematologylibrary.orgFrom

and VA1-2/JA11*01 (patient no. 57; M.W.W., unpublishedresults, August 2004).

Our study was initiated based on the theory that the uniqueand often very lineage-restricted cytopenias associated withLGL leukemia may be a result of the lineage specificity of thetarget antigens recognized by clonal TCR. Consequently, it maybe possible that in patients with similar presentations andmatching HLA restriction elements, identical or highly homolo-gous clonotypes will be found. Clearly, given the estimatedphysiologic variability of CDR3 regions within the VB chain,13

such an event may be rare but it would prove the nonrandomorigin of clonal transformation in LGL leukemia. Based on ananalogous theory, similarity of the rearranged immunoglobulin G(IgG) genes in B-cell clones found in chronic lymphocyticleukemia (CLL) suggests a nonrandom nature of transformation.78

Clearly, the restriction of TCR by HLA makes the principle of theT-cell recognition more complex. Nevertheless, numerous reportsconfirmed homology of clonotypes recognizing specific antigens inindividuals matched for HLA elements.7,29,30,32,34 A specific TCRcan recognize epitopes derived from the same protein andpresented in the context of different yet genotypically similarHLA types.79

In analogy to these findings, identical expanded clonotypeswere found in 2 patients. Based on the extent of the physiologicTCR repertoire and the extremely low frequency of sharedclonotypes identified between healthy controls, the coincidentalfinding of common clonotypes is unlikely although it cannot betotally excluded. It suggests that these clones may not evolverandomly but occur in the context of an initially oligoclonal/polyclonal immune response directed against identical or highlysimilar antigenic targets. This conclusion is also supported bythe identification of sequence sharing between the 2 otherimmunodominant clonotypes and nonexpanded clonotypes foundin other LGL leukemia cases. It is likely that these minorclonotypes are the remnants of an initial polyclonal response.The low frequency of these clones may be due to “dilution” byexpanding semiautonomous LGL clones. Consequently, theLGL clone may have evolved not from a total possible pool ofCTL memory cells but from a selected population participatingin a polyclonal immune response. Detection of high levels ofsimilarity between clonotypes within a given VB repertoire ofindividual patients also supports our conclusions that severalclones recognizing similar or identical peptides exist prior to theexpansion of immunodominant clones. Occasional presence ofclonotypes that are identical in their AA sequences but distinctin their nucleotide sequences may have resulted from selectionof individual clones with identical antigenic affinity. In suchcases the effective frequency of the corresponding TCR may bea sum of 2 independently rearranged clones. In the previousreport,63 ClustalW analysis80 was used to assess the similaritybetween the clonotypic sequences. Since peptide binding isaffected through the physical AA properties and the tertiarystructure of CDR3, we believe that the analysis of physicochem-ical properties of AAs performed in the current study is moreappropriate than linear homology comparisons in 20-AA code.

Sequencing of clonotypes prior to and following successfulcytotoxic therapy also revealed that the clonotypes found afterretraction of the LGL clone show similarity to the immunodom-inant clonotype. All of these observations have to be viewed inthe context of the extremely low frequency of identical clono-types (both expanded as well as minor) shared by healthyindividuals in whom a relatively low homology between ran-domly sequenced clonotypes was also found. Association ofLGL leukemia with typical pathologies is consistent withclonotype sharing. The initial autoimmune process leading tothe specific manifestations may have provided a signal for theexpansion of a clone that through a certain, yet not specified,molecular event induces uncontrolled proliferation. In ourprevious study, a smaller cohort of patients was analyzed andseveral similar and 2 identical clonotypes were identified.63

Based on these preliminary results, we expected a larger numberof shared clonotypes to be found. Due to the higher samplingsize, the frequency of clonotype sharing established in thecurrent report is more precise.

Identification of highly homologous or identical clonotypesis remarkable. One could argue that PCR-based carryover couldexplain such a striking result. To avoid this possibility, inaddition to general contamination precautions, in each PCR set,different VB families were amplified and sequenced. Foranalysis, the VB-specific primer sequence was used as aninternal control to verify patient-specific sequences. Moreover,the finding of homologies between sequences reported in theliterature and detected in our laboratory further illustrates thatthe isolation of identical sequences is not due to a contamination

Table 3. Similarity between LGL-associated clonotypes

Patient VB CDR3 JB% clonal

frequency

27 13 ASSVPGGEQYF 2-7 Minor

50 13 ASSVPGGEQYF 2-7 48

54 13 ASSVPGGEQYF 2-7 43

27 13 ASSYYREPQHFGD 1-5 Minor

50 13 ASSYYREPQHFGD 1-5 Minor

27 13 ASSYTFQGETQY 2-5 Minor

54 13 ASSYTFQGETQY 2-5 29

27 18 ASSPGTGLDGYTF 1-2 Minor

49 1 ASSPGTGRNQPQHFGD 1-5 89

50 13 ASRDGAGEQYF 2-7 Minor

28 13 ASRDRAGN 2-6 Minor

44 21 ASRSGTGAHEQYF 2-7 Minor

28 13 ASRSGTGSYNEQFF 2-1 Minor

23 3 ASSGEETQYF 2-3 Minor

33 18 ASSGETDTQYF 2-3 Minor

62* 13 ASSLAGGPDTQYF 2-3 Minor

47 6 ASSLAGTGGGEQYF 2-7 Minor

47 6 ASSLAGTSSTDTQYF 2-3 Minor

35 6 ASSLGSYNSPQHFDG 1-5 Minor

63* 21 ASSLGSYNSPLH 1-6 Minor

62* 13 ASSLLKPTSDTQYF 2-3 Minor

12 14 ASSLLTKTGSYEQYF 2-7 81

35 6 ASSLVGDPGELF 2-2 Minor

47 6 ASSLVGGRPSYNEQFF 2-1 Minor

27 18 ASSPPVRARDTQYF 2-3 Minor

33 18 ASSPPVRSRAQY 2-5 Minor

18 17 ASSWGGAGELFF 2-2 Minor

29 5 ASSWGGRGEQFF 2-1 Minor

65* 15 ATSDLGQGASGELF 2-2 Minor

47 15 ATSDLGQQETQY 2-5 Minor

42 15 ATSDLGEKVGNEQFF 2-1 Minor

47 24 ATSRDLAGEKLF 1-4 Minor

35 24 ATSRDLASGRETQYF 2-3 Minor

47 24 ATSRDLRGEKLF 1-4 Minor

41 24 ATSRDLRTQETQY 2-5 Minor

VB indicates variable beta family; CDR3, complementarity-determining region 3;JB, joining beta family with first 3 variant AAs; minor, nonexpanded (single) clonotype.

Italics indicate identical AAs.*Patients in whom a major “immunodominant” clonotype was not determined,

therefore not included in Table 1.

2776 WLODARSKI et al BLOOD, 15 OCTOBER 2005 � VOLUME 106, NUMBER 8

For personal use only. by guest on December 18, 2013. bloodjournal.hematologylibrary.orgFrom

Figure 4. Examples of TCR repertoire similaritieswithin individual LGL patients. Freq indicates clono-typic frequency as the number of identical clonotypesequences divided by total number of clones; and VB-JB,restriction of a given variable and joining beta chain of theTCR. Solid boxes indicate nucleotide exchanges result-ing in AA exchange (the affected AAs are highlighted inyellow); dashed boxes and lines, nucleotide exchangesthat do not lead to AA exchanges. ‡Total of 11 similarexchanges were found and were not included in theanalysis, due to the possibility of Taq-polymerase–generated errors. Of interest, in patient 44 an immunodom-inant clonotype has 2 “supporting” clonotypes that differin only 1 amino acid of the CDR3 sequence. Those 2clonotypes were translated from 2 different nucleotidesequences each (*). In both cases the underlying se-quences differed in 2 nucleotides that did not affect theAA sequence. Although we postulate that single nucleo-tide exchanges reflect Taq-polymerase reading errors, inthis case the accumulation of nucleotide substitutionsindicates that the 5 shown clones in patient 44 may haveresulted from independent rearrangement events.

Table 4. Identical clonotypes in LGL and other conditions

Patient Diagnosis VB-JB CDR3 % clonal frequency

27 LGL VB13-JB2.7 ASSVPGGEQYF Minor

50 LGL VB13-JB2.7 ASSVPGGEQYF 48

54 LGL VB13-JB2.7 ASSVPGGEQYF 43

409 HD VB13-JB2.7 ASSVPGGEQYF Minor

418 HD VB13-JB2.7 ASSVPGGEQYF Minor

27 LGL VB18-JB2.3 ASSAVGGTDTQYF Minor

142 AHSCT VB18-JB2.3 ASSAVGGTDTQYF 40

573 AHSCT VB18-JB2.3 ASSAVGGTDTQYF 25

217 AHSCT VB17-JB2.3 ASSIAVRLASTDTQYF 30

4 LGL VB17-JB2.3 ASSIAVRLASTDTQYF 94

McFarland et al73 HIV-1 VB13-JB1.2 ASSPITGTGSYGYTF 7

27 LGL VB13-JB1.2 ASSPITGTGSYGYTF Minor

Beck et al74 GvHD VB13-JB2.2 ASSSETSGRNTGELFF 25

27 LGL VB13-JB2.2 ASSSETSGRNTGELFF Minor

27 LGL VB13-JB2.7 ASSYGQEHYEQYF Minor

419 HD VB13-JB2.7 ASSYGQEHYEQYF Minor

418 HD VB13-JB2.1 ASSYSGSHNEQFF Minor

54 LGL VB13-JB2.1 ASSYSGSHNEQFF 21

258 AHSCT VB15-JB2.5 ATSDLTWGETQY 25

574 AHSCT VB15-JB2.5 ATSDLTWGETQY 60

242 AHSCT VB15-JB2.5 ATSDLTWGETQY 27

32 LGL VB15-JB2.5 ATSDLTWGETQY 26

VB-JB indicates restriction of the variable beta and joining beta region of the TCR; LGL, large granular lymphocytic leukemia; minor, nonexpanded clonotype; HD, healthydonor; AHSCT, allogeneic hematopoietic stem cell transplantation; and GvHD, graft-versus-host disease.

CLONAL CTL RESPONSES IN LGL LEUKEMIA 2777BLOOD, 15 OCTOBER 2005 � VOLUME 106, NUMBER 8

For personal use only. by guest on December 18, 2013. bloodjournal.hematologylibrary.orgFrom

artifact. Identical sequences from different patients found in ourlaboratory were confirmed by repeated independent sequencing.Many of the immunodominant clonotypes in individual patientswere invariably found on multiple time points throughout theclinical course and patients with common clonotypes shared atleast one HLA allele that could produce a display of identicalantigenic peptides. Finally, in our preliminary studies we haveestablished optimal conditions for transformation to excludethat a higher frequency of certain clones is a result of bacterialdivision rather than preponderance of the specific amplificationproduct in the sample. By minimizing the posttransformationphenotypic expression time, outgrowth of bacterial clones isavoided and the sequence distribution of the transformationproduct likely represents the composition of the PCR amplicons.The results obtained from healthy adult donors and cord bloodcontrols clearly demonstrate that replicate sequences are onlyrarely encountered.

In addition to information about clonotype sharing andstructure, our study provides an estimate of the frequency ofclonal CTLs and the degree of decrease in the TCR repertoirevariability. In some extreme cases, up to 95% of peripheral CTLrepertoire is clonal, a surprising finding in view of the lack ofimmunodeficiency among patients with LGL leukemia. Undernormal circumstances, even the most predominant clones that

most likely correspond with the immune responses to ubiquitousantigens comprise only up to 1.4% of the whole repertoire,suggesting that the finding of an extreme expansion of anindividual clone is a significant pathologic sign. The individualexpansions varied in their contribution to the TCR repertoireamong the patients and did not correlate with the severity ofcytopenia. It is possible that killing efficiency may also dependon (1) the affinity of the clonal TCR to the antigen and (2) thefrequency and distribution of the antigens in target tissues.However, hematologic responses were associated with a de-crease in clonal dominance. (In cases in which the initiallyexpanded clonotype was not found by sequencing, its persis-tence was confirmed by clonotypic PCR, consistent with thehigh relapse rate of LGL leukemia.)

We have compared LGL clonotypes with each other and withclonotypes found in healthy controls. Clonotypes specific formalignant clones were not encountered to a great extent in healthyindividuals. Previously, we have demonstrated using clonotypicPCR that some of the LGL clonotypes may also be found at verylow levels in controls. When the disease-associated clonotypesfound in our study (and those described by others) were cross-referenced against a clonotypic database containing nearly 5000clonotypes sequenced in our laboratory and described in the

Table 5. Examples of a comparison of LGL-associated clonotypes based on the physicochemical AA properties (8-AA substitution)

Patient VB-JB CDR3/20AA code CDR3/8AA substitution

53 VB18-JB2.7 ASS-PPVGDR -SYEQYF BEE-BBBDGH -EAGEAA

4 VB17-JB2.3 ASS-IAVRLA -STDTQYF BEE-BBBHBB -EEGEEAA

50 VB13-JB2.7 ASS-VPGG -EQYF BEE-BBDD -GEAA

54 VB13-JB2.7 ASS-VPGG -EQYF BEE-BBDD -GEAA

18 VB3-JB2.1 ASS-PLGAGV -YNEQFF BEE-BBDBDB -AEGEAA

27 VB18-JB2.7 ASS-PAGGSG -SYEQYF BEE-BBD DED -EAGEAA

12 VB14-JB2.7 ASS-LLTKTG -SYEQYF BEE-BBEHED -EAGEAA

29 VB5-JB2.7 ASS-PLRG -SYEQYF BEE-BB H D -EAGEAA

55 VB18-JB2.5 ASS-PNI -ETQY BEE-BEB -GEEA

19 VB2-JB2.3 ASS-PTLRDRGR -TDTQYF BEE-BEBHGHDH -EGEEAA

46 VB17-JB2.1 ASS-TFG -EQFF BEE-EAD -GEAA

44 VB21-JB1.1 ASS-SY -NEAFF BEE-EA -EGBAA

A redundant classification scheme was applied for the comparison of the TCR-CDR3 region, based on the structural, chemical, charge, and functional properties of theamino acids. For substitution code, see “Sequence analysis.”

VB-JB indicates VB and JB regions used by the immunodominant clonotype; CDR3/20AA code and CDR3/8AA, CDR3 region sequence in 20-AA and 8-AA code,respectively; and Jun, invariant amino acids of the JB region. Italic amino acids indicate the beginning of the JB region and include the invariant JB-specific amino acids.Underlined type indicates AA differences.

Table 6. Examples of a comparison of LGL-associated clonotypes based on the physicochemical AA properties (7-AA substitution)

Patient VB-JB CDR3/20AA code CDR3/7AA substitution

7 VB6-JB2.1 ASS-LASI -SYNEQFF BUU-BBUB -UJZZZJJ

41 VB24-JB2.5 ATS-LGTL -QETQYF BUU-BBUB -ZZUZJJ

8 VB23-JB1.1 ASS-LGNR -EVAFF BUU-BBZX -ZBBJJ

52 VB1-JB1.1 ASS-AGQD -TEAFF BUU-BBZZ -UZBJJBZ

58 VB13-JB1.5 ASS-YGGG -QPQHFGD BUU-JBBB -ZPZXJ

55 VB13-JB2.3 ASS-YVGV -DTQYF BUU-JBBB -ZUZJJ

18 VB3-JB2.1 ASS-PLGAGV -YNEQFF BUU-PBBBBB -JZZZJJ

27 VB18-JB2.7 ASS-PAGGSG -SYEQYF BUU-PBBBUB -UJZZJJ

46 VB17-JB2.1 ASS-TFG -EQFF BUU-UJB -ZZJJ

44 VB21-JB1.1 ASS-SY -NEAFF BUU-UJ -ZZBJJ

14 VB2-JB2.7 SAS-GLLAGGA -SYEQYF UBU-BBBBBBB -UJZZJJ

24 VB2-JB2.7 SAS-FLTGLAG -EQYF UBU-JBUBBBB -ZZJJ

A redundant classification scheme was applied for the comparison of the TCR-CDR3 region, based on the structural, chemical, charge, and functional properties of theamino acids. For substitution code, see “Sequence analysis.”

VB-JB indicates VB and JB regions used by the immunodominant clonotype; CDR3/20AA code and CDR3/7AA, CDR3 region sequence in 20-AA and 7-AA code,respectively; and Jun, invariant amino acids of the JB region. Italic amino acids indicate the beginning of the JB region and include the invariant JB-specific amino acids.Underlined type indicates AA differences.

2778 WLODARSKI et al BLOOD, 15 OCTOBER 2005 � VOLUME 106, NUMBER 8

For personal use only. by guest on December 18, 2013. bloodjournal.hematologylibrary.orgFrom

literature in association with various conditions, identical clono-types derived from patients who underwent allogeneic BMT andsome LGL leukemia patients were found. This finding is notsurprising, as clonal T-cell expansions have been reported in BMTpatients61,81,82 and a common mechanism may be involved. T-cellclonality after BMT results from the physiologic immune reconsti-

tution that is shaped by various triggering antigens, includinginfectious (eg, cytomegalovirus [CMV]) and allogeneic epitopes.Shared/homologous clonotypes in LGL and after allogeneic BMTmay provide insight into the nature of target antigens; for example,minor HLA antigen may drive the corresponding clonotypes afterBMT,83 and theoretically such self -“minor” antigens could betargeted in LGL leukemia. This theory is particularly attractive asclonotype sharing was found in the context of the presence ofmatching HLA antigens.

Identification of immunodominant clonotypes has not onlyseveral pathophysiologic implications but also clinical appli-cations. Clonotypic sequences may be used to monitor thefrequency of malignant T-cell clones using, for example,quantitative PCR. Such an approach may also work for poly-clonal responses in which clonotypes derived from most sig-nificantly expanded T-cell clones may serve as markers forspecific autoimmune processes or for responses to vaccines orinfectious agents. Finally, the availability of the clonotypicsequences may allow for the measurement of natural andpathologic anti-idiotypic responses as well as clonotype-basedanti-idiotypic vaccination as a potential therapeutic modality forT-cell malignancies.

Acknowledgment

M.W.W. would like to thank his former mentors Dr MartinDigweed and Dr Ilja Demuth for their exceptional skills intraining young scientists and Dr Hans-Dieter Volk for hiscontinued guidance.

References

1. Lamy T, Loughran TP Jr. Current concepts: largegranular lymphocyte leukemia. Blood Rev. 1999;13:230-240.

2. Kuzushima K, Hoshino Y, Fujii K, et al. Rapid de-termination of Epstein-Barr virus-specific CD8(�)T-cell frequencies by flow cytometry. Blood. 1999;94:3094-3100.

3. Gillespie GM, Wills MR, Appay V, et al. Functionalheterogeneity and high frequencies of cytomega-lovirus-specific CD8(�) T lymphocytes in healthyseropositive donors. J Virol. 2000;74:8140-8150.

4. Bousso P, Kourilsky P. A clonal view of alphabetaT cell responses. Semin Immunol. 1999;11:423-431.

5. Callan MF, Tan L, Annels N, et al. Direct visualiza-tion of antigen-specific CD8� T cells during theprimary immune response to Epstein-Barr virus Invivo. J Exp Med. 1998;187:1395-1402.

6. Maryanski JL, Jongeneel CV, Bucher P, CasanovaJL, Walker PR. Single-cell PCR analysis of TCR rep-ertoires selected by antigen in vivo: a high magnitudeCD8 response is comprised of very few clones. Im-munity. 1996;4:47-55.

7. Mandruzzato S, Rossi E, Bernardi F, et al. Largeand dissimilar repertoire of Melan-A/MART-1-specific CTL in metastatic lesions and blood of amelanoma patient. J Immunol. 2002;169:4017-4024.

8. Klein J, Sato A. The HLA system: first of twoparts. N Engl J Med. 2000;343:702-709.

9. Klein J, Sato A. The HLA system: second of twoparts. N Engl J Med. 2000;343:782-786.

10. Loughran TP Jr. Clonal diseases of large granularlymphocytes. Blood. 1993;82:1-14.

11. Gorski J, Yassai M, Zhu X, et al. Circulating T cellrepertoire complexity in normal individuals andbone marrow recipients analyzed by CDR3 sizespectratyping: correlation with immune status.J Immunol. 1994;152:5109-5119.

12. Pannetier C, Even J, Kourilsky P. T-cell repertoirediversity and clonal expansions in normal andclinical samples. Immunol Today. 1995;16:176-181.

13. Arstila TP, Casrouge A, Baron V, et al. Diversity ofhuman alpha beta T cell receptors. Science.2000;288:1135.

14. Silins SL, Cross SM, Krauer KG, et al. A func-tional link for major TCR expansions in healthyadults caused by persistent Epstein-Barr virusinfection. J Clin Invest. 1998;102:1551-1558.

15. Maini MK, Gudgeon N, Wedderburn LR, Rickin-son AB, Beverley PC. Clonal expansions in acuteEBV infection are detectable in the CD8 and notthe CD4 subset and persist with a variable CD45phenotype. J Immunol. 2000;165:5729-5737.

16. Woitas RP, Sippel M, Althausen EM, et al. Differ-ential expansion of T-cell receptor variable betasubsets after antigenic stimulation in patients withdifferent outcomes of hepatitis C infection. Immu-nology. 2002;106:419-427.

17. Risitano AM, Maciejewski JP, Green S, et al. In-vivo dominant immune responses in aplasticanaemia: molecular tracking of putatively patho-genetic T-cell clones by TCR beta-CDR3 se-quencing. Lancet. 2004;364:355-364.

18. Zeng W, Maciejewski JP, Chen G, Young NS.Limited heterogeneity of T cell receptor BV usagein aplastic anemia. J Clin Invest. 2001;108:765-773.

19. Plasilova M, Risitano A, Maciejewski JP. Applica-tion of the molecular analysis of the T cell recep-tor repertoire in the study of immune-mediatedhematologic disease. Hematol J. 2003;8:173-181.

20. Padovan E, Casorati G, Dellabona P, et al. Ex-pression of two T cell receptor alpha chains: dualreceptor T cells. Science. 1993;262:422-424.

21. Davis MM, Bjorkman PJ. T-cell antigen receptorgenes and T-cell recognition. Nature. 1988;334:395-402.

22. Borg NA, Ely LK, Beddoe T, et al. The CDR3 re-gions of an immunodominant T cell receptor dic-tate the ‘energetic landscape’ of peptide-MHCrecognition. Nat Immunol. 2005;6:171-180.

23. Garboczi DN, Ghosh P, Utz U, et al. Structure ofthe complex between human T-cell receptor, viralpeptide and HLA-A2. Nature. 1996;384:134-141.

24. Hedrick SM, Engel I, McElligott DL, et al. Selec-tion of amino acid sequences in the beta chain ofthe T cell antigen receptor. Science. 1988;239:1541-1544.

25. Kalergis AM, Nathenson SG. Altered peptide li-gand-mediated TCR antagonism can be modu-lated by a change in a single amino acid residuewithin the CDR3 beta of an MHC class I-restrictedTCR. J Immunol. 2000;165:280-285.

26. Goyarts EC, Vegh Z, Kalergis AM, et al. Point mu-tations in the beta chain CDR3 can alter the T cellreceptor recognition pattern on an MHC classI/peptide complex over a broad interface area.Mol Immunol. 1998;35:593-607.

27. Engel I, Hedrick SM. Site-directed mutations inthe VDJ junctional region of a T cell receptor betachain cause changes in antigenic peptide recog-nition. Cell. 1988;54:473-484.

28. Huang JC, Ober RJ, Ward ES. The central resi-dues of a T cell receptor sequence motif are keydeterminants of autoantigen recognition in murineexperimental autoimmune encephalomyelitis. EurJ Immunol. 2005;35:299-304.

29. Sourdive DJ, Murali-Krishna K, Altman JD, et al.Conserved T cell receptor repertoire in primaryand memory CD8 T cell responses to an acuteviral infection. J Exp Med. 1998;188:71-82.

Figure 5. Clonotypic expansions and disease course. Changes in the size of LGLclone were recorded prior to and after oral cytoxan therapy resulting in remission in 6patients. (A) Clonotype expansions within CD8 cell population (percent of VB familywithin CD8� population multiplied by % of identical sequences within a VB family). (B)Comparisons of redundancy (defined as percent of the pathogenic clone within VBfamily) before (u) and after (o) therapy and diversity (defined as percent of differentclones within all clones sequenced). (C-D) Examples of the changes of clonal sizeexpressed as percent of the total CD8 spectrum during the course of disease in 2patients with LGL leukemia associated with pure red cell aplasia (u). Lines representthe hemoglobin levels (�) and absolute reticulocyte counts (�). Tx and arrowsindicate transfusions; Hg, hemoglobin; and Ret, reticulocytes. Error bars represent 2 SD.

CLONAL CTL RESPONSES IN LGL LEUKEMIA 2779BLOOD, 15 OCTOBER 2005 � VOLUME 106, NUMBER 8

For personal use only. by guest on December 18, 2013. bloodjournal.hematologylibrary.orgFrom

30. Peggs K, Verfuerth S, Pizzey A, et al. Character-ization of human cytomegalovirus peptide-spe-cific CD8(�) T-cell repertoire diversity following invitro restimulation by antigen-pulsed dendriticcells. Blood. 2002;99:213-223.

31. Hirokawa M, Matsutani T, Saitoh H, et al. DistinctTCRAV and TCRBV repertoire and CDR3 se-quence of T lymphocytes clonally expanded inblood and GVHD lesions after human allogeneicbone marrow transplantation. Bone MarrowTransplant. 2002;30:915-923.

32. Bourcier KD, Lim DG, Ding YH, et al. ConservedCDR3 regions in T-cell receptor (TCR) CD8(�) Tcells that recognize the Tax11–19/HLA-A*0201complex in a subject infected with human T-cellleukemia virus type 1: relationship of T-cell finespecificity and major histocompatibility complex/peptide/TCR crystal structure. J Virol. 2001;75:9836-9843.

33. Matsumoto I, Tsubota K, Satake Y, et al. CommonT cell receptor clonotype in lacrimal glands andlabial salivary glands from patients with Sjogren’ssyndrome. J Clin Invest. 1996;97:1969-1977.

34. Sensi M, Salvi S, Castelli C, et al. T cell receptor(TCR) structure of autologous melanoma-reactivecytotoxic T lymphocyte (CTL) clones: tumor-infil-trating lymphocytes overexpress in vivo the TCRbeta chain sequence used by an HLA-A2-restricted and melanocyte-lineage-specific CTLclone. J Exp Med. 1993;178:1231-1246.

35. Lamy T, Loughran TP Jr. Clinical features of largegranular lymphocyte leukemia. Semin Hematol.2003;40:185-195.

36. Bible KC, Tefferi A. Cyclosporine A alleviates se-vere anaemia associated with refractory largegranular lymphocytic leukaemia and chronicnatural killer cell lymphocytosis. Br J Haematol.1996;93:406-408.

37. Rose MG, Berliner N. T-cell large granular lym-phocyte leukemia and related disorders. Oncolo-gist. 2004;9:247-258.

38. Berliner N, Duby AD, Linch DC, et al. T cell recep-tor gene rearrangements define a monoclonal Tcell proliferation in patients with T cell lymphocy-tosis and cytopenia. Blood. 1986;67:914-918.

39. Langerak AW, van den BR, Wolvers-Tettero IL, etal. Molecular and flow cytometric analysis of theVbeta repertoire for clonality assessment in ma-ture TCRalphabeta T-cell proliferations. Blood.2001;98:165-173.

40. Lima M, Almeida J, Santos AH, et al. Immunophe-notypic analysis of the TCR-Vbeta repertoire in98 persistent expansions of CD3(�)/TCR-alpha-beta(�) large granular lymphocytes: utility in as-sessing clonality and insights into the pathogene-sis of the disease. Am J Pathol. 2001;159:1861-1868.

41. Greer JP, Kinney MC, Loughran TP Jr. T cell andNK cell lymphoproliferative disorders. In: Hema-tology. Washington, DC; American Society of He-matology; 2001;259-281.

42. Bowman SJ, Hall MA, Panayi GS, Lanchbury JS.T cell receptor alpha-chain and beta-chain junc-tional region homology in clonal CD3�, CD8� Tlymphocyte expansions in Felty’s syndrome. Ar-thritis Rheum. 1997;40:615-623.

43. Kwong YL, Wong KF. Association of pure red cellaplasia with T large granular lymphocyte leukae-mia. J Clin Pathol. 1998;51:672-675.

44. Ergas D, Tsimanis A, Shtalrid M, Duskin C, Ber-rebi A. T-gamma large granular lymphocyte leu-kemia associated with amegakaryocytic thrombo-cytopenic purpura, Sjogren’s syndrome, andpolyglandular autoimmune syndrome type II, withsubsequent development of pure red cell aplasia.Am J Hematol. 2002;69:132-134.

45. Go RS, Lust JA, Phyliky RL. Aplastic anemia andpure red cell aplasia associated with large granu-lar lymphocyte leukemia. Semin Hematol. 2003;40:196-200.

46. Sood R, Stewart CC, Aplan PD, et al. Neutrope-nia associated with T-cell large granular lympho-cyte leukemia: long-term response to cyclospor-ine therapy despite persistence of abnormal cells.Blood. 1998;91:3372-3378.

47. Starkebaum G, Loughran TP Jr, Gaur LK, DavisP, Nepom BS. Immunogenetic similarities be-tween patients with Felty’s syndrome and thosewith clonal expansions of large granular lympho-cytes in rheumatoid arthritis. Arthritis Rheum.1997;40:624-626.

48. Liu JH, Wei S, Lamy T, et al. Blockade of Fas-dependent apoptosis by soluble Fas in LGL leu-kemia. Blood. 2002;100:1449-1453.

49. Tanaka M, Suda T, Haze K, et al. Fas ligand inhuman serum. Nat Med. 1996;2:317-322.

50. Epling-Burnette PK, Liu JH, Catlett-Falcone R, etal. Inhibition of STAT3 signaling leads to apopto-sis of leukemic large granular lymphocytes anddecreased Mcl-1 expression. J Clin Invest. 2001;107:351-362.

51. Epling-Burnette PK, Loughran TP Jr. Survival sig-nals in leukemic large granular lymphocytes. Se-min Hematol. 2003;40:213-220.

52. Melenhorst JJ, Sorbara L, Kirby M, Hensel NF,Barrett AJ. Large granular lymphocyte leukaemiais characterized by a clonal T-cell receptor rear-rangement in both memory and effector CD8(�)lymphocyte populations. Br J Haematol. 2001;112:189-194.

53. Kook H, Risitano AM, Zeng W, et al. Changes inT-cell receptor VB repertoire in aplastic anemia:effects of different immunosuppressive regimens.Blood. 2002;99:3668-3675.

54. Hingorani R, Monteiro J, Furie R, et al. Oligo-clonality of V beta 3 TCR chains in the CD8� Tcell population of rheumatoid arthritis patients.J Immunol. 1996;156:852-858.

55. Gonzalez-Quintial R, Baccala R, Pope RM, Theo-filopoulos AN. Identification of clonally expandedT cells in rheumatoid arthritis using a sequenceenrichment nuclease assay. J Clin Invest. 1996;97:1335-1343.

56. Hafler DA, Duby AD, Lee SJ, et al. Oligoclonal Tlymphocytes in the cerebrospinal fluid of patientswith multiple sclerosis. J Exp Med. 1988;167:1313-1322.

57. Monteiro J, Hingorani R, Peroglizzi R, Apatoff B,Gregersen PK. Oligoclonality of CD8� T cells inmultiple sclerosis. Autoimmunity. 1996;23:127-138.

58. Akashi K, Shibuya T, Taniguchi S, et al. Multipleautoimmune haemopoietic disorders and insidi-ous clonal proliferation of large granular lympho-cytes. Br J Haematol. 1999;107:670-673.

59. Bowman SJ, Bhavnani M, Geddes GC, et al.Large granular lymphocyte expansions in pa-tients with Felty’s syndrome: analysis using anti-Tcell receptor V beta-specific monoclonal antibod-ies. Clin Exp Immunol. 1995;101:18-24.

60. Molad Y, Okon E, Stark P, Prokocimer M.Sjogren’s syndrome associated T cell largegranular lymphocyte leukemia: a possible com-mon etiopathogenesis. J Rheumatol. 2001;28:2551-2552.

61. Trentin L, Imberti L, Zambello R, et al. Detectionof identical T-cell clonotype expansions in boththe donor and recipient after allogeneic bonemarrow transplantation. Br J Haematol. 1999;106:119-127.

62. Matsutani T, Yoshioka T, Tsuruta Y, et al. Determi-nation of T-cell receptors of clonal CD8-positiveT-cells in myelodysplastic syndrome with ery-throid hypoplasia. Leuk Res. 2003;27:305-312.

63. O’Keefe CL, Plasilova M, Wlodarski M, et al. Mo-lecular analysis of TCR clonotypes in LGL: aclonal model for polyclonal responses. J Immu-nol. 2004;172:1960-1969.

64. Semenzato G, Zambello R, Starkebaum G, Os-

himi K, Loughran TP Jr. The lymphoproliferativedisease of granular lymphocytes: updated criteriafor diagnosis. Blood. 1997;89:256-260.

65. Herling M, Khoury JD, Washington LT, et al. Asystematic approach to diagnosis of mature T-cellleukemias reveals heterogeneity among WHOcategories. Blood. 2004;104:328-335.

66. Starkebaum G. Chronic neutropenia associatedwith autoimmune disease. Semin Hematol. 2002;39:121-127.

67. Bennett JM, Catovsky D, Daniel MT, et al. Pro-posals for the classification of the myelodysplas-tic syndromes. Br J Haematol. 1982;51:189-199.

68. van Dongen JJ, Langerak AW, Bruggemann M, etal. Design and standardization of PCR primersand protocols for detection of clonal immuno-globulin and T-cell receptor gene recombinationsin suspect lymphoproliferations: report of theBIOMED-2 Concerted Action BMH4-CT98–3936.Leukemia. 2003;17:2257-2317.

69. Lefranc M-P, Lefranc G. The T cell receptorFactsBook. London, UK: Academic Press; 2001.

70. The International Immunogenetics InformationSystem. IMGT/V-QUEST. http://imgt.cines.fr/cgi-bin/IMGTdnap.jv?livret0&OptionhumanTcR.Accessed April 2004.

71. The International Immunogenetics InformationSystem. http://imgt.cines.fr/. Accessed April 2004.

72. Stryer L. Biochemistry. 3rd ed. New York, NY:W. H. Freeman and Company; 1988.

73. McFarland EJ, Harding PA, Striebich CC,MaWhinney S, Kuritzkes DR, Kotzin BL. ClonalCD8� T cell expansions in peripheral blood fromhuman immunodeficiency virus type 1-infectedchildren. J Infect Dis. 2002;186:477-485.

74. Beck RC, Wlodarski M, Gondek L, et al. Efficientidentification of T-cell clones associated withgraft-versus-host disease in target tissue allowsfor subsequent detection in peripheral blood. Br JHaematol. 2005;129:411-419.

75. Davey MP, Starkebaum G, Loughran TP Jr.CD3� leukemic large granular lymphocytes uti-lize diverse T-cell receptor V beta genes. Blood.1995;85:146-150.

76. Dhodapkar MV, Li CY, Lust JA, Tefferi A, PhylikyRL. Clinical spectrum of clonal proliferations ofT-large granular lymphocytes: a T-cell clonopathyof undetermined significance? Blood. 1994;84:1620-1627.

77. Kjer-Nielsen L, Clements CS, Purcell AW, et al. Astructural basis for the selection of dominant al-phabeta T cell receptors in antiviral immunity. Im-munity. 2003;18:53-64.

78. Tobin G, Thunberg U, Karlsson K, et al. Subsetswith restricted immunoglobulin gene rearrange-ment features indicate a role for antigen selectionin the development of chronic lymphocytic leuke-mia. Blood. 2004;104:2879-2885.

79. Tomiyama H, Yamada N, Komatsu H, HirayamaK, Takiguchi M. A single CTL clone can recognizea naturally processed HIV-1 epitope presented bytwo different HLA class I molecules. Eur J Immu-nol. 2000;30:2521-2530.

80. European Bioinformatics Institute. ClustalW.http://www.ebi.ac.uk/clustalw/. Accessed April2004.

81. Mohty M, Faucher C, Vey N, et al. Features oflarge granular lymphocytes (LGL) expansion fol-lowing allogeneic stem cell transplantation: along-term analysis. Leukemia. 2002;16:2129-2133.

82. Au WY, Lam CC, Lie AK, Pang A, Kwong YL. T-cell large granular lymphocyte leukemia of donororigin after allogeneic bone marrow transplanta-tion. Am J Clin Pathol. 2003;120:626-630.

83. Goulmy E, Pool J, van den Elsen PJ. Interindi-vidual conservation of T-cell receptor beta chainvariable regions by minor histocompatibility anti-gen-specific HLA-A*0201-restricted cytotoxic T-cell clones. Blood. 1995;85:2478-2481.

2780 WLODARSKI et al BLOOD, 15 OCTOBER 2005 � VOLUME 106, NUMBER 8

For personal use only. by guest on December 18, 2013. bloodjournal.hematologylibrary.orgFrom