From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
Analysis
of the
Erythroid
Phenotype
and
By Th.
the
Effect
Papayannopoulou,
E
STABLISHED
ERYTHROLEUKEM
IA cell
lines
capable
of constitutive
and induced
globin
expression
are valuable
tools for the investigation
of cellular
and molecular
mechanisms
regulating
ment
which,
We
have
globin
expression
and
erythroid
previously
described
the
establish-
of a human
erythroleukemia
like K562 cells,2 are capable
sion.
Detailed
characterization
file of HEL
cells3
of the
disclosed
erythroid-specific
surface
of other
nonerythroid
cytic
lineage)
that
line,
the HEL
cells,’
of human
globin
expressurface
that
these
markers
and,
cell lineages
have been the
antigenic
cells
pro-
display
several
in addition,
features
(ie, platelet/megakaryosubject
of several
recent
studies”2
However,
only preliminary
data
have been presented
to date on the erythroid
phenotype
of HEL cells.’
In
the
ization
present
study,
of globin
we report
produced
detailed
by HEL
induction
by several
inducing
the availability
of antiglobin
bodies
(MoAbs)
we were
both
HEL
and
globins
at basal
ers.
The
certain
type
augmentation
of HEL
and
comparative
two
lines,
exploited
Lines
as assessed
through
with
of their
K562
unique
level
in
attended
erythroid
by
pheno-
the use of monoclonal
The present
cells indicate
analysis
that these
characteristics
and
their
complement
each other as models
to be
on the regulation
of globin expression.
Culture
AND
METHODS
Methods
HEL cells have been in continuous
their establishment
in 1980. Three
S. Kurachi,
different
in our
sublines
laboratory
since
of K562
a 1987
by
cells
From the Department
ofMedicine,
Divisions
ofllematology
and
Medical
Genetics,
University
of Washington,
Seattle;
and the
Department
ofPharmaceutical
Chemistry.
University
of California
Francisco.
Submitted
July
Address
reprint
ment
ofMedicine,
26. 1985; accepted July 29. 1987.
requests
to Th. Papayannopoulou,
Division
ton, Seattle,
WA 98195.
© 1987 by Grune & Stratton,
0006-4971/87/7006-0009$3.00/0
1764
ofHematology.
Inc.
University
MD, Departof
Washing-
and R. Nelson
& Stratton.
Grune
Inc.
were used for comparison:
clone C-l6, obtained
from Dr M. Fukuda’3; K562 cells from Dr J. Hansen
(K562-H);
and K562(S)
cells
from Dr G. Rovera.’4 Both K562 and HEL cells were maintained
in
RPMI
1640 medium
supplemented
with 10% fetal calf serum,
pyruvate,
and antibiotics
in a humidified
5% CO2 incubator.
Cells
were split and transferred
into fresh medium
at biweekly
intervals.
All cultures
were tested
periodically
to ensure
the absence
of
mycoplasma.
To obtain cell doubling
time in suspension
cultures,
daily cell counts in a hemocytometer
were done. To derive clones
from HEL or K562 cells, two systems
of cloning were used: the
limited-dilution
method in microtiter
wells or cloning in methylcellulose (MC)
semisolid
media and subcloning
of individual
clones
initially lifted from MC plates. Cloning of HEL and K562 cells was
also carried out in plasma clot cultures,
but these cultures
were used
primarily
for either benzidine
staining or immunofluorescence
staining ofcolonies
(see the following sections).
During
log-phase
growth
(about
2 x 105/celis/mL),
a given
inducer (Tablel ) was added at various doses to the culture niedia,
and aliquots of cells were taken for testing at several intervals
after
the addition
of inducer.
The effect of various
inducers
on heme
synthesis
was assessed
by benzidine
staining
and total hemoglobin
determination,
whereas the effect on globin was evaluated
by globin
chain analysis and by anti-globin
immunofluorescence
labeling. The
effect on surface markers
was evaluated
by surface immunofluorescence labeling using antibodies
with known specificities.
Benzidine
Staining
For evaluation
of heme or
cells, benzidine
staining
was
fixed cytocentrifuge
smears.
method described
by McLeod
hemoglobin
induction
in HEL or K562
used in either cell suspension’5
or in
For the latter,
an adaptation
of the
et al’6 was used.
Analysis
Hemoglobin
determination
in lysates of induced
and uninduced
cells was carried
out by a spectrophotometric
method,’7
whereas
separation
at San
Variation
contrast
to parental
cells. significant
amounts
of a globin.
This
clone differed
from K562 cells by the absence
of any
globin
expression.
thus demonstrating
the independent
regulation
of the
two
embryonic
chains.
a and . Changes
in the expression
of
several
surface
markers
specific
for erythroid
cells were found to
accompany
the globin accumulation
in these
cells. and some of
these changes
appeared
to be inducer
specific.
Thus. the unique
globin and nonglobin
phenotypic
properties
of HEL cells and their
subclones
make them valuable
cellular
models complementary
to
the existing
K562 cells for studying
regulatory
aspects
of crythroid-spec’sflc
proteins.
Hemoglobin
culture
Clonal
Induction
of the various
by the inducwas
surface
specificity.
studies
by virtue
and
at the cellular
expression
in the
erythroid
MATERIALS
Cell
to study,
in globin
phenotype,
for studies
composite
able
changes
cells,
with
characterand after
agents.
Furthermore,
through
chain-specific
monoclonal
anti-
K562
cells,
the distribution
conditions
and its modulation
coordinate
antibodies
biochemical
cells before
Cells:
of Inducers
B. Nakamoto,
The erythroid
phenotype
of HEL cells. before and after the addition
of a variety
of inducers.
was assessed
at the cellular and biochemical level.
Among
14 inducers
used. #{246}-aminolevulinic acid (i-ALA)
was identified
as the most optimal
inducer
of heme
and globin
synthesis
in HEL cells. The relative
synthesis
of globin
chains
produced
by HEL cells. mainly ‘y and a chains with traces
of a and
chains.
was not influenced
by the majority
of the inducers
used.
However.
6-ALA and bromodeoxyuridine
did increase
the relative
synthesis
of a and a chains
respectively.
Subcloning
experiments
revealed
heterogeneity
in the constitutive
expression
of a globin;
however,
the latter was inducible
in all clones
by either hemin or
&-ALA.
One rare clone
of HEL cells was found
to produce.
in
differentiation.
of HEL
of
isoelectric
the
focusing
modifications.
hemoglobin
Hemoglobin
staining.
For definitive
gels, we isolated
their
giobin chain isoelectric
hemoglobin
Globin
species
in polyacrylamide
present
gels
bands
were
identification
of
constituent
chains
focusing
gels by
was
achieved
as described’8
with
by
slight
visualized
by benzidine
hemoglobin
bands in our
(see the next section)
in
using sliced bands from
gels.
Chain
Analysis
For globin chain separation,
(for an eight-I 2-hour period)
Blood.
lysates from 3H-leucine-iabeled
were subjected
to isoelectric
Vol 70, No 6 (December),
1987:
cells
focusing
pp 1764-1772
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
ANALYSIS
OF THE
ERYTHROID
PHENOTYPE
OF
HEL
according
to a method described
by Rhighetti
et al’9 and
by us.#{176}
Before isoelectric
focusing,
globin was purified
haptoglobin
binding.2’
Under our conditions
of isoelectric
all human chains (s, ‘, y, fi, a) can be distinctly
separated
other. After isofocusing,
the gels were fixed, treated with
(DuPont,
Boston), dried, and subjected
to fluorography
tometry as described.#{176}
Immunofluorescence
With
Antiglobin
Chain
1765
CELLS
as adapted
through
focusing,
from each
En3Hance
and densi-
smears
prepared
from uninduced
and induced
in methanol
and reacted
with antiglobin
chain
antibodies22 in an indirect
immunofluorescence
assay using antimouse
IgG (F(ab’)2)
conjugated
to fluorescein
isothiocyanate
(FITC) as a second antibody.
Two MoAbs, one against y chains and
one specific for
chains, were used in a I : I ,000 dilution of ascites
fluid. A polyclonal
antibody
monospecific
for chains was also used.
This antibody
was made by using Hb Portland
as an immunogen,
and it was kindly provided to us by Dr D. Chui.23
Immunofluorescence
Several previously
characterized
MoAbs were used in an indirect
immunofluorescence
assay for evaluating
the membrane
reactivity
of induced v uninduced
cells. These antibodies
include antiglycophorin A (R-10 and F-l l), EP-l and EP-2,24 anti-i (Den),
anti-I
(Ma),25 anti-H
(BE2 and Ulex europaeus
I),25 and antitransferrin
receptor antibodies
0KT92’ and BE/2527 as well as a recent series of
antierythroid
MoAbs generated
by us.22 Labeled
cells were viewed
under a Zeiss epifluorescence
microscope
through
the appropriate
filters, or they were subjected
to fluorometric
analysis
in a fluorescence-activated
cell sorter (Ortho system 50H Cytofluorograf
with a
2150 computer,
Westwood,
MA).
of HEL
cells was done in MC media. Single colonies
grown in MC were subcloned
at least twice to ensure a single-cell
origin of the clones. Subcloning
of K562 cells was carried out by the
limited-dilution
method.
After cloning,
the isolated
clones were
expanded
in suspension
cultures,
and they were induced and evaluated for globin expression
and benzidine
positivity,
as indicated
for
parental
cells.
Subcloning
Commitment
Assay
To test whether a brief exposure
to inducers
leads to an irreversible decision
for terminal
differentiation
in the absence
of the
inducer,
a two-step commitment
assay was performed
as described
previously
for K562 cells or MEL cells.#{176}
Briefly, after a one-day
exposure
of HEL and K562 cells to an optimal concentration
of an
inducer,
the cells were washed to remove the inducer, counted,
and
plated (at I ,000 to 3,000 cells/mL)
in Iscove’s modified Dulbecco’s
medium (IMDM)
or RPMI 1640, 30% fetal calfserum
(FCS), and
0.8% MC. Colony number
as well as the proportion
of benzidinepositive colonies#{176}
was determined
at day 5 postpiating.
RESULTS
Induction
ofHeme
Uninduced
positive
cells
were between
the various
HEL
inducers
addition
cells
Synthesis
HEL cells showed
only occasional
(Table
I ), and hemoglobin
levels
0.2 and 0.4 pg hemoglobin/cell.
yielded
on benzidine
postivity
of b-aminolevulinic
the
most
optimum
acid
benzidinein these cells
The effect of
is shown
in Table
(#{244}-ALA) to growing
induction,
and
as assessed
hemoglobin
0.4 to 3 to 4 pg hemoglobin/cell.
caused
contrast
analysis;
Most
a slight
increase
to 6-ALA,
had
toporphyrin-IX)
in significant
culture.
in
an
and
It is notable,
of the other
inducers
and,
on
cells,
three
which resulted
or four days in
however,
that
in
the
metalloporphyIX, or pro-
hemolysates,
however,
to one
the total
ie, from
benzidine
positivity
early
adverse
effect
of HEL cells. The
Co-protoporphyrin
solutions
less toxic.
up
and
to tenfold,
were very toxic to HEL
death
during
the first
cell
Hemin
much
HEL
cells
were
are
far
more sensitive
than K562 cells to the toxic effects
of hemin.
The effect
of hemin
on new heme
synthesis
could
not be
assessed
with
because
uptake
confidence
when
using
the
of nonspecific
positivity,
of exogenous
hemin.
Among
tested,
there
including
an earlier
passage
was no significant
variability
either
before
or after
induction.
the other
Apart
and
wet
However,
of serum
induction.
The
tested
was
bovine
serum
follows:
20%
serum
FCS
one clone,
tested.
inducer,
influence
HEL-R,
level
level
of
of the
10% FCS
> horse
the
the
property
>
(HEL-92),
positivity,
sera
>
newborn
globin
species
serum.
Synthesis
previously
shown
that
by uninduced
HEL
traces
of hemoglobin
embryonic
hemoglobins
globin
F were present.
determining
the chain
fractions
disclosed
hemoglobin
a mixture
to
to the
clones
positivity,
especially
respect,
differed
from
potentiating
> human
ofGlobin
produced
I). Only
seemed
induction
as
It was
used
stain
due
HEL
of HEL cells
in benzidine
parental
HEL cells or subclones
from the presence
of the specific
type
benzidine
presumably
the various
showed
significant
levels of benzidine
after induction
(Table
2) and, in that
Induction
Subcloning
1 . The
positivity
third
of the cells became
benzidine-positive,
hemoglobin
content
increased
from three-
proliferation
and viability
rins (Zn-protoporphyrin-IX,
Antibodies
Cytocentrifuge
cells were fixed
Surface
by benzidine
the
cells
main
is Bart’s
hemoglobin
fractions
comigrating
(Fig
with
(Gower
I, Gower
II) or with hemoThis
phenotype
was confirmed
by
composition
of the major
hemoglobin
in two-dimensional
gels.
This
type of analysis
that
the fraction
migrating
in the position
of
Portland
is composed
not of -y and chains
but of
of acetylated
and nonacetylated
G-y and A-1’ chains
(Fig 1). Furthermore,
as expected
globin
chain
isofocusing
of
from hemoglobin
3H-leucine-labeled
analysis,
lysates
showed
either
and a variable
trace or undetectable
levels of #{128}
and
chains
amount
of a chains.
Although
the major
globin
are ‘y chains,
species
as in fetuses
of Gy. The
because
pattern
the proportion
contrasts
with the one present
predominance
of G7 chains
embryonic
globin
chains
(Figs
At a cellular
uninduced
HEL
clonal
antibody,
positive
chains
in K562 cells in which
there
and
abundant
synthesis
3-5).
antibody
for
cells seen. An
to t chains
was not available
to assess
of t globin.
The effect of the various
synthesis
is
of
level, a small
but significant
proportion
of
cells was labeled
with anti--y
chain
monowhereas
only rare cells were
found
to be
with a polyclonal
and monospecific
(Fig 2A). In no instance
were fl-positive
antibody
distribution
globin
of Gy to A’y is not
there is an excess of A-y chains
instead
of globin
chain
synthesis
in HEL
cells
is shown
in Table
1 and
Figs
the cellular
inducers
on
3 to 5. y chain
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
1766
PAPAYANNOPOULOU
Table
1 . Effect
of Different
Inducers
on the Benzidine
Days of
Agents
Treatment
Concentration
IX
50
Co-Protoporphyrin-IX
smol/L
50 imol/L
Viability
(%)
Benzicine+
1%)
100
>90
Hemin
<1.0
3-10
7
26
60
0.4
<1.0
<2.0
5
49
83
0.5
13
52
-
-
3
64
95
-
-
6
5
10
-
7
90
90
50 imol/L
Hemolysate
gHb/dL
3
100
95
2.2
10-30
Mmol/L
3
100
94
< 1-2.0
40-60
7
100
92
5-20
60-80
3
75
84
0.24
<1.0
88
0.0
6.57
x lO5mol/L
0.5%
butyrate
-
63-97
0.8
6
-
77
0.3
3
mmol/L
3
14
0.4
1.1
33
59
1.4
6
50
60
0.8
3
48
49
6
16
7
105mo1/L
4
88
96
5.0
i#{248}-mol/L
4
34
0.2
15-21
2.4
x
1O’mol/L
Ara-C
3.5
x iO
100
3
mol/L
1x1Omol/L
5.Ox
50
positivity
-
-
-
x
lO6mol/L
3
58
92
1.3
39.8
1.62
x
lO5mol/L
3
88
91
2.0
40.3
3
91
90
0. 1
6
93
1.0
18
93
0.6
36.0
1 .5 x 1
mol/L
1 mmol/L
-
dimethyl
sultoxide;
Ara-C,
cytosine
is given, it indicates
values from multiple
by immunofluorescence
-
3
DM50.
globin
induction
with
3H-leucine
0.6
3.25
L-Ethionine
a range
39.0
10-15
x
12
e/.Jn
1-5
3.25
D
Abbreviations:
Rare
88
Bleomycin
Actinomycin
3
1 mmol/L
0.5
BrDU
20-50
500
DMSO
Sodium
-
(96)
0.45
h-ALA
Hydroxyurea
Globin+
(%)
10
25 imol/L
Zn-Protoporphyrin-IX
of HEL Cells
Positivity
No. of Cells
(Percentage
of Control)
None
Protoporphyrin
and ‘y-Globin
ET AL
was
used
100
arabinoside;
BrDU,
bromodeoxyuridine.
experiments.
as a screening
with all inducers,
and labeling
ofcell
was carried
out only in the most
of
lysates
potent
not
shown).
Of interest,
the
addition
of BrDU
in HEL
cells
but not in K562 cells induced
the appearance
of a significant
proportion
of adherent,
elongated
cells (data
not shown).
inducers
identified
by the screening
assay.
The majority
of
the inducers
appeared
to somewhat
augment
-y globin synthesis. However,
only few of them
appeared
to influence
the
specific
globin
pattern
of HEL cells. Thus, hemin and s-ALA
Because
of the demonstrated
cellular
heterogeneity
in
globin
and a presumed
similar
heterogeneity
in e globin
expression,
subcloning
of HEL
cells was carried
out in the
hope of obtaining
globin
phenotypes
that
differed
signifi-
consistently
described
cantly
tested
increased
previously
for
the
K562
proportion
of
cells,3’ whereas
a-chains,
5-azacytidine
as
and BrDU
increased
the relative
expression
of e chains.
The
effect on t globin
was mostly
evident
in a subclone
of K562
cells,
C-16,
chains
BrDU
which
does
not
synthesize
in the uninduced
state
(Fig
and sodium
butyrate
increase
t globin
in HEL-92,
whereas
Table
2.
K562(S)
Cellular
Hemin
(97)
6-ALA
Distribution
0-it
3-9
13-28
15-45
<1.0
70
35
1.3
81
34-67
61
10
4.0
77
7.0
-
8.0
18.5
independent
experiments.
68-91
found
of a chain
and After
35
2.0
27-55
2.5
HEL
clone
treated
for
to be distinctly
subclone,
level of
expression
compared
fi Globin+
HEL-R
0
-
0
difdesigchain
Induction
Cells (%)
34.0
1.1
several
Before
Globin+
-
However,
one
cells
repeatedly
was
levels
3-6
12
from
Cells
K562
Cells (%)
higher
HEL-92
43.0
values
in HEL and
and
K562(S)
four to six days after treatment.
indicate
Globins
expression
HEL-R
24
-
of Individual
contamination
Among
a total of 20 clones
was the one concerning
the
from the parental
population.
This
HEL-R,
is characterized
by a high
HEL-92
<1.0
BrDU
data
mycoplasma
ferent
nated
cells.
variation
(data
not shown).
from
parental
HEL
K562(S)
51
-
5 and
a chains
obtained
HEL-R
93
42
(Fig
‘y Globin+
-
Butyrate
CeIls tested
of
of
(96)
Are-C
tRanses
of t
Cells (%)
HEL-92
30
None
significant
#{244}-ALAdid not
Benzidine+
lnducer
levels
3). Also, the addition
the relative
synthesis
from the parental
in detail,
the only
Cells (%)
K562(S)
HEL-92
0.4
0
Rare
0
0
-
0
2-4
1.5
HEL-R
0
-
48
<5
41.0
47
-
0
1.5
0
0
34
36.0
-
-
0
-
0
0
39
49.0
18
-
0
1.3
0
0
-
-
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
ANALYSIS
OF THE
ERYTHROID
PHENOTYPE
Globin
Yoc-{’
Chain
OF HEL CELLS
1767
IEF
-
cz
EGyA7
Yoc
N
4-
Ay
Gy-
4
.
HEL
Bromodeoxyuridine
#{149}( Uninduced
.
I It
CB HEL
1f
+
K-562
(equine)
Hemin
(bovine)
+Hnin
5-Azocytidine
III
(C-16)
Hemin
5-Azocytidine
II
N
I
#{149}
Hemin(RL)
i
Bromodeoxyurldlne
#{149}t
a
Fig 1 .
Hemoglobin
(Hb) and globin chain isoelectric
focusing
of
(K562-H)
and HEL lysates
(3H-Ieucine
labeled).
Individual
hemoglobin
species
separated
in hemoglobin
isoelectric
focusing
gels (Hb-IEF)
were
sliced
and run in noxidet
P40-urea
gels to
analyze
their
constituent
globin
chains
(globin
chain
IEF gels).
Major
Hb bands from HEL cells are Bert’s and Bart’s
acetylated
(B-ac).
which
migrates
in the position
of Hb Portland.
Embryonic
globins (Gower
I. Gower
II. Hb-X) as well as Hb F are minor bands in
this HEL subline (severely
a chain deficient).
although
there were
considerably
more
in HEL-R.
Hb-IEF.
first-dimension
Hb gels
stained
with benzidine;
globin chain IEF. second-dimension
gels
showing
radiolabeled
globin chain bands; CB. Cord blood lysete;
K562
acetylated
EGYAYYOC
Fig 3.
Effect of various
inducers
on the relative
expression
of
globin chains in K562 (C-i6)
and HEL cells. Note the increase
in a
chains
by hemin
and the relative
increase
in t chains
by 5azacytidine
and BrDU.
(Hemin
[RI] is hemin crystallized
by Dr R.
labbe’,
University
of Washington.)
Globin Chain IEF
yac,
21 2)
with
parental
cells
(Fig
However,
5).
absent,
and this feature
as well
A’y over Gy chains
distinguishes
cells
(Fig
5).
In
addition
to
chain
as the increased
this subclone
the
high
levels
inducibility
(Table
2).
by
the
same
inducers
used
for
Anti-y
Anti-c
Anti-c
Anti-$
A
HEL
B
K562
of
t globin
production,
this clone was characterized
by high constitutive
levels of heme
and total globin
synthesis
(ie, higher
basal
levels of y globin and benzidine-positive
cells) and by its high
‘y chains.
Fig 2.
(A) Immunofluorescent
labeling
of &-ALAinduced
HEL cells by monoclonal
anti-’y chain antibody
conjugated
to FITC and by monospecific
anti-c antibody
plus antirabbit
IgG-FITC.
Preparations
are of similar
cellular
densities
and photographed
at the same magnification
( x 31 2). Note the high proportion
of y + cells
and the very low proportion
of +
cells within this HEL
population.
(B) Immunofluorescent
labeling
of i-ALAinduced
K562 (C-i 6) cells with anti-i
antibody
and with
monoclonal
antiantibody
conjugated
to FITC.
Although
‘y globin can be detected
in the great majority
of induced
K562 (C-16)
cells (see Table 2). a smaller
proportion
of cells is c-positive.
and rare cells are
p6-positive.
This positivity
is attributable
to the presence of 6 rather
than
chains (see text).
is virtually
proportion
of
from
K562
parental
cells
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
PAPAYANNOPOULOU
1768
a
the
CGyAy
E
number
11*1
colonies
differed
3ALA
generated
from the
comparison.
(S)
No Butyrote
Na Butyrate
.
t
but
only
Globin
chain
Commitment
The
duced
GyAy
isofocusing
of 3H-leucine.-Iabeled
K562(S)
Maturation
that
late
higher
a certain
terminal
maturation.
ment
to terminal
enhanced
proportion
hemoglobin
under
as
To test
maturation
conditions
of HEL
if they
whether
can
of clonal
produce
colonies
cells
tends
the unincells sugto accumu-
spontaneously
undergo
spontaneous
commitalso
be observed
or
growth,
we evaluated
from
were
HEL
with
maturation
from MC
lysates
globin
K562
3ALA
+
l-IEL-R
BALA
+
+
BALA
K562(S)
cells,
but colonies
In K562
a reddish
Cord Blood
with
plates,
color
and
Colonies
the rest
of
of
compact
colonies
cob(data
assessed
by
the
benzidine-positive
there
many
with
were
more
apparent
rare,
small
colonies
with
spontaneous
of reddish
color)
were plucked
with tritiated
leucine,
and their
(see the previous
section).
Of note,
present
in any of these spontaneously
whether
ration
tested
commitment
preliminary
exposure
experiments,
to the inducer
we have
followed
decision.
and
Parallel
were again
colonies.
To
differentiation/matu-
by inducers,
assay
as
absence
of the inducer
are adequate
conditions
ment
no 13 chains
maturing
to terminal
can be influenced
in a commitment
selective
described
inducers
were
before.29
In
determined
by plating
by evaluation
to detect
any
experiments
that 24-hour
in MC in the
of colonies
at day
influence
in commitwith
K562
were also carried
out. The data
3. An increase
in the uniformly
and in the mixed
benzidine-positive
cells
under
obtained
are
benzidinecolonies
in K562
cells by these
inducers
was confirmed
with these
experiments.
However,
in HEL cells, at best, partial
commitment
by the same
inducers,
as judged
by the assay
used,
could
be observed.
colonies
in
Of interest,
HEL
that
virtually
cells
were
part
of the
only
of
all benzidine-positive
the
mixed
progeny
type,
becomes
which
terminally
committed.
ofErythroid
Surface
been previously
treatment
of
Table
3.
presented.3
HEL
cells
Induction
Antigens
Cell Line
K562(S)
No Inducer
membrane
with K562
antigenic
cells has
It was also shown previously
with
I 2-0-tetradecanoylphorbol-
of Commitment
Benzidine+
Fig 5.
Isoelectric
focusing
of uninduced
end induced
Hleucine-labeled
cell lysates from HEL and HEI-R.
The features
that
distinguish
this clone from the parental
line are the presence
of
high levels of
chain. the increased
levels of a chains.
and its
higher overall heme and globin inducibility.
colonies,
type,
in morphology
Subcloning
was
A detailed
characterization
of the
profile
of HEL cells and its comparison
+ BALA
round
of this
were used for globin
chain
isoelectric
focusing.
The
phenotype
of these spontaneously
maturing
HEL
or
colonies
was similar
to the ones obtained
in subclones
Induction
HEL
and
for
and by benzidine
positivity
of
no fully hemogbobinized
(red)
(any appearance
plates
and labeled
indicated
HEL
characteristic
cells cultured
reminiscent
colonies.
maturation
positivity.
were
of
and
were
assay.
and colonies
The morphology
to generate
mostly
true for noncompact
color
were
up to 4%.
these conditions
shown
in Table
positive
colonies
HEL-’92
and
Spontaneous
colonies
test
presence
of y globin-positive
cells among
HEL
cells and the rare benzidine-positive
gests
compact
of HEL
of a red
in situ. There
benzidine
induced
by h-ALA and sodium
butyrate
and
by 6-ALA.
Note the difference
in
chain
K562(S)
and C-i 6 induced
by the same
the a chain increase
by -AIA
in both K562
uninduced
cells in Fig 3).
to Terminal
cells
third
shown).
colonies
and K562 (C-i 6) cells
of HEL cells induced
expression
between
inducer
(butyrate)
and
and HEL (compare
with
K562
appearance
colonies
cells
3
one
cells was
in K562
compact
colonies
tended
nies, and the same was
not
84L4
CBf
a
in a clonal
from
HEL
ones observed
being diffused
or scattered
human
granulocytic-macrophage
Is
C-16{
HEL{
clones
HEL
cells were
cloned
in MC media,
evaluated
six to ten days
postplating.
I
Fig 4.
of benzidine-positive
ET AL
Colonies
to Terminal
(%) After
a 24-h
Maturation
Exposure
to
6-ALA
(0.5 mmol/L)
Sodium Butyrate
(0.5 mmol/L)
Are-C
(5 x iO
mol/L)
29.5
32.8
52.0
46.5
HEL-R
6.8
31.5
20.8
12.0
HEL-92
2.9
11.6
13.6
#{149}Meanof two
that
replicate
MC plates
counted
5.0
on day 5 postplating.
5
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
ANALYSIS
OF
1 3-acetate
THE
ERYTHROID
(TPA)
reduced
PHENOTYPE
the
OF
abundance
erythroid
markers
(ie, glycophorin,
same
time,
the cells changed
their
noninducible.4
induction
In
in HEL
surface
erythroid
HEL
this
study,
we
cells
was
accompanied
of
gbobin)
behavior
some
while,
and
examined
1769
CELLS
Anti-I
of the
BB-2
at the
became
whether
globin
by changes
in their
antigens.
Glycophorin.
Uninduced
and
s-ALA-induced
HEL
cells were
treated
with
two MoAbs
against
glycophorin,
R-lO,
and F-i 1 and then with antimouse
F(ab’)2-IgG
(or
1gM)
conjugated
Before
to FITC
induction,
positive
with
R-i0.
shift
fluorescence.
Although
of the
with
R-lO,
cells;
was
great.
change
highly
a very
however,
low
after
observed
of
the
mag-
to findings
with
reactive
in
intensity
reproducible,
intensity
6).
were
resulted
greater
In contrast
was
although
shows
uninduced
not
(Fig
all cells
by b-ALA
toward
shift
was
latter,
to analysis
virtually
induction
profile
this
change
erythroblasts,3
toward
changes
changes
of the
a significant
F-i 1 . The
subjected
noted,3
However,
a rightward
nitude
and
as previously
antibody
with
normal
of fluorescence
induction,
in
a significant
shift
higher
fluorescence
intensity
is found (Fig 6). Similar
were
produced
with
hemin
induction,
but
the
were greater
and more
reproducible
with b-ALA
Fig 7.
Enhancement
of expression
of I antigen
in HEL cells
after
induction
by butyrate
and BrDU.
Both inducers
yielded
a
weak induction
of HEL cells. and the isolated
I antigen
increase
is
partly attributable
to cessation
of cell proliferation
as indicated
by
control
nondividing
cells (HEI cells with antitransferrin)
and by
parallel
changes
in another
antigen
(present
in all myeloid
and B
lymphoid
cells) detected
by antibody
BB-2 (x- and y-axes are as in
Fig 6).
induction.
il-antigens.
occasional
antigenic
HEL
with
hemin,
two
inducers
b-ALA,
were
on I antigen
caused
cells
expressed
i antigen,
cell reacted
with anti I serum.5
The
expression
was tested
after induction
sodium
chosen
expression
only
minor
butyrate,
because
in K562
changes
and
of their
cells.32
BrDU.
Hemin
only
an
of ii
cells
The
reported
in I expression.
and butyrate
increased
the expression
Because
these
two agents,
ie, sodium
and
behavior
of HEL
last
influence
and
However,
b-ALA
BrDU
of I antigen
(Fig 7).
butyrate
and BrDU,
cause a significant
at the doses
used,
proliferative
status
reduction
in the proliferation
of the cells
we examined
whether
changes
in the
of the cells could
be responsible
for the
effect.
for
To
control
proliferation
inhibition,
we
cultured
HEL cells in media
containing
antitransferrin
for six days.
Such
cells,
as independent
experiments
showed,
failed
to
proliferate,
but they remained
viable
for that period
of time.
As seen
an
in Fig
increased
likely
that,
at least
Anti-EP-1
HEL
of I antigen.
in part,
the
cells
Therefore,
increase
showed
it is very
in I antigen
expres-
sion after
sodium
butyrate
and BrDU
is due to changes
in
their proliferative
status,
although
a direct
induction,
especially
by BrDU
(Fig 7), cannot
be entirely
excluded.
The
interpretation
of these changes
is somewhat
supported
by the
information
that anti-I-positive
K562 cells (separated
by a
cell sorter)
showed
with anti-I-negative
of interest
Anti-Glycophorin
7, antitransferrin-treated
expression
that
dramatically
enhanced
the
another
unrelated
increased
upon
addition
with
the cessation
labeled
The
with
Ulex europaeus
great
anti-H
time compared
In addition,
it is
antigen,
butyrate
BB-2,
which
treatments,
of proliferation
of antitransferrin
H antigen.
tively
an increased
doubling
cells (data
not shown).
is
is also
of HEL
cells
by
or TPA.4
majority
MoAb
I. No significant
of HEL
BE225
changes
and
were
cells
with
are
posi-
the
lectin
observed
with
induction.
Other
described
antibodies
of normal
antibodies
increased
Fig 6.
HEL cells labeled with antiglycophorin
MoAbs
(R-iO and
F-i 1) before
and after
induction
by #{212}-ALA.Before
induction
significant
reactivity
with MoAb R-i 0 and much less so with F-il
was observed.
After
induction.
however,
reactivity
with
both
MoAbs,
especially
with F-i 1 #{149}was increased.
as indicated
by a
rightward
shift in fluorescence
intensity.
Enhancement
of expression of Ep-i antigen
in HEL end K562 cells after #{212}-ALA
induction.
Dotted
peaks are negative
control
samples
labeled with an irrelevent first antibody
of the same idiotype
as Ep-i and with the same
second antibody
(x-axis.
relative
cell number;
y-axis. fluorescence
intensity).
erythroid
surface
antigens.
We have
recently
the characterization
of two antierythroid-specific
(anti-EP-1
and anti-EP-2)
raised through
the use
erythroid
cells
as immunogens.24
Using
these
in K562 cells, we observed
that their reactivity
is
after gbobin induction.24
Similar
experiments
with
b-ALA-induced
ment can also
6). Furthermore,
HEL
cells
with
Augmentation
first three days
HEL
cells have shown
be observed
in HEL cells
we
have
tested
the
that
after
reactivity
a set of four
new antierythroid
in reactivity
of induction
was
(data
Transferrin
receptor.
The
iron in the developing
erythroid
of the most active
hemoglobin
again
observed
not shown).
this enhanceinduction
(Fig
of induced
monocbonals.28
during
the
increase
in transferrin-bound
cell correlates
with the stages
synthesis.
We have examined
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
PAPAYANNOPOULOU
1770
the
reactivity
with
two
BE/25).
sion
or
b-ALA
of uninduced
and
antitransferrin
The
data
indicate
the
density
induction
of
cells’ proliferation
that of control
cells,
reactivity
reflect
the cells
that
rate
during
the
some
K562
or K562
cells
(OKT9
and
increase
in the
receptors
occurs
or HEL
(doubling
increase
an
HEL
antibodies
transferrin
of either
may
induced
receptor
cells.
expresafter
Because
in the
traffic
of iron
into
induction.
HEL
of the basal
cells
interest.
and
performed
Among
inducible
in this
several
erythroid
study
disclosed
inducers
used,
b-ALA
was surprising.
Augmentation
sis after b-ALA
addition
was observed
initial
effects
on the
proliferative
phenotype
several
the
points
and
of
syntheof any
cell
viability.
that
cells.
remarkable
This
cell
lines
b-ALA
to b-ALA
was
also
a potent
responsiveness
addition
inducer
of these
may
indicate
that
in K562
two
human
in both
lines
about
small number
present
study
only
by a MoAb
(kindly
but
readily
detected
suggest
that
from
provided
those
as implied
b mRNA
either
the
required
for
by
these
between
90%)
tions,
not
accompanied
by both HEL
the proportion
c-positive
population
virtually
K562(S)-support
previous
evidence
ies, showed
that
embryonic
Subsequent
ing
the
(25%)
all
clones
and
through
globin
virtual
globins
of
earlier
(y
and
development.36
cells coexpress-
understanding
regulated
and
of control
provides
or that
to
different
cis inhibition
with
in the
few
specific
of
exceptypes
of
were
noted
(ie,
these were shared
noted variations
in
or globins
in K562
reconciled
by the findings
presented
in a and t production
by individual
fact
that
some
of
chains
over
is augmented
a cellular
mechanisms.
does
not
the
inducers
the rest.
after induction
expression
as it occurs
take
place.
favor
the
of HEL
and
of the
in murine
Thus,
terminal
erythro-
of normal
nucleated
erythroid
cells.6
Nevertheless,
despite
the absence
of a coordinated
expression
of the entire
erythroid
maturation
program
upon induction,
a few nonglobin
changes
were noted, and these most likely are indicative
of an
attempt
form
for further
presumably
6A), and
recognized
increase
BrDU
in I antigen
and butyrate
cells have
increased
maturation;
more
the expression
by MoAbs,
enriched
of some
is enhanced
thus,
glycophorin
in carbohydrate
attains
groups
a
(Fig
erythroid-specific
antigens,
(Fig 6B). In addition,
an
expression
was noted
(Fig 7), as previous
with two inducers,
studies
with K562
pointed
out.32 It is of interest
that, in our studies,
I expression
was also correlated
with decreased
proliferative
activity
and
a decreased
number
of divisions
in
both HEL
and K562
cells.
However,
the relevance
of this
finding
to physiological
increments
in I antigen
expression
during
erythroblast
maturation
is unclear.
realized
leukemic
transformation.
Furthermore,
the
of HEL-R
expressing
high levels of t in the
of globin
suggests
that the two embryonic
independently
for further
phenotypes
are
the
or no
globin
do exist during
to fetal
erythropoie-
developmental
seem
out in
protein
profile
this notion.
This is of interest
in view of
which,
based on gbobin biosynthetic
studglobin
expression
is confined
only to the
during
normal
indicated
that
studies39’42
minimum
coexpression
of fetal and embryonic
gbobins
in
K562 cells may not be necessarily
aberrant
but a
their
phenotype
absence
are
the double
in ‘y gbobin-HEL-R
fetal (-y) and embryonic
(t and )
transitional
period
of embryonic
reflection
cells,
cells,
of b
and
in these cells.
synthesis
was,
and K562 cells. Previously
of various
hemoglobins
the
cells,
leukemic
in K562
b chains,22
changes
in nonglobin
proteins
have
been
pointed
previous
studies
of K562
cells,45
and the overall
profile
of induced
cells does not resemble
the protein
in the same
cells (data
not
production
of
gbobin
by a
positive
cell population
studies,
however,
sis.37.38 Thus,
HEL-R
and
model
cells
labeling
experiments
and the significant
is
studies,
for b expression
by changes
cells#{176}’”can be easily
here, ie, the heterogeneity
and
of gbobin chains
and
that, in contrast
to the
expression
of embryonic
The
globin.
Some
consistent
trends,
however,
preferential
increase
in a or e chains),
and
and
) is restricted
to a subpopulation
of cells. This
however,
does not seem to synthesize
exclusively
globins
and seems to overlap
with cells expressing
The numerical
relationships
of rny-positive cells (over
and
and ‘)
shown),
human
synthesis
and
by these
expression43
i3 globin
may be operative
Induction
of hemoglobin
effects.
Studies
on the cellular
distribution
the subcloning
experiments
suggested
widespread
expression
of ‘y gbobin,
globins
population,
embryonic
.‘ygbobin.
the
with
previous
factors
maturation
pathway
underlie
both
by Dr Jensen).4’
not 13 globin,
within
synthetic
A
of b
an
presence
used
a well-coordinated
protein
program,
may
HEL
was
K562
heme
in both
in which
an
with b chains
of some
globin
lines35
hemoglobin
reacting
production
Although
the
absent
a small number
and,
in another,
of /3-positive
cells found
in K562
is of interest.
These
cells were
early steps in heme biosynthesis
are rate limiting.
Furthermore,
because
b-ALA
is not among
the potent
inducers
of
murine
erythroleukemia
cells,33’34
biochemical
differences
murine
is virtually
and they were not found
in subsequent
studies
anti-fl
chain
antibody
that does not crossreact
of
effectiveness
in hemoglobin
in the absence
capacity
of
Other
inducers
showed
fewer effects
on the augmentation
of
hemoglobin
synthesis,
and this was accompanied
by more
toxicity
(cessation
of proliferation
and cell viability).
It was
of interest
detected
globins
although,
in one study,
copies
were detected39
claim
made.#{176}The
cells in the
globin
DISCUSSION
Analysis
of adult
K562 cells
mRNA
unsubstantiated
the
times)
early on is similar
to
in antitransferrin
receptor
increase
Expression
and
and
ET AL
ACKNOWLEDGMENT
We are grateful
to Dr David Chui for his gift of anti-c antibody
and to Dr R.H. Jensen for anti-a monoclonal
antibody.
We also
thank Drs M. Fukuda,
G. Rovera,
and J. Hansen
for the original
donation
of various
K562 lines and Doreen
Loomans
for skillful
secretarial
assistance.
REFERENCES
1 . Martin
P, Papayannopoulou
Th: HEL
erythroleukemia
cell line with spontaneous
expression.
Science 216:1233,
1982
cells: A new
and induced
human
globin
2. Rutherford
leukaemic
cells
haemin. Nature
TR, Clegg
JB, Weatherall
Di:
synthesize
embryonic
haemoglobin
208:164,
1979
K562 human
in response
to
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
ANALYSIS
OF THE
ERYTHROID
PHENOTYPE
1771
OF HEL CELLS
3. Papayannopoulou
Th, Yokochi T, Nakamoto
B, Martin P: The
surface
antigen
profile of HEL cells, in Stamatoyannopoubos
G,
Nienhuis
AW (eds): Globin Gene Expression
and Hematopoietic
Differentiation.
New York, Liss, 1983, p 277
4. Papayannopoulou
Th, Nakamoto
B, Yokochi
T, Chait A,
Kannagi
R: Human
erythroleukemia
cell line (HEL)
undergoes
a
7. Frankel SR. Walloch J, Roli KH, Bondurant
MC, Villanueva
R, Weil SC: Carbonic
anhydrase
is aberrantly
and constitutively
expressed
in both human and murine
erythroleukemia
cells. Proc
tially in culture.
I. Studies
in cultures
of adult peripheral
blood
BRie.
Blood 58:969, 1981
21. Tsapis A, Hinard
N, Testa J, Dubart
A, Vainchenker
W,
Rouyer-Fessard
Ph, Beuzard
T, Rosa J: Globin
chain
affinity
chromatography
on Sepharose-haptoglobin:
A new method of study
of hemoglobin
synthesis
in reticulocytes,
in bone marrow,
and in
colonies of erythroid
precursors.
Eur J Biochem
1 12:513, 1980
22. Stamatoyannopoulos
G, Farquhar
M, Lindsley
D, Brice M,
Papayannopoulou
Th, Nute PE: Monoclonal
antibodies
specific for
gbobin chains. Blood 61:530, 1983
23. Chung
SW, Wong SC, Clarke
BJ, Patterson
M, Walker
WHC,
Chui DHK:
Human
embryonic
-globin
chains
in adult
patients
with a-thalassemia.
Proc Natl Acad Sci USA 81:6188,
1984
24. Yokochi T, Brice M, Rabinovitch
PS, Papayannopoulou
Th,
Stamatoyannopoulos
G: Monoclonal
antibodies
detecting
antigenic
determinants
with restricted
expression
on erythroid
cells: From the
erythroid-committed
progenitor
level to the mature
erythroblast.
Natl Acad
B1ood63:1376,
drastic
macrophage-like
shift
with
TPA.
Blood
62:832,
1982
5. Kannagi
R, Papayannopoulou
Th, Nakamoto
B, Cochran
T,
Yokochi
T, Stamatoyannopoulos
G, Hakomori
S: Carbohydrate
antigen profiles of human erythroleukemia
cell lines HEL and K562.
Blood62:1230,
1983
6. Larson
AE, Papayannopoubou
Th: HEL and K562 cells:
Analysis
of proteins
by two-dimensional
gel electrophoresis
and
comparison
to erythroid
and nonerythroid
cells. Exp Hematol
13:129, 1985
Sci USA
82:5175,
1985
8. Drexler HG, Gaedicke
G, Minowada
J: Erythroleukemia
cell
lines HEL and K562: Changes
in isoenzyme
profiles and morphology during induction
of differentiation.
Hematol
Oncol 4:163, 1986
9. Tabilio A, Rosa JP, Testa U, Kieffer N, Nurden
AT, DelCanizo MC, Breton-Gorius
J, Vainchenker
W: Expression
of platelet
membrane
glycoproteins
and a-granule
proteins
by a human crythroleukemia
cell line (HEL).
Eur Mol Biol Organ 3:453, 1984
10. Papayannopoubou
Th, Raines
E, Collins
5, Nakamoto
B,
Tweeddale
M, Ross R: Constitutive
and inducible
secretion
of
platelet-derived
growth factor analogs by human leukemic cell lines
coexpressing
erythroid
and megakaryocytic
markers.
J Clin Invest
79:859,
1987
1 1. Poncz M, Surrey S, LaRocco
P, Weiss MJ, Rappaport
EF,
Conway TM, Schwartz
E: Cloning and characterization
of platelet
factor 4 cDNA
derived
from a human erythroleukemia
cell line.
Blood 69:219, 1987
12. Lopez JA, Chung DW, Fujikawa
K, Hagen FS, Papayannopoubou Th, Roth GJ: Cloning of the alpha-chain
of human platelet
glycoprotein
Ib: A transmembrane
protein with homology
to leucine
rich a2-glycoprotein.
Proc Natl Acad Sci USA 84:561 5, 1987
13. Fukuda M: K562 human leukaemic
cells express fetal type (i)
antigen
on different
glycoproteins
from circulating
erythrocytes.
Nature
285:405,
1980
14. Cioe L, McNab A, Hubbell
HR. Moo P. Curtis P. Rovera G:
Differential
expression
of the gbobin genes in human
leukemia
K562(S)
cells induced
to differentiate
by hemin or butyric
acid.
Cancer Res 41:237, 1981
15. Orkin S, Swan D, Leder P: Differentiation
in erythroleukemia cells and their somatic
hybrids.
Proc Natl Acad Sci USA
72:98, 1975
16. McLeod
plasma
culture
vitro: Quantitative
DL, Shreeve
MM, and Axelrad
AA: Improved
system
for production
of erythrocytic
colonies
in
assay method for CFUe. Blood 44:517, 1974
17. Rutherford
TR, Weatherall
DJ: Deficient
heme
the cause of noninducibility
of hemoglobin
synthesis
erythroleukemia
cell line. Cell 16:415, 1979
synthesis
as
in a Friend
18. Basset P. Beuzard
Y, Garel MC, Rosa J: Isoelectric
focusing
of human hemoglobin:
Its application
to screening,
to the characterization of 70 variants,
and to the study of modified
fractions
of
normal hemoglobins.
Blood 51:971, 1978
19. Righetti
PG. Gianazza
E, Gianni AM, Comi
Ottolenghi
5, Secchi C, Rossi-Bernardi
L: Human
separation
by isoelectric
focusing.
J Biochem Biophys
1979
20. Papayannopoulou
Th, Nakamoto
nopoulos G: Globin synthesis
in erythroid
P. Giglioni
B,
globin chain
Methodsl:45,
B, Kurachi
5, Stamatoyanbursts that mature sequen-
1984
25. Young WW, Portoukalian
J, Hakomori,
SI: Two monocbonal
anticarbohydrate
antibodies
directed
to glycosphingolipids
with a
lacto-N-glycosyl
type II chain. J Biol Chem 256:10967,
1981
26. Trowbridge
IS, Omary MB: Human cell surface glycoprotein
related to cell proliferation
is the receptor
for transferrin.
Proc NatI
Acad Sci USA 78:3039,
1981
27. Omary
MB, Trowbridge
IS, Minowada
J: Human
cellsurface glycoprotein
with unusual properties.
Nature 286:888,
1980
28. Cotner T, Das Gupta A, Papayannopoulou
Th, Stamatoyannopoulos G: Characterization
of a novel form of transferrin
receptor
preferentially
found in normal erythroid
cells. Blood (submitted)
29. Rowley PT, Ohlsson-Wilhelm
BM, Farley BA: K562 human
erythroleukemia
cells demonstrate
commitment.
Blood 65:862,
1985
30. Gusella
J, Geller
R, Clarke
B, Weeks
V. Housman
D:
Commitment
to erythroid
differentiation
by Friend erythroleukemia
cells:
A stochastic
analysis.
Cell
9:221,
1976
3 1 . Rowley PT, Ohlsson-Wilhelm
BM, Rudolph
NS, Farley BA,
Kosciolek B, LaBella 5: Hemin preferentially
stimulates
synthesis
of
alpha-globin
in K562 human erythroleukemia
cells. Blood 59:1098,
1982
32.
Tonthat
antigen
33.
Testa
U,
Henri
A,
Bettaib
A,
Titeux
M,
Vainchenker
W,
H, Docklear
MC, Rochant
H: Regulation
of i- and Iexpression
in the K562 cell line. Cancer Res 42:4694,
1982
Sassa
S: Control
of heme
biosynthesis
in erythroid
cells,
in
Rossi GB (ed):
In Vivo and In Vitro Erythropoiesis:
The Friend
System. Amsterdam,
Elsevier North Holland,
1980, p 219
34. Laskey J, Ponka P, Schulman
HM: Control of heme synthesis
during Friend cell differentiation:
Role of iron and transferrin.
J Cell
Physiol 129:185, 1986
35.
Chang
CS,
erythroleukemic
65:939,
Sassa
cells:
5: b-Aminolevulinate
An
dehydratase
immunologically
distinct
in human
enzyme.
Blood
1985
36. Peschle C, Mavilio F, Cave A, Migliaccio
G, Migliaccio
AR,
Salvo G, Samoggia
P. Petti 5, Guerriero
R, Marinucci
M, Lazzaro
D, Russo G, Mastrobenardino
G: Haemogbobin
switching
in human
embryos:
Asynchrony
of - a and
-‘
y switches
in primitive
and
definitive
erythropoietic
lineage. Nature 3 13:235, 1985
37.
Yagi
M,
Gelinas
R, Elder
JT,
Peretz
M,
Papayannopoubou
Tb, Stamatoyannopoulos
G, Groudine
M: Chromatin
structure
and
developmental
expression
of the human a-globin
cluster.
Mol Cell
Biol6:1l08,
1986
38. Stamatoyannopoulos
G, Constantoulakis
P. Brice M, Kurachi 5, Papayannopoulou
Th: Coexpression
of embryonic,
fetal, and
adult gbobins in erythroid
cells of human embryos:
Relevance
to the
cell-lineage
models ofglobin
switching.
Dcv Biol 123:191, 1987
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
PAPAYANNOPOULOU
1772
39. Dean A, Ley Ti, Humphries
RK, Fordis M, Schechter
AN:
Inducible
transcription
of five globin genes in K562 human leukemia
cells. Proc Natl Acad Sci USA 80:55 15, 1983
40. Lyon SB, Buonocore
L, Miller M: Naturally
occurring
meth-
Hemenway
T, Melis M, Anagnou
NP, Najfeld
V: Adult
human globin genes are expressed
following
chromosomal
into MEL cells. Proc Natl Acad Sci USA 82:780, 1985
ylation
CB, Lozzio
BB: K562 human
embryonic
globin gene expression.
inhibitor:
DNA
hypomethylation
and
hemoglobin
synthesis
in human K562 cells. Mob Cell Biol 5:1759, 1987
41 . Stanker
LH, Branscomb
G, Vanderloom
M, Jensen
RH:
Monocbonal
antibodies
recognizing
single amino acid substitution
in
hemoglobin.
i Immunol
136:4174,
1986
42. Enver T, Zhang
i, Anagnou
NP, Stamatoyannopoulos
G,
Papayannopoubou
Th: Diverse patterns
of DNA methylation
and
globin gene expression
in several human erythroleukemic
cell lines:
Implications
for gbobin
Gene Dcv (submitted)
43. Papayannopoubou
gene
Th,
switching
Lindsley
and
stem
cell
D, Kurachi
differentiation.
5, Lewison
K,
44.
Rowley
PT,
Ohlsson-Wilhelm
BM,
Wisniewski
leukemia
cell passages
Leuk Res 8:45, 1984
ET AL
and fetal
transfer
L,
Lozzio
differ
in
45. Horton,
MA, Cedar SH, Edwards
PAW: Expression
of red
cell specific determinant
during differentiation
in the K562 erythroleukemia
cell line. Scand J Haematol
27:23 1 , 1981
46.
Fukuda M, Fukuda MN, Papayannopoulou
Th, Hakomori
5:
Membrane
differentiation
in human
erythroid
cells: The unique
profile of cell surface
glycoproteins
expressed
in erythroblasts
in
vitro from three ontogenic
stages. Proc Natl Acad Sci USA 77:3474,
1980
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
1987 70: 1764-1772
Analysis of the erythroid phenotype of HEL cells: clonal variation and the
effect of inducers
T Papayannopoulou, B Nakamoto, S Kurachi and R Nelson
Updated information and services can be found at:
http://www.bloodjournal.org/content/70/6/1764.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.