1. Art and Diseño

Fetal Calf Serum Contains Activities That Induce Fetal Hemoglobin in Adult
Erythroid Cell Cultures
By P. Constantoulakis, B. Nakamoto, Th. Papayannopoulou, and G. Stamatoyannopoulos
Cultures of peripheral blood or bone marrow erythroid
progenitorsdisplay stimulated production of fetal hemoglobin. We investigated whether this stimulation is due to
factors contained in the sera of the culture medium.
Comparisons of y/y
@ biosynthetic ratios in erythroid
colonies grown in fetal calf serum (FCS) or in charcoal
treated FCS IC-FCS) showed that FCS-grown cells had
significantly higher y/y
@ ratios. This increase in globin
chain biosynthesiswas reflected by an increase in relative
amounts of steady-state y-globin mRNA. In contrast to its
effect on adult cells, FCS failed to influence 7-chain
synthesis in fetal burst forming units-erythroid (BFU-E)
colonies. There was a high correlation of y-globin expression in paired cultures done with C-FCS or fetal sheep
serum. Dose-responseexperiments showed that the induction of 7-globin expression is dependent on the concentration of FCS. These results indicate that FCS contains an
activity that induces y-globin expression in adult erythroid
progenitor cell cultures.
0 1990 by The American Society of Hematology.
T
activities that can induce H b F production in cultures of adult
erythroid progenitors.
From the Divisions of Medical Genetics and Hematology, Department of Medicine. University of Washington. Seattle.
Submitted September 25,1989; accepted January 3,1990.
Supported by Grant No. HL 20899 from the National Institutes
of Health, Bethesda. MD.
Address reprint requests to G. Stamatoyannopoulos, MD, Dr Sei.
Professor of Medicine. Medical Genetics, RG-25. University of
Washington. Seattle. WA 98195.
The publication costs of this article were defrayed in part by page
charge payment. This article must therefore be hereby marked
“advertisement”in accordance with 18 U.S.C.section 1734 solely to
indicate this fact.
0 1990 by The American Society of Hematology.
0006-4971/90/7509-0113$3.00/0
Cells and cultures. Peripheral blood was obtained from hematologically normal adult volunteers and people with sickle cell anemia,
&thalassemia, hereditary persistence of fetal hemoglobin (HPFH),
or other conditions (Diamond-Blackfan,juvenile chronic myeloid
leukemia (JCML), hemoglobin H (HbH), a-thalassemia, etc).
Patients with @-thalassemiawere homozygotes of varying degrees of
severity (polytransfused cases of Cooley’s anemia or thalassemia
intermedia). The blood was collected with phlebotomy in sterile
preservative-free heparin. Mononuclear cells were obtained by
centrifugation of the blood over Hypaque-Ficoll (Lymphoprep,
Nyegaard, Norway), washed in phosphate-buffered saline (PBS),
and incubated in a-media containing 5% FCS in plastic Petri dishes
for 1 hour at 37OC, in order to remove adherent cells (monocytes and
macrophages). After low speed centrifugation, the cells were resuspended in Iscove’s modified Dulbecco’s medium (IMDM) with 5%
FCS and used for cultures (see below).
Fetal liver and yolk sac samples were obtained from dead
abortuses after maternal consent. The tissues were rinsed well, finely
minced and the cells were mechanically dispersed by vigorous
pipeting, counted, and used for culture. Bone marrow was aspirated
into heparinized syringes from the iliac crest of the donors. Bone
marrows from normal non-hemoglobinopathic persons were mainly
obtained from donors of the bone marrow transplantation program.
Bone marrow from sickle cell patients was obtained during steady
state and also during treatment with cytosine arabinoside(Ara-C) or
hydroxyurea or recombinant erythropoietin.Separation of mononuclear cells was done as described above. In fetal liver, yolk sac, or
bone marrow experiments,total mononuclear cells (without removal
of adherent cells) was used for culture.
After cell counting, the cell suspensions were adjusted to suitable
plating concentrations. Cells were inoculated (2 to 5 x io5 cells/
mL) in either plasma clot or methylcellulose. When plasma clotsz2
were used, the following components were employed: beef embryo
extract (BEE; Grand Island Biological Co, Grand Island, NY) 10%;
bovine serum albumin (BSA; Intergen, Purchase, NY) 10%;
2mercaptoethanol (Sigma Chemical Co, St Louis, MO)
mol/L; FCS (Rehatuin; Intergen) 30%; FSS or plasma (collected
from fetuses between 80 and 120 days of gestation) 30%; human AB
serum 10%;recombinant human erythropoietin (Genetics Institute,
Cambridge, MA) 0.2 U/mL; and bovine citrated plasma (BCP;
Irvine Scientific, Santa Anna, CA) 10%. In the methylcellulose
cult~res,’~
the
, ~ following
~
components were used: methylcellulose
(Fisher Scientific, Fair Lawn, NJ) 0.9%; human AB plasma 10%;
erythropoietin 2.0 U/mL; FCS, BSA, and 2-mercaptoethanol,all in
concentrationsas above.
Five differentlots of FCS from two vendors were used during the
+
+
HE INDUCTION OF fetal hemoglobin (HbF) in adult
erythroid cell cultures was first documented when
erythroid colonies were analyzed for globin chain expression,
either with fluorescent antibodies or by globin chain
biosynthesis.’ Several studies subsequently described the
activation of H b F in culture,’-I4 but there have also been
discrepancies in experimental results.2.3 One of the initial
interpretations of such discrepancies was that the culture
media may differ in their capacity to stimulate HbF.” We
have attributed the activation of H b F in culture to a
“deficient” culture environment, which leads to a high
incidence of premature commitment of erythroid progenitors.I6 Terasawa et al” and Ogawa’*suggested that H b F
was induced by the burst promoting activity present in the
culture medium. In contrast to cultures done with fetal calf
serum (FCS), there is low y-globin expression in cultures
done with fetal sheep serum (FSS).’9*20Rosenblum et a1
reported that y-globin expression in culture can also be
substantially reduced when the serum is treated with activated charcoal.*’
In the studies reported here we examine the effect of sera
on fetal globin expression. We observed a stimulation of H b F
production in FCS-containing cultures of adult erythroid
progenitors. This stimulation is reflected by higher y-globin
m R N A accumulation in FCS cultures. Treatment of FCS
with activated charcoal results in a consistent decrease of
y/y /3 ratios in peripheral blood or bone marrow erythroid
colonies, but not in fetal burst forming unit-erythroid (BFUE) colonies. Furthermore, dose-response experiments show
that the induction of y-globin expression by FCS is concentration-dependent. These results indicate that FCS contains
+
1862
MATERIALS AND METHODS
Blood, Vol 75, No 9 (May 1). 1990: pp 1862-1869
1863
HBF-INDUCING ACTIVITY
6-year period of this study. The FCS was used either untreated or
after treatment (twice for 30 minutes each) with activated charcoal
(Norit “A”; Fisher Scientific) at a concentration of charcoal 10
mg/mL FCS. Thecharcoal was removed by centrifugation (lO,OOOg),
and the FCS was sterilized by filtration (0.22 pm filter; Millipore,
Bedford, MA). In the dialysis studies, both FCS or charcoal-treated
FCS (C-FCS), were dialyzed against PBS at 4OC for 12 to 16 hours,
using Spectra/Por tubing membranes (Spectrum Medical Industries
Inc, Los Angeles, CA) in various molecular weight cut off pore sizes.
In several experiments, increasing concentrations of FCS were
added on top of 30% C-FCS or FSS used in the culture media. The
final volume of media per culture plate was kept at 1.15 mL by
adjusting the concentrations of the stock solutions of BSA and
methylcellulose and by suspending the cell pellet in serum (C-FCS
or FSS) instead of IMDM.
Cultures were incubated in a highly humidified 37OC incubator
and continuously flushed with 5% CO,. Erythroid colony forming
units (CFU-E, 8 to 64 mature erythroblasts) were evaluated in
benzidine-hematoxylin stained plasma clots at day 3 to 5 in culture.
BFU-E colonies (usually greater than 100 and up to several
thousand cells per colony) were evaluated in either plasma clot or
methylcellulose cultures at 10 to 16 days in culture.
Globin synthesis. The methodologies used for globin chain
biosynthesis of culture-derived cells have been described
CFU-E-derived colonies were labeled at day 5 with [3H]-leucine
(Amersham, Arlington Heights, IL) for 12 to 16 hours. At the end of
the incubation period, the plasma clots were washed with normal
saline and the cell pellets were either analyzed immediately or stored
in liquid nitrogen. Globin biosynthesis in BFU-E-derived colonies
was carried out in bursts plucked from methylcellulose cultures,
usually on culture days 14 to 16. Individually lifted bursts were
pooled and incubated with [’HI-leucine (300 to 400 pCi/mL), and
after washing, the cells were either analyzed for globin chains or
stored in liquid nitrogen.
To determine the levels of globin chain synthesis, the plasma
clot-derived colonies were lysed with 10% Nonidet P-40 and 0.01%
KCN for 1 hour on ice and centrifuged (14,OOOg for 2 minutes).
After addition of nonradioactive HbA and HbF carriers, cell lysates
(2 to 4 pL) were mixed with 25 pL of denaturing buffer (8 mol/L
urea, 10% @-mercaptoethanol, and 2% NP-40), and isoelectric
focusing was carried out as previously described.” After focusing,
the gels were fixed in 20% trichloroacetic acid (TCA) in 30% ethanol
for 5 to 6 hours, incubated in EN-3HANCE (New England Nuclear,
Boston, MA) for 1 to 2 hours, dried on a slab gel dryer, and exposed
on preflashed Kodak X-OMAT-VR film. Relative densities of the
globin chain bands (a,@, Gy, and Ay) were determined on a Gelman
automatic computing densitometer.
RNA studies. Total cellular RNA was isolated using the method
of Karlinsey et a1.28 RNAse protection assay was performed as
described by Zinn et aIz9; 1 pg total RNA was used for each
hybridization reaction with y- and @-globinspecific RNA probes.
For the blot hybridization method, the cells were prepared as
follows: dilutions of 50.000, 10,000, and 2,000 cells were washed
twice in PBS, and then the cell pellet was resuspended in cold TE
buffer (10 mmol/L Tris, 0.1 mmol/L EDTA) and 20 units of
RNAse inhibitor was added (RNasin; Promega, Madison, WI). The
cells were then lysed with a 5% NP-40 solution, and after centrifugation, the cytoplasm was transferred to tubes containing 20x SSC
(0.15 mol/L NaCI, 0.015 mol/L trisodium citrate) and 37%
formaldehyde. After denaturing at 65OC for 15 minutes, the sample
was immediately processed for “spot blotting,” or stored at -7OOC
for later analysis.
The “spot blotting” technique of Kafatos et aI3’ was employed
using the Schleicher & Schuell (Keene, NH) Minifold I1 apparatus,
as previously described.” The samples were diluted to a final volume
of 300 pL with 15x SSC and loaded on prewet (15x SSC)
Schleicher & Schuell nitrocellulose paper under vaccum. After
drying, the nitrocellulose was baked in a vacuum oven at 8OoC for 2
hours. After a 3-hour prehybridization in a 65OC water bath, the
blots were hybridized with 50 ng of in vitro transcribed radioactive
(a-”P-uridine triphosphate [UTP]) RNA probes. For detection of
@-globinmRNA, the 0.7 kb Pst I/EcoRI fragment containing the 3’
end of the human genomic @-globingene was subcloned in antisense
orientation in a T7-plasmid vector. For the detection of y-globin
mRNA, the 0.6 kb EcoRIIHindIII fragment of the 3‘ end of the
human Gy-globin gene was subcloned in an Sp6-plasmid vector.
Hybridization was allowed to take place between the homologous
RNA sequences for at least 16 hours, followed by extensive washing
under stringent conditions (2x SSC with 0.1% sodium dodecyl
sulfate [SDS] at 65OC six times, followed by 0 . 1 SSC
~ with 0.1%
SDS at 65OC two times). After exposing the blot on a Kodak
X-OMAT-AR film at -7OoC, the relative amounts of y- and
@-globinmRNAs were calculated by both scanning in a densitometer and by directly measuring the radioactivity (counts/min) of
every slot in a liquid scintillation @-counter.
RESULTS
Induction of y-chain synthesis in peripheral blood BFU-E
cultures. Experiments were designed in a pair-wise fashion
so that within each pair all conditions were identical, except
that in one set of cultures, untreated FCS was used while in
the other set of cultures, C-FCS was used. When charcoaltreated FCS was used in culture, cloning efficiency of normal
peripheral blood BFU-E was 1.57-fold higher (n = 39) and
of normal bone marrow, BFU-E was 1.84-fold higher
(n = 13) compared with FCS controls. Results of globin
biosynthesis are shown in Fig 1.
In normal BFU-E cultures (n = 34), consistently higher
y/y + 0 ratios were found when cells were grown in FCS as
compared with C-FCS. On the average, the y/y + (3 ratios
were 3.6-fold higher in the FCS cultures than in C-FCS
cultures (range, 1.5-fold to 6.5-fold). Similar results were
obtained in cultures of cells from patients with sickle cell
disease; in 20 experiments, there was a 3.2-fold mean
increase in y/y + @ ratios (range, 1.3-fold to 5.7-fold). A
significant induction of y-globin expression in FCS cultures
was also observed in cultures of homozygous @-thalassemia
BFU-E; in 18 experiments, there was a 2.2-fold mean
increase in y/y + @ ratios. Five of six @-thalassemia
heterozygotes displayed an increase in y/y + @ ratios in FCS
cultures (Fig 1). Six cases of HPFH (deletion or non-deletion
mutants) showed an average y/y + B increase of 2.9-fold in
the presence of FCS. The “other conditions” in Fig 1 include
cases of Diamond Blackfan syndrome and HbH disease. The
only condition that consistently failed to respond to FCS by
an increment of y/y + @ ratio was J C M L (data not shown).
Effect of FCS on globin mRNA accumulation. Steady
state of y/y + @ globin mRNA ratios were determined in six
experiments of normal peripheral blood BFU-E cultures. In
one experiment, a n RNAse protection assay was used, while
in five experiments we used a blot hybridization assay. A
twofold to ninefold increase in y/y + @ mRNA ratios was
observed when FCS was used instead of C-FCS in the culture
medium (Table 1). Calculations of the total globin mRNA
counts (@+ y) per cell (from the counts incorporated by
hybridization) showed similar degrees of globin mRNA
1864
CONSTANTOULAKIS ET AL
x
0.4
x
\
+
0.2
Fig 1. Comparisonof y/y fl ratios in peripheral blood BFU-E colonies grown in FCS or C-FCS.
fl ratios from paired cultures of the
The y/y
same cells are connected with a line. Note that in
almost all experiments there are higher 'y/y
fl
ratios
cultures, suggesting that
_ in the FCS-grown _
~
+
+
0
C-FCS
FCS C-FCS
Normal
~
FCS
Sickle Cell
Anemia
C-FCS
FCS
~C-FCS
FCS C-FCS
Homozygous Heterozygous
P-thalassemia B-thalassemia
FCS
_
C-FCS
HPFH
accumulation in the cells grown in FCS or in C-FCS (Table
1).
Effect of FCS on bone marrow erythroid progenitors.
The effect of FCS on y-globin production was also examined
in bone marrow-derived colonies. The y/y + P ratios in FCS
cultures were 2.3-fold higher than in C-FCS cultures in
normal bone marrow BFU-E (range, 1.5-fold to 3.1-fold)
and 1.8-fold higher in CFU-E (range, 1.1-fold to 2.6-fold)
(Fig 2). The increase in y/y + ratio was fourfold in sickle
cell anemia BFU-E (range, 1.9-fold to 8.6-fold) and 2.4-fold
in the CFU-E (range, 1S-fold to 2.9-fold). A similar increase
of y/y + ratios in FCS cultures was also noted in cultures
of bone marrow progenitors of homozygous P-thalassemics
(one experiment), heterozygous P-thalassemics (two experiments), and in six experiments using cells from other
disorders.
Effect of FCS on fetal origin BFU-E. To examine the
effect of FCS on y/y + P ratios of fetal erythroid progenitors, paired cultures of fetal liver or yolk sac origin BFU-E
were done in FCS or C-FCS. As shown in Table 2, in seven
experiments using fetal liver cells and two experiments using
yolk sac cells, there was essentially no difference in y/y + P
ratios between FCS and C-FCS cultures.
Comparison of y/y + /3 ratios in C-FCS-or FSS-grown
erythroid cultures. Paired cultures were done to compare
y/y + P ratios in C-FCS or FSS plasma-grown BFU-E
FCS
Other
Conditions
FCS contains a charcoal-removable activity that
stimulates 7-globin expression.
colonies. Figure 3 presents results using adult peripheral
blood BFU-E (Fig 3A) or adult bone marrow BFU-E (Fig
3B). A significant correlation was obtained in both; correlation coefficients were r = .75 in 40 peripheral blood BFU-E
comparisons and r = .85 in 38 bone marrow BFU-E
comparisons.
The increase in y/y + p ratios by FCS is concentrationdependent. The following experiments were done to test
whether the effect of FCS on the y/y + p ratio is quantitative. First, adult peripheral blood BFU-E cultures were
performed in medium containing increasing concentrations
of FCS (5%, lo%, 15%, 20%, 25%, 30%, and 40%; all
cultures were done in the presence of 10% adult human AB
plasma). Colonies (usually at culture days 14 to 16) were
plucked from the cultures, pooled, and used for globin
biosynthesis. Control cultures were done using increasing
concentrations (from 5% to 40%) of C-FCS. While there was
no change in the y/y + ratios in the C-FCS cultures, a
clear increase was observed in response to the increase of
serum concentration in the FCS cultures (Fig 4).
Since there are differences in growth between cultures
done with low or with high serum concentrations, we repeated the experiments using conditions that minimized
differences in progenitor cell growth. Increasing concentrations of FCS (from 0% to 30%) were added to media that
already contained 30% C-FCS or 30% FSS. Peripheral blood
Table 1. Globin mRNA Accumulation in Peripheral Blood BFU-E Cultures
C-FCS
Experiment
y-mRNA.
(c/min/cell)
P-mRNA*
(c/min/cell)
1
2
3
4
5
207
180
308
106
140
6,789
6,210
29,348
17,654
6,860
Abbreviation: c, counts.
*Means of measurements of triplicate slots.
FCS
y
+ B mRNA
r/y
+
Ic/min/celll
Ratio
y-mRNA*
(c/min/cell)
6,996
6,390
29,656
17,760
7,000
0.02
0.03
0.01
0.006
0.02
1,044
1,020
2,620
408
259
P-mRNA.
(c/min/cell)
6,102
6,490
25,020
12,325
6,2 16
+
P mRNA
(c/min/cell)
Y/Y -t 13
7,146
7,510
27,640
12,733
6,475
0.15
0.14
0.09
0.04
0.04
y
Ratio
1865
HBF-INDUCING ACTIVITY
0.70.6-
t
Normal
Sickle cell onemia
t
0 0.5._
c
0
0.4 -
Q
0.3
y / y + P in C-FCS
y / y + P in C - F C S
0.1Fig 3. (A) Peripheral blood cultures and ( 6 ) bone marrow
cultures. Correlation of y/y B ratios in paired cultures grown in
FSS or in C-FCS. Correlation coefficients were for (A) r = .75, n =
40;(B) r = .85, n = 38.
+
C-FCS
FCS
C-FCS
BFUe
FCS
C-FCS
CFUe
FCS
BFUe
C-FCS
FCS
CFUe
+
Fig 2. Comparison of y/y
@ ratios in cultures of normal
bone marrow cells (persons without hemoglobinopathy)and cells
of individuals with sickle cell anemia (at steady-state or during
treatment with cell cycle drugs). Note the consistently higher
ratios in FCS-grown BFU-E or CFU-E colonies.
y/y
+
BFU-E from patients with homozygous j3-thalassemia (Fig
5A) or heterozygous HPFH (Fig 5B) were used for these
studies. In both types of cultures, there was an FCS concentration-related increase in y/y + j3 ratios. As Fig 5B shows,
plateau levels of y/y + j3 synthesis were reached a t 15% to
20% FCS.
In the third series of experiments, increasing concentrations of FCS (from 0% to 30%) were added to cultures grown
in the presence of a range of concentrations of FSS (from 0%
to 25%). A clear relationship between FCS concentration
and levels of y/y + j3 ratio was again observed (Fig 6).
Other studies. To test whether late addition of FCS in
culture affects y-globin synthesis, peripheral blood mononuclear cells were grown in C-FCS media, and on day 8,
colonies were collected and divided in two parts. One was
transfered into another C-FCS-containing plate, while the
other part was transfered to an FCS-containing plate. In the
C-FCS to C-FCS transfer, the y/y + j3 ratio was 0.04 on day
13 and 0.04 on day 15. In the C-FCS to FCS transfer, the
y/y + j3 ratio was 0.11 on day 13 and 0.13 on day 15. These
data suggest that FCS can induce y-globin expression in late
erythroid progenitors. Similar conclusions can be drawn
Table 2. y/y
from the FSS and FCS transfer experiments published
before.”
In three experiments, BFU-E cultures were done in
C-FCS media, and colonies were collected on day 13. The
pooled cells were subsequently incubated for 16 hours in
biosynthesis labeling mediaz6 containing either FCS or CFCS, and y/y + j3 ratios were determined. The y/y + j3
biosynthetic ratios for C-FCS and FCS 16 hour-incubations
were, respectively: experiment 1, 0.03 and 0.10; experiment
2, 0.02 and 0.11; experiment 3, 0.04 and 0.12. In one
experiment, colonies were collected on day 11, and cells were
processed as above, but instead of biosynthesis, mRNA
y/y + j3 ratios were estimated as described in Materials and
Methods. The y/y + j3 mRNA ratios were 0.01 and 0.125,
respectively, for the C-FCS and FCS 16 hour-incubations.
These data suggest that the “factor” present in FCS can
0.31
-
FCS
-Control
(C-FCS)
0
.c
cr“ 0.2
Q
+
x
+ B Ratios in Fetal BFU-E Cultures
rlr + 8 Ratio
Fetal Liver
1
2
3
4
5
6
7
Yolk Sac
1
2
C-FCS
FCS
0.93
0.99
0.91
0.93
0.94
0.95
0.95
0.93
0.99
0.97
0.93
0.89
0.88
0.93
0.90
0.91
0.88
0.88
10
0
20
30
40
Serum Concentration (%)
+
Fig 4. Effect of FCS concentration on 7/y B ratios. Peripheral blood BFU-E from a normal person ware cultured in the
concentrations of FCS shown in the horizontal axis, and y/y
/3
ratios were determined in the colonies formed. Control cultures
were done in C-FCS. Note that there is a concentration-dependent
increase in the 7 / 7 B ratio in the FCS-grown cells but not in the
C-FCS-grown cells.
+
+
1866
CONSTANTOULAKIS ET AL
0.6
-
0.4
-
-
-
0.2
day 15
day 18
day 20
b-4
0
.-
0
Q
+
h
\ 0.5h
0.4
(against PBS using membranes with molecular weight cut off
ranging from 2 Kd to 25 Kd) was added to cultures
containing C-FCS, and the y/y + 3/ ratios in BFU-E colonies
were determined. The presence of the y/y + /3 stimulating
activity in the dialysates (Table 3) suggests that the molecular size of this factor is greater than 25 Kd. Finally, gel
filtration chromatography of FCS resulted in elution of the
HbF stimulating activity with the large molecular weight
fractions. In these experiments, the tubes of chromatographic eluent were pooled (in four pools), the pools were
concentrated by ultrafiltration to the initially applied to the
column or FCS volume, and 0.2 mL aliquots were added in
peripheral blood BFU-E cultures (done using cells from the
same person), and effects on globin biosynthesis were assessed. In one G-200 Sephadex experiment, the highest
y/y + fl ratio was obtained when the concentrate of the pool
10
20
30
FSS 0 %
FSS 15%
1
-
0.4
0.21
0.3-
-day
17
0.2-
FSS 20%
0.1Q
+
0-
1
I
I
I
0
10
20
30
I
FCS Conc. (%I
0.4
o-2
L
0
+
Fig 5. Effect of FCS concentration on y/y @ synthesis. Cells
were grown in medium containing 30% FSS (A) or 30% C-FCS (E).
In these media, various concentrations of FCS (from 0%t o 30%. as
shown in the horizontal axis) were added. Panel A is an experiment
with homozygous @-thalassemia BFU-E; in panel E, HPFH heterozygous cells were used. Note the FCS concentration-dependent
increase of y/y
/3 ratios in both experiments. In panel E. the
y/y j3 ratio plateaus at 15% FCS.
+
+
induce y-globin expression even when it acts at the level of
early erythroblast (or on late progenitors maturing during
the 16 hour-incubation period).
Preliminary results suggest that the FCS activity that
increases the y/y + /3 ratio has a relatively high molecular
weight. First, treatment of FCS with BSA-coated or dextrancoated charcoal did not result in the loss of the HbF
stimulating activity (Table 3). Since charcoal particles
presaturated with large molecules such as BSA or dextran
can only adsorb small molecules,32this result suggests that
the factor in FCS is a large molecule. Second, dialyzed FCS
IO
0
20
30
10
0
20
30
FCS Concentration
+
@ ratio. In this
Fig 6. Effect of FCS concentration on y/y
experiment, homozygous &thalassemia EFU-E were grown in
cultures using the concentrations of FCS shown in the horizontal
axis. These concentrations of FCS were added t o media containing
the concentration of FSS shown inside each panel. Note the FCS
6 ratios. Plateau
concentration-dependent increase in y/y
levels of y/y @ ratios are reached a t about 15% t o 20% FCS.
+
+
1867
HBF-INDUCING ACTIVITY
Table 3. Globin Synthesis in Peripheral Blood BFU-E Cultures in the Presence of FCS Submitted to Various Treatments
Treatment of FCS
~
Emeriment
None
0.35
0.17
0.26
0.17
0.12
0.17
0.18
0.12
0.64
Values are y/y
Dextran
BSA
Charcoal
Charcoal
Charcoal
0.09
0.03
0.08
0.07
0.03
0.02
0.04
0.03
0.43
-
-
0.13
0.15
-
Dialysis
2,000
3,500
0.30
0.12
0.25
0.14
0.12
0.22
0.18
0.16
0.15
0.10
-
-
0.15
0.17
0.15
0.12
0.15
0.13
-
-
-
-
-
0.58
0.56
-
7,000
13,000
25,000
0.20
-
-
-
0.20
0.17
0.12
0.14
0.22
0.16
0.11
0.2 1
0.15
0.12
-
-
-
0.60
0.59
0.58
-
+ j3 ratio.
corresponding to 100 Kd to 200 Kd was added in culture. In
three G-100 Sephadex experiments, four pools of eluent
corresponding to over 60 Kd (pool I), 20 Kd to 60 Kd (pool
II), 5 Kd to 20 Kd (pool III), and 0.5 Kd to 5 Kd (pool IV)
were collected. The y/y + @ ratios in cultures were
(means + S D of the three experiments): 0.116 f 0.002 for
pool I; 0.094 + 0.002 for pool 11; 0.051 0.001 for pool 111;
and 0.056 f 0.001 for pool IV. The y/y + @ ratio in the
C-FCS control cultures was 0.053 f 0.002.
DISCUSSION
Our data show that serum from calf fetuses contains an
activity (or activities) that increases the relative synthesis of
H b F in erythroid cultures. This activity induces y-globin
synthesis in peripheral blood BFU-E and in bone marrow
BFU-E and CFU-E cultures of normal persons and of
individuals with a variety of hematologic conditions. The
effect of the activity on H b F is mediated through a relative
increase in steady-state y-globin mRNA. The effect of FCS
on y-globin expression is quantitative: in several experiments, there was an increment in y/y + @ ratios when the
concentration of FCS was raised in culture. While there was
a consistent induction of y-globin synthesis in the adult
BFU-E cultured in FCS, there was essentially no effect in
BFU-E cultures of fetal origin. These results confirm previous observations2’that FCS contains a charcoal removable
factor(s) that induces H b F in adult BFU-E cultures.
The possibility that the sera used in culture contain
activities that can affect fetal globin expression was first
realized when it was observed that, in contrast to the cultures
done with FCS, cultures done with FSS had low y-globin
p r o d u ~ t i o n . ’ While
’ ~ ~ ~ FSS decreased y/y + @ ratios in adult
BFU-E cultures, there was no such effect in cultures of fetal
BFU-E.’9.20.33Most interestingly, FSS inhibited y-globin
biosynthesis in cultures of H P F H or &@-thalassemia
BFU-E.33.34These results suggest that FSS contains a “hemoglobin switching activity,” which inhibits y-globin expression
and enhances P-globin expression. By culturing cells in FSS
and subsequently lifting the colonies and subculturing them
in FCS (or the opposite), we showed that this “switching
activity” was acting on cells of early as well as late
erythropoiesis.20 By using single cell clones, splitting them in
two parts a t an early stage of their development (less than 50
cells); and subculturing each part in either FCS or FSS, we
showed that sibling clones can have either high or low y/y +
@ ratios, depending on the environment in which the clones
grow (ie, whether they are grown in FCS or FSS20.34).Does
FSS contain an inducer for y to @ switching, as we have
postulated, or does it simply lack the y-globin inducer that is
present in FCS?
Results of comparisons of globin biosynthesis in paired
FSS or C-FCS cultures support the possibility that FSS
cultures display low y/y + @ synthesis because they lack the
inducer present in FCS (Fig 3). Also, like FSS, C-FCS does
not affect y/y + @ ratios on fetal BFU-E cultures. The
significant inhibition of y-globin expression in FSS-grown
HPFH and @-thalassemia BFU-E culture^^^'^^ supports the
existence, in FSS, of an activity that inhibits y expression
and increases @ expression. It is possible that fetal sera
contain two types of activities: a y-globin inducer like the one
removed by charcoal treatment of FCS and a y-globin
inhibitor that is revealed by the effect of FSS on HPFH
mutants.
Our studies provide insights on whether FCS induces
y-globin expression by acting at the transcriptional or at the
post-transcriptional levels. We have compared globin gene
expression in FSS and FCS cultures by analyzing y- and
@-globin gene methylation, y- and @-globin gene DNAse 1
hypersensitivity, and y- and @-globin gene tran~cription.~’
These studies showed that the differences in y-globin biosynthesis between FCS or FSS cultures reflect differences at the
level of methylation and the degree of DNAse 1 sensitivity of
the y genes.35 Run off transcription assays showed that
differences in y/y + @ biosynthetic ratios reflect differences
at the level of y-gene t r a n ~ c r i p t i o n In
. ~ ~this article, we show
that the differences in y/y + @ globin biosynthetic ratios
between FCS and C-FCS cultures reflect differences a t the
level of steady-state globin mRNA. We conclude that the
factor@)contained in FCS induce y-globin expression through
an effect a t the transcriptional rather than the posttranscriptional level.
How does FCS exert its effect at the cellular level? Various
mechanisms can be proposed. First, the factor in FCS may
produce in culture an environment of “stress erythropoiesis”
leading to “premature commitment” of progenitors, a condition thought to result in activation of fetal globin expression.”
CONSTANTOULAKIS ET AL
1868
Second, FCS may increase the y/y + 0 ratio by affecting the
maturation of erythroblasts. Since there are higher synthesis
levels of y-globin in immature erythroblast^:^.^^-^^ decreased
erythroblast maturation is expected to be associated with
higher y/y + 0 ratios. The data on mRNA accumulation in
FCS and C-FCS cultures (Table 1) are incompatible with
this interpretation. Third, it is possible that the factor in FCS
directly induces y-globin expression on progenitor cells or
early erythroblasts. The previously reported cell transfer
experiments of FSS or FCS grown cell^*^.-'^ and the transfer
experiments described herein are compatible with this interpretation.
REFERENCES
1. Papayannopoulou Th, Brice M, Stamatoyannopoulos G: Stimulation of fetal hemoglobin synthesis in bone marrow cultures from
adult individuals. Proc Nat Acad Sci USA 73:2033, 1976
2. Housman D, Clarke B, Hillman D, Alter B, Forget B, Nathan
D: Control of HbF synthesis in adult erythroid precursor cells:
Evidence for different programs for bone marrow and peripheral
blood precursors, in Stamatoyannopoulos G, Nienhuis AW (eds):
Cellular and Molecular Regulation of Hemoglobin Switching. Philadelphia, PA, Grune & Stratton, 1979, p 351
3. Clarke B, Nathan D, Alter B, Forget B, Hillman D, Housman
D: Hemoglobin synthesis in human BFUe and CFUe derived
erythroid colonies. Blood 54:805,1979
4. Papayannopoulou Th, Brice M, Stamatoyannopoulos G: Hemoglobin F synthesis in vitro: Evidence for control at the level of
primitive erythroid stem cells. Proc Natl Acad Sci USA 74:2923,
1977
5. Ogawa M, Kidoguchi K, Karam J: Preferential synthesis of
HbF in culture by human erythropoietic precursors in the marrow
and blood, in Stamatoyannopoulos G, Nienhuis AW (eds): Cellular
and Molecular Regulation of Hemoglobin Switching. Philadelphia,
PA, Grune & Stratton, 1979, p 153
6. Kidoguchi K, Ogawa M, Karam J, Martin AG: Augmentation
of fetal hemoglobin (HbF) synthesis in culture by human erythropoietic precursors in the marrow and peripheral blood: Studies in sickle
cell anemia and nonhemoglobinopathic adults. Blood 52:1115, 1978
7. Messner H, Fauser A: Distribution of HbF in individual
colonies by RIA: A lead toward human stem cells, in Stamatoyannopoulos G, Nienhuis AW (eds): Cellular and Molecular Regulation of
Hemoglobin Switching. Philadelphia, PA, Grune & Stratton, 1979,
p 379
8. Fauser A, Messner H: Fetal hemoglobin in mixed hemopoietic
colonies (CFU-GEMM) erythroid bursts (BFUe), and erythroid
colonies (CFUe): Assessment by RIA and immunofluorescence.
Blood 54:1384,1979
9. Papayannopoulou Th, Nakamoto B, Buckley J, Kurachi S,
Nute PE, Stamatoyannopoulos G: Erythroid progenitors circulating
in the blood of adult individuals produce fetal hemoglobin in culture.
Science 199:1349, 1978
10. Stamatoyannopoulos G, Papayannopoulou Th: Fetal hemoglobin and the erythroid stem cell differentiation process, in Stamatoyannopoulos G, Nienhuis AW (eds): Cellular and Molecular Regulation of Hemoglobin Switching. Philadelphia, PA, Grune & Stratton,
1979, p 153
11. Papayannopoulou Th, Nute PE, Kurachi S, Stamatoyannopoulos G: Consistent activation of fetal hemoglobin synthesis in
cultured adult bone marrow cells. Blood 51:671, 1978
12. Vainchenker W, Testa U, Hinard N, Beuzard Y, Dubart A,
Tsapis A, Monplaisir N, Rouyer-Fessard P, Rosa J: Hemoglobin
synthesis in 7-day and 14-day-old erythroid colonies from the bone
marrow of normal individuals. Hemoglobin 453, 1980
13. Huisman THJ, Reese AL, Gravely ME, Wilson JB, Webber
B, Felice AE: Adult and fetal hemoglobin production in erythroid
colonies from subjects with @-thalassemiaor with hereditary persistence of fetal hemoglobin (HPFH). Hemoglobin 4:449, 1980
14. Peschle C, Migliaccio G, Covelli A, Lettieri F, Migliaccio
AR, Condorelli M, Comi P, Pozzoli ML, Giglioni B, Ottolenghi S,
Cappellini MD, Polli E, Gianni AM: Hemoglobin synthesis in
individual bursts from normal adult blood: All burst and subcolonies
synthesize Gyand Ay-globinchains. Blood 56:318, 1980
15. Nienhuis AW: Discussion: Hemoglobin switching conference,
in Stamatoyannopoulos G, Nienhuis AW (eds): Cellular and Molecular Regulation of Hemoglobin Switching. Philadelphia, PA, Grune
& Stratton, 1979, p 362
16. Papayannopoulou Th, Nakamoto B, Kurachi S, Kurnit D,
Stamatoyannopoulos G: Cell biology of hemoglobin switching 11.
Studies on the regulation of fetal hemoglobin synthesis in human
adults, in Stamatoyannopoulos G, Nienhuis AW (eds): Hemoglobin
in Development and Differentiation. New York, NY, Liss, 1981, p
263
17. Terasawa T, Ogawa M, Porter PN, Golde DW, Goldwasser
E: Effect of burst-promoting activity (BPA) and erythropoietin on
hemoglobin biosynthesis in culture. Blood 56:1106, 1980
18. Ogawa M: Human hemoglobin switching in culture. Am J
Hematol9:127,1980
19. Papayannopoulou Th, Kurachi S, Nakamoto B, Zanjani E,
Stamatoyannopoulos G: Hemoglobin switching in culture: Evidence
for a humoral factor that induces switching in adult and neonatal but
not fetal erythroid cells. Proc Natl Acad Sci USA 79:6579,1982
20. Stamatoyannopoulos G, Nakamoto B, Kurachi S, Papayannopoulou Th: Direct evidence for interaction between human erythroid
progenitor cells and a hemoglobin switching activity present in fetal
sheep serum. Proc Natl Acad Sci USA 805650,1983
21. Rosenblum B, Strahler J, Hauash S, Whitten C, ButkunasPuskorius R, Roberts A: Peripheral blood erythroid progenitors from
patients with sickle cell anemia: HPLC separation of hemoglobins
and the effect of a HbF switching factor, in Stamatoyannopoulos G,
Nienhuis AW (eds): Experimental Approaches for the Study of
Hemoglobin Switching. New York, NY, Liss, 1985, p 397
22. McLeod DL, Shreeve MM, Axelrad AA: Improved plasma
culture system for production of erythrocytic colonies in vitro:
Quantitative assay method for CFUe. Blood 44517, 1974
23. Iscove NN, Sieber F: Erythroid progenitors in mouse bone
marrow detected by macroscopic colony formation in culture. Exp
Hematol3:32, 1975
24. Papayannopoulou Th, Kalmantis Th, Stamatoyannopoulos G:
Cellular regulation of hemoglobin switching: Evidence for inverse
relationship between fetal Hb synthesis and degree of maturity of
human erythroid cells. Proc Natl Acad Sci USA 76:6420, 1979
25. Lingrel JB, Borsook H: A comparison of amino acid incorporation into the hemoglobin and ribosomes of marrow erythroid cells
and circulating reticulocytes of severely anemic rabbits. Biochemistry 2:309,1963
26. Papayannopoulou Th, Kurachi S, Brice M, Nakamoto B,
Stamatoyannopoulos G: Asynchronous synthesis of HbF and HbA
during erythroblast maturation. 11. Studies of Gy, Ay, and 0 chain
synthesis in individual erythroid clones from neonatal and adult
BFUe cultures. Blood 57531, 1981
27. Righetti PG, Gionazza E, Gianni AM, Comi P, Giglioni B,
Ottolenghi S, Sechi C, Bernardi R: Human globin chain separation
by isoelectric focusing. Biochem Biophys Methods 1:45,1979
28. Karlinsey J, Stamatoyannopoulos G, Enver T: Simultaneous
purification of DNA and RNA from small numbers of eukaryotic
cells. Anal Biochem 180:303, 1989
HBF-INDUCING ACTIVITY
29. Zinn K, DiMaio D, Maniatis T: Identification of two distinct
regulatory regions adjacent to the human &interferon gene. Cell
342365, 1983
30. Kafatos F, Jones W, Efstratiadis A: Determination of nucleic
acid sequence homologies and relative concentrations by a dot
hybridization procedure. Nucleic Acid Res, 7:1541, 1979
31. Tilsty T: Gene amplification, in Schimke R T (ed): Cold
Spring Harbor, NY, Cold Spring Harbor Laboratory, 1981, p 231
32. Gottlieb C, Lau K, Wasserman L, Herbert V:Rapid charcoal
assay for intrinsic factor, gastric juice unsaturated B12 binding
capacity, antibody to I F and serum unsaturated B12 binding
capacity. Blood 25:875, 1965
33. Stamatoyannopoulos G, Papayannopoulou Th, Nakamoto B,
Kurachi S: Hemoglobin switching activity, in Stamatoyannopoulos
G, Nienhuis AW (eds): Globin Gene Expression and Hematopoietic
Differentiation. New York, NY, Liss, 1983, p 347
34. Papayannopoulou Th, Tatsis B, Kurachi S, Nakamoto B,
Stamatoyannopoulos G: A haemoglobin switching activity modulates hereditary persistence of foetal haemoglobin. Nature 308:71,
1984
35. Groudine M, Peretz M, Nakamoto B, Papayannopoulou Th,
1869
Stamatoyannopoulos G: The modulation of Hb F synthesis in adult
BFUe cultures reflects changes in y gene transcription and chromatin structure. Proc Natl Acad Sci USA 83:6887, 1986
36. Chui D, Wong S, Eukin M, Patterson M, Ives R: Proportion
of HbF synthesis decreases during erythroid cell maturation. Proc
Natl Acad Sci USA 77:2757,1980
37. Wood W, Jones R: Erythropoiesis and hemoglobin production:
A unifying model involving sequential gene activation, in Stamatoyannopoulos G, Nienhuis AW (eds): Hemoglobin in Development
and Differentiation. New York, NY, Liss, 1981, p 243
38. Dover G, Boyer S: Quantitation of hemoglobins within
individual red cells: Asynchronous biosynthesis of fetal and adult
hemoglobin during erythroid maturation in normal subjects. Blood
56:1082, 1980
39. Peschle C, Migliaccio A, Migliaccio G, Lettieri F, Maguire
Y, Coudovelli M, Gianni A, Ottolenghi S, Giglioni B, Pozzoli M,
Comi P: Regulation of Hb synthesis in ontogenesis and erythropoietic differentiation: In vitro studies on fetal liver, cord blood, normal
blood or marrow, and blood from HPFH patients, in Stamatoyannopoulos G, Nienhuis AW (eds): Hemoglobin in Development and
Differentiation. New York, NY, Liss, 1981, p 359