Download Evidence for Antigenic Selection of Large Granular Lymphocytes in

From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
Evidence for Antigenic Selection of Large Granular Lymphocytes
in a Patient With Wiskott-Aldrich Syndrome
By Alessandra Sottini, Alessandra Bettinardi, Eugenia Quiros-Roldan, Alessandro Plebani, Paolo Airo,
Daniele Prirni, and Luisa lmberti
It is now recognized that CD3' large granular lymphocyte
( L W prolierations may be clonally derived from their normal CD3+LGL+ counterpart, but the nature of the pressure
responsible for the prolieration of these cells remains unclear. We approachedthis problem by analyzing the diversity
of the T-cell receptor repertoire of LGLdeveloped in different
clinical settings. Two of our patients had typical lymphoproliferative disorders. The third case was much more unusual,
as the LGL proliferation was associated with a Wiskott-Aidrich syndrome. Our data relative to the patients with the
lymphoproliferative disorders only suggest that these LGL
were clonally expanded. The data relative to the patient with
Wiskott-Aldrich syndrome were more unexpected, as the T-
cell repertoire of the LGL appeared to have common features
with that of the other T-cell populations analyzed. These
LGL were characterized by the clonal expansion of a few
TCRBV segments that shared common amino acid motifs in
the junctional region of the T-cell receptor. This common
pattern of junctional diversity associated with different
TCRBV segments is, therefore, consistent with a strong ongoing antigenic selection process,possibly related to the
pathogenesis of Wiskott-Aldrich syndrome. Furthermore,
the finding that thesame TCRBV segments were also highly
expanded among other T-cell subpopulations questions the
malignant nature of this LGL proliferation.
0 1995 by The American Society of Hematology.
A
the apparentlyconflictingresults
concerning themonoclonal4"' or p o l y ~ l o n anature
l ~ ~ ~of
~~
LGL expansionscan be
conciliated by proposing that the initial stage of the disease
is characterized by polyclonal expansions, from which oligoclonal populations can subsequently be selected as the
consequence of antigenic pressure by unknown ligandsx In
accordance with thishypothesis, Kasten-Sportsset
have
recently reported that the LGL population of a patient, characterized by the monoclonal expression of one TCRBV segment, useda common TCRAV chain that was associated
with a multiplicity of N region additions and TCRAJ segment rearrangements. This result has been interpreted as the
action of an immune selection-mediated process, occurring
at an early stageof cell maturation that may have contributed
to the selection of this particular clone.
Recently,
patient
a
with Wiskott-Aldrich syndrome
(WAS) who presented with lymphocytes with morphologic
characteristics of LGL wasreferred to the Department of
Paediatrics of the Spedali Civili of Brescia, Italy. WAS is
an X-linked disorder characterized by severe thrombocytopenia, eczema,andprofound immunodeficiencyinvolving
~ ' ~ ~ ~ WAS is associated
both B and T l y m p h o ~ y t e s . Although
with an increased risk of malignancies, particularly lymphoreticular tumors andleukemia,?" LGL proliferations are probably rare in primary immunodeficiencies, as they have been
so far detected only in a patient with Ataxia Telangiectasia.?'
The availabilityof LGL developed in an unusualclinical
context provided a unique opportunity for further defining
thenature of theselectivepressureresponsible
forLGL
expansions. In this study, therefore, we analyzed, at the molecular level, the TCRBV repertoire of the LGL population
expanded in the peripheral blood of this patient. Our data
demonstrate that the repertoire of all T-cell populations analyzed are characterized by the clonal expansion of different
TCRBV segments that share common motifsatthe junctional regions, a condition that clearly distinguishes this patient from the others studied.
CENTRALQUESTIONconcerningthe
nature of
lymphoproliferative diseaseof large granular lymphocytes (LGL)is whether it represents a neoplasticor a reactive
disorder. Because oftheirgenerally
chronicand indolent
clinical course, LGL proliferations were for many years believed not to be monoclonal leukemia. The discovery of the
genomic organization of the T-cell receptor (TCR)',2 andthe
consequent possibility of defining T-cell clonality using
TCR
probes in Southern blot analysis? have,however,greatly
contributed to the revisionof this belief.It is now recognized
that the majority of CD3+ LGL are clonally derived from
their normal counterpart: but the nature of the pressure responsiblefortheexpansion
ofthese
cellsremains unclear. The demonstration of TCR clonal rearrangements of
LGL4"" does not necessarily imply a neoplastic transformation, as oligoclonal T-cell populations havebeendetected
inseveral human disease^""^ and in patientswith severe
combined immunodeficiency (SCID) and
maternal
engraftment.'* Furthermore,clonal T-cellpopulationshave
been found within the CDVCD45RO' subset" and in elderly2"
normal
human
lymphocytes,
as
well in
as
CD8+CD57' T cells prepared from normal individuals and
rheumatoidarthritis patients." These findings demonstrate
that in vivo selective pressures can also be responsible for
the expansion of clonal or oligoclonal cell populations. Thus,
Fromthe Terzo Laboratorio Analisi and Clinical Immunology:L..
Consorzio per le Biotecnologie, Spedali Civili, Brescia; andthe
Department of Paediatrics, University of Genova, Istituto G. Gaslini,
Genova. Ita1.y.
Submitted February 6, 1995; accepted May 9, 1995.
Supported by Sorin Biomedica (Saluggia, Italy), Consiglio Nazionale delle Ricerche (Grant No. 94.01328.PF39), and Istituto
Superiore di Sanita (VIII Progerto AIDS).
Address reprint requests to Daniele Primi, PhD, Consorzio per
le Biotecnologie, Laboratorio di Biotecnologie, P. le Spedali Civili,
I , 25123 Brescia, Italy.
The publication costs of this article were defrayed in part by page
charge payment. This article must therefore be hereby marked
"advertisement" in accordance with 18 U.S.C. section 1734 solely to
indicate this fact.
0 1995 by The American Society of Hematology.
0006-4971/95/8606-0002$3.00/0
2240
PATIENTS AND METHODS
Patients. Patient FC, a whiteboy,was born in May 198.5 after
8 months of age,thechild was
anuncomplicatedpregnancy.At
hospitalized and was found to have widespread petechiae, ecchymoBlood, Vol 86, No 6 (September 15), 1995: pp 2240-2247
15
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
TCR REPERTOIRE IN2241
LGL OF WAS PATIENT
Table 1. Immunologic Evaluation of Patients
Patients (sex/age lyrl)
Controls*
FC (MM
1986
Lymphocytes (/pL)
CD2 (%)
CD3 (%l
CD4 (%)
CD8 (%)
58CD51 (%)
CD16 (%)
CD19 (%)
TCRAB (%l
TCRGD (%)
58
3,400
ND
77
31
33
10
ND
8
ND
ND
1991
1994
2,565
84
55
19
9
61
25
2,900
73
38
12
55
4
10
42
0.3
FLR
(M/2)
IT
(F/63)
FS
(M/68)
Children (1 to 9
yrs old)
Adults (20 to
50 yrs old)
1,420
48
44
4,018
85
84
10,296
99
99
4,068 2 1,391
80 2 6
71 2 7
2,382 -C 914
78.5 +- 6.5
74 2 7
14
18
92
64
1
5
ND
2
7
26
30
16
6
2
8
ND
ND
<l
<l
ND
ND
40 2 7
-C 5
<l0
11 c 5
18 2 6
65 2 8
<6
45 +- 7.5
30 C 10
14 2 11
13 2 7
6-c2
61 2 8
<8
<5
1200
15 -c 10
250-560t
LGL
ND
ND
YO
/!JL
56
1,624
51
1,4622,411
1
14
60
90
9,266
Abbreviation: ND, not determined.
* Data represent the mean values 2 SD obtained with lymphocytes from 10 children and 20 adults.
t See Pandolfi et al?
sis, extensive atopic dermatitis, and a small platelet count of 30 X
1 0 9 L On the basis of clinical and laboratory data, a diagnosis of
WAS was made. The platelet count decreased over time until 5 years
of age when, after a splenectomy, platelet number (80 X 109/L to
140 X 109/L) and size increased. In February 1991, he developed
arthritis of the right knee, and serum IgM-rheumatoid factor and
antiplatelet antibodies were detected. Treatment with nonsteroidal
antiinflammatory drugs led to clinical improvement, but positivity
for the rheumatoid factor persisted. At that time, the patient presented
with a very high number of circulating LGL (56%).
Patient FLR was a boy with WAS but without LGL proliferation,
while patients IT and FS had classical, long-lasting, lymphoproliferative disorders characterized by the expansion of CD8+LGL+.Patients
IT and FS did not show splenomegaly, lymphoadenopathy, or hepatomegaly, and they were negative for rheumathoid factor, antinuclear
antibodies, and the Coomb's test. However, patient IT presented at
the time of the diagnosis with multinodular goiter and antithyroperoxidase and antithyreoglobulin antibodies. Both patients have not
required therapy and have remained relatively asymptomatic until
now. The immunophenotypes of all patients are listed in Table 1,
while the two-color fluorescence analysis performed on patient FC
at the time of the present study is summarized in Table 2.
Using light microscopy, the morphology of LGL was similar in
all cases. The LGL were medium-sized cells with moderate to abundant pale blue cytoplasm containing azurophilic granules. The cells
were phosphatase- and beta-glucuronidase-positive.
Table 2. Two-Color Fluorescence Analysis
of Lymphocytes of Patient FC
Cell Subsets
CD8'CD57'
CD8+CD57CD8-CD57'
CD8-CD57CD4TD57'
CD4+CD57CD4-CD57'
CDb-CD57-
% of Total Lymphocytes
4.6
62.3
Lymphocytes preparation. Blood samples from patient FC were
obtained on two different occasions, while a single sample was
obtained from patients E R , IT, and FS. Samples from children and
adult normal healthy donors were included in this study as control
material. Peripheral blood mononuclear cells were obtained after
Ficoll Hypaque gradient centrifugation, and different lymphocyte
subpopulations were prepared using magnetic microspherical beads
coated with specific monoclonal antibodies (MoAbs) and DETACHaBEAD (Dynabeads N-450; Dynal, Oslo, Norway), according
to the manufacturer's instructions. Four different subpopulations
were obtained from the blood of patient FC (CD8+CD57',
CD8'CD57-, CD8-CD57'.
and CD8-CD57-), while the cells of
patients FLR and IT were separated in CD8+ and CD8- subsets. As
CD8+LGL+represented the vast majority of the total T-cell population of patient FS, no cell separation was performed in this sample.
The purity of cell preparations was analyzed by cytofluorimetry after
each step of the preparations.
Cytofuorirnetric analysis. Purified, phycoerythrin (PE)- or fluorescein isothiocyanate (FITC)-conjugated anti-CD2, -CD3, -CM,
-CD8, -CD16, and -CD19 MoAbs were purchased from Ortho Diagnostic Systems (Raritan,NY), while anti-CD57 and anti-TCRAB
MoAbs were obtained from Becton Dickinson (San Jose, CA). The
anti-TCRGD MoAb was from Cell Diagnostics (Cambridge, MA).
Cells were analyzed using a flow cytometer (FACScan; Becton Dickinson, Erembodegem-aalst, Belgium) equipped with an argon-ion
laser. Cells were stained according to the manufacturer's instructions. Briefly, 5 X lo5to 1 X lo6cells were incubated for 30 minutes
at 4°C with the FITC-conjugated MoAbs or, for double-fluorescence
analysis, with PE- andFITC-conjugated MoAbs. The flow cytometer
was gated to include only viable cells. Using logarithmic amplificationof fluorescence intensity, 1 X IO4 cells were accumulated for
histograms.
Preparation of RNA, cDNA synthesis, and arnpliJication of cDNA
by polymerase chain reaction (PCR). Total RNA was prepared by
the guanidinium thiocyanate-phenol-chloroform method from total
lymphocytes or lymphocyte subpopulations. One microgram of total
RNA was used to synthesize the first strand of the TCRB chainspecific complementary DNA (cDNA) using the RiboClone cDNA
Synthesis System (Promega Corp, Madison, WI) and a primer specific for TCRBC1 and TCRBC2 genes (PcDNA: 5' GGG CTG CTC
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
SOTTlNl ET AL
2242
Table 3. Distribution of LGL Among Purified Cell Populations
LGL
31
350
Absolute No./pL'
Cell Subsets
%
CD8'CD57'
85
1,090
CD~*CD~~
10
80
6
35
cm~~57'
CD8-CD57-
Extrapolated from the relative percentage of LGL distribution.
were purified by cutting the band with the expected size from NuSieve GTG low-melting agarose 2.5% gel (FMC BioProducts. Rockland, ME) and eluting the melted gel through an ion-exchnnFe resin
column (Qiagen tip S: Qiagen Inc. Chatsworth. CA). Purified DNA
fragments were ligated to a pCR I1 vector (Invitrogen Corp.San
Diego. CA). Plasmids were grown in INVaF' modified competent
E.sc/~crichicrroli cells in LB agar plates. and single plaques were
picked up and expanded according to the manufacturer's instructions
(TA Cloning Kit: Invitrogen). To verify the presence of the correct
insert, recombinant plaques were tested byPCR with the abovedescribed TCRBV-specilic family and TCRBC primers. TCRBVI-.
TCRBV2-. TCRBV 13sI TCRBV I3S2-. TCRBV14-. TCRBV IS-.
TCRBV 17-.and TCRBV20-positive plaques were selected. and
plasmid DNA was purified using QlAwell X Plasmid Kit (Qiagen)
and sequenced withan Automated Laser Fluorescent ALF DNA
Sequencer (Pharmacia LKB. Uppsala. Sweden) using the AutoRead
Sequencing Kit (Pharmacia). In the case of the TCRBV14 chain,
both universal and reverse primers were used. Sequences were compared with published data relative to TCRBV. TCRBD. TCRBJ. and
TCRBC segments.'.".'"
-.
CTT GAG GGG CTG CGG 3'). The cDNA was then subjected to
enzymatic amplification using a second human TCRBC primer @AI:
S' CCC ACT GTG CAC CTC CTT CC3') and a TCRBV degenerated primer [Vpd: S' ACG TGA ATT CT(GT) T(ACT)(CT) TGG
TA(CT)(AC)(AG)(AT) CA 3'1, which was designed to amplify
TCRB chain rearrangements containing virtually all the known human TCRBV genes." The original sequence of this primer was
modified by adding at the S' end an EroRI restriction site. The PCR
products are about 400 base pairs long: they contain one half of the
V region gene and extend through the V-D-J junction to the TCRBC
region. The cycles ofPCR were performed under the following
conditions: denaturation at 93°C for I minute, annealing at 52°C for
I minute. and extension at 72°C for I minute. The last cycle extension was performed at72°C for 7 minutes. The specificity of the
total amplified products was analyzed using the previously described
colorimetric method and biotinylated TCRBV-specific probes.'' The
relative percentage of expression of each of the TCRRV segments
analyzed was calculated by normalizing the optical density (OD)
value of each individual TCRBV segment [OD(i)l with respect to
the sum of the OD values of all 26 TCRBV chains. as follows:
o/c
of Expression = -x 100
1
I-
I
Subsequently, the TCRBV chains of interest were amplified by PCR
using TCRBV-specific family primers (TCRBVI: S' GCA CAA
CAG TTC CCT GAC TTG CAC3': TCRBV2: S' TCA TCA ACC
ATG CAA GCC TGA CCT 3': TCRBV 13sI: S' CAA GGA GAA
GTC CCC AAT 3'; TCRBV13S2: S' GGT GAG GGT ACA ACT
GCC 3': TCRBV14: S' GTC TCT CGA AAA GAG AAGAGG
AAT 3': TCRBVIS: S' ACT GTC TCT CGA CAG GCA CAG
GCT 3'; TCRBV17: S' CAG ATA GTA AAT GAC TTT CAG 3':
and TCRBVZO: S' AGC TCT GAG GTG CCC CAG AAT CTC 3')
and a TCRBC oligonucleotide (BAl).
Clonin'qcrnrl.sc,clrrcwcin,y of PCR products. The PCR products
RESULTS
TCRRV IISP o f T-cell sul~p(~pr:lrtiorIs
obtained ,from the
patient with LGL prol(feration and WAS. The initial morphologic and phenotypic characterization of the lymphocytes
of patient FC suggested that the totality of CD8- cells were
LGL. Because a fraction of CD8' lymphocytes were CDS7-.
we initially studied the repertoire of the CDg'CDS7- subset.
We observed that these lymphocytes were characterized by
the
preferential
use of TCRBV13SI.
TCRBV13S2.
TCRBV17. and TCRBV22 segments (data not shown). To
understand whether this peculiar pattern of TCRBV use was
a characteristic of LGL expressing the typical CD8TDS7'
phenotype only. we extended
the
analysis to the
CD8'CDS7-,CD8-CDS7'.
andCD8-CDS7-subpopulations prepared from a second sample of blood. obtained 3
months later.Detailedtwo-colorimmunophenotyping
was
undertaken by flow cytometer to determine the purity of each
cellseparation.Afterbeadsdetachment.
869 of purified
CD8' cells(98% viable and99%pure)
coexpressed the
CD3 marker; 83% of these isolated CD8- lymphocytes were
CD8'CDS7', and only 1.7% were CD4'CDS7*. The CD%
cell preparation still contained 1 1 % of contaminating cells
expressing the CD8 marker, at low density. These contaminating cells, however. were CD3-. The CD8TDS7-prepa-
I
-
N
~
'p.
m
.
m
ro
+
-
N
~
~
v,
2
"
TCRBV CHAINS
m
~
=
=
y
,
N
Fig 1. TCRBV expression of
CDE+CD57' (L).
CD8'CD57- ID),
and CDEXD57' ( C ) subpopulations obtained from patient FC.
Data are expressed as percentage of the colorimetricsignal obtained with the individual
~
r
~
~
m
TCRBV-specific
probes,
calculated as described in Patients
and Methods.
~
~
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
TCRREPERTOIRE
IN LGL
2243
OF WAS PATIENT
-
L
v1
W
E
201
15
S
ni
Fig 2. TCRBV expression of
CD8' (NI and CD8- (W) subpopulations obtained from patient
perFLR. Data as
are expressed
centage of the colorimetric signe1 obtained with the individual
TCRBV-specific probes.
g
I
v )) .vn
v
l
ration was contaminated by 10% of cells coexpressing CD8
and CD57 markers, while separated CD8-CD57- cells contained only 1% of CDSTD57' lymphocytes. The distribution of LGL in the different cell preparations is shown in
Table 3. Surprisingly, we found that the TCRBV13S1,
TCRBV13S2, TCRBV17, and TCRBV22 segments were
dominantly expressed notonly on the CD8+CD57+ cells,
but also on the CDVCD57- and CDSTD57+ subsets, and,
furthermore, we also detected the expansion of the
TCRBV24 segment in two of the three cell populations analyzed and an increased expression of TCRBV14 in
CD8-CD57+ lymphocytes (Fig I). We did not obtain any
amplification product from CD8-CD57- cells, probably because the number of recovered cells, after the two negative
separations by magnetic beads, was below the detection capability of the degenerated PCR assay. The abnormal expansion of only few TCRBV chains in patient FC did not appear
to be selectively related to WAS, because the TCRBV profile
of CD8+ and CD8- lymphocytes prepared from a second
patient with this immunodeficiency, in the absence of LGL
proliferation (Fig 2), did not deviate significantly from that
of normal individuals.
The nature (monoclonal or polyclonal) of the TCRBV
expansions observed in patient FC was determined by nucleotide sequencing of the V-D-J junctions of the expanded
segments. More precisely, we analyzed the junctional regions of TCRBV13S2 cDNA clones derived from
CD8+CD57+,CD8'CD57-, and CDKCD57+ lymphocytes
and of TCRBV17 clones prepared from CD8+CD57+ and
CD8+CD57- cells. We also sequenced TCRBV13S1 clones
obtained from purified CD8+CD57+ lymphocytes and
TCRBV14 chains derived from CD8+CD57+ and
CD8-CD57+ cells. The majority of TCRBV13S2 sequences
derived from both CD8+CD57+and CD8+CD57- cells used
the same TCRBJ segment and the same NDN junctional
sequence, suggesting that TCRBV13S2 enrichment is due
to the expansion of a dominant clone (Fig 3, left). Interestingly, six of the eight TCRBV13S2 clones derived from the
CDKCD57' cells contained an NDN region identical to that
of the dominant CD8TD57' clone, with the only exception
being a C-to-A nucleotide substitution that resulted in a replacement of aspartic acid by alanine. Furthermore, the clone
derived from CD8-CD57+ cells and bearing the TCRBJ2S 1
I
I
I
I
,..
I
~
"TT
I
7
T
T
FT
,.,
TCRBV CHAINS
region had a junctional sequence closely related to the identical TCRBJ2S7 clones found in the CD8+CD57+ and
CDS'CD57- lymphocyte subsets. Because the CDS'CD57cell population contained only 10% of LGL, these data suggest that, in this patient, the expansion of closely related and
still differentiating clones expressing TCRBV13S2 may not
be selectively restricted to LGL.
The analysis of TCRBV17 segments derived from both
CD8+CD57+and CDVCD57- T-cell subsets also revealed
a pattern of rearrangement that defined the dominant presence of transcripts encoding the same TCRBJlSl segment
and the identical junctional region (Fig 3, right). Similar
molecular analysis of the TCRBV13S 1 segment confirmed
the expansion of a dominant clone in the CD8+CD57+lymphocytes. Strikingly, the TCRBV13S1 clone contained, at
the first three positions of the NDN region, the same prolineserine-serine motif that characterized the junctional region
of the TCRBV17 transcript. Thus, although these clones use
different TCRBV and TCRBJ segments, it is likely that they
may have been selected by the same, or closely related,
dominant T-cell epitope. To determine whether oligoclonality was also detected in cells expressing less dominant
TCRBV segments, we also analyzed TCRBV14 transcripts
derived from CDS+CD57+ and CDS-CD57+ cells.
TCRBV14 transcripts derived from CD8'CD57+ cells were
characterized by polyclonal V-D-J rearrangements, and there
was no evidence of clonal selection. The sequence analysis
of TCRBV14 transcripts expressed by CDS-CD57+ cells,
on the other hand, revealed a completely different pattern of
rearrangements that defined the dominant presence of a
group of sequences encoding the same TCRBJ2S5 segment
and a stop codon in the third hypervariable (CDR3) region.
Because this nonfunctional transcript was found to be highly
represented in the CD8-CD57+ lymphocytes, it is likely that
it represents the product of the nonfunctionally rearranged
allele of the dominant clone, detected in the same subset and
expressing the TCRBV13S2 segment.
TCRBV use in patients with LGL proliferation not associatedwith WAS. To establish whether the unusual LGL
clonal expansions detected in patient FC are typical of this
clinical setting or reflect a more general property of LGL
transformation, we extended the analysis of the TCR repertoire to two additional patients affected by a lymphoprolifer-
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
SOTTlNl ET AL
2244
TCRBV13S2
CDB+CD57+
CDB*CD57+
TSVYFCASS
PGLED
FFG
TSVYFCASS LGGPGHG EQYFG
151
257
(6/7)
1
TAFYLCASS
NTEAFFG
PSS
1s1 (10/10)
TAFYLCASS
NTEAFFG
PSS
1s1
(7/7)
TSVYFCASS
EQYFG
PSSRN
2S7
l7/7)
TSLYFCASS
TSLYFCASS
TSLYFCASS
TSLYFCASS
TSLYFCASS
1s2
1s2
2s1
1
1
2
CD8*CD57-
CDB'CDS 7 -
TSVYFCASS
PGLED
FFG
TSVYFCASS LGGPGHGEQYFG
1S1
257
(7/8)
1
CDB'CD57'
TSVYFCASS
PGL
EAFFG
TSVYFCASS
LGHL
SNQPQHFG
TSVYFCA
LGGPR
NEQFFG
1S1
1S5
ZS1
(6/8)
1
1
TCRBV13Sl
CDB'CD57'
TCRBV14
CDB'CD57'
RGF
NYGYTFG
MLGAGNLN
GYTFG
SMDMD-R SYNEQFFG
LSSGGASLHEQYFG
RPRASGW
YEQYFG
257
257
1
2S5
(7/7)
1
CDB-CD57+
TSLYFCASS L-GWIR*'
ETQYFG
Fig 3. (Left) Junctional TCRBV13SP sequences obtained from CD8+CD57+. CD8+CD57-, and CD8-CD57' subpopulations prepared from the
lymphocytes of patient FC. (Right) JunctionalTCRBV17, TCRBV13S1, and TCRBV14 sequences obtained from differentT-cell subpopulations
prepared from lymphocytes of the patient with
WAS. Amino acid sequences were deduced from nucleotide sequences and are shown using
standard one-letter code. Only the last nine amino acids of the TCRBV segments and the first5' amino acids of TCRBJ chains are shown. -,
Stop signal; *, changes in the reading frame; ', clones that were sequenced with both universal and reverse primers; n, number of clones
sequenced.
monoclonal in the first subset (20 identical V-D-J sequences)
and polyclonal in the second (Fig S). The sequence analysis
of another randomly chosen TCRBV segment, such as
TCRBV2, confirmed the polyclonality of CD8+ cells bearing
this molecule. In conclusion, the strict monoclonality of the
LGL proliferations observed in these two patients contrasts
sharply with the more complex pattern of clonal expansions
detected in the patient with WAS, and collectively, the data
underscore the complexity of this lymphoproliferative disorder.
ative disorder but not by WAS. The PCR analysis of lymphocytes from patient FS showed a dramatic expansion of the
TCRBVIS segment, which is expressed at a very low level
in lymphocytes obtained from normal subjects, as well as
an increase of the TCRBVl segment (Fig 4). The analysis
of cDNA clones revealed the presence of only one rearrangement in TCRBVlS transcripts, while TCRBVl
clones were characterized by a polyclonal pattern of V-D-J
junctions (Fig 4).
Similar results were obtained with patient IT. In this case,
due to coexistence of CD8+ LGL and normal CD4+ T cells
in the blood circulation, we analyzed the TCRBV chain use
by separated CD8+ and CD8- cell populations. TCRBV20
cDNA clones derived from CD8+ and CD8- cells were
DISCUSSION
In this study we analyzed the diversity of the TCRBV
repertoire of LGL developed in different clinical settings.
CDB'TCRBV15'
"1
a
t4
TCRBVNDN
TEAFFG 1S1
SALYFCASS HLD
NTEAFFG 1S1
SALYFCASS M G G G
SALYFCASS VDNRR
YGYTFG 152
EKLFFG 154
SALYFCASS W A G G G A
SALYFCASS KDPWV
QPQHFG 1S5
SALYFCASS VESGGA
YEQYFG 2S7
SALYFCASS VESGGP
YEQYFG 257
TCRBJAT
"
C
n
R BYKU
2
SALYFCATS
SGLAQ
NTGELFFG
2S2
(10/10)
1
1
1
1
1
1
IS
T
TCRBV CHAINS
Fig 4. TCRBV expression of lymphocytes (NI prepared from patient SF. Data are expressed as percentage of the signal obtained with the
individual TCRBV-specific probes. Results are compared with the average expression of each TCRBV segment in healthy individuals, expressed
as mean ? SD (W). Junctional TCRBVl5 and TCRBVl amino acid sequences were deduced from nucleotide sequences and are shown using
standard one-letter code. Only the last nine amino acids of the TCRBV chains and the first 5' amino acids of TCRBJ segments are shown. n,
number of clones sequenced.
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
TCRREPERTOIRE
IN LGL OF WAS PATIENT
Fig 5. TCRBVexpression of
CD8+ IN) and CD8- 1. subpopulations prepared from patient IT.
Data are exprewd as percentage of the signal obtained with
the individual TCRBV-specific
probes. Junctional TCRBVP and
TCRBVZO amino acid sequences
were deduced from nucleotide
sequences and are shown using
standard one-letter code.Only
the last nine amino acids of the
TCRBVchains and the first 5'
amino acids of TCRBJ segments
are shown.
CDB+TCRBVZO+
CDB*FCRBVP+
TCRsVaaL
"tCRaJ
A12 DSSFYICSAEDQAVRGTEAFFG1S1
NTEAFFG1S1
A19 DSSFYICSA RFT
A5
DSSFYICSA
PRRQN
SNQPQHFG
1S5
A14 DSSFYICSA
RDWEN
YEQYFG 2S7
A21 DSSFYICSA RDMASRHHF
YEQYFG 2S7
Two of our patients had typical lymphoproliferative disorders characterized by the expansion of CD8+LGL+. The
third
case was much more unusual, as the expansion of the LGL
subset was associated with WAS. The long-lasting presence
of LGL in this patient excludes the presence of a transient
lymphocytosis with LGL features, which is usually associated with viral infections?'.40 The total number of circulating
LGL in this patient was constantly below 2,0OO/pL, ie, the
cut-off value currently accepted as diagnostic criteria.' However, clonal LGL proliferations have also been detected in
patients with lower absolute numbers of circulating LGL.4,""
WAS is an X-linked recessive immunodeficiency characterized by eczema, bloody diarrhea, recurrent
and increased risk of autoimmune diseases!2 Because T-cellmediated autoimmune disorders have been postulated to be
involved in LGL expansion: we reasoned that the repertoire
analysis of this patient could be extremely rewarding for
understanding the putative role of the immunologic selection
process in establishing this proliferative syndrome.
Our data relative to patients FS and IT, affected by a
classic lymphoproliferative disorder, confirm andextend previous studies suggesting that at least some LGL proliferations are clonally derived from the pool of the normal lymphocyte~.~
In these two patients, we observed that the
circulating CD8+ lymphocyte subsets were characterized by
the dominant expression of one TCRBV segment, TCRBV 15
and TCRBV20, respectively, with monoclonal rearrangements. The other cell populations analyzed had a
polyclonal pattern of V-D-J junctions. These data are, therefore, consistent with a classical model of neoplastic clonal
evolution, due to the combined action of genetic and environmental factors on a single cell.
The data relative to the patient with WASare more intriguing and collectively imply that the natural history of LGL
proliferation can be more complex than previously appreciated. The TCRBV repertoire of this patient was characterized
by the expression of few TCRBV segments only; however,
these transcripts were dominantly expressed on T-cell subpopulations with different phenotypes. Surprisingly, we
found that an almost identical and monoclonal TCRBV13S2
transcript was dominantly expressed by both CD8+CD57+
and CD8-CD57+ cells. The only feature that differentiated
the two transcripts was the substitution of aspartic acid with
alanine at the NDN joining region. It is possible, therefore,
-
2245
SGFYLCAWSGGTTFAEQYFG257(20/20)
CD8-TCRBVZO'
tio T c a a v - I I P K "
A6
A7
SGFYLCAWS
DRGD
TEAFFG
1S1
TEAFFG1S1
SGFYLCAWS ALSG
SGFYLCAWS QRTAK NTEAFFG1S1
SGFYLCAWS QTGS SNQPQHFG1S5
SGFYLCAWS ARGGK ETQYFG2S5
SGFYLCAWS DLGVT
QYFG 2S7
TCRBV CHAINS
that the selection of a limited numberof TCRBV chain
molecules had occurred in still differentiating CD8- and
CD4- cells, before the differentiation event that leads to the
acquisition of the LGL phenotype. Relevant to this point is
the observation that CD4-CD8- TCRAB+ cells, which are
capable of recognizing bacterial antigens, are often oligoclor~al.~~."
Alternatively, the expression of this TCRBV segment
can be a common feature of different clones selected after
their acquisition of the CD8+ and CD8- phenotype. However, the finding that the monoclonal TCRBV transcripts
were dominantly expressed by both CD8+CD57+ and
CD8+CD57- lymphocytes further supports the hypothesis
of the differentiative capability of the originally selected
cells.
A more trivial explanation for these results is that the
clonality of the CD8- cells was due to CD8+ contaminants
present in the CD8- preparation. This possibility, however,
is unlikely because CD8+ low density cells were present
only in small numbers in the CD8- cell population. Furthermore, all CD8+ cells contaminating the CD8- preparation
did not express CD3 marker, and, therefore, they probably
do not productively transcribe TCR segments.
Themost surprising finding of this study was that the
dominant and monoclonal TCRBV 17 and TCRBV 13s1 transcripts expressed by patient FC had closely related junctional
sequences. This is highly unlikely to occur by chance alone,
if one considers the very large number of possible combinations of V-D-J segments and N additions. Therefore, this
common pattern of junctional diversity is consistent with a
strong antigenic selection process, possibly due to the particular clinical setting of the patient. Based on the structure of
immunoglobulins, current models of TCR predict the existence of three major CDRs (CDRl, CDR2, CDR3) on both
TCRA and TCRB chains.45 The CDR3 loop is formed by
the joining of the V-D-J segments and interacts predominantly with determinants on the bound peptide of the antigen/
major histocompatibility complex (MHC).45-47
Thus, our data
suggest that, although the dominant TCRBV17 and
TCRBV13S1 clones use different TCRBV segments, they
have both been selected by a common T-cell epitope whose
recognition is strictly dependent on the conformational structure of the CDR3 loop. The presence of a proline residue
in the first position of the NDN regions of the dominant
TCRBV13S1, TCRBV13S2, and TCRBV17 transcripts may
also not be causal.
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
2246
SOlTINI ET AL
Our data are, therefore, not compatible with an oncogenic
chain (To)gene rearrangements demonstrates the monoclonal nature
of T-cell chronic lymphoproliferative disorders. Blood 67:247, 1986
event, but rather suggest the existence of receptor-mediated
6. Pelicci P-G, Allavena P, Subar M, Rambaldi A, Pirelli A, Di
selective pressures. The nature of the antigen responsible for
Bello M, Barbui T, Knowles DM 11, Dalla-Favera R, Mantovani A:
the selection of the observed clonal dominance canonly
T-cell receptor (a,P, y ) gene rearrangements and expression in
be speculated at this time. The association between LGL
normal and leukemic large granular lymphocytednatural killer cells.
proliferation and autoimmunity has long been suspected, alBlood 70:1500, 1987
though retroviruses have been implicated as the original
7. Chan WC, Dah1 C, Waldmanu T, Link S, Mawle A, Nicholson
pathway of antigen
The hallmarks of WAS are
J, Bach FH, Bongiovanni K, MCCue PA, Winton EF: Large granular
abnormal cell-surface cytoarchitecture:’ abnormalities in the
lymphocyte proliferation: An analysis of T-cell receptor gene arpattern of glycosylation of cell-surface protein^:^.^' and derangement and expression andthe effect ofin vitro culture with
inducing agents. Blood 71:52, 1988
fective transmembrane ~ignaling.~’
The cytoarchitectural de8. Pandolfi F, Zambello R, Cafaro A, Semenzato G: Biologic and
fect may, therefore, induce autoimmune phenomena that may
clinical heterogeneity of lymphoproliferative diseases of peripheral
evolve into either malignancies or aberrant clonal expanmature T lymphocytes. Lab Invest 67:274, 1992
sions. However, we also cannot disregard the possibility that
9. Dhodapkar MV, Li C-Y, Lust JA, Tefferi A,PhylikyRL:
the association between the observed clonal dominance and
Clinical spectrum of clonal proliferations of T-large granular lymWAS is only fortuitous. Kasten-Sportb et al,26 for instance,
phocytes: A T-cell clonopathy of undetermined significance? Blood
have recently proposed that LGL expansion may represent
84: 1620, 1994
a T cell-mediated autoimmune disorder that involves recog10. Davey MP, Starkebaum G, Loughran TP Jr: CD3’ leukemic
nition by the TCR of the clonal T cells of an antigen on
large granular lymphocytes utilize diverse T-cell receptor Vp genes.
a protein required for granulopoiesis or erythropoiesis. In
Blood 85:146, 1995
11. Moebius U, Mams M, Hess G , Kober G, Meyer zum
addition, oligoclonal evolution may also depend on a number
Buschenfelde K-H, Meuer SC: T cell receptor gene rearrangements
of exogenous microbial or viral agents characterized by the
of T lymphocytes infiltrating the liver in chronic active hepatitis B
presence of dominant T-cell epitopes. For instance, the huand primary biliary cirrhosis LPBC): oligoclonality of PBC-derived
man TCR repertoire, developed against definedinfluenza
T cell clones. Eur J Immunol 20:889, 1990
nuclear protein epitopes, has been shown to be o l i g ~ c l o n a l . ~ ~
12. Oksenberg JR, Panzara MA, Begovich AB, Mitchell D, Erlich
Similarly, dominant selection of an invariant T-cell antigen
HA, Murray RS, Shimonkevitz R, Shemt M, Rothbard J, Bernard
receptor is observed in patients with chronic hepatitis C,
CCA, Steinman L: Selection for T-cell receptor Vp-Dp-JP gene
among intrahepatic T cells specific for the protein NS4 of
rearrangements with specificity for a myelin basic protein peptide
hepatitis C virus.”
in brain lesion of multiple sclerosis. Nature 362:68, 1993
In conclusion, our results establish that, at least in our
13. Pantaleo G, Demarest JF, Soudeyns H, Graziosi C, Denis F,
Adelsberger J W , Borrow P, Saag MS, Shaw GM, Sekaly RP, Fauci
patient, LGL proliferation was due to an antigenic selection
AS: Major expansion of CD8* T cells with a predominant VP usage
process responsible for the expansion of a limited number
during the primary immune response to HIV. Nature 370:463, 1994
of T-cell clones, characterized by the expression of different
14. Forman JD, Klein JT, Silver RF, Liu MC, Greenlee BM,
TCRBV segments butwith remarkably similar structural
Moller DR: Selective activation and accumulation of oligoclonal
constraints in the CDR3 regions. The observation that the
VP-specific T cells in active pulmonary sarcoidosis. J Clin Invest
TCRBV segments were also highly enriched in other T-cell
94: 1533, 1994
subpopulations raises important questions concerning the
15. Grom AA, Thompson SD, Luyrink L, Passo M, Choi E, Glass
malignant nature of this LGL proliferation.
DN: Dominant T-cell-receptor p chain variable region Vp14+ clones
ACKNOWLEDGMENT
We thank Prof Alberto Albertini for his enthusiasm and for continuous support; Drs Alessandra Tucci, Alberto Zambruni, and Luigi
D. Notarangelo for providing the control patients’ samples; and Dr
Duilio Brugnoni for the immunologic characterization of thepatients
with WAS.
REFERENCES
1. Kronenberg M, Siu G, Hood LE, Shastri N: The molecular
genetics of the T-cell antigen receptor and T-cell antigen recognition.
Annu Rev Immunol4:529, 1986
2. Marrack P, Kappler J: The T cell receptor. Science 238:1073,
1987
3. Waldmann TA, Davis MM, Bongiovanni KF, Korsmeyer SJ:
Rearrangements of genes for the antigen receptor on T cells as
markers of lineage and clonality in human lymphoid neoplasms. N
Engl J Med 313:776, 1985
4. Loughran TP Jr: Clonal diseases of large granular lymphocytes.
Blood 82:1, 1993
5. Foa R, Pelicci P-G, Migone N, Lauria F, Pizzolo G, Flug F,
Knowles DM 11, Dalla-Favera R: Analysis of T-cell receptor beta
on juvenile rheumatoid arthritis. Proc Natl Acad Sci USA 90: 1 1104,
1993
16. Hara H, Morita M, Iwakt T, Hatae T, Itoyama Y, Kitamoto
T, Akizuki S, Goto I, Watanabe T: Detection of human T lymphotrophic virus type I (HTLV-I) proviral DNA and analysis of the T
cell receptor Vp CDR3 sequences in spinal cord lesions of HTLVI-associated myelopathy/tropical spastic paraparesis. J ExpMed
180:831, 1994
17. O’Hanlon T P , Dalakas MC, Plotz PH, Miller W :Predominant TCR-(r/p variable and joining gene expression by muscle-infiltrating lymphocytes in the idiopathic inflammatory myopathies. J
Immunol 152:2569, 1994
18. Sottini A, Quiros-Roldan E, Notarangelo LD, Malagoli A,
Primi D, Imberti L: Engrafted maternal T cells in a SCID patient
express TCRBV segments characterized by a restricted V-D-J junctional diversity. Blood 85:2105, 1995
19. Hingorani R, Choi I-H, Akolkar P, Gulwani-Akolkar B, Pergolizzi R, Silver J, Gregersen PK: Clonal predominance of T cell
receptors within the CD8’ CD45RO’ subset in normal human subjects. J Immunol 1515762, 1993
20. Posnett DN, Sinha R, Kabak S, Russo C: Clonal population
of T cells in normalelderly humans: The T cell equivalent to “benign
monoclonal gammapathy”. J Exp Med L79:609, 1994
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
TCRREPERTOIRE
IN LGL OF WASPATIENT
21. Mortey J, Batliwalla F, Hingorani R, Furie R, Chartash E,
Gregersen PK. Clonal expansion in CD8+ T cells is primarily found
in the CD57 subset. Arthritis Rheum 375375, 1994
22. Semenzato G, Pandolfi F, Chisesi T, De Rossi G, Pizzolo G,
Zambello R, Trentin L, Agostini C, Dini E, Vespignani M, Cafaro
A, Pasqualetti D, Giubellino MC, Migone N. Foa R: The lymphoproliferative disease of granular lymphocytes. A heterogeneous disorder ranging from indolent to aggressive conditions. Cancer
60:2971, 1987
23. Oshimi K Granular lymphocyte proliferative disorders: Report of 12 cases and review of the literature. Leukemia 2:617, 1988
24. Loiseau P, Divine M, Le Paslier D, Marolleau J-P, Farcet JP,
Flandrin G, Cohen D, Degos L, Sigaux F, Reyes F: Phenotypic and
genotypic heterogeneity in large granular lymphocyte expansion.
Leukemia 1:205, 1987
25. Zambello R, Trentin L, Agostini C, Francia di Celle P, Francavilla E, Borelli A, Cerutti A, Siviero F, Foa R, Semenzato G:
Persistent polyclonal lymphocytosis in human immunodeficiency virus-l infected patients. Blood 81:3015, 1993
26. Kasten-Sport&sC, Zaknoen S, Steis RG, Chan WC, Winton
EF, Waldmann TA: T-cell receptor gene rearrangement in T-cell
large granular leukocyte leukemia: Preferential Va but diverse J a
usage in one of five patients. Blood 83:767, 1994
27. Rosen FS, Cooper MD, Wedgwood RJP The primary immunodeficiencies (first of two parts). N Engl J Med 311:235, 1984
28. Rosen FS, Cooper MD, Wedgwood RIP The primary immunodeficiencies (second of two parts). N Engl J Med 311:300, 1984
29. Fischer A: Primary T-cell immunodeficiencies. Curr Opin
Immunol 5569, 1993
30. Ammann A J , Hong R: Disorders of the T-cell system, in:
Stirhm ER (ed): Immunologic Disorders in Infants and Children.
Philadelphia, PA, Saunders, 1989, p 257
3 1. Duhrsen U, Uppenkamp M, Uppenkamp I, Becher R, Engelhard M, Konig E, Meusers P, Meuer S, Brittinger G: Chronic T cell
leukemia with unusual cellular characteristics in ataxia telangectasia.
Blood 68577, 1986
32. Martin R, Howell MD, Jaraquemada D, Flerlage M, Richert
J, Brostoff S, Long EO, McFarlin DE, McFarland HF: A myelin
basic protein peptide is recognized by cytotoxic T cells in the context
of four HLA-DR types associated with multiple sclerosis. J Exp
Med 173:19, 1991
33. Bettinardi A, Imberti L, Sottini A, Primi D: Analysis of amplified T cell receptor Vp transcripts by a non-isotopic immunoassay.
J Immunol Methods 146:71, 1992
34. Toyonaga B, Yoshikai Y, Vadasz V, Chin B, Mak TW: Organization and sequences of the diversity, joining, and constant region
genes ofthe human T-cell receptor p chain. Proc Natl Acad Sci
USA 82:8624, 1985
35. Concannon P, Pickering LA, Kung P, Hood L: Diversity and
structure of human T-cell receptor ,&chain variable region genes.
Proc Natl Acad Sci USA 83:6598, 1986
36. Kimura N, Toyonaga B, Yoshikai Y, Triebel F, Debre P,
Minden MD, Mak TW: Sequences and diversity of human T cell
receptor p chain variable region genes. J Exp Med 164:739, 1986
37. Kimura N, Toyonaga B, Yoshikai Y, DuR-P,Mak
TW:
Sequences and repertoire of the human T cell receptor a and p chain
variable region genes in thymocytes. Eur J Immunol 17:375, 1987
38. Ferradini L, Roman-Roman S, Azocar J, Michalaki H, Triebel
2241
F, Hercend T: Studies on the human T cell receptor a/@ variable
region genes. 11. Identification of four additional Vp subfamilies.
Eur J Immunol 21:935, 1991
39. Rosenberg WMC, Moss PAH, Bell JI: Variation in human T
cell receptor V s and Jp repertoire: Analysis using anchor polymerase chain reaction. Eur J Immunol 22:541, 1992
40. Pandolfi F, Loughran TP Jr. Starkebaum G, Chisesi T, Barbui
T, Chan WC, Brouet JC, De Rossi G, McKenna RW, Salsano F,
Herrmann F, vanoostveen J W , Schlimok G, Cafaro A, Zambello R,
Garcia Rodriguez MC, Geisler CH, Pizzolo G, Steis RG, Brisbane
JU, Kadin ME, Mantovani A, Tagawa S, Fauci S, Gastl G, Palutke
M, Proctor SJ, Pross HF, Mancini P, Aiuti F, Semenzato G: Clinical
course and prognosis of the lymphoproliferative disease of granular
lymphocytes. A multicenter study. Cancer 65:341, 1990
41. Berliner N, Duby AD, Linch DC, Murre C, Quertermous T,
Knott LJ, Azin T, Newland AC, Galvin MC, Seidman JG: T cell
receptor gene rearrangements define a monoclonal T cell proliferation in patients with T cell lymphocytosis and cytopenia. Blood
67:914, 1986
42. Preston SJ, Buchanan W:Rheumatic manifestations of immune deficiency. Clin Exp Rheumatol 7547, 1989
43. Dellabona P, Casorati G, Friedli B, Angman L, Sallusto F,
Tunnacliffe A, Roosneek E, Lanzavecchia A: In vivo persistence of
expanded clones specific for bacterial antigens within the human T
cell receptor alp CD4-8- subset. J Exp Med 177:1763, 1993
44. Porcelli S, Yockey CE, Brenner MB, Balk SP: Analysis of
T cell antigen receptor (TCR) expression by human peripheral blood
CD4-CD8- a@ T cells demonstrates preferential use of several V@
genes and an invariant TCR a chain. J Exp Med 178:1, 1993
45. Chothia C, Boswell DR, Lesk AM: The outline structure of
the T-cell ap receptor. EMBO J 7:3745, 1988
46. Matis LA: The molecular basis of T-cell specificity. Annu
Rev Immunol 8:65, 1990
47. Jorgensen J,Reay PA, Ehrich EW, Davis MM: Molecular
components of T-cell recognition. Annu Rev Immunol 10:835, 1992
48. Le Deist F, de Saint Basile G, Coulombel L, Breton-Gorius
J, Maier-Redelsperger M, Beljorde K, Bremard C, Griscelli C: A
familial occurrence of natural killer cell-T-lymphocyte proliferation
disease in two children. Cancer 67:2610, 1991
49. Higgins EA, Siminovitch KA, Zhuang DL, Brockhausen I,
Dennis J W Aberrant 0-linked oligosaccharide biosynthesis in lymphocytes and platelets from patients with the Wiskott-Aldrich syndrome. J Biol Chem 266:6280, 1991
50. Piller F, Le Deist F, Weinberg KI, Parkman R, Fukuda M:
Altered 0-glycan synthesis in lymphocytes from patients with
Wiskott-Aldrich syndrome. J Exp Med 173:1501, 1991
51. Simon HU, Mills GB, Hashimoto S, Siminovitch KA: Evidence for defective transmembrane signaling in B cells from patients
with Wiskott-Aldrich syndrome. J Clin Invest 90:1396, 1992
52. Lehner PJ, Wang ECY, Moss PAH, Williams S, Platt K,
Friedman SM, Bell JI, Borysiewicz L K Human HLA-A0201-restricted cytotoxic T lymphocyte recognition of influenza A is dominated by T cells bearing the Vp17 gene segment. J Exp Med 181:79,
1995
53. Minutello MT, Pileri P, Unutmaz D, Censini S, Kuo G,
Houghton M, Brunetto MR, Abrignani S: Compartmentalization of
T lymphocytes to the site of the disease: Intrahepatic CD4' T cells
specific for the protein NS4 of hepatitis C virus in patients with
chronic hepatitis C. J Exp Med 178:17, 1993
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
1995 86: 2240-2247
Evidence for antigenic selection of large granular lymphocytes in a
patient with Wiskott-Aldrich syndrome
A Sottini, A Bettinardi, E Quiros-Roldan, A Plebani, P Airo, D Primi and L Imberti
Updated information and services can be found at:
http://www.bloodjournal.org/content/86/6/2240.full.html
Articles on similar topics can be found in the following Blood collections
Information about reproducing this article in parts or in its entirety may be found online at:
http://www.bloodjournal.org/site/misc/rights.xhtml#repub_requests
Information about ordering reprints may be found online at:
http://www.bloodjournal.org/site/misc/rights.xhtml#reprints
Information about subscriptions and ASH membership may be found online at:
http://www.bloodjournal.org/site/subscriptions/index.xhtml
Blood (print ISSN 0006-4971, online ISSN 1528-0020), is published weekly by the American
Society of Hematology, 2021 L St, NW, Suite 900, Washington DC 20036.
Copyright 2011 by The American Society of Hematology; all rights reserved.