1. No category

From www.bloodjournal.org by guest on December 22, 2014. For personal use only.
Dexamethasone Recruitment of Self-Renewing Osteoprogenitor Cells in Chick
Bone Marrow Stromal Cell Cultures
By N. Kamalia, C.A.G. McCulloch, H.C. Tenebaum, and H. Limeback
Bone marrow stromal cells are a mixed population that
contribute to the formation of the hematopoietic microenvironment. The osteogenic lineage includes populations of
cells that, in culture, form discrete nodules of mineralized
tissue when grown in the presence of ascorbic acid and
sodium p-glycerophosphate. We have used nodule formation
to assay for the self-renewal capacity of osteoprogenitor
cells in chick bone marrow cultures. To examine the regulatory influence of dexamethasone (Dx), first subcultures were
grown continuously or split 1 : l at repeated subculture. Cells
in continuous culture exhibited less than two population
doublings, while cellular proliferation and alkaline phosphatase area were inhibited by lo-’ mol/L Dx. Cells in split
(redistributed) cultures exhibited up t o 14 population dou-
blings and cellular proliferation was also inhibited by Dx. In
contrast with continuous cultures, redistributed cultures
treated with Dx had increased alkaline phosphatase area and
15-fold larger amounts of mineralized tissue formation than
controls. Osteogenesis was sustained for up to four subcultures and the ratio of mineralized tissue area to alkaline
phosphotase positive cell area was at most 0.55. These data
indicate that the osteogenic lineage of bone marrow stromal
cells contains self-renewing progenitors that are recruited by
Dx in culture and that at a maximum, only 55% of the alkaline
phosphatase-positivecell population contributes to osteogenesis.
0 1992 by TheAmerican Society of Hematology.
H
may enhance osteogenic metabolism, but the direct influence that these factors exert on BMSC-hematopoietic
interactions is not known. Supplementation of culture medium with hydrocortisone is known to accelerate osteogenesis’’ and also hematopoiesis, perhaps in part by augmenting
the development of the hematopoietic microenvir~nment.~~.~
Thus, hydrocortisone may influence hematopoiesis by regulating the size and function of BMSC populations, including
the osteogenic lineage.
To examine the regulatory influence of dexamethasone
(Dx) on the self-renewal capacity of osteoprogenitor cells,
we have measured the formation of mineralized bone
nodules in BMSC cultures that were replated repeatedly to
induce proliferation of the osteoprogenitor cells. These
methods sort out the progeny of osteoprogenitor cells at
each subculture, thereby permitting study of self-ren e ~ a l , ”an
~ ’essential
~
characteristic of putative osteogenic
stem cells in BM. We have also studied the expression of
alkaline phosphatase as a phenotypic marker for cells of the
osteogenic lineage25’27
and its regulation by Dx. The results
show specifically the existence of osteogenic lineages in
mixed adherent marrow cell cultures that include progenitor cell populations reliant on Dx for growth and differentiation.
EMATOPOIESJS is dependent on the support and
cellular interactions provided by bone marrow stromal cells (BMSC)’.’ that contribute to the production of
soluble factors and extracellular matrix formation.6-8Although poorly understood in vivo, the production of regulatory factors and matrix components by BMSC is essential
for maintenance of hematopoiesis in long-term BM cultures.’ The study of BMSC has been complicated by the
heterogeneity of the cell populations and the lack of specific
markers for the different cell lines.’ Consequently, the
number and hierarchy of cell lineages in BMSC is not well
understood nor are the factors that regulate the proliferation and differentiation of cells within these lineages.
Several stromal cell lines have been introduced’~Lo-12
that
facilitate the study of BMSC-derived hematopoietic regulatory factorsI3 and the contribution of stromal cells to the
hematopoietic microenvironment. In addition, the recent
development of colony assays that measure the numberL4
and self-renewal capacityL5of osteogenic progenitor cells in
mixed adherent cell cultures has permitted specific study of
the osteogenic lineage in BMSC16 and the proliferation of
progenitors in vitro.”
The microenvironment created by BMSC in vitro is
believed to be essential for hematopoietic stem cell proliferation and differentiati~n‘~.’~~~;
however, little is known
about the intimate relationship between BM and its supporting tissue-bone. Previous reports have shown that several
factors including estradiol,” bone morphogenetic protein,15
transforming growth factor+ (TGF-P),’” and activin-AZ1
From the Faculty of Dentistry, University of Toronto; and the
Samuel Lunenfield Research Institute, Mount Sinai Hospital, Toronto, Ontario, Canada.
Submitted June 6,1991; accepted September 9, 1991.
Supported by MRC Operating Grant No. M-9870.
Address reprint requests to C.A.G. McCulloch, DDS, PhD, Room
430, 124 Edward St, Toronto, Ontario, Canada M5G IG6.
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 1992 by The American Society of Hematology.
0006-4971/92/7902-0003$3.00/0
320
MATERIALS AND METHODS
Cell isolation and culture. BM was obtained aseptically from
femora and tibias of 17-day-old chick embryos by pressureinjection of cell culture medium (see below) containing 5x
antibiotics into the medullary cavity. The expelled marrow was
dispersed into single cell suspensions by repeatedly aspirating the
cells through the needle. For each experiment, the BMs from 10
embryos were pooled, and seeded into three 75 cm2 tissue culture
flasks (Falcon; Becton Dickinson, Mississauga, Ontario, Canada)
containing a-minimal essential medium with ribosides and deoxyribosides (a-MEM + DNA + RNA), 30% fetal bovine serum (FBS;
Flow Laboratories, McLean, VA), ascorbic acid 50 pg/mL, 10
mmol/L sodium P-glycerophosphate (Sigma Chemical Co, St
Louis, MO), and antibiotics (penicillin G at 100 pg/mL, Sigma;
gentamycin at 500 pg/mL, GIBCO, Burlington, Ontario, Canada;
and amphotericin B at 0.3 kg/mL, Sigma). Stromal cells were
enriched by allowing 1 day for cell attachment in primary culture
and then unattached cells, dead cells, and debris were washed off.
This procedure effectively reduces the concentration of monocytes/
t?lood, Vol79, No 2 (January 15). 1992: pp 320-326
From www.bloodjournal.org by guest on December 22, 2014. For personal use only.
DEXAMETHASONE RECRUITMENT OF OSTEOGENIC CELLS
macrophages (nonspecific esterate positive) to less than 5% and
depletes lymphoid cells. Attached cells were subcultured by
trypsinization (0.01% trypsin in citrate saline), pooled, subdivided
1:1, and cultured in medium with 15% FBS, or in the same medium
supplemented with Dx (Sigma) to a final concentration of lo-’
mol1L Dx. This concentration of Dx is not toxic for chick cells and
previous workz2 has shown that it is the optimal dosage for
stimulating osteogenesis in chick bone culture. Pilot experiments
also indicated that cultures treated with
mol1L Dx provided
the maximum contrast with controls in the study of regulation of
the BMSC osteogenic lineage. Cell cultures were maintained at
37°C in a humidified atmosphere consisting of 95% air plus 5%
CO,. The medium was changed every 2 to 3 days and the
morphologic appearance of cultured cells was monitored every 2 to
3 days by phase contrast microscopy.
To assess the dependence of mineralized tissue formation on
time, experiments were conducted on continuous first subcultures
of chick BMSC. One half of the cells from each flask were
electronically counted (model ZM, Coulter Counter; Coulter
Electronics, Hialeah, FL) and were seeded at a density of 2.4 x lo4
cellsicm’ into five 24-well multi-well plates (Flow), coated with 1%
gelatin (British Drug House [BDH], Toronto, Ontario, Canada).
Pilot experiments showed that cell attachment of chick BMSC to a
plastic substrate was poor, consistent with the findings of Greenberg et al.= Gelatin was used to promote cell attachment and we
have found previously that the coating did not favor selective
adherence of macrophages or monocytes staining positively for
nonspecific esterase.*’ The cell densities used in these experiments
were found in pilot experiments to be sufficient for production of
mineralized tissue and that could be accurately assessed by
automated image analysis (see Results). Cells in first subculture
were grown continuously in multi-well plates for a period of 3,4,5,
6, and 7 weeks (hereafter termed continuous cultures) and terminated by fixation.
To detect the ability of the progeny of osteoprogenitor cells to
produce bone, the remaining one half of the cells from primary
cultures were seeded into a new T-75 flask at a density of 2.4 x lo4
cellsicm’, and subcultured 1 week later. At each subculture, one
half of the cell population was seeded into a new T-75 flask at 2.4 x
lo4cellslcmZand the remaining one half was seeded into wells of a
24-well multi-well plate also at 2.4 X lo4cells/cm2.This procedure
was repeated with identical plating densities at each subculture to
produce a split ratio of 1:l at each subculture. Cultures redistributed in this manner for 4 consecutive weeks (redistributed cultures;
hereafter subcultures 2, 3, 4, and 5) were used to dilute osteoprogenitor cells at each subculture and thereby assess the self-renewal
capacity of these cells. Cells from each subculture were seeded into
multi-well plates and grown for 3 weeks, so that the initial plating
density and total culture age of redistributed cultures was identical
to that of continuous cultures.
Alkaline phosphatase (AP). All cultures were terminated by
fixation in ice-cold 3% paraformaldehyde (pH 7.4) for 30 minutes
and were analyzed for AP activity. Cultures were stained with 0.1%
wt/vol Fast Blue BB salt (Sigma) and 0.03% wt/vol Naphthol AS
Phosphate (Sigma) in 0.2 mol/L Tris buffer (pH 9.0) for 40 minutes
and washed in running tap water for 30 minutes.” The area of cells
stained for AP activity was assessed by image analysis using a light
microscope (Orthoplan, Leitz, Germany) equipped with a MTI-65
video camera containing a Newvicon camera tube (Dage-MTI;
Michigan City, IN), a drawing tube, and a computerized morphometric system (Bioquant, Nashville, TN). Three microscopic fields
(400 mmz each) per culture and eight cultures for each time period
were analyzed. The threshold used for imaging AP cell area was
adjusted to measure cells with a wide spectrum of AP activity and
was kept constant for all measurements (threshold, 50 units). AP
321
activity per cell was estimated using video densitometry methods.”
Briefly, using microdensitometry, the activity of AP is directly
proportional to the density of staining.w Staining of 15 individual
cells was measured in each culture with an automated image
analyzer. Only cells with density values above threshold (50 units)
were measured. Ten measurements per culture were made from
each of eight cultures per time period.
Mineralization. To estimate the amount of mineralized tissue
formed in each culture, fixed cultures were stained with the
calcium stain Alizarin Red S (Eastman Kodak Co, Rochester, NY)
pH 4.2 for 5 minutes and then washed with distilled water. Alizarin
Red (AR) area was measured in the same cultures as the AP
activity, but with a 620130 nm interference filter in the optical path
to eliminate signal due to the Fast Blue salt. In this manner,
measurements of mineralized tissue area could be directly compared with earlier estimates of A P activity in each culture. Eight
cultures per time period were analyzed.
Cell number. Estimation of cell number was evaluated by first
destaining cultures with weak acid (70% ethanol, 1% HC1) to
eliminate the AR stain. Cell nuclei were then stained with 10
pg1mL propidium iodide (Sigma) containing 0.01% nonidet in
phosphate-buffered saline (PBS) and 100 pg/mL RNAse for 1
minute and washed in PBS. Cells were measured with a microscope
fluorimeter (Leitz MVP-SP) at 530120 nm excitation and an
emission monochromator setting of 640/6 nm. For each culture,
three fluorimeter measurements were made in randomly chosen 2
mm2microscope fields and direct cell counts were also made at the
same time. The cell number was estimated from photometer
counts.
Data analysis. AP area, AR area, and photometer count data
from each well were entered into a computer and maintained in
register. Data were normalized for cell number by estimates
derived from photometer counts made on each culture. AP density
per cell was not normalized. Statistical analyses were performed
with SAS (version 6.03; SAS institute, Cary, NC). Three-way
factorial analysis of variance was performed using time of culture,
drug treatment (Dx, no Dx), and culture type (continuous, redistributed) as factors. To assess the relation between photometer counts
of propidium iodide-stained nuclei and actual counts of cell
number, Pearson’s correlation was computed. To determine the
relation between the number of AP-positive cells and mineralization, the ratio of AR area to A P area was computed.
RESULTS
Cultures. BMSC in first to fourth subculture initially
formed clusters containing cells with a predominantly
cuboidal-polygonal cell morphology (Fig 1A) that subsequently expanded into clusters of cells that formed mineralized tissue. Cells with the spindle-shaped morphology of
fibroblasts were found peripheral to cell clusters and in
some instances were apparently contiguous with the cell
clusters. However, in continuous cultures, fibroblastic cells
predominated. Small aggregates of phase contrast dense
bone-like nodules were observed as early as day 8 and were
associated with cell clusters. These cell aggregates were
found to increase in size and opacity with time. Mineralized
nodules stained for calcium with ARS were easily observed
with the naked eye after 3 weeks of culture (Fig 1B).
Nodules were observed at a different focal plane than that
of the adjacent cell monolayer, thus indicating their threedimensional morphology. The formation of nodules was in
part dependent on initial plating density because little or no
bone formation occurred in low-density cultures ( < lo4
From www.bloodjournal.org by guest on December 22, 2014. For personal use only.
322
KAMALIA ET AL
Fig 1. (A) Phase contrast micrograph (original magnification
x 104) of chick BMSC at first subculture showing clusters of cuboidal cells after 3 days of culture.
Note fibroblastic cells located peripheral t o cell cluster. (B)ARstained bone nodule that was
mineralized after 3 weeks of culture (original magnification x76).
Nodules were readily observed
by naked eye alone and in this
low power photomicrograph the
nodule is at a higher plane of
focus than the cells that surround the cell cluster and contribute t o its formation. Second subculture of chick BMSC grown in
the presence of 10 e mol/L Dx,
ascorbic acid, and sodium pglycerophosphate. (C) Cell clusters in
chick BM stained positive for AP
appear as dark masses after 21
days in second subculture (original magnification x 104). Many of
the cell clusters were not associated with mineralized nodules.
cells/cm2), while at higher cell densities ( > 10s cells/cm2)
there was complete coverage of the culture dish with
mineralized tissue. Cultures that were seeded at 2.4 x lo4
cells/cm' formed discrete nodules that were readily measured by image analysis. Nodules increased in size for l to 2
weeks and then stopped growing. Cell clusters surrounding
the nodules were always positivcly stained for AP but many
AP-positive cells were not associated with nodules (Fig 1C).
Cell proliferation. Estimates of cell numbers in individual wells by fluorimetry showed a linear relation between
photometer counts and actual visual counts of cell nuclei
(I? = .79) and, because much larger proportions of each
well were sampled by fluorimetry, photometer counts were
used to estimate overall cell number in each well. In both
continuous and redistributed cultures, cells were innoculated at 2.4 x lo4cells/cm' and grown in multi-well plates to
confluence. Computation of cumulative population dou-
bling levels (CPDL) from Coulter counts of cultures in T-75
flasks and from photometer counts of cells in multi-well
plates indicated that, after growth for up to 7 weeks in
24-well plates, the whole BMSC population had undergone
only 2.6 CPDL without Dx and approximately 2.0 CPDL
with Dx in continuous cultures (Table 1). There was a 30%
increase of CPDL from 3 to 7 weeks in both Dx-treated and
control cultures. In redistributed cultures, thc CPDL were
twofold to fivefold higher than continuous cultures at
termination. Dx significantly reduced the CPDL of the
whole BMSC population at all sample times (Table 1;
P < .001) for both continuous and redistributed cultures.
Mineralized tissue formation. The area of mineralized
tissue stained with AR and normalized to photometer
counts was used to estimate osteogenic activity per cell (Fig
2). Cells in continuous culture had not proliferated before
seeding in multi-well plates (Table 1) and during the course
of the culture underwent a maximum of only 2.6 CPDL.
These cultures showed increased mineralized tissue area
per cell between 3 and 4 weeks but exhibited no significant
change thereafter. The increased mineralized tissue area
was the result of increased numbers of nodules and not an
increase of individual nodule areas. During the increase of
mineralized tissue area, cell numbers increased only 10%.
At 4 weeks, osteogenesis was significantly less in Dx-treated
continuous cultures (Fig 2; P < .05),while thereafter there
was no significant difference (P > .05) between Dx-treated
and control cultures.
Redistributed cultures split 1:l at subculture were used
to dilute osteoprogenitor cells. This technique evenly sorts
the progcny of osteoprogenitor cell divisions between flasks
as the whole BMSC population expands.'' Between subcultures 1 and 2, cells in Dx-treated cultures had undergone
2.4 population doublings and exhibited 15-fold increases of
mineralized tissue compared with subculture 1. At subcultures 3 and 4 when the whole BMSC had undergone 4.4 and
6.8 population doublings, respectively, mineralized tissue
Table 1. Cumulative Population Doubling Levels of BMSC Cultures
+ Dx
At
Plating
No Dx
At
Termination
Continuous cultures (subculture 1)
Culture age (wk)
3
0
0
4
5
0
6
0
7
0
Redistributed cultures
Subculture no.
2
0.2
3
1.6
4
4.2
5
7.4
At
Plating
At
Termination
1.4
1.1
1.o
2.1
1.9
0
0
0
0
0
2.1
2.2
2.2
2.4
2.6
2.4
4.4
6.8
9.6
0.8
2.8
7.6
10.8
4.2
5.9
10.5
13.7
The coefficients of variation of estimated CPDL were 10% to 15%. Dx
(lO-'mol/L) significantly reduced CPDL in continuous and redistributed
cultures (P < ,001) at all times. CPDLs at plating were estimated from
Coulter counts. CPDLs at termination were estimated from sums of
Coulter counts and photometer count estimates of cell numbers.
From www.bloodjournal.org by guest on December 22, 2014. For personal use only.
323
DEXAMETHASONE RECRUITMENT OF OSTEOGENIC CELLS
n
0
I
= o
e-
e=
6 -
a= E
E
-
50-
0
40-
Q -
a-
T
40-
T
20-
lo4
Culture
5
6
3
7
nge (weeks)
-
o
4
2
formation decreased to 50% and 25% of that formed at
subculture 2 (Fig 2). After 9.6 doublings at subculture 5,
there was no detectable osteogenesis. Redistributed cultures grown without Dx exhibited very low osteogenic
activity at all subcultures despite extensive cell growth (up
to 13.7 CPDL for whole BMSC population).
AP. The area of AP activity normalized to cell counts
was used as an estimate of the relative proportion of
AP-positive cells in each culture (Fig 3). The area of
AP-positive cells in continuous cultures was generally
reduced by Dx and decreased sharply after 4 weeks. In
contrast, Dx strongly increased the area of AP-positive cells
in redistributed cultures, particularly at subcultures 2 and 3
(P < .02). Thereafter, Dx-treated cultures exhibited large
reductions of AP activity with subsequent subcultures. In
addition to the proportion of AP-positive cells, AP density
was measured to provide estimates of the amount of
enzyme activity per cell in the most heavily stained cells in
each culture (Fig 4). Cells with optical densities greater
than 50 U were measured. Dx treatment significantly
(P < .001) decreased AP activity in both continuous and redistributed cultures. In continuous cultures, AP activity increased with time, while in redistributed cultures there was
a progressive decrease of activity from subcultures 2 to 5.
To assess the phenotype of the cells capable of mineralized tissue formation, the ratio of AR-stained tissue area to
(weeks)
50
4
3
5
Subculture Number
Subculture Number
Fig 2. Histograms of mean (tSEM) area of AR-stained nodules
normalized t o cell number. Continuous first subcultures grown for up
t o 8 weeks (top panel) or redistributed cultures (lower panel) replated
and split 1:l every week from subcultures 2 through 5. Cultures
grown in presence or absence of lo-*mol/L Dx. Continuous cultures
grown in absence of Dx exhibited significantly (P < .05) more mineralized tissue area only at 4 weeks; there was no significant difference at
other times (P> .05; N = 6 cultures; 1 culture exhibited nonspecific
AR staining due t o large-scale cell death). Redistributed cultures
exhibited significantly more (P <~.001) mineralized tissue formation
at 2 t o 4 weeks in the presence of Dx.
7
6
5
Culture Age
Fig 3. Histograms of mean area of cells (tSEM) stained for AP
activity and measured by image analysis. AP activity normalized t o
cell counts in each culture. Continuous first subculture (top panel) and
redistributed cultures (lower panel) of chick BMSC.
AP area was computed (Fig 5 ) . In redistritked cultures,
the ratio was always less than 0.55 and was largest at
subculture 2 in Dx-treated cultures. Therefore, less than
55% of the AP-positive cell clusters in the whole BMSC
population contributed to osteogenesis in these cultures,
3
--
4
5
Culture Age
7
6
(weeks)
300-
- 0
2
3
4
5
Subculture Number
Fig 4. Histograms of mean (+.SEM) AP activity (in density units)
per cell in most densely staining cells of continuous first subcultures
(top panel) or redistributed cultures (lower panel). Note that the
staining density of cells with the most AP activity was estimated from
microdensitometric measurement of histochemically stained cells.
From www.bloodjournal.org by guest on December 22, 2014. For personal use only.
KAMALIA ET AL
324
continuously for up to 7 weeks during the active synthesis
phase of their lifetime.35Alternatively discrete and separate
populations of osteoblasts with a shorter synthetic lifetime
0.8
L
E
than 4 weeks could become recruited sequentially into a
2 0.6
mineralized tissue formation phase. We observed during
al
the course of continuous cultures that the area of individual
& 0.4
nodules increased rapidly over a period of 1 to 2 weeks and
a
E
0.2
then slowed considerably, consistent with previously obtained
morphometric data in osteogenic tissue culture.36
0.0
2
3
4
5
Nodule numbers increased while the areas of individual
Subculture Number
nodules remained the same. Therefore, the increased
mineralized tissue formation up to 4 weeks is best explained
Fig 5. Mean ratios (&EM) of area of AR-stained mineralized tissue
to AP-positive cell area in redistributed cultures. Note that at a
by the existence of discrete and separate populations of
maximum, less than 70% of AP cells are associated with mineralized
functional osteoblasts with an active lifetime of approxitissue formation (at subculture 2).
mately 1 to 2 weeks. These cells, or their immediate
precursors, may be recruited sequentially during the formation of nodules.
although all cell clusters surrounding mineralized nodules
In contrast with continuous cultures in which mineralized
were AP positive. Cultures permitted to grow for up to 7
tissue formation is not dependent on extensive proliferaweeks also contained clusters of AP-positive cells that did
tion, measurement of nodule formation in redistributed
not form nodules. This finding indicated that the nonminercultures detects the ability of proliferating osteoprogenitor
alizing AP cells were indeed nonosteogenic and not simply
cells and their progeny to produce bone-like tis~ue.’~~’’
given inadequate time to form nodules.
Subculturing at split ratios of 1:l evenly distributes both
DISCUSSION
osteoprogenitor cells and their progeny at each subculture
and
creates conditions for proliferation of the whole BMSC
Osteoprogenitor cells in chick BMSC give rise to colonies
population including osteoprogenitor cells. At first subculof AP-positive cells, some of which appear to form mineralture and at 0 population doublings, BMSC exhibited low
ized nodules very similar to that of rat BMSC.I6The area of
mineralized tissue formation either in the presence or
these nodules is proportional overall to their number and
absence of Dx. After 2.4 doublings of the whole BMSC
provides a direct estimate of the relative proportion of
population, mineralized tissue formation in redistributed
functional osteoprogenitor cells in each culture.” Thus, the
cultures increased up to 15-fold in the presence of Dx,
formation of mineralized tissue in chick BMSC provides a
thereby indicating the existence of a separate population of
specific assay to study the regulation of the osteogenic
Dx-dependent progenitor cells. In contrast with continuous
lineage of BMSC by Dx and directly demonstrates the
cultures, the progeny of Dx-dependent cells in redistribexistence of osteoblastic lineages in adherent marrow cell
uted cultures produce up to 15-fold more mineralized tissue
cultures.
area than control cultures, an observation consistent with
From experiments using cells at first subculture and
previous data from chick periosteal tissue
and rat
grown continuously for up to 7 weeks, mineralized nodules
calvarial cell culture.”
were produced by cell populations in chick BMSC at high
The Dx-dependent osteoprogenitor cells appear to cycle
plating densities and at very low CPDL, independent of the
at the same time as the whole BMSC population because
presence of Dx. Dx-treated cultures exhibited sequential
the initial increase of mineralized tissue formation was
increases of mineralized tissue area up to 7 weeks, indicatsynchronous with the increases of whole population douing that osteoprogenitor cells or functional osteoblasts
blings at subculture 2. However, unlike the whole BMSC
maintain their capacity to produce bone even after propopulation, which exhibited up to 10 population doublings,
longed cell culture. Proliferation in first subculture was
the osteoprogenitor cells exhibited limited self-renewal.
density limited and the cells proliferated minimally before
After more than seven population doublings, osteogenesis
forming mineralized tissue. Even lower proliferation was
was no longer detectable. While it is quite possible that the
observed in cells cultured under identical conditions but in
osteoprogenitor cells in chick BMSC cycle more rapidly
the continuous presence of
mol/L Dx, and these cells
than the whole population, previous reports of rat calvarial
showed about 50% less osteogenic activity at 4 weeks, a
finding consistent with several in
and in ~ i t r o ~ ~cells15
, ~ ~do not support this view, and data from rat BMSC”
also indicate that osteoprogenitor cells have a limited
studies. Our data indicate that Dx inhibits mineralized
capacity for self-renewal.
tissue formation by cell populations specifically requiring
The phenotype of the osteogenic lineage in chick BMSC
limited proliferation before forming mineralized tissue up
appears to be restricted to expression of AP, because all
to 4 weeks. The mechanism of action of Dx in the regulation
cells surrounding nodules were AP positive. Indeed, the
of osteogenesis is not well understood, but in both Dx and
methods used in this report permit accurate quantitation of
control continuous cultures there was an increase of minerAP activity in cells that are committed to osteogenesis by
alized tissue area between 3 and 4 weeks. Two possible
virtue of the colocalization of AP and AR staining. Howexplanations could account for this observation. First,
ever, while this enzyme has been shown previously to be a
functional osteoblasts could produce mineralized tissue
t
a
1.0
From www.bloodjournal.org by guest on December 22, 2014. For personal use only.
325
DEXAMETHASONE RECRUITMENT OF OSTEOGENIC CELLS
good marker for the osteogenic lineage,z5-27.37,38
our results
also indicate a marked functional heterogeneity of APpositive cells. We observed that less than 55% of the AP cell
clusters were associated with nodules and that the nonosteogenic AP cell clusters would not form mineralized tissue,
even if grown for 7 weeks. These data are consistent with
previous work3' indicating that, in BM, cells contributing to
granulopoiesis as well as reticular cells may also express this
enzyme. While AP is not restricted to cells of the osteogenic
lineage, AP activity and osteogenesis appear to be coregulated by Dx. For example, in redistributed cultures treated
with Dx, AP area and mineralized tissue area both peaked
at subculture 2 and AP activity was greatest also at
subculture 2. Redistributed cultures without Dx treatment
exhibited much lower mineralized tissue area and AP area.
Further, in continuous cultures, Dx inhibited AP activity,
AP area, and mineralized tissue formation at all culture
times. These data clearly show a tight coupling of AP
expression and mineralized tissue formation in BMSC
osteogenic cell populations and also show that Dx inhibits
AP enzyme activity irrespective of plating density and
population doubling level.
Taken together, these findings indicate that Dx recruits a
population of Dx-dependent proliferating osteoprogenitor
cells that are capable of a limited number of self-renewing
mitoses and whose progeny can produce bone-like tissue
after 3 weeks in culture. Thus, Dx appears to directly
influence the proliferation and differentiation of cells with
some of the characteristics of osteogenic stem cells. Owen
et aI3' have found that formation of AP-positive colonies
was stimulated by hydrocortisone. However, our use of a
colony assay that separates the functional activity of osteoblasts from cells with the phenotypic marker of AP activity
permits the specific measurement of osteogenic lineages
and the regulation of self-renewal capacity by Dx. Dx also
appears to actually inhibit mineralized tissue formation and
AP activity per cell in populations of osteogenic cells that
do not divide. Thus, Dx appears to exert differential
regulatory effects on the hematopoietic microenvironment
at least in part by enhancing the proliferation of Dxdependent osteoprogenitor cells and by inhibiting the
osteogenic activity of nonproliferating osteoblastic cells. Dx
may also exert a regulatory effect on other cell lineages in
the BMSC populations that contribute to the hematopoietic microenvironment by stimulating the proliferation of
fat and cartilage precursor cells.4oHowever, the use of a
colony assay that specifically detects the proliferative and
functional activity of only the osteogenic lineage of cells has
permitted selective study of Dx regulation on a single
lineage of BMSC that contributes to the hematopoietic
microenvironment. These methods could be applied directly to study the spatial association of hematopoietic
colonies with mineralizing cell clusters and to determine
whether cytokine augmentation of nodule formation is
coregulated with hematopoiesis.
ACKNOWLEDGMENT
We thank G. Kulkarni for the statistical analysis.
REFERENCES
1. Dexter TM, Allen TD, Lajtha LG: Conditions controlling the
morphogenetic protein-2 stimulates osteoblastic maturation and
proliferation of haemopoietic stem cells in vitro. J Cell Physiol
inhibits myogenic differentiation in vitro. J Cell Biol 113:681, 1991
91:355,1977
12. Zipori D, Tamir M: Stromal cells of hemopoietic origin. Int J
2. Sore11 JN, Weiss L: Cell interactions between hematopoietic
Cell Cloning 7:281,1989
and stromal cells in the embryonic chick bone marrow. Anat Rec
13. Zipori D: Regulation of hemopoiesis by cytokines that
restrict options for growth and differentiation. Cancer Cells 2:205,
197:1,1980
3. Weiss L Haemopoiesis in mammalian bone marrow, in
1990
Microenvironments in Haemopoietic and Lymphoid Differentia14. Bellows CG, Aubin JE: Determination of numbers of
tion. Ciba Fdn Symp, vol84. New York, NY, Elsevier, 1981, p 5
osteoprogenitors present in isolated fetal rat calvaria cells in vitro.
4. Dexter TM: Stromal cell associated haemopoiesis. J Cell
Dev Biol133:8, 1989
Physiol Suppl 1237,1982
15. Bellows CG, Heersche JNM, Aubin JE: Determination of
5. Perkins S, Fleischman RA: Hematopoietic microenvironthe capacity for proliferation and differentiation of osteoprogeniment: Origin, lineage, and transplantability of the stromal cells in
tor cells in the presence and absence of dexamethasone. Dev Biol
long-term bone marrow cultures from chimeric mice. J Clin Invest
140:1332,1990
81:1072,1988
16. Maniatopoulos C, Sodek J, Melcher AH: Bone formation in
6. Song ZX, Shadduck RK, Innes DJ, Waheed, Quesenberry PJ:
vitro by stromal cells obtained from bone marrow of young adult
Hematopoietic factor production by a cell line (TC-1) derived from
rats. Cell Tissue Res 254:317,1988
adherent murine marrow cells. Blood 66:273,1985
17. McCulloch CAG, Strugurescu M, Hughes F, Melcher AH,
7. Zipori D, Lee F: Introduction of interleukin-3 gene into
Aubin JE: Osteogenic progenitor cells in rat bone marrow stromal
stromal cells from the bone marrow alters hemopoietic differentiapopulations exhibit self-renewal in culture. Blood 72:1906, 1991
tion but does not modify stem cell renewal. Blood 71:586, 1988
18. Bentley SA: Studies of bone marrow stromal cells in vitro, in
8. Long MW, Williams JL, Mann KG: Expression of human
Myelofibrosis and the Biology of Connective Tissue. New York,
bone-related proteins in the hematopoietic microenvironment. J
NY,Liss, 1984, p 179
Clin Invest 86:1387, 1990
19. Ernst M, Schmid CH, Froesch ER: Enhanced osteoblast
9. Owen M: Marrow stromal stem cells. J Cell Sci Suppl 10:63,
proliferation and collagen gene expression by estradiol. Proc Natl
1988
Acad Sci USA 85:2307,1988
10. Benayahu D, Kletter Y, Zipori D, Wientroub S: Bone
20. Sporn MB, Roberts AB: Autocrine growth factors and
marrow-derived stromal cell line expressing osteoblastic phenotype
cancer. Nature 313:745,1985
in vitro and osteogenic capacity in vivo. J Cell Physiol 140:1, 1989
21. Centrella M, McCarthy TL, Canalis E: Activin-A binding
11. Yamaguchi A, Katagiri T, Ikeda T, Wozney JM, Rosen V,
and biochemical effects in osteoblast-enriched cultures from fetal
Wang EA, Kahn AJ, Suda T, Yoshiki S: Recombinant human bone
rat parietal bone. Mol Cell Biol 11:250,1991
From www.bloodjournal.org by guest on December 22, 2014. For personal use only.
326
22. Tenenbaum HC, Heersche JNM: Dexamethasone stimulates osteogenesis in chick periosteum in vitro. Endocrinology
117:2211,1985
23. Greenberger JS: Sensitivity of corticosteroid-dependent,
insulin-resistant lipogenesis in marrow preadiopcytes of mutation
diabetic-obese mice. Nature 275752,1978
24. Greenberger JS, Davisson PB, Cans PJ: Corticosteroid
dependent differentiation of human marrow pre-adipocytes in
vitro. In Vitro 152323,1979
25. White MP: Alkaline phosphatase: physiological role explored in hypophosphatasia, in Peck WA (ed): Bone and Mineral
Research. New York, NY, Elsevier, 1989, p 175
26. Rodan GA, Rodan SB: Expression of the osteoblastic
phenotype, in Peck WA (ed): Bone and Mineral Research, vol 2.
New York, NY, Elsevier, 1984, p 244
27. Wuthier RE, Register TC: Role of alkaline phosphatase, a
polyfunctional enzyme, in mineralizing tissues, in Butler WT (ed):
Chemistry and Biology of Mineralized Connective Tissues. Birmingham, AL, Ebsco Media, 1985, p 113
28. Greenburg BR, Wilson FB, Woo L Granulopoietic effects
of human bone marrow fibroblastic cells and abnormalities in the
“granulopoietic microenvironment.” Blood 58:557,1981
29. Tsuji T, Hughes FJ, McCulloch CAG, Melcher AH: Effects
of donor age on osteogenic cells of rat bone marrow in vitro. Mech
Ageing Dev51:121,1990
30. Tenenbaum HC, McCulloch CAG, Fair C, Birek C: The
regulatory effect of phosphates on bone metabolism in vitro. Cell
Tissue Res 257555,1989
KAMALIA ET AL
31. Baylink DJ: Glucocorticoid-induced osteoporosis. N Engl J
Med 309:306,1983
32. Burkhardt P: Corticosteroids and bone: A review. Hormone
Res 2059,1984
33. Canalis EM: Effects of glucocorticoids on type I collagen
synthesis, alkaline phosphatase activity and deoxyribonucleic acid
content in cultured rat calvariae. Endocrinol 112:931,1983
34. Chyun YS, Kream BE, Raisz LG: Cortisol decreases bone
formation by inhibiting periosteal cell proliferation. Endocrinology
114:477, 1984
35. McCulloch CAG, Heersche JNM: Lifetime of the osteoblast
in mouse periodontium. Anat Rec 222:128, 1988
36. McCulloch CAG, Tenenbaum HC: Dexamethasone induces
proliferation and terminal differentiation of osteogenic cells in
tissue culture. Anat Rec 215:397,1986
37. Fedarko NS, Bianco P, Vetter U, Gehron Robey P: Human
bone cell enzyme expression and cellular heterogeneity: Correlation of alkaline phosphatase enzyme activity with cell cycle. J Cell
Physiol144:115,1990
38. Owen ME, Cave J, Joyner CJ: Clonal analysis in vitro of
osteogenic differentiation of marrow CFU-F. J Cell Sci 87:731,
1987
39. Westen H, Bainton DF: Association of alkaline phosphatasepositive reticulum cells in bone marrow with granulocytic precursors. J Exp Med 150919,1979
40. Grigoriadis AE, Heersche JNM, Aubin JE: Differentiation
of muscle, fat, cartilage, and bone from progenitor cells present in
a bone-derived clonal cell population: Effect of dexamethasone. J
Cell Biol 106:2139,1988
From www.bloodjournal.org by guest on December 22, 2014. For personal use only.
1992 79: 320-326
Dexamethasone recruitment of self-renewing osteoprogenitor cells in
chick bone marrow stromal cell cultures
N Kamalia, CA McCulloch, HC Tenebaum and H Limeback
Updated information and services can be found at:
http://www.bloodjournal.org/content/79/2/320.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.