Download The Murine Interleukin-3 Receptor a Subunit Gene

From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
The Murine Interleukin-3 Receptor a Subunit Gene: Chromosomal
Localization, Genomic Structure, and Promoter Function
By lkuko Miyajima, Lee Levitt, Takahiko Hara, Mary A. Bedell, Neal G. Copeland, Nancy A. Jenkins,
and Atsushi Miyajima
The interleukin-3 receptor (IL-3R) is composed of (Y and p
subunits, members of the class I cytokine receptor family.
Here we describe isolation andcharacterization of the chromosomal gene for the mouse IL-3R (Y subunit (mlL-3Ra).
Whereas the human IL-3Ra gene is tightly linked with the
granulocyte-macrophage colony-stimulating factor receptor
(Y subunit (GM-CSFRa) gene in the pseudoautosomal region
of the X and Y chromosomes, the mlL-3Ra gene (113ra) is
located in the proximal region of mouse chromosome 14,
separated from the mouse GM-CSFRa gene, which is on
chromosome 19. The mlL-3Ra gene spans about 10 kb and
is divided into
12 exons. All the exon-intron
boundaries possess the splicing junction consensus sequences (5’GTAG3 ’ 1, and the wholegenomic structure is similar t o those
of the previously reported class I cytokine receptor genes.
There are t w o major transcription initiation sites that are
located at 215 and 188 nucleotides upstream ofthe initiator
codon. The promoter region is GC-rich and contains potential binding sites for GATA, Ets, c-myb,, Spl, Ap-2, and G-C
boxes, but not a typical TATA or CAAT sequence. A fusion
gene containing 0.8 kb of the 5 ’ noncoding sequence linked
to the firefly luciferase gene directed the transcription in
mouse mast cells but not in fibroblasts or T cells, suggesting
that this promoterfunctions in a cell type-specific manner.
Further sequential deletion of the 5’ region suggests t w o
potential regulatory regionsfor transcriptionof the mlL-3Ra
gene.
0 7995 b y The American Society of Hematology.
H
shared by the three receptors.* The high affinity receptors
for IL-3, IL-5, and GM-CSF are composed of two subunits,
a and p, both of which are members of the class I cytokine
receptor family. The a subunits (IL-3Ra, IL-SRa, and GMCSFa) are cytokine-specific and bindtheir specific ligand
p
withlowaffinity
by themselves, whereasthecommon
subunit (pc,originally termed KH97) isrequired for high
affinity binding as well as signaling by these three receptors
in the human. Thus, pc is responsible for common functions
of these cytokines, and expression of the a subunits is responsible for specificity to cytokines.’ In contrast to the human receptors, there are two distinct
p subunits in the mouse.
Likethe humanreceptors, themousehas three cytokinep subunit (pc,originally
specific a subunits and the common
termedAIC2B). In addition, mouseIL-3(mIL-3) specific
p subunit
originally
termed
AIC2A),
which
is
91%
identical to p, at theaminoacid level, is presentin the
binds mIL-3 with low affinity andinteracts
mouse.’..’ pIL3
mGMwithonly themIL-3Ra5 but not witheitherthe
CSFRa6 or the mIL-5Ra7 subunit. However, no clear functional difference hasbeen found betweenthe two distinct
high affinity mIL-3Rs formedwith either pcor plLl.’ The two
p subunit genes are closely linked on mouse chromosome
15, and their genomic organization as well as surrounding
sequences arewell conserved,’ suggesting that the twogenes
arose by duplication.
As receptor expression is a prerequisite for cytokine response and IL-3 interacts with an early multipotential hematopoietic progenitor, the IL-3R expression may be an excelfor hematopoietic development. In fact,the
lentmarker
IL-3R expression seems to be restricted mainly to myeloid
lineages. By using antibodies against receptor subunits,
we
have previously shown that IL-3a and pc are widely distributed in myeloid cell lineages, as well as early progenitors
in human bone marrow and cordblood cells. In the lymphoid
compartment, IL-3R expression is found in a minor fraction
of cells with B-cell marker CD19, but not in cells with Tcell marker CD3.’ To study the developmental and cell-type
specific expression of the IL-3R, we have characterized the
chromosomal gene (113ra)for the mIL-3Ra subunit and analyzed the transcriptional activity of its promoter.
EMATOPOIESIS isan excellent model system tostudy
the mechanism of cell proliferation and differentiation, because a pluripotent stem cell differentiates through a
series of committed progenitors towardterminally differentiated cells of each blood cell type. Each process is regulated
by cytokines that are produced
by a wide variety of cell
types, either constitutively or inducibly in response to hematologic or immunologic stimulation, and exhibit pleiotropic
functions.’ Among those cytokines, interleukin-3 (IL-3), IL5, andgranulocyte/macrophage colony-stimulating factor
(GM-CSF) are major hematopoietic cytokines that are secreted by activated T cells and mast cells and play an important role in hematopoiesis associated with the inflammatory reactions. Whereasthesethreecytokinesshow
no
significant aminoacidsequencehomology,
the genesare
tightlylinked onthesamechromosome,
theexon-intron
structures are similar, and they exhibit similar functions in
common target cells.’
Common biologic function exhibited by these three cytop subunitthatis
kinesisnow
explained by the common
From the DNAX Research Institute of Molecular and Cellular
Biology, Palo Alto, CA; and the Mammalian Genetics Laboratory,
ABL-Basic Research Program, National Cancer Institute-Frederick
Cancer Research and Development Center, Frederick, MD.
Submitted January 18, 1994; accepted October 28, 1994.
Supported in part by the National Cancer Institute, Department
of Health and Human Services, under Contract No. Nol-CO-74101
with ABL. M.A.B. was supported by a fellowship fromthe Foundation for Advanced Cancer Studies, Inc. DNAX Research Institute i s
supported by Schering-Plough.
L.L. is presently at the Department of Hematology, Stanford University School of Medicine, Stanford, CA.
Address reprint requests to Atsushi Miyajima, PhD. DNAX Research Institute of Molecular and Cellular Biology, 901 California
Ave, Palo Alto, CA 94304.
The publicationcosts 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.
0006-4971/95/8505-0026$3.00/0
1246
Blood, Vol 85, No 5 (March l),
1995:
pp 1246-1253
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
1247
MURINE IL-3 RECEPTOR a SUBUNIT GENE
MATERIALSANDMETHODS
Cell culture. A mouse mast cell line, PT18" was cultured in
RPM1 1640 medium supplemented with 10% heat-inactivated fetal
calf serum (FCS), 50 pmoUL 2-mercaptoethanol (2-ME), and 50 U/
mL mGM-CSF. A mouse fibroblastic cell line, NIH3T3, was cultured
in Dulbecco's modified Eagle's medium (DMEM) with 5% FCS.
An IL-2-dependent T cell line, CTLL2," was maintained in RPM1
1640 medium containing 10% FCS, 50 pmol/L 2-ME, and 1 0 0 U/
mL mIL-2.
Isolation and subcloning of genomic clones. A Balb/c sperm
genomic library in Charon 4A (provided by Dr Mark Davis, Stanford
University, Stanford, CA) and an adult Balbk liver genomic library
in A SP6n7 (Clonetech, Palo Alto, CA) were screened with the 1.3kb fragment encoding the entire IL-3Ra. Furthermore, a C129 mouse
genomic library in AEMBL was screened with a 0.6-kb NCOI-Sma
I fragment of cDNA. The probes were labeled with [a-'*P]dCTP by
the T7 Quick Prime kit (Pharmacia, Alameda, CA) to a specific
activity of greater than 4 X 10' c p d p g . Plaques were transferred
to a nitrocellulose filter and hybridized for 18 hours at 42°C in
hybridization solution (6X SSPE, 50% formamide, 5X Denhardt
solution, 100 pg/mL yeast tRNA, 0.2% sodium dodecyl sulfate
[SDS]). Positive plaques were purified, and phage DNA was prepared by a standard plate lysate procedure. The mIL-3Ra gene was
subcloned into pUCl8 for further characterization and sequencing.
Characterization of cloned DNA fragment and DNA sequencing.
The subcloned DNA fragments were mapped by restriction enzyme
analyses, andthe exons were localized by Southern hybridization
using mIL-3Ra cDNA as a probe. Supercoiled plasmid DNA was
alkaline-denatured and sequenced by the dideoxy chain termination
method using Sequenase (United States Biochemical Corp, Cleveland, OH) for all the intron-exon junctions and short introns. The
sizes of introns were determined by polymerase chain reactions
(GeneAmp PCR reagent kit: Perkin Elmer Cetus, Norwalk, CT)
using several oligonucleotide primers corresponding to mIL-3a
cDNA. The reaction mixture including 1 ng template DNA and 100
pmol/L each primer were incubated in a thermal cycler (Perkin Elmer
Cetus) for 30 cycles (denatured for 1 minute at 9 4 T , annealed for
2 minutes at 5 0 T , elongated for 3 minutes at 72°C). Products were
analyzed by electrophoresis on a 1.6% agarose gel in the presence
of ethidmm bromide.
Southern hybridization analysis of mouse genomic DNA. Genomic DNA (10 p g ) from liver of adult Balbk (Clonetech) and C129
SV mice were digested to completion by the restriction endonucleases (EcoRI, BarnHI, and HindIII). After electrophoresis on a 1%
agarose gel, the DNA was treated with 0.25 mom HC1, alkalinedenatured (0.5 m o m NaOH, 1.5 mol/L NaCI), neutralized (0.5 mol/
L Tris HC1, pH 7.5; 1.5 m o m NaCl), transferred to a Hybond-N
nylon membrane (Amersham, Arlington Heights, IL) using 20x
SSPE, crosslinked by UV transilluminator (Stratacross-linker; Stratagene, La Jolla, CA), and then hybridized with the 32P-labeled1.3kb mIL-3Ra cDNA.
5'-rapid amplification cDNA ends (5'-RACE) assay. PolyA+
RNA was prepared from PT18 cells by the Fast Track mRNA Isolation kit (Invitrogen Corp, San Diego, CA). The d L - 3 R a cDNA
was reverse transcribed from 10 pg of polyA+ RNA together with
an 18-mer oligonucleotide primer complementary to a sequence of
the third exon (ST104; 5 'GCTCAGTGTGTAGTGGGCGGG3').
The cDNA purified by agarose gel electrophoresis was then ligated
with AmpliFINDER anchor according to the manufacturer's protocol
(Clonetech). The cDNA including 5 '-end was amplified by polymerase chain reaction (PCR) as follows. One tenth of the ligation product
was amplified using a set of a 38-mer anchor primer and an antisense
18-mer primer (ST-125; S'GAGTGGGATGCCAGGAGCCCC3')
within the second exon, under the following condition: 35 cycles of
94°C for 1 minute, 57°C for 1 minute, 74°C for 3 minutes, and eight
more cycles with additional enzyme. The first amplified product was
then diluted one tenth and further amplified using another nested
antisense primer (ST-143; 5 'GCCTCGGAGGCCTGGGGGCGG3')
under the same condition except for annealing temperature at 55°C.
The final amplified product was treated with T4 DNA polymerase
to create blunt ends. After heat inactivation of polymerase, the DNA
was cleaved with EcoRI, as the anchor sequence includes an EcoRI
site. The resultant DNA fragments were purified by agarose gel
electrophoresis and spin column (Microspin S-200: Pharmacia) and
then subcloned into the EcoRI-Sma I sites of pUCl8. The cloned
DNAs were sequenced to identify the site(s) of 5 '-cDNA ends.
Interspec$c backcross mapping. Interspecific backcross progeny were generated by mating (C57BL/6J X Mus spretus)Fl females
and C57BL/6J males as described.'* A total of 205 Nzmice were
used to map the 113ra locus (see Results for details). DNA isolation,
restriction enzyme digestion, agarose gel electrophoresis, Southern
blot transfer, and hybridization were performed as de~cribed.'~
All
blots were prepared with Zetabind nylon membrane (AMF-Cuno,
Meriden, CT). The probe, a 1.3-kb Xho I fragment of mIL-3Ra
cDNA (SUT-l), was labeled with [a"P]dCTP using a random
primed labeling kit (Boehringer Mannheim, Indianapolis, IN); washing was performed to a final stringency of 1 .OX SSCP, 0.1% SDS,
65°C. A major fragment of 7.6 kb was detected in Bgl 11-digested
C57BW6J DNA, and a major fragment of 11 .0 kb was detected in
Bgl 11-digested M sprerus DNA. The presence or absence of the
11.0-kb M spretus-specific Bgl I1 fragment was followed in backcross mice.
A description of the probes and restriction fragment length polymorphisms (RFLPs) for the loci linked to 113ra, including retinoic
acid receptor beta (Rarb, formally referred to as Hap), plasminogen
activator urokinase (Plau), and SP-A pulmonary surfactant protein
( S f t p l ) , has been reported previo~sly.'~.'~
Recombination distances
were calculated as described16 using the computer program Spretus
Madness. Gene order was determined by minimizing the number
of recombination events required to explain the allele distribution
patterns.
Transient expression assays. For the analysis of promoter activity, the fragment containing the 5' noncoding region of the mIL3Ra gene was fused to the firefly luciferase gene. The luciferaseexpression vector pGL2-Basic Vector (Promega Corp, Madison, WI)
was used. Cells (5 X 10' PT18 cells, 5 X IO6 CTLL-2 cells, lo6
NIH3T3 cells) were transfected with supercoiled plasmid DNA (40
pg for FT18,40 pg for CTLL-2,30 p g for N1H3T3) by electroporation. At 18 hours after transfection, the luciferase activity was assayed using the luciferase assay kit (Promega Corp, Madison, WI).
The SRa promoter (hybrid promoter containing SV40 and HTLVLTR)-luciferase fusion gene was used as a positive control.
RESULTS
Chromosome mapping. The mouse chromosomal location of the IL-3Ra gene (I13ra) was determined by interspecific backcross analysis using progeny derived from matings
of [(C57BL/6J X M spretus)Fl X C57BL/6J] mice. This
interspecific backcross meiotic mapping panel has been
typed for over 1,500 loci that are well distributed among all
the autosomes as well as the X chromosome.'* This indicates
that, on average, markers are spaced about 1 centiMorgan
(CM) apart over the genome. To place a new marker, an
RFLP is established by Southern blot analysis of the two
parental strains in the cross, and the backcross mapping panel
is then typed for the informative RFLP. The segregation
patterns of the new marker are then compared with known
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
1248
MlYAJlMA ET AL
segregation patterns of all other markers. Gene order is determinedby minimizing the number of double or multiple
crossovers required to explain the new probe distribution.
C57BU6J and M spretus DNAs were digested with several enzymes and analyzed by Southern blot hybridization
for informative RFLPs using a mouse IL-3Ra cDNA probe.
An 11.O-kb M spretus Bgl I1 RFLP (see Materials and Methods) was used to follow the segregation of the I13ra locus
in backcross mice. The mapping results indicated that Il3ra
is located in the proximal region of mouse chromosome 14
linked to Rarb, Plau, and Sftpl.Although 161 mice were
analyzed for every marker and are shown in the segregation
analysis (Fig l), up to 185 mice were typed for some pairs
0 0
11 7
0 2
of markers. Each locus was analyzed in pairwise combinations for recombination frequencies using the additional data.
The ratios of the total number of mice exhibiting recombinant chromosomes to the total number of mice analyzed for
each pair of loci and the most likely gene order are as
follows: centromere-Rarb-0/174-I13ra-2/185-Plau-19/18314
S'pl.
The recombination frequencies (expressed as genetic
distances in centiMorgans [CM] 2 the standard error) are
(Rarb, IZ3ra)-l.l 2 0.8-Pluu-10.4
+- 2.3-Sftpl.No
recombinants were detected between Rurb and I13ra in l74
animals typed in common, suggesting that the two loci are
within 1.7 CM of each other (upper 95% confidence limit).
Rarb
3p24
We have compared our interspecific map of chromosome
I13ra
lOq24qter
Plau
14 with a composite mouse linkage map that reports the map
location of many unclonedmouse mutations.'haI13ru mapped
in a region of the composite map that lacks mouse mutations
with a phenotype that might be expected for an alteration in
10.4
l - 1
this locus (data not shown).
Isolation of chromosomal gene for the mIL-3Ra gene.
Of 5 X IO5 independent phage plaques of the Balb/c sperm
library screened with the 1.3-kb cDNA (SUT-l) fragment
1Oq21-q24
of mIL-3Ra as a probe, one positive clone (Xsut-l) was
obtained. Using the same probe, three positive clones (Xsut3, -4, and -6) were obtained from 2 X 10' plaques of the
Balb/c liver library. These four clones were plaque-purified,
and the X DNA were analyzed by restriction enzyme mapping and Southern hybridization. Three clones (Xsut- I , -3,
and -4) were overlapped, covering the 3 ' region of the gene,
Fig 1. 113ramaps in the proximal region of mouse chromosome 14.
113rawas placed on mouse chromosome 14 by interspecific backcross
while one clone (Xsut-6) contained the 5 ' portion (Fig 2).
analysis. The segregationpatterns of N3ra and flanking genes in 161
These four clones covered most of the cDNA sequence of
backcross animalsthat were typed for all loci are shown at the top
mIL-3Ra, except for the short fragment located in the middle
of the figure. Forindividual pairs of loci, more
than 161 animals were
of cDNA including the exon V. To isolate the DNA for this
the chromosome identityped (see Results). Each column represents
missing portion, 10' plaques of the C129 mouse genomic
fied in the backcross progeny that was inherited from the (C57BLI
6J x M spretus)Fl parent. Thesolid boxes representthe presence of a
library were screened with the 0.6-kb Sma I-NCOl fragment
C57BL/6J allele, and open boxes represent
the presence ofMspretus
of mIL-3Ra. Only one positive clone (Xsut-7) was isolated,
allele. The number ofoffspring inheriting each type of chromosome
and it contained the sequence that was missing in the other
is listed at the bottom of each column. A partial chromosome 14
clones (Fig 2 ) .
linkage mapshowing the location of 113ra in relation to linked genes
is shown at the bottom of the figure. Recombination distances beTo confirm that the cloned DNA fragments represented
tween loci in CMare shown to the left of the chromosome, and the
the mIL-3Ra gene locus, mouse genomic DNA from the
positions of loci in human chromosomes, where known, areshown
liver of Balb/c and C129 mice was digested with EcoRI,
to the right. References for the human mappositions of loci mapped
BamHI, or HindIII and analyzed by Southern blotting using
in this study canbe obtained from Genome DataBase (QDB), a computerizeddatabase of human
linkage information maintainedby The
the mIL-3Ra cDNA as aprobe. The EcoRI or HindIII cleavWilliam H. Welch Medical Library of The Johns Hopkins University
age generated a single fragment of approximately 18 kb and
(Baltimore, MDLU It should be noted that, while the region of the
approximately 13 kb, respectively, whileBamHI cleavage
mouse chromosome 14 where113ra is placed is homologous to huproduced two fragments of 7.5 kb and4 kb (data not shown).
man chromosome 3p24, the human 113ra resides in the pseudoautosomal region of X and Y chromosomes, as described in Results.
The 10-kb cloned chromosomal gene of mIL-3Ra, which
om o m
m 0 U.
0.
Om
B O BO O B
B O BO BO
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
MURINE IL-3 RECEPTOR
l 2
3
4 5
(Y
1249
SUBUNIT GENE
6 70910
U
1112
nn
\
B
P S I
B
P P
AsUnr
P
R
HB
1 kb
U
"
"
Fig 2. Genomic structure of the mouse IL-3Ra gene. Exons are
shown byboxes. Solid boxes indicate the noncodingregion, and
open boxes are coding exons. The restriction sites for & d l , Pot 1,
Stu 1, Hindlll, and € d l are shown as B, P, St, H, and R, respectively.
The clones used for the analysis are shown by bars.There is a single
BamHl site present in exon 4, but no Hindlll or € d lsite within about
10 kb of mlL3Ra gene from exons 1 to 12. Clones h u t 3 and h u t 6
are derived from the Balb/c, and Asut7 is derived from C129. The
dashed lines represent A arms.
spans 12 exons, has no recognition site for EcoRI or HindIII
and has a single site for BamHI (Fig 2), whereas an EcoRI
site, a HindIII site, and two BamHI sites are located nearby
in the surrounding regions, which indicates that one copy of
the mIL-3Ra gene is present per genome and is covered by
the cloned DNA fragments.
Structure of the mIL-3Ra gene. The overall structure of
the mIL-3Ra gene was determined by Southern hybridization, PCRs, and DNA sequencing: PCR and restriction enzyme analyses were performed to determine intron length
using the cloned A DNAs and the subcloned plasmid DNAs.
All exon-intron borders and several smaller introns were
sequenced entirely. As shown in the schematic diagram of
the gene and restriction map (Fig 2), the estimated length of
the gene from the transcription start site to the polyadenylation site is approximately 10 kb. The gene contains 11 introns, and all the sequences of intron-exon junctions conform
with the consensus sequences of eukaryotic splice junctions
(Table 1).
Exon I encodes 5 "untransiated sequence, and exon I1
encodes more 5 "untranslated sequence and the signal peptide. The extracellular domain is encoded by seven exons
(exons I11 to IX), the transmembrane domain by a single
exon (exon X), and the cytoplasmic domain by two exons
(exons XI and XII). Exon XI1 also contains the 3 "untranslated sequences.
Identijcation of the transcriptional srart site(s). To identify the functional promoter region that regulates mIL-3Ra
gene expression, we attempted to map the 5'-end of the
transcript by S 1 nuclease protection assay and primer extension. However, the results were rather ambigous, probably
due to the presence of the G/C rich sequences in the 5'
region. Therefore, we mapped the transcriptional start site(s)
using the 5 ' RACE assay. First, the 5 '-ends of the mRNA
were reverse transcribed from polyA+ RNA of PT18 cells.
The cDNA was amplified and subcloned into pUC18. Plasmid DNA from 13 individual clones was isoIated, and the
cloned DNA fragments were sequenced. Six clones had the
same 5 '-end at 215 nucleotides, and seven clones had the
5 '-end at 188 nucleotides upstream from the initiation codon
(Fig 3), suggesting that the transcription starts at these two
positions.
Sequences of the S'janking regions. Although there are
no conventional TATAA and CCAAT sequences in the 5 'flanking region of the mIL-3Ra gene (Fig 3), the potential
promoter region carries several noteworthy features. First,
the overall G + C content (65%) is unusually high. There
are several GC boxes,I7 Spl binding sites (G/TG/AGGCG/
TG/AG/AG/T)," AP-2 binding sites (CCCA/CNG/CG/CG/
C),'8 and a G-stretch of a consensus sequence (CS) for erythrocyte-specific binding protein BGP1.I9 Second, an A T
stretch of 21-mer is present and includes a CS (C/TA/GGTTTCA/GC/"TC/TC/TN) for interferon CS binding protein
ICSBP.20Third, two known recognition sequences are present tandemly: GATA (A/TGATAA/G)2' andEts (G/CA/
CGGAA/TGT/C)," two Myb sequences (T/CAACG/TG)'*
that compose a short palindrome, CaGTTGCAACgG. Moreover, there is a 21-bp potential palindromic sequence (GaGGCCCCGgcgCGGGGCCaC) just upstream of the GATAEts sequences. An NFKB site (GGA/GA/GAT/GT/CCCCC/
T)" is also found at about -350 bp.
The promoter region contains several repetitive sequences,
including some CS for known transcription factors: three
GC boxes, three y-interferon responsive elements (CA/TGG/
+
Table 1. Exonllntron Boundaries
No.
Exon Size (bp)
Donor
1
2
3
4
5
6
7
8
9
10
11
12
135
178
156
123
130
233
119
30
115
103
67
ACTCAG
GAT CTG G
GTC TGG
GAT GAT G
GAT GTG CG
CGC GCA G
AAC CAG
TAC AAT
GGC CTC G
TGG TGG AG
GAG ATG
Intron Size Ikb)
gtccggtccg-0.5-accatgacag
gtaggtgtct-0.6-ttgacaacag
gtgaggggca-0.8-gaccccgcag
gtgggtgagg-0.25-aggcccacag
gtgggtttta-0.7-ctcgcaccag
gtgagggcct-0.4-ccgcccccag
CTG
gtgcggggtg-0.12-ctgcccccag
gtgagttggg-0.09-gcccaccccag
gtgagtgggt-0.8-cttcccgcag
gtgagggcct-3.0-gtccccacag
gtgagggggc-0.27-ctctctgcag
Acceptor
GTC
GAACCA
CT GCG GTC
GCT GAC
GT GGC GAC
G CTC GGT
AG
AGC TCC
GTC TCC
TC TGC CCC
G AAG TCG
GTC GTG
All exon/intron boundaries were sequenced. Spaces in the splicing donor and acceptor sequences indicate reading frame of the protein.
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
1250
MlYAJlMA ET AL
C.T...G(;AEiCACAGG ACAAATTCTG
AAAAACGGAA ACGGAAATAA GGCCGCCTGT GAGGCAT$W$$GC$TATCC
,TGGGCGGGGT,GGGGCCGGCC TCCTAAGCACJ,
GC-BOX
CIFTGTGACG
TCCGCTGTGA
PVUII
GCTCTGCTC
CAGAGTGGGG GGGGGGGGGG GGGATGAAGA CAGGGCAGGG
-A0
BOP 1
TCCGCTCACA AGTTAAGTTC CGTGGGTCTC AATGGGATGC CTTCAGGCAT
*..***.*
GGATGGCAGG TAGTGAACAA CCGCTACAAA
e.*.*
GCAGGGGTAT GATGGC&G.Cx. ....FTCAGTGACT GCGGGGTAAA GAGCGTGAGG
-
C G A G T ~ A G TT&AACG&T
GACAGGAGAG CGACTTCCTG TCTGCGGCAG
GACGTCACGF CCGCTCGCCC GTGACTGACGGGAACGACTT
CCAGTTTCAT
ICSBP
-250
- 4
TCCCTCCCGC GTGACTCCCG A G G X C ~ ~ ~ C " ~ C G G G G C $ C TATC%AGGA
-50
BGCTCAA
TCTCGGGACC ATGGGCGCGG T ~ G G G C G G GGF GGGGCTCT
' GTTCCGACCC
'
GGGGTCACG GCAGGGTCFCAGECCTCTGCCCTGTAAGCA
+l
E
GC-Box
t10
CCAGACCCFC GAGGACAGGA
E%
+50
*
AE~ACGCCG
.&f"
transfected cell lysates was measured 18 hours after transfection. As shown in Fig 4, luciferase activity was clearly detected
in
PT18 but
not
in
CTLL-2 or NIH3T3 cells
transfected with the Pst-luc plasmid. The plasmid with the
Pst I-Stu I fragment inan opposite direction (shownas a
Pst-luc-) generated a significantlylow level of luciferase
activity, suggesting an orientation dependency of thepromoter. The results also suggest that the 0.8-kb fragment is
able to direct transcription in a cell type-specific manner.
To locate the promoter function within the 0.8-kb Pst IStu I fragment, the 0.4-kb Pvu 11-Stu I and the 0. 17-kb Eag
I-Stu I fragments were also fused to the reporter gene, and
the resultant plasmids, Pv-luc and E-luc (Fig 4), were used
to evaluate the promoter function. Pv-luc directed production
of luciferase in PT18 cells, but not in CTLL or NIH3T3
cells. The luciferase expressed in PT18 by Pv-luc was about
half of that produced by Pst-luc. In contrast, luciferase produced by E-luc in PT18 was about the same level as that
produced by Pst-Iuc- (Fig 4). These results suggest that multiple DNA elements may be required for the transcription of
the IL-3Ra gene in hematopoietic cells.
CCAGCGCCGT GGGCGCCACT
,.."""2
....At+++
++++++++++
Fig 3. Nucleotidesequence ofthe 5"flanking region of the mouse
IL-3Ra gene. The transcriptional start sites determined by 5' RACE
are indicated by arrowheads. The numbering (bp) in sequence is
based onthe longer 5'-cDNA end as +l.
A sequence of the previously
isolated cDNAis marked with pluses. The potential binding sites
for various transcription factors are shown. An arrow indicates the
apparent palindromicstructure. NF-IL6-like sequences (see Results)
are marked by asterisks. The Pst 1, Pvu II, Eag I, and Stu I sites used
for the luciferase constructs in Fig4 are also indicated.
TAAGT/C),22five AP-2 binding sites, and eight GCF CS
(C/GCGC/GCIGC/GC)." An imperfect direct repeat ofan
1 l-bp motif, GCTGTGACGTCCGCTGTGAGCTCexists at
around -570, and a sequence of 5 'GCGCCGTGGGCG3 '
is present at around +80 and -610. In addition, the sequence
(T/G)(G/A)(A/T)GGCAG, which is similar to the NF-IL6
CS (TT/GNNGNAAT/G)," is repeated six times between
-320 and -470 (Fig 3).
Cell type-specijic function of the promoter. IL-3Ra
mRNA is expressed in various hematopoietic cell lines, including mast cells, myeloid progenitors, pro-B cells, and
early progenitors, and in nonhematopoietic cells such as ES
Cells5.23 and endothelial
but
not
in
mature T cells,
mature B cells, and fibroblasts.' To study whether the 5'flanking region of the mIL-3Ra gene has transcriptional activity, the 0.8-kb Pst I-Stu I DNA fragment was fused to the
reporter gene, the firefly luciferase gene. Pst I and Stu I are
located in the 5 '-flanking region (Figs 2, 3, and 4). This
fragment contains one of the transcription start sites mapped
by the S'RACE assay. The resultant plasmid, Pst-luc, was
transfected into the mouse mast cell line PT18, the T-cell
line CTLL-2, and the fibroblastic cell line NIH3T3. As a
control, pME-luc, which has the SRa promoter (ie, a hybrid
promoter between SV40 and HTLV-LTR25), was also
transfected into these cell lines. Luciferase activity in the
DISCUSSION
The genes for IL-3, IL-5, and GM-CSF are tightly linked
on human chromosome 5 and mouse chromosome 1 l.' In
contrast, while the human IL-3Ra (fL3RA)and GM-CSFRa
(CSF2RA) genes are also tightly linked on the pseudoautosothe ILmal region of the human X and Y chromosomes,2h-28
5Ra gene (fL5R)is on 3p24-3~26.~'
All the mouse a subunit
genes are located on different chromosomes: mIL-3Ra
(IZ3ru), mGM-CSFRa (Csfzra),'" and mIL-5Ra (fL5r)'' are
on mouse chromosomes 14, 19, and 6, respectively. Interestingly, however, the proximal region of mouse chromosome
14, where 113ru and Rarb genes reside, shares regions of
homology with human chromosome 31324, and humanRARB
has been placed on 3p24. Although the human IL3RA gene
is located on psuedoautosomal regions of X and Y chromosome~,~'.~*
the human fL5-R gene resides on 3 ~ 2 4 . ~While
'
this is an interesting finding, it is unknown whether this is
due to a result of recombination between these loci or is a
simple coincidence.
The mIL-3Ra gene spans about 10 kb and is divided into
12 exons. As shown in Fig 5, the overall structure of the gene
is similar to that of previously reported cytokine receptors,
including human and mouse erythropoietin receptors (h/m
E ~ o R ) , ' human
~ . ~ ~ IL-2 receptor P subunit (hIL-2RP)," mIL4R," h/m IL-7R," human G-CSF receptor (hG-CSFR)," and
two mouse P subunits, AIC2A and AIC2B.s In general, the
gene encoding the class I cytokine receptor motif is divided
into four exons, except for hIL-2RP and mIL-3Ra (Fig S ) .
In IL-2RP, one of the two exons encoding the conserved
cysteine residues is split into two exons." In contrast, there
is a small exon of 30 nucleotides in front of the exon for
the WSXWS motif in the mIL-3Ra gene. It would be interesting to see whether a unique small exon is also present in
the GM-CSFRa and IL-5Ra genes.
When linked to the promoterless firefly luciferase gene,
the 0.8-kb Pst I-Stu I fragment, including a transcription
initiation site and a 14-bp 5"untranslated region, directed
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
1251
MURINE IL-3 RECEPTOR n SUBUNIT GENE
Xhol
I
Pstl
Pvull
I
I
Eagl Stul
I
-1640
murine IL-3Ra
I
0
+l4
start
-
Luciferase
-776
-394
I
Pst-luc
P
Pv-luc
-160
-
1
E-IUC
Relative Luciferase Activity (SRa=l)
0
Pst-luc+
0.1
0.2
0.3
0.4
a
i
NIH3T3
Pst-lucPv-luc
Fig 4. Promoter activity of the 5"flanking region
of themlL3Ra gene. The 0.8-kb Pst I-StuI, 0.4-kb Pvu
Il-Stu 1, or 0.15-kb Eag I-Stu I fragment containing
IL-BRa gene was fused
the 5"flanking region of the
t o the luciferase gene as described in Results. Pstluc' and Pst-luc- indicate
the correct and reverse orientationofthe
fragments, respectively. Plasmid
DNA was transfected intoNIH3T3 (solid bars), CTLL2 (shaded bars), or PT18 cells (open bars) and the
luciferase activity in cell extracts was determined at
18 hours after transfection. The results are shown
as relative luciferase activity; ie, luciferase activity
obtained with SRa-luc (raw value of luciferase activity/100 p g cell extract protein) is regarded as1.0.
Values represent the average of three t o four independent experimentswith standard deviation.
E-Iuc
Pst-luc+
0CTLL-2
Pst-lucPv-luc
E-Iuc
Pst-luc+
0PT18
Pst-lucPv-luc
E-IUC
the transcription in mouse mast cells but not in either mouse
fibroblasts nor T cells in an orientation-dependent manner
(Fig 4). In addition, another IL-3-dependent hematopoietic
cell line, OTT-l , also expressed luciferase after transfection
of Pst-luc at a level comparable to PT18 (data not shown).
These results indicate that the Psr I-Sru I fragment contains
the promoter that functions in a hematopoietic cell-specific
manner. In addition, there are at least two regions of positive
regulatory elements within 0.8 kb, as the 0.4-kb f v u IISru I fragment produced less luciferase than the f s r I-Stu I
fragment (Fig 4). Although there is no typical TATAA and
CCAAT sequence, TAATAA and GCAAT are present at
-200 and - 140, respectively. Instead, like many housekeeping genes, the most proximal region to the start site is GCrich (GC content is 70%). Although there are many potential
recognition sites for transcription factors as described above
(Fig 3 ) , the role of these elements for cell type-specific expression is unknown. Identification of the cis elements for
cell type-specific transcription is important to understand the
regulation of the IL-3Ra.
Because IL-3 interacts with very early multipotential hematopoietic progenitors, the IL-3 receptor must be present
on these cells. In fact, the early human hematopoietic progenitors with a hematopoietic stem cell marker CD34' express
IL-3Ra andIL-3RDsubunits.') The 0 subunit RNA is not
present in mouse embryonic stem (ES) cells. and it appears
on day 7 of in vitro development after initiation of embryoid
body culture. Likewise, 0 RNA is not detectable in blastocysts but is induced byin vitro culture of blastocysts. Timing
of the appearance of the 0 subunit RNA is consistent with
the onset ofhematopoiesis.".3"J" In contrast, the IL-3Ra
transcript is detectable in ES cells as well as blastocysts.'3
However, the IL-3Ra expression is clearly tissue-specific
andis restricted mainly to myeloid lineages.'.' Thus, transcription of the IL-3Ra and IL-3RP subunit genes appears
to be differentially regulated during development. Consistent
with the differential regulation, there is no significant similarity of the 5' upstream sequences between the mIL-3Ra
andthe mlL-3RP genes (AIC2A and AIC2B),* although
the GATA motif, frequently found in hematopoietic specific
genes, is present in both genes.
Another interesting feature of the 5 ' upstream sequence of
the mIL-3Ra gene is a high frequency of CpG dinucleotide.
Whereas CpG dinucleotides are usually found one fifth as
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
MlYAJlMA ET AL
1252
26
42
63
45
53
52
45
57
52
30 19 35
353
oo~moo81oo
19 52
41
43
77
39 10 38
33
22 18
frequently as GpC in mammalian DNA (CpG:GpC ratio is
0.2);' the CpG:GpC ratio in the region between - I to -200
of the IL-3Ra gene is as high as 1.0. Interestingly, like IL3Ra. the mouse pim-l gene, one of the hematopoietic specific genes, also contains a high G C content and has a high
CpG:GpC ratio (1.0) in the corresponding region." In contrast, the CpG:GpC ratio of the S ' flanking regions of the 0
subunit genes between -1 and -200 is 0.27.' AsDNA
methylationofteninhibits
gene expression andCpG
is
known to be a common target of DNA methylation,"3 high
frequency of CpG in the S'-flanking regionmayindicate
gene inactivation by methylation. If methylation occurs in a
cell type-specific manner during differentiation, it may explain why ES cells express IL-3Ra RNA but the differentiatednonmyeloid cellsdo not.5"." However,exogenously
introduced unmethylated plasmid DNAs containing themIL3Ra-luciferase fusion genedid not expressluciferase in
NIH3T3 cells. Thus. cell type-specific methylation,
if it is
present. does not solely explain the cell type-specific function of the mlL-3Ra promoter. Presumably, cell type-specific factors are required for regulated expression of mIL3Ra. either positively or negatively. Identification of such
transcription factors is critical to understanding the mechanism of IL-3 receptor expression.
ACKNOWLEDGMENT
Wethank DJ. Gilbert and D.Thompson for excellent technical
assistance, D. Robison for oligonucleotide synthesis,and G. Budget
for preparation of the manuscript. We also thank Drs T. Iwamoto,
D. Gorman. E. Masuda,and N. Araiforhelpfuldiscussionand
critical reading of the manuscript.
Fig 5. Structures of the cytokine receptors. Schematic representationsof the members of the class I
cytokine receptor family, whose genomic structure
has been published. Boxes represent exons. Hatched
boxes indicate signal sequences. Exons encoding
the
transmembrane domains are shown as solid boxes.
CC and WS indicate the exonsencoding the conWS motif, respectively.
served cysteine residues and
Stippled boxes in the most 5'-side represent exons
that carry a signal peptide. The G-CSF receptor contains additional motifs in the extracellular domain,
and only the exons encodingthe class I cytokine receptor motif are shown. Although the exons encoding the N-terminal and C-terminal domains also contain noncoding
regions,
the figure shows
only
protein-coding regions. Amino acid residues corresponding to each exon are indicated underneath.
REFERENCES
I . Arai K, Lee F, Miyajima A, Miyatake S, Arai N. YokotaT:
Cytokines:Coordinators ofimmuneandinflammatoryresponses.
Annu Rev Biochem S9:783, 1990
2. Miyajima A. Hara T. Kitamura T Common subunits of cytokine receptors and the functional redundancy of cytokines. Trends
BiochemSci 17378, 1992
3.Itoh N. Yonehara S, Schreurs J, Gorman DM, Maruyama K,
Ishii A. Yahara 1. Arai K, Miyajima A: Cloning of an interleukin3 receptor: A memberof a distinct receptor gene family. Science
247324, 1990
4. GormanDM, ltoh N, Kitamura T, Schreurs J, Yonehara S,
Yahara I, Arai K, Miyajima A: Cloning and expression of a gene
encoding an interleukin-3 receptor-like protein: Identificationof another member of the cytokine receptor gene family. Proc Natl Acad
SciUSA 875459. 1990
S. Hara T, Miyajima A: Twodistinctfunctionalhighaffinity
receptors for mouse IL-3. EMBO J IO:1875, 1992
6. Park LS, Martin U, Sorensen R, Luhr S . Morrissey PJ, Cosman
D, Larsen A: Cloning of the low-affinity murine granulocyte-macroa
phagecolony-stimulatingfactorreceptorandreconstitutionof
high-affinity receptor complex. Proc Natl AcadSciUSA 89:4295,
I992
S, Yamaguchi N.
7. Takaki S , Mita S, KitamuraT.Yonehara
Tominaga A. Miyajima A, Takatsu K: Identification of the second
subunit of the murine interleukin-S receptor: Interleukin-3 receptorlike protein, AICZB is a component of the high-affinity interleukinS receptor.EMBO J 10:2833, 1991
8. Gorman DM, Itoh N, Jenkins NA, Gilbert DJ. Copeland NG.
Miyajima A: Chromosomal localization and organization of the murinegenesencodingthe
p subunits(AIC2AandAICZB) of the
interleukin 3, granulocytelmacrophage colony-stimulating factorand
interleukin 5 receptors. J Biol Chem 267: 15842, 1992
9. Sato N. Caux C, Kitamura T, Watanabe Y, Arai K, Banchereau J,
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
1253
MURINE IL-3 RECEPTOR a SUBUNIT GENE
Miyajima A . Expression and factordependent modulation of the L - 3
receptor subunits on human hematopoietic cells. Blood 82752, 1993
IO. Pluznik DH, Tare NS, Zatz MM, Goldstein AL: A mast/
basophil cell line dependent on colony stimulating factor. Exp Hemato1 10:211, 1982
Brunner KT: Gen11. Cerottini J-C, Engers HD, MacDonald
eration of cytotoxic T lymphocytes in vitro. I. Response of normal
and immune mouse spleen cells in mixed leukocyte cultures. J Exp
Med140:703, 1974
12. Copeland NG, Jenkins NA: Development and application of
a molecular genetic linkage map of the mouse genome. Trends Genet
7:113, 1991
13. Jenkins NA: Organization, distribution, and stability of endogenous ecotropic murine leukemia virus DNA sequences in chromosomes of Mus musculus. J Virol 43:26, 1982
14. Ceci JD, Kingsley DM, Silan CM, Copeland NG, Jenkins
NA: An interspecific backcross linkage map of the proximal half of
mouse chromosome 14. Genomics 6:673, 1990
15. Moore KJ, D’Amore-Bruno MA, Korfhagen TR, Glasser SW,
Whitsett JA, Jenkins NA, Copeland NG: Chromosomal localization
of three pulmonary surfactant protein genes in the mouse. Genomics
12:388, 1992
16. Green EL: Linkage, Recombination, and Mapping. Genetics
and Probability in Animal Breeding Experiments. New York, NY,
MacMillan, 1981, p 77
16a. Davisson MT, Roderick TH, Hillyard AL, Doolittle DP:
GBASE database, maintained at The Jackson Laboratory, Bar Harbor, ME
17. Briggs MR, Kadonaga JT, Bell SP, Tjian R Purification and
biochemical characterization of the promoter-specific transcription
factor, Spl. Science 234:47, 1986
18. Faissts S, Meyer S: Compilation of vertebrate-encoded transcription factors. Nucleic Acids Res 20:3, 1992
19. Lewis CD, Clark SP, Felsenfield G, Gould H: An erythroidspecific protein that binds to the poly(dG) region of the chicken
beta-globin gene promoter. Genes Dev 22363, 1988
20. Driggers PH, Ennist DL, Gleason SL, Mak WH, Marks MS,
Levi BZ, Flanagan JR, Appella E, Ozato K: An interferon gammaregulated protein that binds to the interferon-inducible enhancer element of major histocompatibility complex class I gene. Proc Natl
Acad Sci USA 87:3743, 1990
21. Orkin SH: GATA-binding transcription factors in hematopoietic cells. Blood 80:575, 1992
22. Yang Z, Sugawara M, Ponath PD, Wessendorf L, Banerji J,
Strominger JL: Interferon gamma response region in the promoter
of the human DPA gene. Roc Natl Acad Sci USA 87:9226, 1990
23. McClanahan T, Dalrymple S, Barkett M, Lee F Hematopoietic growth factor receptor genes as markers of lineage commitment
during in vitro development of hematopoietic cells. Blood 83:2903,
1993
24. Colotta F, Bussolino F, Polentarutti N, Guglielmetti A, Sironi
M, Bocchietto E, De Rossi M, Mantovani A: Differential expression
of the common beta and specific alpha chains of the receptors for
GM-CSF, IL-3 and IL-5 in endothelial cells. Exp Cell Res 206:31 I ,
1993
25. Takebe Y, Seiki M, Fujisawa J, Hoy P, Yokota K, Arai K,
Yoshida M, Arai N: SR alpha promoter: An efficient and versatile
mammalian cDNA expression system composed of the simian virus
40 early promoter and the R-U5 segment of human T-cell leukemia
virus type 1 long terminal repeat. Mol Cell Biol 8:466, 1988
Pritchard M,
26. Gough N M , GearingDP,NicolaNA,BakerE,
HR.
Callen DF, Sutherland GR: Localization of the human GM-CSF
receptor
gene to the X-Y pseudoautosomal region. Nature 345:734, 1990
27. Kremer E, Baker E, D’Andrea RJ, Slim R, Philips H, Moretti
PAB, Lopez AF, Petit C, Vada MA, Sutherland GR, Goodall GJ:
A cytokine receptor gene cluster in the X-Y pseudoautosomal region? Blood 82:22, 1993
28. Milatovich T, Kitamura T, Miyajima A, Francke U: Gene
for the alpha subunit of the human interleukin 3 receptor (IL3RA)
localized to the X-Y pseudochromosomal region. Am J Hum Genet
53:1146, 1993
29. Isobe M, Kumura Y, Murata Y, Takaki S, Tominaga A, Takatsu K, Ogita Z: Localization of the gene encoding the a subunit
ofhuman interleukin-5 receptor (IL5RA) to chromosome region
3p24-3p26. Genomics 14:755, 1992
30. Disteche CM, Brannan CI, Larsen A, Adler DA, Schorderet
DF, Gearing D, Copeland NG, Jenkins NA, Park LS: The human
pseudoautosomal GM-CSF receptor a subunit gene is autosomal in
mouse. Nature Genet 1:333, 1992
31. Gough NM, Rakar S: Localization of the IL-5 receptor gene
to the distal half of murine chromosome 6 using recombinant inbred
strains of mice. Genomics 122355, 1992
32. Youssoufian H,Zon LI, Orkin SH, D’Andrea AD,Lodish
H F Structure and transcription of the mouse erythropoietin receptor
gene. Mol Cell Biol 10:3675, 1990
33. Noguchi CT, Bae KS, Chin K, Wada Y, Schechter AN, Hankins D: Cloning of the human erythropoietin receptor gene. Blood
78:2548, 1991
34. Maouche L, Tournamille C, Hattab C, Boffa G, Cartron J-P,
Chretien S: Cloning of the gene encoding the human erythropoietin
receptor. Blood 78:2557, 1991
35. Shibuya H, Yoneyama M, Nakamura Y, Harada H, Hatakeyama M, Minamoto S, Kono T, Doi T, White R, Taniguchi T: The
human interleukin-2 receptor p chain gene: Genomic organization,
promoter analysis and chromosomal assignment. Nucleic Acids Res
18:3697, 1990
36. Wrighton N, Campbell L, Harada N, Miyajima A, Lee F: The
murine interleukin-4 receptor gene: Genomic structure, expression
and potential alternative splicing. Growth Factors 6:103, 1991
37. Pleiman CM, Gimpel SD, Park LS, Harada H, Taniguchi T,
Ziegler SF: Organization of the murine andhuman interleukin-7
receptor genes: Two mRNAs generated by differential splicing and
presence of a type I-interferon inducible promoter. Mol Cell Biol
11:3052, 1991
38. Set0 Y, Fukunaga R, Nagata S: Chromosomal gene organization of the human granulocyte colony stimulating factor receptor. J
Immunol 148:259, 1992
39. Schmitt RM, Bruyns E, Snodgrass HR: Hematopoietic development of embryonic stem cells in vitro: Cytokine and receptor gene
expression. Genes Dev 5:728, 1991
40. Keller G, Kennedy M, Papayanopoulou T, Wiles MV: Hematopoietic commitment during embryonic stem cell differentiation in
culture. Mol Cell Bioi 13:473, 1993
41. Edwards YH: CpG islands in genes showing tissue-specific
expression. Philos Trans R SOCLond [Biol] 326:207, 1990
42. Selten G, Cuypers HT, Boelens W, Robanus-Maandag E,
Verbeek J, Doman J, van Beveren C, Berns A: The primary structure
of the putative oncogene pim-l shows extensive homologywith
protein kinases. Cell 46:603, 1986
43. Cedar H: DNA methylation and gene activity. Cell 53:3, 1988
44. Genome Data Base, The William H. Welch Medical Library,
Johns Hopkins University, Baltimore, MD
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
1995 85: 1246-1253
The murine interleukin-3 receptor alpha subunit gene: chromosomal
localization, genomic structure, and promoter function
I Miyajima, L Levitt, T Hara, MA Bedell, NG Copeland, NA Jenkins and A Miyajima
Updated information and services can be found at:
http://www.bloodjournal.org/content/85/5/1246.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.