/ . Embryol. exp. Morph. Vol. 40 pp. 159-166, 1977
Printed in Great Britain © Company of Biologists Limited 1977
159
Cell growth and differentiation of murine
extra-embryonic fetal and adult hematopoietic
tissues in diffusion chamber cultures
By SHEILA R. WEINBERG, 1 MARGARET E. CHOVANIEC 1
AND FREDERICK STOHLMAN, JR. (deceased)
From the Saint Elizabeth's Hospital of Boston
SUMMARY
Studies were conducted with diffusion chambers (DC) filled with cell suspensions from
different CFX murine hematopoietic tissues: adult peripheral blood; adult tibial marrow;
day 17-5 of gestation fetal liver, spleen and thymus; day 14-5 of gestation fetal liver; day 10-5
of gestation yolk sac. After an initial decrease in DC cell numbers on day 2 of culture, growth
of each cell group continued, but, at different rates. ATM had the highest growth ratio and
FT-D17-52 had the lowest. The growth rates for APB and FL-D17-5 were similar. FS-D17-5
and FL-D14-5 cultures did not recover from the day 2 values (i.e. FL-D14-5 DC values on
day 13-14 of culture were half that recorded on day 2). The YS-D10-5 DC cell numbers continued to increase throughout the 14 days of study. The profile of cellular elements from the
DCs did not reflect the original cell suspensions. The predominant cell type recovered from
peripheral blood cultured for 14 days was the macrophage. By day 10-14 of culture, the
populations of cells harvested from the fetal tissue DC groups were similar to that of tibial
marrow. Both proliferative and mature granulocytes, and macrophages were the predominant
cell types. The yolk-sac pattern of cytodifferentiation recorded on day 7—14 was unlike that
of the other groups. These DC cultures were comprised of mainly macrophages and plasma
cells.
INTRODUCTION
Previous studies from our laboratory (Weinberg & Stohlman, \916a, b) have
demonstrated the existence of a unique population of hemopoietic stem cells
in the yolk sac of a 10-5-day developing mouse with hemopoietic characteristics
unlike that reported for other hematopoietic stem cells of different origin. A
murine yolk-sac cell suspension has been shown to be: (1) a population of stem
cells which differentiate into immunocompetent plasma cells and myeloid cells
when cultured in diffusion chamber (DC) cultures (Weinberg & Stohlman,
1976 a), (2) genetically strain-specific for the expression of immunocompetent
1
Authors'1 address: Saint Elizabeth's Hospital of Boston, 736 Cambridge Street, Boston,
Massachusetts 02135, U.S.A.
2
Abbreviations used: APB-adult peripheral blood; ATM-adult tibial marrow; FTD17-5-fetal thymus, day 17-5 of gestation; FS-D17-5-fetal spleen, day 17-5 of gestation;
FL-D17-5-fetal liver, day 17-5 of gestation; FL-D14-5-fetal liver, day 14-5 of gestation;
YS-D10-5-yolk sac, day 105 of gestation.
160
S. R. WEINBERG, M. E. CHOVANIEC AND F. STOHLMAN
features (Weinberg & Stohlman, 1976 a), (3) not responsive to humoral factors
responsible for stem cell proliferation and myeloid differentiation (Weinberg &
Stohlman, 1977), (4) maintained in long-term in vivo DC cultures without any
change in the differentiation of the cultured cell population (Weinberg & Stohlman, 1977) and (5) less radioresistant than adult tibial marrow cells (Weinberg
& Stohlman, 19766).
The present series of experiments were designed to use the in vivo DC culture
system to evaluate the growth rate and pattern of cytodifferentiation of CFi
murine tissues obtained during the successive stages of hemopoietic development. The data obtained from these studies provide further information about
the yolk-sac stem cell population, regarding the life span and primary seeding
potential of embryonic hematopoietic organs via the embryonic circulation.
MATERIALS AND METHODS
Mice
All experimental animals were virgin female CFX mice (Carworth Farms,
Rockland County, N.Y.) maintained on a diet of mouse pellets from Charles
River and tap water (containing 0-5 ml per liter of a diluted Dakin's solution 1 part in 1000, and 0-5 g per liter of neomycin sulphate) ad libitum. Mice at
10-12 weeks of age were used for mating, and those at 12-16 weeks of age
served as donors for tibial marrow and as recipients for millipore diffusion
chambers. CFX x CFX matings during a 24-h period were considered as day 0 of
gestation, and the pregnancies were surgically interrupted at the day points
selected for investigation.
Media
The cell suspensions were prepared in a chilled media that consisted of: a
balanced salt solution - T C Hanks' solution with calcium and magnesium
(Difco Laboratories, Detroit, MI), 75 units per ml of penicillin and 75 fig per ml
of streptomycin (without preservatives, Flow Laboratories, Rockville, Md.)
and 10% (v/v) fetal calf serum (Difco Laboratories, control nos. 583809,
608445 and 620901, millipore filtered immediately before use).
EXPERIMENTAL DESIGN
A comparative study was conducted with adult, fetal and extra-embryonic
hematopoietic tissues implanted into millipore diffusion chambers (DC). For
14 days of culture, cell growth and differentiation of the cell suspensions were
examined. The technical manipulations followed for the DC cultures (with
millipore filter membranes, 0-22 ju,m porosity) have been previously described
(Weinberg & Stohlman, 1976 a).
Cell growth and differentiation of mouse tissues
161
Adult tissues - peripheral blood and tibial marrow
Ten 10-5-day pregnant mice were bled via cardiac puncture and a cell suspension was prepared containing 5 x 105 nucleated cells with erythrocytes per 0-1 ml
to fill each DC. For each of two experiments both tibias from 15 mice were
flushed out with media. The marrow cells collected were washed and placed into
the DC (5 x 105 cells per 0-1 ml).
Fetal tissues - liver, spleen and thymus
On day 17-5 of gestation, the fetuses from 6 to 11 litters were aseptically
collected in a manner similar to that followed in obtaining yolk sacs (Weinberg
& Stohlman, 1976a). The spleen was a defined small pale pink organ closely
attached to the curvature of the stomach and in contact with the tail of the
pancreas (Owen, Raff & Cooper, 1975). The mouse embryo thymus was not as
defined in structure as the spleen or liver, and great care must be taken in order
to dissect out the entire tissue. Single cell suspensions of each fetal tissue were
prepared using mechanical manipulations previously outlined (Weinberg &
Stohlman, 1976a) and chambers were filled: 3-23 x 105 thymus cells per 0-1 ml,
15 x 105 spleen cells per 0-1 ml, or 15 x 105 liver cells per 0-1 ml. On day 14-5
of gestation, 4-7 litters were dissected free from maternal tissues and the fetal
livers were collected. A cell suspension with 15xl0 5 cells per 0-1 ml was
prepared and placed in chambers.
Extra-embryonic tissue - yolk sac
The yolk sacs were collected at 10-5 days of gestation and following the
protocol for making a cell suspension (Weinberg & Stohlman, 1976 a), each
chamber received 15 x 105 cells per 0-1 ml.
Each of the fetal and extra-embryonic studies was repeated at least three
times, with at least six chambers per group for each day point to evaluate the
cells harvested.
RESULTS
After an initial decrease in DC cell numbers on day 2 post-implantation,
growth of each cell group continued, but at different rates (Figs. 1, 2). ATM
had the highest growth ratio (i.e. by day 10, twice as many cells as implanted
were harvested from the DC cultures), and FT-D17-5 had the lowest. The growth
rate of APB was similar to that of FL-D17-5, with an increase to day 7-8 that
was double the values for day 2, followed by a slight decrease to day 13-14,
which was higher than the day 2 values.
FS-D17-5 and FL-D14-5 did not appear to recover from the initial decrease
in cell numbers noted on day 2 in all of the cultures. In contrast to FL-D14-5
cultures, the same number of cells were harvested from FS-D17-5 DC cultures
throughout the study to day 13-14. The FL-D14-5 growth rate continued to
162
S. R. WEINBERG, M. E. CHOVANIEC AND F. STOHLMAN
Yolk-sac,
Peripheral blood
• Granulocyie
Macrophage
u
10
14
Days of culture
Fig. 1. Growth and differentiation of CFx peripheral blood, tibial marrow and
10-5-day yolk-sac cell suspensions, cultured in millipore diffusion chambers implanted into CFi hosts. The values represent the average total numbers of nucleated
cells (±S.E.M.) and mean percentages of each cell type per chamber at various times
after implantation.
Table 1. Differential distribution of different CFX murine hematopoietic cell
suspensions implanted into millipore diffusion chambers
Granulocytic
131 o crnci
j laMHa
A
Tissue
n*
Peripheral blood (10 micef)
(59)
Tibial marrow
Fetal thymus
D17-5 gestation
(73)
Fetal spleen
(260)
D17-5 gestation
Fetal liver
D17-5 gestation
(80)
Fetal liver
D14-5 gestation
059)
Yolk sac
D10-5 gestation
(434)
Mature
(%)
erythroid
(%)
1 66
16-94
10-83
36-44
0
22-94
0
006
87-49
22-94
0
0-75
2-50
400
13-50
700
7100
1-50
11-34
44-34
21-80
0-30
22-84
0
1019
415
77-96
015
7-26
0
7-08
010
88-23
003
4-42
0
11-31
40-64
25-29
2-48
2011
Proliferative
(%)
Macrophage Lymphocyte
(%)
(%)
cell
(%)
011
* n = number of specimens collected and pooled.
t Peripheral blood obtained from 10-5-day pregnant mice.
decrease and by day 13-14, the numbers of cells harvested were half the value
of that noted on day 2. The growth rate of YS-D10-5 was unlike that of the
other tissues and suggested a similarity to the ATM. The growth ratio of day
13-14 was five times as great as that of day 2.
The types of cellular elements harvested from the diffusion chambers did not
reflect the original cell suspensions reported in Table 1 (i.e. cell population
profiles of ATM, FS-D17-5 and YS-D10-5 were similar; no significant differences were observed between day 17-5 and day 14-5 cell suspensions of fetal
liver, except that FL-D17-5 appeared to obtain slightly more granulocytic
elements; and ATM and FT-D17-5 suspensions had higher lymphocyte
Cell growth and differentiation of mouse tissues
163
lxlO6
g Granulocyte
Erythrocyte
H Macrophage
Lymphocyte
Plasma cell
lxlO3
lxlO5
Days of culture
Fig. 2. Growth and differentiation of CFX fetal thymus, fetal spleen and fetal liver at
.17-5 days of gestation, and fetal liver at 14-5 days of gestation. Each cell suspension
was cultured in diffusion chambers implanted into CFX hosts. The values represent
the average total numbers of nucleated cells (±S.E.M.) and mean percentages of each
cell type per chamber at various times after implantation.
percentages compared to the other tissues). Day 14 APB-DC cultures contained
mostly macrophages, and few granulocytes and lymphocytes (Fig. 1). The
profile of cells in the DC cultures by day 10-14 after implantation of tibial
marrow (Fig. 1) was similar to that of the fetal tissues (Fig. 2) (i.e. both proliferative and mature granulocytes, and macrophages were the predominant cell
types). However, the degree of granulocytic activity in the FL-D17-5 cultures
was significantly higher compared to the FL-D14-5 cultures (Fig. 2). The most
dramatic change in all of the DC cultures was the significant decrease in the
numbers of nucleated erythroid cells by day 7. The pattern of cytodifferentiation
recorded on day 7-14 for the YS-D10-5 was unlike that of any of the other
hematopoietic tissues cultured in the DC (Fig. 1). The macrophages increased
in number by day 7 and continued to be a predominant cellular element.
Plasma cells and lymphocytes began to rapidly proliferate and by day 14 of
culture, various stages of differentiation of the plasmacytoid cell line were
observed. Binucleated plasma cells were also observed on occasion.
DISCUSSION
The experiments outlined in this report were concerned with comparing the
growth rate and pattern of cytodifferentiation of diffusion chamber cultures
of fetal blood cell forming tissues obtained at different days of development
to extra-embryonic tissue and adult medullary tissues.
164
S. R. WEINBERG, M. E. CHOVANIEC AND F. STOHLMAN
Fetal liver was studied on day 14-5 and 17-5, since evidence exists which
indicate a transition in hematopoietic activity on day 15: a change in the
stabilization of hemoglobin synthesis (Djaldetti, Chui, Marks & Rifkind, 1970),
and a difference in the ability to respond to erythropoietin in vitro (Cole,
Hunter & Paul 1968; Cole & Paul, 1966; Paul, Conkie & Freshney, 1969;
Tarbutt & Cole, 1970). The blood cell differential distribution of the fetal liver
cell suspension was notably different on day 17-5 compared to that of day 14-5
(i.e. FL-D14-5 was almost entirely erythroid in nature, and FL-D17-5 reflected
an increase in granuloctyopoiesis).
Fetal spleen has been identified as an active site of hematopoiesis by day 15-5—
16 of gestation (Bessler, Nothi, Fishman & Djaldetti, 1976) containing colonyforming units (Pozzi, Andreozzi & Silini, 1972) and most of the identifiable
blood cellular elements (Djaldetti, Bessler & Rifkind, 1972). The contamination
of pancreatic tissue elements on the fetal spleens does not appear to have any
influence on diffusion chamber cultures or organ cultures (Owen et al. 1975).
Regardless of the means of pretreatment to the cells prior to implantation
into DC or the type of treatment to the hosts of the DC, murine yolk-sac
hemopoietic cells demonstrate a unique pattern of cell differentiation (Weinberg
& Stohlman, 1976 a). Cytochemical staining studies, electron microscopy
studies (Weinberg & Stohlman, 1976a) and immunofluorescent staining studies
(unpublished) have indicated that the predominant cell types harvested from
YS-D10-5 DC after 14 days of culture are: macrophages, lymphoid cells and
plasma cells.
The results reported here concerning fetal and adult hematopoietic tissues in
DC are in agreement with others. In vivo DC cultures of FL-D14-5 and FLD17-5 differentiate into proliferative and mature granulocytes and macrophages
(Symann et al. 1976; Vilpo & Vilpo, 1976). The growth rate of fetal liver obtained at day 14-5 and day 17-5 were significantly different and is in accordance
with other erythropoietic changes noted on day 15 (reviewed by Rifkind, Bank
& Marks, 1974).
The harvesting of predominantly granulocytic cellular elements and macrophages from DC containing adult murine bone marrow or peripheral blood
reflects the same pattern of differentiation obtained by other investigators
(reviewed by Stohlman, Quesenberry & Tyler, 1973; Benestad & Breivik, 1972).
Consequently, it appears that the potential of hematopoietic stem cell differentiation reflected in the DC culture system for fetal and adult hematopoietic
tissues is significantly different from extra-embryonic tissue.
The embryonic hemopoietic stem cell, morphologically identified by Jones
(1970), has been shown to be multipotential (Metcalf & Moore, 1971; Owen,
1972). Its developmental capabilities to differentiate into lymphoid or myeloid
cell lines are thought to be determined by the microenvironment provided by
each of the embryonic blood cell forming organs (e.g. fetal liver, fetal spleen,
fetal thymus and fetal bone marrow). Evidence also exists which indicate the
Cell growth and differentiation of mouse tissues
165
yolk-sac blood islands to be the source of the primary stem cells which seed
the embryonic hematopoietic tissues via the circulation (Metcalf & Moore,
1971; Moore & Metcalf, 1970; Owen, 1972). Although, Rifkind and his colleagues have suggested that the appearance of stem cells in the fetal liver is de
novo, and ' without the mediation of blood borne stem cells from more primitive
loci of hematopoiesis' (Rifkind, Chui & Epler, 1969; Rifkind et al. 1974).
These observations allow us to conclude that the yolk sac has a population
of hemopoietic stem cells which is also found in each of the other fetal blood
cell forming tissues, and another stem cell compartment which has a predetermined life span and is preprogrammed to provide immunological protection
during early embryogenesis.
The authors wish to thank Mr Donald Howard for his technical assistance. This investigation was supported in part by grants from the National Heart and Lung Institute HL5600
and HL7542.
REFERENCES
H. B. & BREIVIK, H. (1972). Murine Haematopoiesis Studied With The Diffusion
Chamber Technique. Norwegian Defence Research Establishment Report no. 61.
BESSLER, H., NOTHI, I., FISHMAN, P. & DJALDETTI, M. (1976). Erythropoietic events in cultured embryonic mouse spleen. Blood 48, 419-424.
COLE, R. J., HUNTER, J. & PAUL, J. (1968). Hormonal regulation of pre-natal haemoglobin
synthesis by erythropoietin. Br. J. Haematol. 14, 477-488.
COLE, R. J. & PAUL, J. (1966). The effects of erythropoietin on haem synthesis in mouse yolk
sac and cultured foetal liver cells. /. Embryol. exp. Morph. 15, 245-260.
DJALDETTI, M., BESSLER, H. & RIFKIND, R. A. (1972). Hematopoiesis in the embryonic
mouse spleen: An electron microscopic study. Blood39, 826-841.
DJALDETTI, M., CHUI, D., MARKS, P. A. & RIFKIND, R. A. (1970). Erythroid cell development
in fetal mice: stabilization of the hemoglobin synthetic capacity. /. molec. Biol. 50, 345358.
JONES, R. O. (1970). Ultrastructural analysis of hepatic haematopoiesis in the foetal mouse.
J. Anat. 107, 301-314.
METCALF, D. & MOORE, M. A. S. (1971). Haemopoietic Cells, pp. 173-271. New York:
American Elsevier.
MOORE, M. A. S. & METCALF, D. (1970). Ontogeny of the haemopoietic system: yolk sac
origin of in vivo and in vitro colony forming cells in the developing mouse embryo. Br. J.
Haematol. 18, 279-296.
OWEN, J. J. T. (1972). The origins and development of lymphocyte populations. In Ontogeny
of Acquired Immunity (ed. R. Port & J. Knight), pp. 35-64. Amsterdam: Associated Scientific Published.
OWEN, J. J. T., RAFF, M. C. & COOPER, M. D. (1975). Studies on the generation of B lymphocyte in the mouse embryo. Eur. J. Immunol. 5, 468-473.
PAUL, J., CONKIE, D. & FRESHNEY, R. I. (1969). Erythropoietic cell population changes
during the hepatic phase of erythropoiesis in the foetal mouse. Cell Tissue Kinet. 2,
283-294.
Pozzi, L. V., ANDREOZZI, U. & SILINI, G. (1972). Colony-forming units in fetal spleen and
peripheral blood. Acta Haemat. 48, 337-346.
RIFKIND, R. A., BANK, A. & MARKS, P. A. (1974). Erythropoiesis. In The Red Blood Cell,
2nd edn, vol. 1 (ed. D. M. N. Surgeon), pp. 51-89. New York: Academic Press, Inc.
RIFKIND, R. A., CHUI, D. & EPLER, H. (1969). An ultrastructural study of early morphogenetic events during the establishment of fetal hepatic erythropoiesis. /. Cell Biology, 40,
343-365.
BENESTAD,
166
S. R. WEINBERG, M. E. CHOVANIEC AND F. STOHLMAN
F., QUESENBERRY, P. J. & TYLER, W. S. (1973). The regulation of myelopoiesis as approached with in vivo and in vitro techniques. Progress in Hematology 68,
259-297.
SYMANN, M., FONTEBUONI, A., QUESENBERRY, P., HOWARD, D. & STOHLMAN, JR., F. (1976).
Fetal hemopoiesis in diffusion chamber cultures. I. The pattern of pluripotent stem cell
growth. Cell Tissue Kinet. 9, 41-49.
TARBUTT, R. G. & COLE, R. J. (1970). Cell population kinetics of erythroid tissue in the liver
of foetal mice. /. Embryol. exp. Morph. 24,429-446.
VILPO, J. A. & VILPO, L. (1976). Growth of haematopoietic cells of mouse fetal liver in diffusion chambers. Ada Haemat. 55, 224-229.
WEINBERG, S. R. & STOHLMAN, JR., F. (1976a). Growth of mouse yolk sac cells cultured in
vivo. Br. J. Haemat. 32, 543-555.
WEINBERG, S. R. & STOHLMAN, JR., F. (19766). Radiosensitivity of hematopoietic stem cells
cultured in diffusion chamber cultures of the murine yolk sac and adult medullary tissue.
Radiat. Res. 68, 84-92.
WEINBERG, S. R. & STOHLMAN, JR., F. (1977). Factors regulating yolk sac hematopoiesis in
diffusion chambers: various types of sera, cyclophosphamide, irradiation and long-term
culture. Expl Hemat. (In the Press.)
STOHLMAN, JR.,
{Received 15 December 1976, revised 28 January 1977)