Download Canine CD34: Cloning of the cDNA and Evaluation

From www.bloodjournal.org by guest on June 16, 2017. For personal use only.
Canine CD34: Cloning of the cDNA and Evaluation of an Antiserum to
Recombinant Protein
By Peter A. McSweeney, Katherine A. Rouleau, Rainer Storb, Laura Bolles, Philip M. Wallace, Mary Beaucharnp,
Ljiljana Krizanac-Bengez, Peter Moore, George Sale, Brenda Sandrnaier, Thierry de Revel, Frederick R. Appelbaurn,
and Richard A. Nash
Increasingly, enriched populations ofhematopoietic progenitors are used in experimental and clinical transplantation
studies. The separation of progenitors is based on the expression of CD34, a marker preferentially expressed on progenitor cells. The dog modelhas been important forpreclinical transplant studies, because it has proven predictive for
outcomes in human hematopoietic stem cell transplantation. To identify and isolate canine hematopoietic progenitors, we have cloned a cDNA encoding a CD34 homologue
from a canine myelomonocytic leukemia cell line, ML2. The
CD34 homologue cDNA predicts an amino acid sequence
that is highly conserved with human and murine CD34 in
the cytoplasmic domain, transmembrane domain, and C-termina1 end of the extracellular domain, but shows considerable divergence from these sequences at theamino-terminal
end ofthe protein. In Western blotting studies, canine CD34
homologue (caCD34) appears t o be a heavily and variably
glycosylated protein with a molecular weight of approxi-
mately 100 kD and shows some tissue-specific differences
in protein mass. To evaluate the expression of caCD34 protein, the extracellular domain of caCD34 was expressed as
an lg fusion protein andused as an immunogent o generate
a rabbit polyclonal antiserum. The antiserum reactedagainst
the fusion protein, against vascular endothelium, and with
three leukemic cell lines. Approximately 1% of canine bone
marrow cells stained brightly withantibodies t o caCD34 and
25- t o 50-fold enriched for colony-formthis population was
ing units-granulocyte-macrophage as compared t o unfractionated marrowmononuclear cells. Thesefindings suggest
that thecanine CD34 homologue is expressed on bone marrow progenitor cells and, thus, that this molecule should
be a valuable marker for identifying and isolating canine
hematopoietic progenitors for experimental hematopoiesis
and stem cell transplantation.
0 1996 by The American Society of Hematology.
C
CD34 may be a ligand for leukocyte L-selectin with a possible in vivo role in vascular adhesion.'," The function of
CD34 is not fully understood and human and murine CD34
have no close homology to other known proteins.
The restricted expression of CD34 has allowed for antibodies to human CD34 and to murine CD34 to be used to
identify and enrich hematopoietic progenitors as determined
by in vitro3.1?.llandin vivo studies.""'In humans, monoclonal antibodies (MoAbs) to CD34 recognize about 1.5% of
marrow mononuclear cells' and CD34-enriched progenitor
cells, derived from marrow or peripheral blood stem cells
( P B X ) , have the capability to fully restore hematopoiesis
after treatment with myeloablative chemoradiotherapy.""'
CD34 selection technology is an active area of clinical researchas a method for purging autografts of tumor or
T-cell-depleting stem cell allografts to prevent graft-versushost disease (GVHD). The dog has a proven role in experimental marrow transplantation and results of canine transplantation studies have predicted findings that have led to
important advances in allogeneic transplantation.'' Because
no marker of early canine hematopoietic progenitors has
previously been reported, we undertook
to
determine
whether the canine homologue for CD34 (caCD34) would
be such a marker. We describe here the cloning of a cDNA
for caCD34 and the characterization of cells expressing this
protein.
D34 IS A TYPE 1 transmembrane protein expressed
primarily on primitive hematopoietic progenitor
and vascular endothelium from many tissues."' Cell
surface expression of CD34 is developmentally regulated
in hematopoiesis' and isinverselyrelatedto
the stage of
differentiation, such that CD34 expression islostbeyond
the committed progenitor stage. This pattern of expression
suggests an important role of CD34 in early hematopoiesis.
The genes for human and murine CD34 have been cloned'-"'
and have significant homology of the genomic structure.
cDNA sequence, amino acid sequence, and predicted protein
structure. Predicted structural features of these proteins suggest properties of an adhesion molecule of the sialomucin
class of molecules" and murine studies showed that vascular
From the Clinical Research Division, Fred Hutchinson Cancer
Reseurch Center, Seattle, WA; the University of Washington, Seattle.
WA; Bristol-Myers Squibb Pharmaceutical Research Institute, Seattle, WA; and the University of California, Davis, CA.
Submitted Februap 26, 1996; accepted May 8, 1996.
Supported in part by Grant No. DK42716from the National Institute of Diabetes and Digestive and Kidney Diseases and Grants No.
CA47748 and CA31787from the National Cancer Institute, National
Institutes of Health, Department of Health and Human Services.
Bethesda, MD. Support was also received from a prize awarded by
the Josef Steiner Krebsstifiung, Bern, Switzerland. P.A.M. is a FeIlow of the Leukemia Society of America. L. K.-B. is supported by the
Department of Experimental Biology and Medicine, Rudjer Boskovic
Institute, Zagreb, Croatia.
Address reprint requests to Peter A. McSweeney, MD, Fred
Hutchinson Cancer Research Center, I124 Columbia St, M318 Seattle, WA 98104.
The publicationcosts of this article were defrayed in part by page
charge payment. This urticle must therefore be hereby marked
"advertisement" in accordance with 18 U.S.C. section 1734 solely to
indicate this fact.
0 1996 by The American Society of Hematology.
0006-497//96/~806-0047$3.00/0
1992
MATERIALS AND METHODS
Cell lines and cell culture. The canine leukemia cell lines MLI , l y
ML2, ML3 (all myelomonocytic), 1390 (CD8' leukemia) (ML2,
ML3, and 1390 were derived from spontaneous canine leukemias
and provided by P. Moore, University of California, Davis, CA),
and CLGL 90 (large granular cell leukemia; M. Wellman, manuscript
in preparation) were maintained in RPMI-1640 medium supplemented with 10% fetal calf serum (FCS), 25 m o V L HEPES, 0.1
mmol/L minimum essential medium (MEM) nonessential amino
acids, I mmol/L sodium pyruvate, and 0.05 m o l n 2-mercaptoethanol. Jugular vein endothelial cells from normal dogs were purchased
Blood, Vol 88, No 6 (September 15), 1996: pp 1992-2003
From www.bloodjournal.org by guest on June 16, 2017. For personal use only.
CLONING OF A DNA FOR CANINE CD34
1993
A
Coding sequence
.......................................
cDNA
0 -
C
0
c
A
B
D
44-3
28-1
54-1
Exon X
B
Fig 1. (A) Map of the cDNA clones for caCD34 in
relationship to t h e proposed cDNA sequence. Gray
segments at each end of the cDNA line represent
sequence obtained from genomic clones. The alternatively spliced exon is shown as exon X from clone
5-41. The cDNA fragments isolated by PCR for subcloning and sequencing were flanked by the primer
pairs (1)34A (A) and 348 (B) and (2) 34C (C) and 34D
(D). (B) Genomic map of caCD34 showing location
of phage clones and predicted exon-intron locations
with exons numbered underneath. The genomic segment 3‘ of the phage clones was determined by restriction analysis of canine genomic DNA. Constructs
made for sequencing t h e 5‘ and 3’ ends of t h e cDNA
are also shown. Known restriction sites are shown
as vertical lines with abbreviations as follows: S,Sa/
I; E, EcoRI; B, BamHI; SM, Sma I; SC, Sac 11.
SM BC E
B
B
BEBE
B
BE
I IIII II I
U
111 I
2 3 4
1
EB
B
I
E
l
- 1 kb
III=
567
a
CD%-27
CD34-14
CD34-5
CD34-2
B 7.0kb
B 5.0 kb
S
S
S-SC
S-
-E
E 3.8kb
1.4kb
SM 1.3 kb
from Endotech (Indianapolis, IN) and cultured according to the manufacturer’s instructions.
Northern Alntting. Total cellular RNA was prepared from cell
suspensions by guanidinum isothiocyanate lysis’“ using RNAzol A
(Texas Biotech, Houston. TX). Fifteen or 30 pg ofRNA was run
on a I % glyoxal denaturing agarose gel and transferred to nylon
membranes. Membranes were probedwith a human CD34 cDNA
probe (kindly provided byDrD. Tenen. Harvard Medical School,
Cambridge. MA) radiolabeled with [a-”P]dCTP (NEN Dupont, Boston, MA) using the random hexanucleotide priming method.”
Reverse trc~nscription-po!\rnercrse clwin reuctinr~ ( R T - f CR).
PCR primers 34A and 348 were designed from sequences of highest
homology when comparing humanand
murine CD34 cDNAs.
Primers were (see Fig I A ) (a) 34A (sense) S’-CCGAATTCGCTCCTTGCCCAGTCTGAGG; (b) 34B (antisense) 5”CCGAATTCCACGTGTTGTCTTGCTGAATGG: (C) 34c (sense) ~ ” T A G A A G (I7T(JTCGAGAAGGATGCGGCGGGCA; and (d) 34D (antisense)
S’-CAGAATTCACTGTGGTAGGAGTAATCAC. EcoRI (A, B,
and D) and Hind111 (C) restriction sites (underlined) were included.
Reverse transcription andPCR were performed as previously described.” Samples were electrophoresed through 2% agarose gels
and analyzed by ethidium bromide staining andor by probing of
Southern blots with caCD34-specific [y-”PIdATP end-labeled oligonucleotides. PCR products were cloned into the EcoRI restriction
site of pBluescript (Stratagene, La Jolla, CA) or directly into
pT7Blue T-Vector (Novapen, Madison, WI)” and sequenced as described below. To extend the S’ end of thecDNA, rapid amplification
of cDNA ends’‘ (S’ RACE) was performed using two commercial
kits (GIBCO BRL [Gaithersburg, MD]and Clontech [Palo Alto,
CA]) as described in the instructions from the manufacturers.
Screening of canine cDNA und genomic phage lihruries. A random-primed cDNA library was made in Lambda ZAP (Stratagene)
frompoly-ARNApurifiedfromtheML2
cell line. Inserts were
cloned nondirectionally into EcnRl restriction sites. Nylon filters
(Amersham. Arlington Heights. IL) lifted from plates containing 1.2
X IO” plaques were probed in buffer containing SO% formamide at
42°C with caCD34-specific probes radiolabeled by random hexamer
priming. Filters were washed at a final stringency of either O.SX or
0.1x SSPE at42°Cand
exposed toKodakXARfilm
(Eastman
Kodak.Rochester. NY) at-70°C for 1 to 3 days. Phage eluates
from positive clones were evaluated for insert size byPCR using
34A and 34B or gene-specific primers (GSP), combined with vectorspecific RNA-polymerase site primers, T3 or T7. Clones selected
for further characterization were converted to plasmids [pBluescript
SK(-)] byanin vivo excision.’5 A canine genomic phage library
made in Lambda Fix I1 (Stratagene) was screened with a cDNA
clone and then with genomic subclones. Clones were characterized
by restriction enzyme mapping. Southern blotting with GSP oligonucleotides. andPCR using primer pairs predicted to be located in
different exons of caCD34. Clones were sequenced on both strands
usingthe dideoxynucleotide method’“ or by PCR-based cycle sequencing withtheTaqDyedeoxy
terminator cycle sequencing kit
(Applied Biosystems. Inc. Foster City. CA) using an Applied Biosystems 373A DNA sequencer. Sequence analysis was performed with
Genepro 4.2 software (Riverside Scientific Enterprises, Bainbridge
Island. WA) and software from Genetics Computer Group (GCG)
Inc (Madison, WI).
froduction of recomhincmr cuCD34. An expression construct
for the extracellular domain of caCD34 was made byPCR using
primers 34F (93 bp; sense, SAGGAAGCTTCTCGAGATGCTGGTCCGCAGGGGCGCGCGCGCAGGGCCCAGGATGCCGCGGGGCTGGACCGCGCTTGCCTGCTCAGTCTGCTG: amino acid
sequence, MLVRRGARAGPRMPRGWTALCLLSLL) with S’
Hind111and Xhn I restriction sites (underlined) and34G (32 bp,
antisense). 5’-CTCCAGATCTGGCTTGCGGGAATAGCTCTGGT
with a BgI 11 restriction site (underlined). The PCR productwas
cloned into an expression plasmid containing a murine Igheavy
chain sequence. A caCD34-murine Ig fusion protein (CD34-Ig) and
a control fusion protein (murine CTLA4Ig containing the same murineIg sequence) were produced by transient expression in COS
cells and purified as previously described.” Sodium dodecyl sulfatepolyacrylamide gel electrophoresis (SDS-PAGE) wasperformed
From www.bloodjournal.org by guest on June 16, 2017. For personal use only.
1994
withprecast 4% to 20% gels (Novex, San Diego, CA) using the
manufacturer’s instructions. Automated N-terminal sequencing was
performed at the Universityof Victoria Microanalytical Centre (British Columbia,Canada)using
an AppliedBiosystemsmodel473
pulsed liquid sequencer.
Immunization of rabbits and production of afJinity-pur$ed untiCD34 polyclonal antiserum (RPacaCD34). A female New Zealand
white rabbit was used to generate a polyclonal antiserum to caCD34
as follows. Primary immunization was with combination
of 2 mg
CD34-Ig in 250 pL phosphate-buffered saline (PBS), 2.50 y L seppic
montanide ISA SO, and RIB1 adjuvant injected subcutaneously in
SO+L aliquots at 10 sites. The animal was boosted at 1 month with
1 mg CD34-Ig, at 6 months with ML2 cells, at 8 months with ML2
cells, and at I O months with a combinationof ML2 cells and canine
endothelial cells from primary culture. Rabbit IgG from serum was
purified on immobilized recombinant protein A (IA-300; Repligen,
Cambridge, MA). The subfraction
of IgGspecificforCD34was
then isolated in two steps. First, IgG reactive with the
IgG tail of
theimmunogenwasremoved
by adsorptiontoacolumn
of the
immobilized control fusion protein (CTLA4-Ig). Second, IgG reactive to CD34-lg was affinity-isolated on a column of immobilized
CD34-lg. Columns were preparedby immobilization of protein onto
CNBr-activated Sepharose 4B using the manufacturer’s instructions
(Pharmacia Biotech, Uppsala, Sweden).
Anti-CD34enzyme-linkedimmunosorbentassay(ELISA).
The
anti-CD34 ELISA was performed as described previously”with the
following modifications. Immunlon 2 (Dynatech) flat-bottom plates
were coated with 3 pg/mL CD34-Ig diluted in 0.05 mollL bicarbon9.6. Boundantibodywasdetected
with a
atebindingbuffer,pH
1:8,000 dilution of horseradish peroxidase-conjugated antirabbitIgG
antibody (Southern Biotechnology, Birmingham, AL).
Westernblorting. CellswerewashedinPBSandlysed
with a
solutionof I % NP40, 1.50 mmol/LNaClin 50 mmollLTris, pH
8.0. Nuclei were removed by I O minutes of centrifugation at 10,OOO
g and lysates were stored at
-70°C. Lysates were electrophoresed
in a reduced Trislglycine SDS 8% polyacrylamide gel (Novex). Protein was transferred to a precut polyvinylidene difluoride (PVDF)
membrane (Novex) using a wet Western transfer apparatus (Novex)
in Tris-glycine transfer buffer. Membranes were blocked
in 5 mg/
mL nonfat dry milk in PBS and then incubated with either
RPacaCD34 or control rabbit serum followed by incubation for I hour
withgoatantirabbitalkalinephosphatase(BoehringerMannheim
Biochemicals, Indianapolis, IN) diluted in blocking solution. Membranes were washed four times with TTBS
(2.5 mmol/L Tris, IS0
mmol/L NaCI, 0.5% Tween-20, pH7.5) and proteins were visualized
with Western blue AP substrate (Promega).
Flow cytometry. The followingantibodies
were used: RPacaCD34, SS” (murine IgGl MoAb, anti-CD44. as apositive control),
31A’” (a murine IgGI, nonreactive with canine hematopoietic cells),
and normal rabbit serum. RPacaCD34 wasused at 10 pg/mL, rabbit
serum at a dilution of 1: 100, and SS and 31A at S yg/mL. A fluorescein isothiocyanate(F1TC)-conjugated polyclonal goat-antirabbit antibody (Caltag, San Francisco, CA) or a phycoerythrin (PE)-conjugated polyclonal goat-antirabbit antibody (Southern Biotechnology)
and an FITC-conjugated goat-antimouse polyclonal antibody (Caltag)were used assecond-stageantibodies.Ficoll-Hypaque-separated bone marrow mononuclear cells (BMMC) and peripheral blood
mononuclear cells adjusted to S to 10 X 10‘ cells/mL were stained
for 20 minutes at 4°C at each stage and washed with PBS/2% horse
serum.Forsomeexperiments,100-pLaliquots
of unfractionated
marrow or blood were stained after the addition of SO y L of PBS/
2% horse serum and washed after each stage with PBS/2% horse
serum, andthen erythrocytes were lysed with hemolytic buffer. Cells
were fixed in 1% paraformaldehyde before analysis. Cell lines and
cultured endothelial cells were incubated with RpacaCD34 ( I O ygl
McSWEENEY ET AL
mL), CD34-lg (100 pg/mL), or a combination of the two. Cells were
thenwashedwithPBS,incubatedwithFITC-conjugatedsecondstage antibodies, and washed with PBS. Flow cytometry
was perFACStar
formed on a FACScan (Becton Dickinson, San Jose,orCA)
(BectonDickinson),andthe
list modedatawereanalyzed
using
Repromansoftware(TrueFactsSoftware
Inc, Seattle, WA) and
Cellquest Software (Becton Dickinson).
Colony-forming unit-grclnulocyte-macrophage (CFU-GM)ussays.
BMMC (2 X 107/mL) were stained with RpacaCD34 as described
above, resuspended in PBSl28 horse serum, and sorted on a FACStar (Becton Dickinson). Lineage depletion of canine BMMC was
performed with the following MoAbs: JD3 (anti-CD8):’ 1 E4 (antiCD4),3’ Tuk4 (antimonocyte; DAKO, Glostrup, Denmark), and DM5
(antigranulocyte),’2 using the magnetic activated cells sorter (MACS;
Miltenyi Biotec. Sunnyvale, CA) separation system as per the manufacturer’sinstructions.Sortedandimmunomagneticallyseparated
cell fractions were washed twice and 500, lo’, S X IO’, or IO4 cells
per plate were assayed for granulocyte-macrophage progenitor cells
( C m - G M ) , as previously described.’’
Immunoperoxidase staining of tissuesections. Normaldog tissues were snap frozen in liquid nitrogen and 6-pm sections were
cut onto glass slides using a Tissue Tek cryostat (Miles Scientific,
Naperville, IL). Air-dried slides were fixed in acetone and in formyl
calcium, and staining was performed using a TechMate I000 Immunostainer (BioTek Solutions, lnc, Santa Barbara, CA). Sections were
treated with 5% goat serum/2% bovine serum albumin (BSA) fraction (Calbiochem, La Jolla, CA), Tween-20, and 0.4% sodium azide.
Staining with RPacaCD34 (2.5 pg/mL) was detected by sequential
applicationsofbiotinylatedgoat-antirabbit(Vector,Burlingame.
CA) and horseradish peroxidase-streptavidin (Zymed, San Francisco.
CA) followed by diaminobenzidine (Polysciences, Warrington, PA)
at 0.5 mg/mL, 0. I % NiCl?, and 0.01 % H?OZ. Slides were counterstained with 0. I % acridine orangelo. 1 % safronin 0 and dehydrated
rapidly through graded alcohols. As negative controls, staining was
performed on canine tissues without primary antibody on
andhuman
lung to assess nonspecific tissue reactivity of RPacaCD34. A rabbit
polyclonal antihuman factor VI11 (Dako, Carpinteria. CA) was used
as a positive control to identify endothelial cells.
RESULTS
Isolation of the cDNA for caCD34. In Northern blotting
studies using ahuman CD34 probe, the ML2 cell yielded
line
a transcript of approximately 2.8 kb, slightly larger than that
from the KG1 myeloidleukemia cell line (human), which
is approximately 2.6 kb.R This transcript was not detected
from MLI or BMMC. In RT-PCR studiesusing primers 34A
and 34B, a strong amplification signal of approximately 450
bp was detected from both KG1 and ML2, whereas weak
signals were detected from MLI and canine bone marrow,
consistentwithNorthernblottingresults.
Sequence of the
PCR fragment isolated from ML2 was highly homologous
with 3’ coding sequenceof human and murine CD34 cDNAs
and this fragment wasused as a canine-specific CD34 probe
to screen the ML2 cDNA library. Initial screening yielded
60 positive clones and, based on PCR analyses of the insert
sizes three clones, 54-1, 44-3, and 28-1 (Fig lA), were further characterized. Clone 54-1 extended 500 bp further 5’
than any other clone and contained a 3‘
135-bp insert not
found in the other two clonesthat suggested the presence of
an alternatively spliced transcript. Clone 54-1, when used to
probe the Northern blot previously described, hybridized to
transcripts from ML-2 and KG-l of identical size to those
From www.bloodjournal.org by guest on June 16, 2017. For personal use only.
CLONING OF A DNA
FOR CANINECD34
that hybridized to the human CD34 probe (data not shown).
Attempts to extend the 5‘ sequence by rescreening the cDNA
library with a 500-bp 5’ cDNA probe from clone 54-1 and
with 5’ RACE were unsuccessful, suggesting the presence
of extensive 5‘ secondary structure in the RNA molecule.
Additional cDNA sequence for the signal peptide and 5‘
untranslated regionwas obtained after screening a canine
genomic library withclone 54-1. Two unique clones, CD3427and CD34-14, together extended from 5’ of predicted
exon 2 to 3‘ of predicted exon 8 (Fig lB), but neither hybridized to the exon 1 oligonucleotide 34E (antisense), 5’-CAGCAGACTGAGCAGGCAGAG, designed from clone 54- 1.
A 3.8-kb EcoRI subclone from CD34-14 wasused to sequence the 3‘ end of the cDNA, including the polyA signal.
After rescreening the library with a 5’ 1.1-kb BamHI fragment from CD34-27, two clones, CD34-2 and CD34-5, hybridized to the oligonucleotide 34E (Fig 1B). Becauseof the
very high GC content in the presumed first exon, deletion
constructs of a 5.0-kb Sal YBamHI subclone from CD34-5
were made to facilitate sequencing (Fig 1B). Also, a 131bp PCR cDNA fragment amplified from ML2 cDNA with
primers 34C and 34D (Fig 1A) was sequenced to determine
the signal peptide nucleotide sequence. Based on restriction
mapping and PCR studies of phage clones, the likely genomic structure of caCD34 was determined (Fig 1B).
The nucleotide sequence of the caCD34 cDNA, derived
from both cDNA and genomic
clones, is shown in Fig 2.
The canine cDNA is 135 bp longer than the human CD34
cDNA, consistent with thetranscript sizes found by Northern
blotting. The transcription start site has not been formally
mapped, butis predicted from computer alignments of canine
sequence with human CD34 cDNA and promoter sequences.
Additional sequence 5’ of theproposedcDNA
sequence
(datanot shown) showedhomology to thehumanCD34
promoter sequence,34and also showed a TATA box motif
34 bp upstream of the proposed start of transcription. Exon
1 (nucleotides 1 through 338) wasextremelyGCrich
(82.5%) and may be able to form stem-loop-stem structures
that have been implicated intranslational regulation. A variant polyadenylation signal, AAUUAA,35started at nucleotide 2730 and is identical to the poly A signal seen in the
human gene. Several AU motifs (AUUUA, AUUUUUUUA)
that may have a role in mRNA stability were found in the
3’ untranslated region.36
Analysis of the predicted amino acid sequence. The
main open reading frame of the cDNA is 1167 bp and began
with the start codon AUG at nucleotide 260 and ended with
the stop codonUGA (Fig 2A). This encodes a protein of
389 amino acids with a predicted molecular weight of 41
kD that has thefeatures of a type one transmembrane protein.
The matureproteinwaspredictedusing
SIGCLEAVE
(GCG) to begin with either the asparagine at residue 32 or
the glutamic acid at residue 34. An extracellular domain of
260 to 262 amino acids is characterized by an amino-terminal
region of approximately 150 amino acids with a high serine
and threonine content and with the potential for extensive
0-linked glycosylation. There are 7 potential N-linked glycosylation sites in the extracellular domain, of which 5 are
found within this N-terminal region. The C-terminal end of
1995
the extracellular domain contains 6 cysteine residues between aminoacids 182 and 247, whichsuggests that disulfide
bonding maybe important for structure of the protein. A
23 amino acid transmembrane domain extending between
residues 294 and 3 16 is predicted based on the presence ofa
hydrophobic domain and homologyto the human and murine
sequences. The cytoplasmic domain is predicted to be 73
amino acids and has two protein kinaseC (PKC) phosphorylation consensus sequences3’ and a potential tyrosine phosphorylation site3’ at amino acid 334 (Fig 2A).
Translation of the sequence from clone 54-1 (Fig2B)
encoded an in-frame early stop codon for themainopen
reading frame that results in a short cytoplasmic domain of
16amino acids, rather than 73 aminoacidspredicted by
sequence from clones 44-3 and 28-1. Competitive RT-PCR
studies were performed to evaluate the presence andrelative
quantity of the two caCD34 transcripts using primers that
flanked sequence encoding both transcripts (Fig 1A). Analysis of the ML2 line, several canine tissues, cultured endothelial cells, and bone marrow (data not shown) showed thatin
each cell type the alternatively spliced transcript represented
no more than 10% of the mRNA for caCD34.
A search of the Genbank and Protein Identification Resource (PIR) databases did not identify any
proteins with
significant homology to caCD34 apart from the two CD34
sequences already reported. A comparison of the amino acid
sequence from dog with those of the human and mouse (Fig
3) showedan overall aminoacidhomologybetweendog
and human of 69% and between dog and mouse of 62%.
Homologies of the cytoplasmic and transmembrane domains
between the three species were high (>86%), whereas homologies in the extracellular domain weremuchlower:
dog:human, 60%; dog:mouse, 55%; and human:mouse,56%.
The overall structure of mature CD34 proteins appears very
similar among these species, with the N-terminal portion of
the extracellular domain being rich in serine and threonine
residues. However, sequence homology is lower in this Nterminal region than further towards the C-terminal end of
the molecule. The C-terminal 100 amino acids of the extracellular domain are cysteine rich, with the number andlocation of the cysteine residues being highly conserved; this
suggests that the structure of these proteins maybevery
similar. The transmembrane and cytoplasmic domains are
highly conserved, including cytoplasmic PKC and tyrosine
phosphorylation motifs, which suggest conservation ofintracellular signalling pathways. The three species all produce an
alternatively spliced variant with an additional exon which,
although different in size and nucelotide sequence, in each
case encodes an identical 4 amino acids before a stop codon
that truncates the intracellular domain.
Expression and analysis of recombinant caCD34. To
prepare antisera to examine the cellular expression of
caCD34 protein, a fusion protein (CD34-Ig) was produced.
Using primers 34F and 34G andclone 54- 1 as PCR template,
a 906-bp fragment was amplified encoding a 99-bp signal
peptide sequence and the 780-bp extracellular domain of
caCD34 (Fig 4A). This fragment was digested with HindIII
and Bgl I1 and subcloned into HindIII and BamHI sites of a
7rLN vectorthat contained cDNA sequence encoding the
From www.bloodjournal.org by guest on June 16, 2017. For personal use only.
McSWEENEY ET AL
1996
A.cccccctcggcctccagggcggcggcaaccccggcccccgctcccgtcccccgcctgcggggctgagccgagcgctcgcg
80
gtggcggcggccaagcggaggggccggccttgccaggaacgcggagggaggggtggggagagacagccagctcgcccacc 1 6 0
ccgctccgggcggagggcggagggcggcgggcggcgggcggcggcgcgtcccggggccgagcgcgtctgtccggagccga 2 4 0
gcggagcggcgcggGAAGGATGCTGGCGGGCAGGGGCGCGCGCGCGGGCGGCGGGCTGCCGCGGGGCTGGACCGCGCTcT
320
M L A G R G A R A G G G L P R G W T A L C 21
GCCTGCTCAGTCTGCTGCCCTTTGGGTTCACAAACACAGAAACCGTGATTACTCCTACCACAGTGCCAACCTCCACAG~ 4 0 0
A
E
T
V
I
T
P
T
T
V
P
T
S
T
E
47
ATAATGTCAGCTGTTTCTGAGAATACATCC~CGGGAAGCCATCACACTAACTCCTTCTGGAACTACCACCCTGTACTC 4 8 0
I M S A V S E N T S K R E A I T L T P S G T T T L Y S 74
TGTCTCTCAAGACAGCAGTGGGACCACAGCAACCATCTCAGAGACTACAGTCCATGTCACATCTACCTCTGAGATCACCC 5 6 0
V S Q D S S G T T A T I S E T T V H V T S T S E I T L 101
TAACGCCTGGGACCATGAACTCTTCTGTTCAGTCGCAGACCTCTTTAGCTATCACGGTATCTTTTACCCC~CCAACTTT6 4 0
T P G T M N S S V Q S Q T S L A I T V S F T P T N F
127
TCAACTTCAAGTGTGACCTTGGAGCCCAGCCTGCTACCTGG~TGGTTCGGATCCCCCCTACAACAGCACCAGCCTTGT
72 0
S T S S V T L E P S L L P G N G S D P P Y N S T S L V 154
GACATCCCCCACGGAATATTATACATCACTTTCTCCTACCCCAAGTAGAAATGACACCCCAAGTACCATCAAGGGAGAAA 8 0 0
T
S
P
T
E
Y
Y
T
S
L
S
P
T
P
S
R
N
D
T
P
S
T
I
K
G
E
I 181
TCAAATGTTCCGGAGTCAAAGAAGTGAAATTGAACCAAGGTATCTGCCTAGAGCTAAATGAGACCTCCAGCTGTGAGGAC 8 8 0
K C S G V K E V K L N Q G I C L E L N E T S S C E D
207
TTTAAGAAAGATAACGAAGACTGACCCAAGTCCTGTGTGAGAAGGAGCCAGCTGAGGCTGGGGCCGGGGTGTGCTC
F
K
K
D
N
E
E
K
L
T
Q
V
L
C
E
K
E
P
A
E
A
G
A
G
V
C
S
960
234
CCTGCTTCTGGCCCAGTCTGAGGTGAGGCCTCACTGCCTGCTGCTGGTCTTGGCCAAC~CAGAACTTTTCAGT~C
1040
L L L A Q S E V R P H C L L L V L A N K T E L F S K L 261
T C C A A C T T C T G A G A A A G C A C C A G T C T G A C C T G A A A A A G C T C C A C 1120
Q L L R K H Q S D L K K L G I R D F T E Q D V G S H
287
CAGAGCTATTCCCGCAAGACCCTGATTGCACTGGTCACCTCAGGGATCCTGCTGGCTGTCTTGGGCACCACTGGTTACTT 1 2 0 0
Q S Y S R K T L I A L V T S G I L L A V L G T T G Y F 3 14
C C T G A T G A A C C G C C G C A G T T G G A G C C C T A C A G G A G A A A G G C G G T G G A G G C C 1280
L M N R R S W S P T G E R L G E D P Y Y T E N G G G Q 341
AGGGCTATAGCTCAGGCCCTGGGGTCTCCCCTGAGGCTCAGGG~GGCCAGTGTGAACCGTGGGCCTCAGGAGAACGGG1 3 6 0
367
G Y S S G P G V S P E A Q G K A S V N R G P Q E N G
ACCGGCCAGGCCACGTCCAGAAACGGCCATTCAGCAAGACAACACATGGTGGCTGATACAGAATTGTGACTCTGGGGGGG1 4 4 0
389
T G Q A T S R N G H S A R Q H M V A D T E L
GAGTAAGGCTGGGCAGGGTCTGGGGAAGGGGGCCCCTCCCAGCACCTGACCACATGCTGCCTCTGTGCTGGAGCTGCCACC 1 5 2 0
ACTTACATTCTAGCCTTTCCTGCTGCACACACCCTCCGAGGCCATTCCTGGGGCCCTGCACTGCACCAGGCCGAGGGGTT1 6 0 0
1680
CTCTCCATCCTGGGGCCCGGGAGGTAACCCCCTACCTTTTATACATTCATCTCACTAAGCCTAGAGTCTGGTCTCCTTTGA
GAAAAGACATGAGGGAGACGTGCCAAAGTATAGAGAAGCTACCAGAGTTGGGGGGGTGGGGGGTGATGATCTCACATCAT 1 7 6 0
1840
TCACGTGTGGGCTTCTTCCCTCTTCCTCCTCTCTGCCTTATT~G~CTTCCAGGGGGAAGCATGGCCTTTTCTGGGC
1920
TACAATGTCCTCCTGGGAGGCTTTGTCTTTTCCTGTGTCTTCCTCATGTCTGTCTCCTCTACTTTAGGGAAACC~GCA
CCTGCTCCTTTGTAATGCTATAGCCAGCAAGACTTGTTGTCTTAAACCGTCTCCCTTGTGCTCACACCAGCTCACTGTGG2 0 0 0
ATTCAGGCAACCGGCTTCCCTCATGCTCTCCGGGCTCCCTGAGCTCCACACCTTCTCCCTGCACCTCTGTGTACAGAC 2 0 8 0
CTGCACTGTTCTCTGGCTGAGCCTGGAACGAGACTCCAAGTTTTG~CAATGTCTTGTGTCTATGTTTGGGAGACAGCAT2 1 6 0
A G G G A T G C G T G G A C A C A T G C G T T C C T A T C T T T G G G G A C A A T C C T T G T C T C T C T G G G G 2240
CTCACAGAGTCTCATCTTGGGCCCCCGTTTCTCCCTGTGAGTCTCAGTGAACGGGACCAAGGGACCAGATCTTGGAGCCA 2 3 2 0
AGCCTCTTGACCCATGCACCTCTGAAGAAGCCCCTCGCTCGAAGGCTAGGTCCTGGCCTTGCCCTCTGATCCTGATGGCT 2 4 0 0
TCCTCCTTCCTCCCTCTGACTCCTGGGTGAGCTGTGGACTCAGACTCCCAGCAGACTCCTTTctgtctcagcctccccga 2 4 8 0
ccccaaccccctcactgttctgtac~cccatatagtcagggcccccgacatctccagaggaccttcatcacaagccatct 2 5 6 0
cctctgtaggtggcccaggttctcatttatttttttaggtattttt~ttttccagaggggtgagcagagatcttggtttc 2 6 4 0
aatgacggttggaaatagaactttccagagataggaagactgggtggattttatttctgaatacaaaaatggtgtgtgta 2 7 2 0
aatactgtaatta~agtgataccgagacacatctgttctgtgtcgctgccccagccaggtgtgtctgaatgccacggcgg 2 8 0 0
2880
tgtccctggtgtcccggtcagacccggccagacttcccaatgatgtggtagagaggggtgaccctggaaagaggtgggcc
2956
catctcgggggatacaggcaaaagccagggtgctgccccttggccaagtgtccctatgggtggggggggttggagg
B.tggagccctacaggagaaaggctggAGCTGGAACCC~ATCGCTCTTCAGGAAG~GGAGTCT~CACATGCAGCTACAA5 5
W S P T G E R L E L E P
CCCCCACTCTCTCCCCCACCCCCCACTGCCTCAATCCCCTGCTTACCAGATAATGCTCCTTTATTTATACACTGTCTAGG 1 3 5
Fig 2. Nucleotide sequence of cenine CD34 with predicted amino acid sequence shown underneath [GenBank accession no. U49457). The
sequence shown in lowercase at each end of thecDNA was determined from
genomic phage clones. The polyadenylation signal is underlined
starting at nucleotide2729. Features of the aminoacid sequence are highlighted as follows. (A)33 amino acid leader peptide (underlined).the
transmembrane domain litalics/underlined), highly conserved cysteine residues (bold), and potential N-linked glycosylation sites INXTIS;
underlined). Potential PKC (double underlined) and tyrosine phosphorylation
sites (bold) present in the cytoplasmic domain are also shown.
(B) Sequence from clone 54-1 that encodes a stop codon (double underlined] that would truncate thecytoplasmic tail after 16 amino acids
(GenBank accession no. U49458). Amino acids encoded by the alternativelyspliced exon are underlined. Sequence from the135-bp alternatively
spliced exon insert is shown in uppercase.
From www.bloodjournal.org by guest on June 16, 2017. For personal use only.
CLONINGCD34
OF A DNA FOR CANINE
1997
Human
Canine
Murine
~
~
K
~
~
--AG""-GGL-""-MQ-H-DT--LLL-WR-V--
~
'
'
~
~ PELPTQGTFS
M
P
R5 0
-------F--TNTETVI-P- TVPTSTEIM"
T-TS"-ISp
-"-"--H*** * * - N W & $ & -
Human
Canine
Murine
EXTRACELLULAR DOMAIN
NVSTNVSYQE TTTPSTLGST SLHPVSQHGN EATTNITETT
VKFTSTSVIT 100
A--E-T-KR- AI-LTPS-T- TLYS---DSS GT-AT-S--- -HV----E-S-P--E-VE- NI-S-IP--- -HYLIY-DSS KT-PA-S--M -N--V--G-P
~
Human SVYGNTNSSV QSQTSVISTV
FTTPANVSTP ETTLKPSLSP GNVSDLSTTS 150
Canine
LTp-TM---- -----UI-- SF--T-F--S SV--E---L- --G--ppyp~Murine
-GS-TPHTFS -P---PTGIL P--SDSI--S -M-W-S--PS INL-HY-Pm
Human TSL*ATSPTK
PYTSSSP*** ***ILSDIKA EIKCSGIREV KLTQGICLEQ
Canine
---*V----E Y---L--TPSm T P - T - - G ------VK-- -LN-""-L
Murine
S-FEM----E --AYT-SSA* ****P-A--G - - - - - - - - - - R-A"-"-L
200
250
Human NKTSSCAEFK KDRGEGLARV LCGEEQADAD AGAQVCSLLL AQSEVRPQCL
Canine
-E----ED-- --m-KLTQ- --EK-p-E-G- - * * - - - - - H- Murine
SEA---E--- -EK--D-IQI ---K-E-E-D ---T"---- -------E--
_______
Fig 3. Comparison of the
amino acid sequences from the
speciesfor which a full-length
CD34 sequence has been
reported. There is high homology
in the cysteine-rich regionof the
extracellular
domain,
the transmembrane domain, and the cytoplasmicdomain, but there is
marked divergence of sequence
in the amino-terminal end of the
molecule.Domains
of
caCD34
were predicted computer
by
analysisoftheproteinstructure
and the homologyto human and
murine proteins. Signal peptide
sequences are shaded.
Alignments were performed using
PILEUP (GCG).
I
Human
Canine
Murine
LLVLANRTEI SSKLQLMKKH QSDLKKLGIL DFTEQDVASH QSYSQKTLIA 300
------K--LF-----LR-- ---------R -------G" ----R
-____
-M----S--L P------E-- ----R----Q S-NK--IG-- ----R-----
H~~~~
Canine
Murine
LVTSGALLAV LGITGYFLMN RRSWSPTGER LGEDPYYTEN GGGQGYSSGP 3 5 0
TRANSMEMBRANE
I
CYTOPLASMIC DOMAIN
-"--I-"--T"""- - - - - _ _ _ _ -_- - _ _ _ _ _ ___ __ _ _ _ _ _ _ _
- - - - - V - - -1 - -T-- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
Human GTSPEAQGKA
SVNRGAQKN'G TGQATSRNGH SARQHWADT EL 385
Canine
WGP------- -----P-E--- - - - - - - - - - -----M----- - 389
Murine
-A-- -T-- - - N-T-- - -E-- - - - - - - - - - - - - - - - - - - - - - - 382
-
= identical to human amino acid
*
=
gap in sequence
hinge, CH2, and CH3 domains of a murine Ig heavy chain.
DNA sequencing verified a correct construct that encoded
amino acids 1 through 26 of the human CD34 signal peptide
and residues 14 to 293 of the predicted caCD34 sequence
fused in frame to the hinge region of murine heavy chain
Cy2a.CD34-Ig wasproduced by transient expression in
COS cells and purified using immobilizedprotein A. The
molecularweight of CD34-Ig determined by SDS-PAGE
under nonreducing conditions was -220 kD, with three distinct species being discernable (Fig 4B).
Upon reducing conditions, 2 diffuse bands were visible at -90 and 115 kD,
consistent with the protein being a homodimer. Because the
predictedmolecularweight
of CD34-Igwas 50 kD, this
suggested that the protein was glycosylated. Treatment of
CD34-Ig with N-glycosidase caused a reduction inmolecular
weight (data not shown), confirming the presence of Nlinked carbohydrate. Automated N-terminal sequencing was
performed to confirm correct processing of the human signal
peptide sequence. This produced a sequence ETVITPXTVPTSXEIMXAV (X = unidentified), which is identical to the
predicted mature caCD34 mature protein for each
ofthe
residues identified. The three unidentified residues at positions 7, 13, and 17 are predicted to be T, T, and S, respec-
-
tively. The failure to identify these residues may be a result
of them bearing carbohydrate residues.
Production and evaluation of an afJiniry-puriJied rabbit
polyclonal antiserum RPacaCD34. Serumfrom arabbit
immunized with CD34-Ig was initially tested for reactivity
to CD34-Igby ELISA, whichconfirmed an immune response
to the fusion protein (data not shown). Subsequent testing
of the affinity-purified anti-caCD34 antiserum(RPacaCD34)
by Western blotting and ELISA analysis using CD34-Ig and
control-Ig showed that an antibody titer at >1:10,000 was
present and that greater than 95% of the antibody reactivity
was directed at CD34 extracellular domain, with little residual activity against the murine Ig tail (data not shown). In
Western blotting studies of leukemic cell lines and cultured
endothelial cells, RPacaCD34 recognized a protein of -90
to 100 kD (Fig 5). A similar sized band was found in ML3,
1390, and CLGL 90 cell lines. Endothelial cells expressed
aslightly lowermolecular weight form of caCD34, suggesting thepossibility
of glycosylation differencesfrom
those in the hematopoietic cell lines.
In flow cytometry analyses using RPacaCD34, there consistently was positive staining of endothelial cells and the
cell lines ML2, ML3, and 1390, which could be blocked by
~
~
From www.bloodjournal.org by guest on June 16, 2017. For personal use only.
McSWEENEY ET AL
1998
97
-
.
L
)
..
"
30
30
21.5
21.5
14.3
14.3
QL"
r).r*Ip
Fig 4. (A) Chimeric plasmid construct for the expression of the
extracellular domain ofcaCD34 used to transfect Cos cells for production of 034-lg.(B) Analysis of CD34-lg by SDS-PAGE and coomassie
staining. CD34-lg is expressed as a homodimer and has a molecular
weight of 95 t o 100 kD under reducing conditions, whereas the predicted molecular weight is 50 kD. There appears to be more than
one
isoform being expressed, most likely due t o variable glycosylation.
A murine IgG monoclonal isrun as a control.
2
3
4
5
?
.
r
RpacaCD34
preincubation of antibody with CD34-Ig (Fig 6), but not with
a control fusion protein (data not shown), indicating that
binding of RPacaCD34 to cell surface epitopes also expressed by CD34-Ig. Endothelial cells, ML3, and 1390 all
expressed high-level caCD34, approximately 1 log of fluorescence greater than ML2. MLI and CLGL did not express
detectable surface CD34. These results were consistent with
PCR-based assays of CD34 mRNA in which strong amplification signals were detected from ML2, ML3, endothelial
cells, and 1390 cells, whereas CD34 expression was weak
or absent from MLI and CLGL 90 (data not shown). Analysis of bone marrow reproducibly showed that RpacaCD34
stained brightly approximately 1% of cells. These were primarily small mononuclear cells withlow side scatter that
were not detected in canine peripheral blood (Fig 7). Cells
were only considered CD34+ if bright staining was found as
compared with other cells. Some dim staining of larger cells
with intermediate side scatter was also observed, and these
cells had the light scattering features of monocytes.
In preliminary experiments, immunohistochemistryof dog
tissues was performed. Positive staining of canine vascular
endothelium with little nonspecificbackgroundreactivity
was found (Fig 8). To further assess nonspecific staining and
whether RPacaCD34 would recognize human endothelial
CD34, the human lung was also tested. There was nostaining
of human lung tissue, indicating that RPacaCD34 did not
contain antibodies that werecrossreactive with humanvascular CD34. Similarly, RpacaCD34 did not stain the human
CD34' leukemia cell line KG1 (data not shown). These
results were somewhat surprising in view of the shared
amino acid homologyseen in the C-terminal end of the
extracellular domains of CD34. The staining pattern in canine lung was very similar to that seen withantihuman factor
VI11 staining of canine lung andhuman lung (data not
-=a
= 250
250
= 148
= 97.4
148
97.4
= 60
60
1 6
6
Control
Fig 5. Western blotting analysis of cell lines and cultured vascular endothelium using RPcrcaCD34 and control rabbk serum. Whole cell
lysates were prepared as described in the Materials and Methods and
15 p L of each was loaded on t o t h egel. Lanes are as follows: 1, CD34l g 10.5 pg); 2, CLGL; 3, 1390; 4, MU; 5, cultured canine endothelial cells. Size markers indicate molecular weight in kilodakons.
From www.bloodjournal.org by guest on June 16, 2017. For personal use only.
1999
CLONING OF A DNA FOR CANINE CD34
by cell sorting and assayed for content of CFU-GM. In two
separate experiments, CD34' cells were 25- to 50-fold enriched for CFU-GM as compared with CD34""" cells, unfractionatedPBMC, or BMMCthathad
undergone a partial
lineage depletion (Table I ) . In experiment 2. the result was
qualitatively the same. but there was substantially better colony growth, probably due primarily to greater experience
usingthepolyclonal
antiserum in flow cytometry experiments and to sample to sample variation in progenitor content.
DISCUSSION
1390
ML-2
Log Fluorescence Intensity
Fig 6. Flow cytometry analysisofCD34expressionon
cultured
endothelial cells and canine leukemia celllinesusingRpcrcaCD34.
Cell lines were stained with RPcrcaCD34 and a second-stage FITCconjugated goat antirabbit polyclonal antibody as described in the
Materials and Methods. In each case, stainingwith RPncaCD34 isthe
shaded plot, the broken line shows staining with antibody that was
preincubated with CD34-lg, and the solid line shows staining with
only the second-stage antibody.
shown), except that staining wasmore intense with RPacaCD34. Intense staining was noted in high endothelial venules from lymph node and. strikingly, in the capillaries of
the myocardium.
Progenitor nssuy. To evaluate whether RpacaCD34
recognized canine hematopoietic progenitors, the brightest
l % of BMMC that stained with RPacaCD34 were isolated
A cDNA library made from the canine leukemia cell line
ML2 was used to clone a cDNA for the canine homologue
for CD34. The isolated cDNA sequence encoded the mature
protein but not the entire signal peptide sequence, and additional S' and 3' cDNA sequence was determined from genomic clones. Attempts to isolate cDNA clones that extended
S' of the signal peptide sequence were unsuccessful, suggesting that the RNA molecule had considerable secondary
structure. Exon I of caCD34 is extremely GC rich (82.5%),
significantly more so than that of exon 1 of the human gene
(62%), and this accounted for failures of reverse transcription
and difficulty in DNA sequencing of this region. The GCrich sequence including and immediately 5' of the signal
peptide sequence canpossiblyform stem-loop-stem structures'" that maybe important in posttranscriptional regulation of protein expre~sion.~"
CpG islands found in the first
exon maybe important in regulating tissue-specific gene
expression by changes in methylation status.MThe genomic
structure of caCD34, including the size of the gene (25 kb
between exon 1 and the polyadenylation signal), and the
proposed introdexon structure are very similar to that of the
human" and murine'') CD34 genes. The promoter region is
currently being sequenced to identify possible control elements that regulate CD34 expression in dogs, which may be
important for subsequent attempts to control gene expression
in canine hematopoietic stem cells.
This is the third CD34 cDNA cloned, and Genbank and
PIR database searches showed no sequences with significant
homology to caCD34 with the exception ofhuman CD34
and murine CD34. Because the function of CD34 remains
undetermined. it was expected that comparisons of the molecule from the three species would be informative as to the
important functional domains of CD34. The predicted protein structure from all three species shows marked homology,
with canine and human CD34 showing greater amino acid
homology than either does to the murine protein. At the Nterminal endof CD34 there appears to have been only limited
evolutionary pressure to conserve specific amino-acid sequence. Extensive 0-linked glycosylation takes place largely
on the N-terminal serine and threonine residues of the human
CD34 molec~~le.~'
Tissue-specific glycosylation differences
may affect the tissue-specific functions of CD34, as suggested by the finding that L-selectin binds CD34
to expressed
on vascular endothelium but not to CD34 expressed on hematopoietic c e k SThe conservation of the cysteine-rich area
in the C-terminal end of the extracellular domain suggests
the presence of critical regions for the extracellular functions
From www.bloodjournal.org by guest on June 16, 2017. For personal use only.
McSWEENEY ET AL
2000
A
B
"0.00
u
e
*
M
0
0
BM
0.02
PB
. .
m-
.
I=100-
.
. .
....
.
.
.
. . . .. .. . . . . .
...
.. , . .... ... 1.... . . .:,. .:.. . . . . . . <
.......... ..?.=:'.;;:.2y.
.....
. . :;. .
. ... .,!'.,
.. .. :. .. .. .. .. .
, : . :3
l
50
W
e
I
::.<r;,;,:
M
BM
'
'
1.02.
25
PB
.
.,C,!. , :
;.
. ".
.i',,..'.'
. . . . .. .. .. . .. .. . . .. . . .
..
.
..
. ....',
. ......
.. . . . ..e. . , , . .' ..
'
d' I=
zoo
D
C
m
IS0
100
Side Scatter
Side Scatter
a50
.. :.. .. .. . .. . .. .. .... .
!?,,?...';.
. * .::. .! . ... . . . .
,..."ti.
.
...
'
. .. . . .
. . .
S.
0
so
0
Side Scatter
Side Scatter
Fig 7. Flow cytometry of canine bone marrow and peripheral blood showing side scatter versus CD34 (log PE) t o illustrate patterns of
staining with RprrcaCD34. Unfractionated bone marrow (C) and peripheral blood (D) were stained with RPacaCD34 followed by a secondstage PE-conjugated polyclonal goat-antirabbit antibody. Red blood cells were lysed with hemolytic buffer. Controls shown for marrow (A)
and peripheral blood(B) were thesecond-stage PE-conjugatedgoat-antirabbit antibody. Debris was gated out before analysis based on forward
light scatter and side scatter properties of the cells. The percentage of CD34' cells (in boxes) is an arbitrary estimate determined by sfdting
a rectangular gate on cells with low side scatter expressing a high level of reactivity t o RPacaCD34. Some nonspecific staining of cells with
low-intermediate side scatter is seen, and thesecells have the forwardversus side scatter characteristics. of monocytes (data not shown).
of the molecule andlor for maintaining tertiary structure.
Cytoplasmic PKC and tyrosine phosphorylation domains
may be potentially critical for intracellular signalling functions of CD34.
An alternatively spliced transcript of CD34 encodes a premature stop codon that truncates the cytoplasmic domain.
Although the size of the alternatively spliced exon differs in
dog, human,"* and mouse,43 in each case an identical truncation of the intracellular domain of CD34 is found. The function of the truncated protein is still unknown, although a role
in progenitor differentiation associated with loss of signalling functions has been proposed."." In mice, the variant
transcript was detected in both hematopoietic and nonhema-
topoietic tissues4' and may contribute up to two thirdsof the
CD34 mRNA, depending on which tissue is examined. In
human CD34' cells, the variant transcript was expressed in
relatively higher proportion in the more differentiated
CD38' subpopulation as compared with the CD38- subpopulation." In several canine tissues, the varianttranscript
comprised less than 10% of the CD34 mRNA, suggesting
possible differences to those observed in murine and human
tissues.
A recombinant fusion protein encoding the extracellular
domain of caCD34 wasproduced once thesequenceand
reading frame of the cDNA was determined. To generate a
disulfide-linked glycosylated molecule, CD34-lgwaspro-
From www.bloodjournal.org by guest on June 16, 2017. For personal use only.
CLONING
OF A DNA FOR CANINE
2001
CD34
I
Fig 8. Immunohistochemistry of canine tissues. Sections were prepared and stainedas described in the Materialsand Methods. (A) Canine
lung-no primary antibody. (B)Canine lung-RPcrcaCD34. (C) Human lung"RPcrcaCD34. (D) Canine lung-antifactor VIII. (E) Canine lymph
node-RPcrcaCD34. (F)Canine myocardium-RPcrcaCD34. (A) through (E)were photographed at low magnification lapproximately 5 0 x ) and
(F) was photographed at medium magnification (approximately
130x1. Slides were photographedwith a Leitz Dialux 20 photomicrograph. In
(B)and (D), the arrows point t o positive staining of pulmonary capillaries. In (E), the arrow points t o a high endothelial venule. In (F), the
arrows pointt o positive staining of vascular endothelium froma myocardial capillary (thin arrow) and from an arteriole (thick arrow).
duced in a mammalian cell line. The expression of caCD34
as an Ig fusion molecule facilitated ease of purification using
protein A columns. Antibodies specific to the caCD34, isolated by a two-step affinity-binding process, showed relatively little nonspecific staining when used for flow cytometryandwereuseful
for initial characterizations of canine
hematopoietic cells. However. MoAbs will be required for
large-scale cell separation for experimental transplantation
studies as well as for more accurate flow cytometry applications.
The positive staining of canine vascular endothelium by
RpacaCD34 was consistent with patterns of CD34 expression that have previously been reported for human and murine tissuesAS Northern blotting studies in mice'" and humans8 showed that CD34 is expressed in many tissues. The
exact cellular origin of these transcripts has not been determined in humans, although murine studies suggest a likely
vascular origin in most instance^.^ In humans, patterns of
CD34 expression have not been determined using a polyclonal antiserum. Because CD34 epitope differences may
From www.bloodjournal.org by guest on June 16, 2017. For personal use only.
2002
McSWEENEY ET AL
Table 1. Analysis of CFU-GM From Sorted CD34' Cells
REFERENCES
Cell No.*
Sample
CD34' 1
CD34- 1
CD34' 2
CD34- 2
BMMC 2
LIN- 3
LIN' 3
LIN- 4
LIN' 4
LIN' 5
LIN' 5
BMMC 3
BMMC 5
o3
1
3.5
0
63
1
6
0
0
0
0
6
0
ND
ND
5
X
103
19
0
1!69
0
9
ND
0
0
0
2
1
ND
ND
104
50
0
304t
Ot
25t
1
0
2-3
0
ND
1
1
9
Values represent average from duplicate plates. Numbers 1 through
5 refer to dogs used in the experiments.
Abbreviations: ND, not done; CD34', sorted caCD34' cells; CD34-,
sorted CD34di" cells; L W , lineage-depleted 6°C;
LIN', lineage-positive BMMC.
* Number of cells plated in experiment.
t Results were calculated on basis of plating 5 x IO4 cells.
exist between different tissue^,^ the use of a polyclonal antiserum (preferably in conjunction with several MoAbs) may
have advantages, and analysis of canine tissues (studies in
progress) will give additional information about the expression patterns of CD34. Possible tissue differences in canine
CD34 epitopes were suggested by the Western blot findings
of different molecular weight isoforms when comparing endothelial cells to several leukemia cell lines.
Our primary purpose for cloning caCD34 was to determine whether CD34 could, as in humans, be used as a marker
for canine hematopoietic progenitors and, if so, to produce
antibodies for studies of canine hematopoiesis. The flow
cytometry patterns of caCD34 expression in bone marrow
are similar to those observed for CD34 in humans and support the possibility that future canine studies using antibodies
to caCD34 will have predictive significance for human studies. The enrichment of CFU-GM in caCD34' BMMC suggests that this marker will also be found on canine hematopoietic stem cells. This will need to be formally proven by
showing long-term engraftment after allogeneic transplantation of purified caCD34+ progenitors. The identification of
markers for canine hematopoietic progenitors would allow
more precise experimental work in the dog in studies that
involve exvivo expansion, peripheral blood allografting,
cord blood transplantation, and gene therapy. The cloning
of the caCD34 gene, as described here, may prove important
in this regard. The recent clonings of a canine CD38 homo10gue~~
and the canine interleukin-3 gene (unpublished preliminary results) may soon lead to development of other
useful reagents for these studies and thus further strengthen
the dog model for use in experimental transplant studies.
ACKNOWLEDGMENT
We thank B. Larson, H. Childs, and M. Kunz for assistance in
preparation of the manuscript.
1. Civin CI, Strauss LC, Brovall C, Fackler MJ, Schwartz JF,
Shaper JH: Antigenic analysis of hematopoiesis. 111. A hematopoietic
progenitor cell surface antigen defined by a monoclonalantibody
raised against KG-la cells. J Immunol 133:157, 1984
2. Katz FE, Tindle R, Sutherland R, Greaves MF: Identification
of a membrane glycoprotein associated with haemopoietic progenitor
cells. Leuk Res 9:191, 1985
3. Andrews RG, Singer JW, Bemstein ID: Monoclonal antibody
12.8 recognizes a 115-kd molecule present on both unipotent and
multipotent hematopoietic colony-forming cells and their precursors.
Blood 672342, 1986
4. Fina L, Molgaard HV, Robertson D, BradleyNJ,Monaghan
P, Delia D, Sutherland DR, Baker MA, Greaves MF: Expression of
the CD34 gene in vascular endothelial cells. Blood 75:2417, 1990
5. Baumhueter S, Dybdal N, Kyle C, Lasky LA: Global vascular
expression of murine CD34, a sialomucin-like endothelial ligand for
L-selectin. Blood 84:2554, 1994
6. Young PE, Baumhueter S, Lasky LA: The sialomucin CD34
is expressed on hematopoietic cells and blood vessels during murine
development. Blood 85:96, 1995
7. Strauss LC, Rowley SD, LaRussa VF, Sharkis SJ, Stuart RK.
Civin CI: Antigenic analysis of hematopoiesis. V. Characterization
of My-l0 antigen expression by normal lymphohematopoietic progenitor cells. Exp Hematol 14378, 1986
8. Simmons DL, Satterthwaite AB, Tenen DG, Seed B: Molecular
cloning of a cDNA encoding CD34, a sialomucin of human hemdtopoietic stem cells. J Immunol 148:267, 1992
9. Satterthwaite AB, Bum TC, Le Beau MM, Tenen DG: Structure of the gene encoding CD34, a human hematopoietic stem cell
antigen. Genomics 12:788, 1992
IO. Brown J, Greaves MF, Molgaard HV: The gene encoding the
stem cell antigen, CD34, is conserved in mouse and expressed in
haemopoietic progenitor cell lines, brain, and embryonic fibroblasts.
Int Immunol 3:175, 1991
11. Baumheter S, Singer MS, Henzel W, Hemmerich S, Renz M,
Rosen SD, Lasky LA: Binding of L-selectin to the vascular sialomucin CD34. Science 262:436, 1993
12. Lu L, Walker D, Broxmeyer HE, Hoffman R, Hu W, Walker
E: Characterization of adult human marrow hematopoietic progenitors highly enriched by two-color cell sorting with My 10 and major
histocompatibility class I1 monoclonal antibodies. J Immunol
139:1823, 1987
13. Krause DS, Takahiko I, Fackler MJ, Smith OM, Collector MI,
Sharkis SJ, May WS: Characterization of murine CD34, a marker for
hematopoietic progenitor and stem cells. Blood 8459 1. 1994
14. Berenson RJ, Andrews RG, Bensinger WI, Kalarnasz D, Knitter G, Buckner CD, Bemstein ID: Antigen CD34'marrow cells
engraft lethally irradiated baboons. J Clin Invest 81:95 I, 1988
15. Berenson RJ, Bensinger WI, Hill RS, Andrews RG, GarciaLopez J, Kalamasz DF, Still BJ, Spitzer G, Buckner CD, Bemstein
ID, Thomas ED: Engraftment after infusion of CD34' marrow cells
in patients with breast cancer or neuroblastoma. Blood 77: 1717,
1991
16. Shpall El, Jones RB, Bearrnan SI, Franklin WA, Archer PG,
Curie1 T, Bitter M, Claman HN, Stemmer SM, Purdy M, Myers SE,
Hami L, Taffs S, Heimfeld S, Hallogan J, Berenson RJ: Transplantation of enriched CD34-positive autologous marrow into breast cancer
patients following high-dose chemotherapy: Influence of CD34-positive peripheral-blood progenitors and growth factors on engraftment.
J Clin Oncol 12:28, 1994
17. Schiller G, Vescio R, Freytes C, Spitzer G, Sahebi F, Lee M.
Wu CH, Cao J, Lee JC, Hong CH, Lichtenstein A. Lill M, Hall J.
Berenson R, Berenson J: Transplantation of CD34+ peripheral blood
From www.bloodjournal.org by guest on June 16, 2017. For personal use only.
CLONING OF A DNA FOR CANINECD34
progenitor cells after high-dose chemotherapy for patients with advanced multiple myeloma. Blood 86:390, 1995
18. Storb R, Thomas ED: Graft-versus-host disease in dog and
man: The Seattle Experience, in Moller G (ed): Immunological Reviews. No. 88. Copenhagen, Denmark, Munksgaard, 1985, p 215
19. Kawakami T, Cain G, Taylor N: Establishment and partial
characterization of a radiation-induced canine monocytic leukemic
cell line (RK9ML-I). Leuk Res 13:709, 1989
20. Chomczynski P, Sacchi N: Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction.
Anal Biochem 162:156, 1987
21. Feinberg AP, Vogelstein B: A technique for radiolabeling
DNA restriction endonuclease fragments tohigh specific activity.
Anal Biochem 132:6, 1983
22. Huss R, Hong DS, McSweeney PA, Hoy CA, Deeg HJ.:
Differentiation of canine bone marrow cells with hemopoietic characteristics from an adherent stromal cell precursor. Proc Natl Acad
Sci USA 92:748, 1995
23. Marchuk D, Drumm M, Saulino A, Collins FS: Construction
of T-vectors, a rapid andgeneral system for direct cloning of unmodified PCR products. Nucleic Acids Res 19:1154, 1991
24. Frohman MA,DushMK, Martin GR: Rapid production of
full-length cDNAs from rare transcripts: Amplification usinga single
gene-specific oligonucleotide primer. ProcNatlAcad
Sci USA
85:8998, 1988
25. Short JM, Fernandez JM, Sorge JA, Huse WD: A ZAP A
bacteriophage A expression vector with in vivo excision properties.
Nucleic Acids Res 16:7583, 1988
26. Sanger F, Nicklen S, Coulson AR: DNA sequencing with
chain-terminating inhibitors. Proc Natl Acad Sci USA 745463, 1977
27. Wallace PM, Johnson JS, Macmaster JF, Kennedy KA, Gladstone P, Linsley PS: CTLA4Ig treatment ameliorates the lethality
of murine graft-versus-host disease across major histocompatibility
complex baniers. Transplantation 58:602, 1994
28. Linsley PS, Wallace PM, Johnson J, Gibson MG, Greene JL,
Ledbetter JA, Singh C, Tepper MA: Immunosuppression in vivo by
a soluble form of the CTLA-4 T cell activation molecule. Science
257:792,1992
29. Schuening F, Storb R, Goehle S, Meyer J, Graham TC, Deeg
HJ, Appelbaum FR, Sale GE, Graf L, Loughran TP, Jr: Facilitation
of engraftment of DLA-nonidentical marrow by treatment of recipients with monoclonal antibody directed against marrow cells surviving radiation. Transplantation 44:607, 1987
30. Denkers E, Badger CC, Ledbetter JA, Bernstein ID: Influence
of antibody isotype on passive serotherapy of lymphoma. J Immunol
135:2183, 1985
3 I . Moore PF, Rossitto PV, Danilenko DM, Wielenga JJ, Raff
2003
RF, Severns E: Monoclonal antibodies specific for canine CD4 and
CD8 define functional T-lymphocyte subsets and high density expression of CD4 by canine neutrophils. Tissue Antigens 40:75, 1992
32. Sandmaier BM, Schuening FG, Bianco JA, Rosenman SJ,
Bernstein I, Goehle S, Storb R, Appelbaum FR: Biochemical characterization of a unique canine myeloid antigen. Leukemia 5:125, 1991
33. Rossbach H-C, Krizanac-Bengez L, Santos EB, Gooley TA,
Sandmaier BM:An antibody to CD44 enhances hematopoiesis in
long-term marrow cultures. Exp Hematol 24:221, 1996
34. Burn TC, Satterthwaite AB, Tenen DG: The human CD34
hematopoietic stem cell antigen promoter and a 3’ enhancer direct
hematopoietic expression in tissue culture. Blood 80:3051, 1992
35. Wilusz J, Pettine SM, Shenk T: Functional analysis of point
mutations in the AAUAAA motif of the SV40 late polyadenylation
signal. Nucleic Acids Res 17:3899, 1989
36. Shaw G , Kamen R: A conserved AU sequence from the 3’
untranslated region of GM-CSF mRNA mediates selective mRNA
degradation. Cell 46:659, 1986
37. Woodget JR. Gould KL, Hunter T: Substrate specificity of
protein kinase C. Use of synthetic peptides corresponding to physiological sites as probes for substrate recognition requirements. Eur J
Biochem 161:177, 1986
38. Patchinsky T, Hunter T, Esch FS, Cooper JA, Sefton BM:
Analysis of the amino acids surrounding sites of tyrosine phosphorylation. Proc Natl Acad Sci USA 79:973, 1982
39. Devereux J, Haeberli P, Smithies 0: A comprehensive set of
sequence analysis programs for the VAX. Nucleic Acids Res 12387,
1984
40. KozakM: Leader length and secondary structure modulate
mRNA function under conditions of stress. Mol Cell Biol 8:2737,
1988
41. Sutherland DR, Watt SM, Dowden G, Karhi K, Baker MA,
Greaves MF, Smart JE: Structural and partial amino acid sequence
analysis of the human hematopoietic progenitor cell antigen CD34.
Leukemia 2:793, 1988
42. Nakamura Y, Komano H, Nakauchi H: Two alternative forms
of cDNA encoding CD34. Exp Hematol 21:236, 1993
43. Suda J, Sudo T, Ito M, Ohno N, Yamaguchi Y, Suda T: Two
types of murine CD34 mRNA generated by alternative splicing.
Blood 79:2288, 1992
44. Fackler MJ, Krause DS, Smith OM, Civin CI, May WS: Fulllength but nottruncated CD34 inhibits hematopoietic cell differentiation of M1 cells. Blood 85:3040, 1995
45. Krause DS, Fackler MJ, Civin CI, May WS: CD34: Structure,
biology, and clinical utility. Blood 87:1, 1996
46. Uribe L, Henthorn PS, Grimaldi JC, Somberg R, Hartnett
BJ, Felsburg PJ, Weinberg K Cloning and expression of the canine
CD38 gene. Blood 86:115a, 1995 (abstr, suppl 1)
From www.bloodjournal.org by guest on June 16, 2017. For personal use only.
1996 88: 1992-2003
Canine CD34: cloning of the cDNA and evaluation of an antiserum to
recombinant protein
PA McSweeney, KA Rouleau, R Storb, L Bolles, PM Wallace, M Beauchamp, L Krizanac-Bengez,
P Moore, G Sale, B Sandmaier, T de Revel, FR Appelbaum and RA Nash
Updated information and services can be found at:
http://www.bloodjournal.org/content/88/6/1992.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.