DEVELOPMENTAL
141,344-352
BIOLOGY
(1990)
Isolation of Embryonic Stem (ES) Cells in Media Supplemented with Recombinant Leukemia Inhibitory Factor (LIF) SHIRLEY PEASE,**~ PAOLA BRAGHETTA, *J DAVID $X&wig
GEARING,?~
DIANNE
GRAIL,+ AND R. LINDSAY WILLIAMS*$~
Institute for Cancer Research, Melbourne Tumor Biology Branch, PO Royal Melbourne Hospital, Victoria 3050, Australia; *European Molecular Biology Laboratory, Meyerhofstrasse 1,690O Heidelberg, Federal Republic of Germany; a$ 7 Walter and Eliza Hall Institute of Medical Research, PO Royal Melbourne Hospital, Victoria 3050, Australia Accepted May 31, 1990
The isolation of pluripotent murine embryonic stem (ES) cells has previously been achieved by coculturing the ES cells with fibroblast feeder cells. In this report we demonstrate that ES cell lines can be isolated from murine 1290~ He blastocysts in the absence of feeder cells in culture medium supplemented with recombinant leukemia inhibitory factor (LIF). Three of the ES cell lines (MBL-1, MBL-2, and MBL-3) were isolated by directly explanting blastocysts, whilst two ES cell lines (MBL-4 and MBL-5) were isolated from blastocysts pretreated by immunosurgery. Three of the ES cell lines contained the Y chromosome (MBL-1, MBL-2, and MBL-5) with a high proportion of the cells displaying a normal diploid karyotype with a modal chromosome number of 40. All of the ES cell lines tested expressed the stem cell markers ECMA-7 and alkaline phosphatase, which were lost on removal of LIF when the ES cells differentiated into a variety of cell types. The full developmental potential of the ES cells was determined by injecting cells from two of the independently derived ES cell lines, MBL-1 and MBL-5, into C57BL/6J blastocysts. A high proportion of the pups born were chimeric as judged by coat pigmentation. Subsequent breeding established that the ES cells had contributed to the germ line. These results demonstrate that feeder cells are not essential for the isolation of pluripotent ES cell lines. o is9o Academic
Press, Inc.
vide the optimal conditions for ES cell isolation and maintenance is crucial. ES cell lines capable of producing germ-line chimeras have been isolated from blastocysts (Doetschman et aZ., 1985; Gossler et aZ., 1986) and blastocysts in implantational delay (Evans and Kaufman, 1981; Bradley et aZ., 1984). The isolation of ES cell lines from morulae has also been described (Eistetter, 1989), however, the ability of such cells to form chimeras has yet to be demonstrated. ES cell lines have also been isolated from embryos which had been treated by immunosurgery to remove extraneous cells (Martin, 1981) or following explanting of whole embryos into culture (Evans and Kaufman, 1981). However all of these techniques use a feeder cell layer of fibroblasts to prevent the differentiation of the ES cells. It has been suggested that the addition of media conditioned by teratocarcinoma cells (Martin, 1981) or buffalo rat liver (BRL) cells (Handyside et ab, 1989) aids the isolation of new ES cell lines, although the presence of feeder cells was apparently still crucial. Once ES cell lines have been established in culture the feeder cells can be removed provided the culture medium is supplemented with medium conditioned by certain tumor cell lines, such as BRL or 5637 cells. This indicates that there is a soluble differentiation-inhibiting activity (DIA) which prevents the differentiation of established ES cell lines (Smith and Hooper, 1987). Recently, we and others discovered that leukemia inhibi-
INTRODUCTION
Embryonic stem (ES)5 cells isolated from murine preimplantation embryos can be maintained as stable cell lines in culture and following the appropriate stimuli in vitro or in viva will differentiate into a variety of cell types. Thus ES cells can be used to study embryonic growth control and differentiation in culture (Doetschman et ah, 1985). Furthermore, ES cells provide an important method for the introduction of both dominant (Williams et ah, 1988a) and recessive (Zijlstra et ah, 1989; Schwartzberg et al, 1989) genetic mutations into mice. In all experiments utilizing ES cells it is desirable that the starting population of ES cells is both karyotypically normal and pluripotent. Therefore a thorough understanding of the factors which are needed to pro’ Present address: Glaxo Group Research Ltd, Greenford Road, Greenford, Middlesex UB6 OHE, UK. a Present address: Instituto di Istologia ed Embriologia, Universita di Padova, Via Falloppio 50, I-35100 Padova, Italy. a Present address: Immunex Corporation, 51 University Street, Seattle, Washington 98101, USA. ’ To whom correspondence should be addressed at the Ludwig Institute for Cancer Research, PO Royal Melbourne Hospital, Victoria 3050, Australia. ’ Abbreviations used: ES, embryonic stem; LIF, leukemia inhibitory factor; BRL, buffalo rat liver; DIA. differentiation-inhibiting activity. 0012-1606/90 $3.00 Copyright All rights
0 1990 by Academic Press, Inc. of reproduction in any form reserved.
344
Isolation of ES Cell Lines in LIF
PEASE ET AL.
FIG. 1. Isolation and culture of MBL ES cell lines. Blastocysts from 129&v He mice were placed in ES cell culture media (see Methods). Within 6 days the inner cell mass form a distinct clump of cells on the adherent trophectoderm cells (a). Phase contrast microscopy of typical MBL ES cell colonies at passage 2 (b). When the ES cells were prevented from adhering in the absence of LIF they formed embryoid bodies (c). Once attached to the dish differentiated cells could be cultured from the embryoid bodies (d).
tory factor (LIF) shared a number of features in common with DIA (Williams et ah, 198813; Smith et aZ., 1988; Gough and Williams, 1989). Using ES cell lines previously isolated on feeders we were able to demonstrate that purified recombinant LIF prevents the difTABLE ISOLATION
OF 129/Sv CONTAINING
1
HE ES CELL LINES IN MEDIA RECOMBINANT LIF
Number of blastocysts Method of isolation
Isolated
Attached
Number of ES cell lines isolated
Explanted Immunosurgery
9 7
9 5
4 2
Note. Blastocysts were either placed directly into ES cell culture media or treated first by immunosurgery as described under Materials and Methods. The number of blastocysts that attached to the cell culture dish and the number of cells with ES morphology that survived the first passage are indicated.
ferentiation of ES cells in a dose-dependent manner (Williams et aZ., 198813; Gearing et aZ., 1989). Importantly, ES cells maintained in LIF retained the ability to form germ-line chimeric mice indicating that LIF is sufficient to maintain the developmental potential of ES cell lines previously isolated and maintained on feeders (Williams et aa, 1988b). However the ability of LIF to replace feeder cells during the isolation of ES cell lines has not been tested. In this report we demonstrate that LIF can substitute for feeder cells in the isolation of ES cell lines and that cell lines isolated under these conditions can subsequently form germ-line chimeras. MATERIALS
Embryo
Isolation
AND METHODS
and Manipulation
Mice were maintained on a 12-hr light/l2-hr dark cycle with a dark cycle midpoint at 1:00 AM. Blastocysts were collected by flushing the uterine horns 4 days after natural mating had taken place (as judged by the pres-
346
DEVELOPMENTALBIOLOGY VOLUME141,199O TABLE
2
KARYOTYPEOFTHEMBL
CELLLINES Chromosome
Cell line
Karyotype
MBL-1 MBLQ MBL-4 MBL-5 Note.
Passage
XY XY x0 XY ES cell chromosome
Modal number (percentage)
number 8 14 11 11
spreads
were
40 40 39 40
prepared
and examined
as described
ence of a vaginal plug on the morning of Day 1). Embryos used to isolate ES cells were obtained by mating inbred 129&v He mice. Host embryos for injecting with ES cells were from inbred C57BLN mice. Pseudopregnant foster mothers were obtained by mating vasectomized outbred ICR mice with (C57BL/6J X SJL) Fl mice and the manipulated embryos transferred to these mice 3 days after mating as described in Hogan et al. (1986). Embryo and ES Cell Culture Embryos were isolated and cultured in M2 and Ml6 culture medium, respectively. ES cell isolation and cul-
05
Maximum
and
minimum
values
for,
z 100 z E 90 $ .-; so E z 5 m
." E E G 0.
(89) (48) (69) (89) under
counts
38
39
40
41
42
1 0 0 0
3 1 20 3
32 14 9 25
0 8 1 2
0 5 0 0
Materials
and Methods.
ture was carried out in ES cell culture medium consisting of Dulbecco’s or Glasgow’s modified Eagles medium supplemented with 15% fetal calf serum, 0.1 mM@-mercaptoethanol and 1000 units/ml (10 rig/ml) recombinant LIF. All tissue culture dishes were preincubated with 0.1% gelatin in phosphate-buffered saline for a minimum of 30 min at room temperature. The fetal calf serum (Sebak GmbH, D-8359 Aidenbach, FRG) and recombinant LIF were tested to ensure that they could support the clonal growth of established ES cell lines. Recombinant human LIF was prepared by expression as a fusion protein with glutathione S-transferase in Escher&a coli, purified on a glutathione-agarose affinity matrix and biologically active LIF released by cleavage with thrombin (Gearing et al, 1989). For chromosome and karyotype analysis metaphase spreads were prepared essentially as described by Robertson (1987) except that the concentration of Colcemid was increased to 0.1 pg/ml.
70
Establishment
60
Blastocysts from 129/Sv He mice were either directly explanted into ES cell culture media or first pretreated by immunosurgery. Immunosurgery was carried out as described in Solter and Knowles (1975). Briefly, blastocysts were cultured for 1 day in Ml6 to allow hatching. The embryos were then treated with rabbit anti-mouse antibodies (gift from S. Hunter) for 20 min; the blastocysts were washed with Ml6 and then treated with guinea pig complement to lyse the trophectoderm. The inner cell mass cells were incubated for a further 24 hr in ES cell media and again treated by immunosurgery.
0 1000
250
60 Units
15 of LIF per
4
0
ml
FIG. 2. Differentiation of MBL ES cell lines. MBL-1, MBL-2, and MBL-5 ES cell lines previously isolated and maintained in culture media containing 1000 units/ml of recombinant LIF for l&21, and 17 passages, respectively, were plated out as single cells (250 cells per 35-mm dish). After 24 hr the media was changed to vary the concentration of recombinant LIF from 0 to 1000 units/ml. After 7 days the majority of the colonies examined consisted of distinct stem or differentiated cell populations (see Fig. 3). The few colonies consisting of mixed cell populations were characterized according to the relative proportions of each cell type.
of ES Cell Lines
Analysis of Stem and Bferentiated
Cell Markers
To test the ability of the ES cells to differentiate in vitro, the cells were plated out and incubated in decreasing concentrations of recombinant LIF. On Day 1 ES cells were plated out onto 35-mm dishes (250 cells per dish) in 1000 units/ml of recombinant LIF. The following day the media was changed to media containing de-
PEASE ET AL.
Isolation of ES Cell Lines in LIF
347
colony morphology and staining for alkaline phosphatase. ES cell colonies incubated with or without 1000 units/ml recombinant LIF were further characterized by immunofluorescence carried out as described in Rudnicki and McBurney (1987) using the ECMA-7 and TROMA-1 monoclonal antibodies (Kemler, 1980; Brulet et cd, 1980). Production
of Chimeric
Mice
Chimeric mice were produced as follows. A single-cell suspension of ES cells was prepared by collection of individual colonies in phosphate-buffered saline, treatment for 5 min in 0.5 mM EGTA, and incubation in trypsinEDTA solution (with 1% chick serum) for a further 5 min at 4°C. Between 15 and 20 ES cells (in Dulbecco’s modified Eagles medium with 10% fetal calf serum, 3000 units/ml DNase 1 [Sigma] buffered in 20 mMHepes [pH 81) were injected into each blastocyst and subsequently transferred into pseudopregnant foster mothers. Chimeric mice were identified by coat pigmentation. RESULTS
Isolation
FIG. 3. Morphology of MBL ES cells in the presence and absence of recombinant LIF. MBL-1~18 ES cells were transferred to media containing 1000 units/ml recombinant LIF (a) or to normal culture media (b) as described in Fig. 2. After 7 days the cultures were stained to detect alkaline phosphatase which is expressed by the stem cells but not their differentiated derivatives (see Methods). Differentiated cells were then stained using hemotoxylin. Compact stem cell colonies could also be distinguished from diffuse differentiated colonies.
creasing concentrations of LIF. Duplicate dishes were prepared for each of the ES cell lines tested. On Day 7 the cells were fixed and the number of colonies containing stem cells estimated by assaying for alkaline phosphatase (Sigma diagnostic kit No. 86, Sigma, St. Louis, MO) which is expressed in the ES cells but not their differentiated derivatives. Differentiated ES cells were counterstained with hemotoxylin. ES cell colonies were classified as stem or differentiated colonies according to
of Embryonic
Stem Cells
Blastocysts were obtained from naturally mated 1291 Sv He mice and ES cell lines isolated by two different methods. In the first method blastocysts were placed in ES cell culture media containing 1000 units/ml (10 ng/ ml) recombinant LIF. Within 48 hr the trophectoderm had attached to the tissue culture dish. After a further 4 days the inner cell mass had formed a distinct clump of cells on the trophectoderm cells (Fig. la). These clumps were picked using a glass capillary, trypsinized to dissociate the clump, and plated into lo-mm diameter tissue culture dishes. In this experiment nine inner cell masses were transferred and this stage was designated as passage 0. Within 6 to 13 days, four of the wells contained 5-8 distinct ES cell colonies (Table 1). One of the ES cell lines was lost after one passage whilst the remaining three ES cell lines, designated MBL-1, MBL-2, and MBL-3, were cultured for further analysis (Fig. lb). The other technique used to isolate ES cells was immunosurgery. Using this method the trophectoderm of hatched blastocysts is destroyed using anti-mouse antibodies and complement. The inner cell mass is then maintained in culture for 24 hr and again treated with anti-mouse antibodies and complement to remove the outer primitive endoderm cells and leave the pluripotent epiblast cells. The following day the remaining cells were then dispersed following trypsinization and plated into ES cell media on lo-mm dishes. After maintaining the dispersed cells for a further 19 to 22 days, stem cell
348
DEVELOPMENTAL BIOLOGY
VOLUME 141,199O
TROMA-1
ECMA-7
LIF
a
No LIF
d FIG. 4. Immunofluorescence of MBL-1 ES cells. MBL-1 ES cells were transferred to media containing 1000 units/ml recombinant LIF (a, b) or normal culture media (c, d) as described in Fig. 2. After 7 days the colonies were fixed and immunofluorescence carried out on the two colony types using the ECMA-7 and TROMA-1 monoclonal antibodies (see Methods).
colonies were identified in two wells (Table 1). These were expanded for analysis and designated MBL-4 and MBL-5 at passage 1. fiflerentiation
of MBL ES Cells in Vitro
In the presence of LIF the ES cells remained as normal stem cells which were cultured for over 25 passages (over 75 days in culture) and displayed no visible changes in their growth characteristics. To investigate the ability of the MBL ES cells to differentiate, MBL-1, MBL-2, and MBL-5 cells were cultured in ES cell medium without LIF in bacterial petri dishes to prevent attachment to the dish. Within 11 days the cells formed simple embryoid bodies and after 26 days, complex cystic embryoid bodies had formed (Fig. lc). When transferred to a normal tissue culture dish the embryoid bodies attached to the surface of the dish and several differentiated cell types became apparent including cardiac and skeletal muscle and fibroblastic cells (Fig. Id).
To investigate the consequence of removal of LIF from the cell culture medium ES cells were plated out as single cells in media containing varying concentrations of LIF. After 7 days, stem cell colonies were identified as compact colonies of small cells which expressed alkaline phosphatase, which is expressed by ES cells but not their differentiated derivatives (Fig. 2). Decreasing the concentration of LIF in the cell culture medium resulted in an increase in the number of cell colonies containing large, flat differentiated cells (Fig. 3). In the absence of LIF over 80% of the colonies were differentiated as assessed by cell morphology and lack of the alkaline phosphatase activity. To confirm that the MBL ES cells displayed normal stem cell markers, immunofluorescence was carried out using the ECMA-7 and TROMA-1 antibodies which recognize a stem-cell-specific surface-antigen and the keratin-like filaments present in differentiated ES cells, respectively (Kemler, 1980; Brulet et al, 1980). Stem
PEASE ET AL.
analysis of MBL-5 FIG. 5. Karyotype chromosome complement.
Iso!.atim of ES Cell Lines in LIF
ES cells. Karyotype analysis of the MBL-5
cells maintained in 1000 units/ml LIF retained their stem cell phenotype as judged by their ability to bind to ECMA-‘7 but not TROMA-1 (Figs. 4a and 4b). Cells cultured in the absence of LIF for 7 days bound TROMA-1 but failed to bind ECMA-7 suggesting that these cells were differentiated (Figs. 4c and 4d). Characterization
of MBL ES Cell Lines
It is generally accepted that only ES cells with normal karyotypes are able to contribute to the germ line of
ES cell line at passage 11 showing a normal XY euploid
chimeric mice. Furthermore it has been demonstrated that XY ES cell lines are more stable than XX ES cell lines in culture (Robertson et cd, 1983). To identify XY ES cell lines DNA was prepared from each of the five MBL cell lines and hybridized with a radioactive labeled probe of Y353B DNA fragment, which specifically recognizes repetitive sequences present on the Y chromosome. Three of the cell lines (MBL-1, MBL-2, and MBL-5) were positive for the Y chromosome (data not shown).
350
DEVELOPMENTAL BIOLOGY TABLE 3a GENERATION OF CHIMERIC MICE WITH MBL-1 ES CELLS
Passage Blastocysts number transferred 14 15 16
17 Total
Mice born
Chimeras
Male chimeras
19 7
9 7
5
3
2
5 20
3 13
4 3 4
2 2
51
32(63%)
16(50%)
9(56%)
Estimated % of ES cell contribution 90,70,60,30,30 40, 40, 20, 20 80, 25, 20 90, 40, 30, 10
Chromosome counts of the ES cells revealed that the three ES cell lines containing the Y chromosome (MBL1, MBL-2, and MBL-5) all had a modal number of 40. Indeed 89% of the MBL-1 and MBL-5 cells had the correct modal number (Table 2). Karyotype analysis by G-banding revealed that the MBL-1 and MBL-5 cells with the modal number of chromosomes had a normal chromosome content with no visible rearrangements (Fig. 5). The MBL-2 ES cells had a high proportion of cells with more than 40 chromosomes; this was not due to one rearrangement but rather to a number of separate mutations including duplication of the Y chromosome, trisomy of chromosome 8 and trisomy of chromosome ‘7. The larger number of mutations in the MBL-2 cell line may reflect the high passage number, passage 14, when the karyotype analysis was carried out. To examine the karyotype of ES cells lacking the Y chromosome the MBL-4 ES cell line was karyotyped. The MBL4 cells were predominantly X0 with a modal chromosome number of 39 (Table 2). The X0 status of the
TABLE 3b GENERATION OF CHIMERIC MICE WITH MBL-5 ES CELLS Passage Blastocysts number transferred
Mice born
Chimeras
Male chimeras
13 17
14 63
1 22
1 8
1 6
Total
77
23 (29%)
9 (39%)
7 (78%)
Estimated % of ES cell contribution 50 4 x 100,60 50, 40, 10
Note. MBL-1 and MBL-5 ES cells isolated and cultured in ES cell media containing 1000 units/ml (10 rig/ml) recombinant LIF, for the indicated passage number, were injected into C57BL/6J blastocysts which were then transferred into pseudopregnant foster mothers. Only foster mothers which became pregnant were counted in estimating the percentage of progeny born. Chimeric pups were identified by the presence of agouti coat pigmentation from ES-derived cells as the cells derived from the C5’7BL/6J blastocysts produce black hairs. The ES cell contribution in individual chimeras was estimated from the proportion of the coat with agouti pigmentation.
VOLUME 141.1890 TABLE 4a GERM-LINE TRANSMISSION OF MBL-1 ES-DERIVED CELLS
Passage number of MBL-1 cells
Estimated % of ES cell contribution
Number of offspring Total
Black
Agouti
14 14 14 15 17
60 30 30 40 40
31 20 21 29 18
7 20 21 29 10
24 0 0 0 8
MBL-4 cells may suggest that this ES cell line was originally XX but had lost one copy of the X chromosome as the partial or complete deletion of one copy of the X chromosome in ES cells which were originally typed as XX has been reported previously (Robertson et aZ., 1983).
Production
of Germ-Line
Chimeras
The most crucial test for the developmental potential of ES cells is their ability to form viable germ-line chimeras. The MBL-1 and MBL-5 ES cell line were selected to make chimeras due to the high number of karyotypitally normal cells in these independently isolated ES cell lines. The MBL-1 and MBL-5 ES cells were injected into preimplantation embryos isolated from C57BL/6J mice at the blastocyst stage. C57BL/6J mice were selected as embryo donors as previous work had demonstrated that these mice were the best strain for the efficient formation of germ-line chimeras when using 129/ Sv derived ES cell lines (Williams et al, 1988b; Pease and Williams, 1990). Since the ES cells were isolated from 129&v He mice which carry the dominant agouti allele and C57BL/6J mice are homozygous recessive at
TABLE 4b GERM-LINE TRANSMISSION OF MBL-5 ES-DERIVED CELLS Passage number of MBL-5 cells
Estimated % of ES cell contribution
Number of offspring Total
Black
13 17 17 17
50 100 10 60
7 11 12 10
6 0 4 10
Agouti 1 10 6 0
Note. To assess the contribution of MBL ES cells, cultured for the indicated passage number, to the germ cells of male MBL-l.C57BL/6J and MBL-5.C57BL/6J chimeras they were mated to C57BL/6J females. The pups born were then examined for agouti coat color which indicates that the germ cells were derived from MBL-1 or MBL-5 ES cells.
PEASEETAL.
351
Isolation of ES Cell Lines in LIF
the same locus it is possible to phenotypically identify chimeric offspring by the presence of agouti coat coloration. Of the offspring analyzed, 50 and 39% of the mice were MBL-1 and MBL-5 chimeras, respectively (Tables 3a and 3b). In individual mice chimerism was variable; with some chimeras agouti coat contribution was as low as 10% whilst in others all of the coat appeared to be agouti and so derived from ES cells. Although all of the chimeras appeared normal some of the mice, including two with a 90% contribution from MBL-l-derived cells, were lost before they could be mated. These losses resulted primarily from infantile diarrhea, a condition to which 129 mice appear to be particularly susceptible. All the remaining male chimaeras were mated to C57BL/6J mice. Two of the five MBL-1 chimeras, which had an ES cell contribution of between 30 and 60%, contained ESderived germ cells as assessed by the presence of agouti pups (Table 4a). Furthermore of the four male MBL-5 chimeras mated to date three had agouti progeny, again indicating that the ES cells had contributed to the germ cells in these mice (Table 4b). DISCUSSION
In this report we have demonstrated that karyotypitally normal ES cell lines can be isolated in the absence of feeder cells if the cell culture media is supplemented with recombinant LIF. These ES cells retain their ability to differentiate in vitro in the absence of LIF. Most importantly the ability of both the MBL-1 and MBL-5 ES cells to form germ-line chimeras indicates that they are developmentally pluripotent. These results clearly eliminate the possibility that feeder cells contribute to the isolation of ES cells either by the production of factors other than LIF or by other more esoteric mechanisms such as the interchange of genetic material between the two cell types. Our ability to isolate ES cells following immunosurgery in the absence of feeders suggests that cell-cell interaction between other embryonic cell types is not critical in the isolation of new ES cell lines. The XY MBL-5 ES cells which were isolated by immunosurgery retained their complete developmental potential as judged by the ability of these cells to form germ-line chimeras. One reason behind our attempts to isolate ES cell lines in the presence of LIF was that under certain circumstances feeder cells produce suboptimal quantities of LIF. We have demonstrated that the amount of LIF present in culture medium conditioned by feeder cells can be as low as 500 units/ml (5 rig/ml) which is below the optimal concentration (1000 units/ml; 10 rig/ml) necessary to culture some ES cells (Williams et ah, 1988b). Moreover, following the addition of fresh medium to
feeder cells, it is likely that the level of LIF in the medium will initially be even lower until the medium has been “conditioned” by the feeder cells. From the studies described here it appears that ES cell lines can be isolated in the presence of LIF at frequencies comparable to that achieved to using feeder cells (Evans and Kaufman, 1981; Martin, 1981; Doetschman et a& 1985). Furthermore the developmental potential of the MBL-1 and MBL-5 ES cells in viva is similar to ES cell lines isolated with feeders (Bradley et al., 1984; Gossler et al., 1986; Williams et aZ., 198813). The ability to isolate and maintain ES cells in the absence of feeders and conditioned media will greatly facilitate studies of ES cells. It should now be possible to carry out detailed studies on ES cell cultures to identify other factors and gene products that influence the control of proliferation and differentiation in these early embryonic cells. We thank Erwin Wagner and Thomas Graf for supporting this project; Rolf Kemler for the ECMA-7 and TROMA-1 antibodies; Susan Hunter for the rabbit anti-mouse antibodies; and David Stoddart and Elaine Major for maintaining the mouse colonies. We are grateful to Phil Wilcox and Natalie Danford for assistance with the chromosome and karyotype analysis and our colleagues at the Ludwig Institute for Cancer Research and Walter and Eliza Hall Institute for advice and critical reading of the manuscript. Work at the Walter and Eliza Hall Institute was supported by the National Health and Medical Research Council (Canberra), the Anti-Cancer Council of Victoria, and AMRAD Corporation (Melbourne). REFERENCES BRADLEY, A., EVANS, M., KAUFMAN, M. H., and ROBERTSON,E. (1984). Formation of germ-line chimaeras from embryo-derived teratocarcinema cell lines. Nature (Lvndm) 309,255-256. BRULET, P., BABINET, C., KEMLER, R., and JACOB, F. (1980). Monoclonal antibodies against trophectoderm-specific markers during mouse blastocyst formation. Proc. Natl. Acad. Sci. USA 77.4113-4117. CAPECCHI, M. R. (1989). Altering the genome by homologous reeombination. Science 244,1288-1292. DOETSCHMAN, T. C., EISTETTER, H., KATZ, M., SCHMIDT, W., and KEMLER, R. (1985). The in vitro development of blastocyst derived embryonic stem cell lines: Formation of visceral yolk sac, blood islands, and myocardium. J. Endnyol. Exp. Morphol. 87,27-45. EVANS, M. J., and KAUFMAN, M. H. (1981). Establishment in culture of pluripotential cells from mouse embryos. Nature (London) 292,154156. EISTETTER, H. R. (1989). Pluripotent embryonal stem cell lines can be established from disaggregated mouse morulae. Dev. Growth mer. 31,275-282. GEARING, D., NICOLA, N. A., METCALF, D., FOOTE, S., GOUGH, N. M., and WILLIAMS, R. L. (1989). Production of leukemia inhibitory factor (LIF) in Escher&&a coli and its use in embryonic stem (ES) cell culture. Biotechnology 7,1157-1161. GOUGH, N. M., and WILLIAMS, R. L. (1989). The pleiotropic actions of Leukaemia Inhibitory Factor. Cancer Cells 1,77-80. GOSSLER, A., DOETSCHMAN, T., KORN, R., SERFLING, E., and KEMLER, R. (1986). Transgenesis by means of blastocyst-derived embryonic stem cell lines. Proc. Natl Acad Sci. USA 83,9065-9069. HANDYSIDE, A. H., O’NEILL, G. T., JONES, M., and HOOPER, M. L.
352
DEVELOPMENTAL BIOLOGY
(1989). Use of BRL-conditioned medium in combination with feeder layers to isolate a diploid embryonal stem cell line. Roux’s Arch. Dev. Biol. 198.48-55. HOGAN, B., COSTANTINI, F., and LACY, E. (1986). “Manipulating the Mouse Embryo, A Laboratory Manual.” Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. KEYLER, R. (1980). In “Progress in Developmental Biology Band 26” (H. W. Sayer, Ed.), p. 175 Fischer, Stuttgart, 1980. MARTIN, G. R. (1981). Isolation of a pluripotent cell line from early mouse embryos cultured in medium conditioned by teratocarcinoma stem cells. Proc Natl. Acad Sci. USA 78,7634-7638. PEASE, S., and WILLIAMS, R. L. (1990). Formation of germ-line chimaeras from embryonic stem cells maintained with recombinant myeloid leukaemia inhibitory factor. Exp. Cell Res. in press. ROBERTSON,E. J. (1987). Embryo-derived stem cell lines. In “Teratocarcinomas and Embryonic Stem Cells, A Practical Approach” (E. J. Robertson, Ed.), pp. 71-112. IRL Press, Oxford. ROBERTSON,E. J., EVANS, M. J., and KAUFMAN, M. H. (1983). X-chromosome instability in pluripotential stem cell lines derived from parthenogenetic embryos. J. Ew&yol. Exp. Mmyhol. 74,297-309. RUDNICKI, M. A., and MCBURNEY, M. W. (198’7). Cell culture methods and induction of differentiation of embryonal carcinoma cell lines. In “Teratocarcinomas and Embryonic Stem Cells, A Practical Approach” (E. J. Robertson, Ed.), pp. 19-49. IRL, Oxford.
VOLUME 141,199O
SCHWARTZBERG,P. L., GOFF, S. P., and ROBERTSON,E. J. (1989). Germline transmission of a c-abl mutation produced by targeted gene disruption in ES cells. Science 246,799-803. SMITH, A. G., and HOOPER, M. L. (1987). Buffalo rat liver cells produce a diffusible activity which inhibits the differentiation of murine embryonal carcinoma and embryonic stem cells. Den Biol 121,1-9. SMITH, A. G., HEATH, J. K., DONALDSON, D. D., WONG, G. G., MOREAU, J., STAHL, M., and ROGERS, D. (1988). Inhibition of pluripotential embryonic stem cell differentiation by purified polypeptides. Nature (London) 336,688-690. SOLTER, D. and KNOWLES, B. B. (1975). Immunosurgery of mouse blastocysts. Proc. Natl. Acad. Sci. USA 72,5099-5102. WILLIAMS, R. L., COURTNEIDGE, S. A., and WAGNER, E. F. (1988a). Embryonic lethalities and endothelial tumours in chimaeric mice expressing polyoma virus middle T oncogene. Cell 52,121-131. WILLIAMS, R. L., HILTON, D. J., PEASE, S., WILLSON, T. A., STEWART, C. L., GEARING, D. P., WAGNER, E. F., METCALF, D., NICOLA, N. A., and GOUGH, N. M. (198813). Myeloid leukaemia inhibitory factor maintains the developmental potential of embryonic stem cells. Nature (London) 336,684-687. ZIJLSTRA, M., LI, E., SAJJADI, F., SUBRAMANI, S., and JAENISCH, R. (1989). Germ-line transmission of a disrupted B,-microglobulin gene produced by homologous recombination in embryonic stem cells. Nature (London) 342,435-438.
141,344-352
BIOLOGY
(1990)
Isolation of Embryonic Stem (ES) Cells in Media Supplemented with Recombinant Leukemia Inhibitory Factor (LIF) SHIRLEY PEASE,**~ PAOLA BRAGHETTA, *J DAVID $X&wig
GEARING,?~
DIANNE
GRAIL,+ AND R. LINDSAY WILLIAMS*$~
Institute for Cancer Research, Melbourne Tumor Biology Branch, PO Royal Melbourne Hospital, Victoria 3050, Australia; *European Molecular Biology Laboratory, Meyerhofstrasse 1,690O Heidelberg, Federal Republic of Germany; a$ 7 Walter and Eliza Hall Institute of Medical Research, PO Royal Melbourne Hospital, Victoria 3050, Australia Accepted May 31, 1990
The isolation of pluripotent murine embryonic stem (ES) cells has previously been achieved by coculturing the ES cells with fibroblast feeder cells. In this report we demonstrate that ES cell lines can be isolated from murine 1290~ He blastocysts in the absence of feeder cells in culture medium supplemented with recombinant leukemia inhibitory factor (LIF). Three of the ES cell lines (MBL-1, MBL-2, and MBL-3) were isolated by directly explanting blastocysts, whilst two ES cell lines (MBL-4 and MBL-5) were isolated from blastocysts pretreated by immunosurgery. Three of the ES cell lines contained the Y chromosome (MBL-1, MBL-2, and MBL-5) with a high proportion of the cells displaying a normal diploid karyotype with a modal chromosome number of 40. All of the ES cell lines tested expressed the stem cell markers ECMA-7 and alkaline phosphatase, which were lost on removal of LIF when the ES cells differentiated into a variety of cell types. The full developmental potential of the ES cells was determined by injecting cells from two of the independently derived ES cell lines, MBL-1 and MBL-5, into C57BL/6J blastocysts. A high proportion of the pups born were chimeric as judged by coat pigmentation. Subsequent breeding established that the ES cells had contributed to the germ line. These results demonstrate that feeder cells are not essential for the isolation of pluripotent ES cell lines. o is9o Academic
Press, Inc.
vide the optimal conditions for ES cell isolation and maintenance is crucial. ES cell lines capable of producing germ-line chimeras have been isolated from blastocysts (Doetschman et aZ., 1985; Gossler et aZ., 1986) and blastocysts in implantational delay (Evans and Kaufman, 1981; Bradley et aZ., 1984). The isolation of ES cell lines from morulae has also been described (Eistetter, 1989), however, the ability of such cells to form chimeras has yet to be demonstrated. ES cell lines have also been isolated from embryos which had been treated by immunosurgery to remove extraneous cells (Martin, 1981) or following explanting of whole embryos into culture (Evans and Kaufman, 1981). However all of these techniques use a feeder cell layer of fibroblasts to prevent the differentiation of the ES cells. It has been suggested that the addition of media conditioned by teratocarcinoma cells (Martin, 1981) or buffalo rat liver (BRL) cells (Handyside et ab, 1989) aids the isolation of new ES cell lines, although the presence of feeder cells was apparently still crucial. Once ES cell lines have been established in culture the feeder cells can be removed provided the culture medium is supplemented with medium conditioned by certain tumor cell lines, such as BRL or 5637 cells. This indicates that there is a soluble differentiation-inhibiting activity (DIA) which prevents the differentiation of established ES cell lines (Smith and Hooper, 1987). Recently, we and others discovered that leukemia inhibi-
INTRODUCTION
Embryonic stem (ES)5 cells isolated from murine preimplantation embryos can be maintained as stable cell lines in culture and following the appropriate stimuli in vitro or in viva will differentiate into a variety of cell types. Thus ES cells can be used to study embryonic growth control and differentiation in culture (Doetschman et ah, 1985). Furthermore, ES cells provide an important method for the introduction of both dominant (Williams et ah, 1988a) and recessive (Zijlstra et ah, 1989; Schwartzberg et al, 1989) genetic mutations into mice. In all experiments utilizing ES cells it is desirable that the starting population of ES cells is both karyotypically normal and pluripotent. Therefore a thorough understanding of the factors which are needed to pro’ Present address: Glaxo Group Research Ltd, Greenford Road, Greenford, Middlesex UB6 OHE, UK. a Present address: Instituto di Istologia ed Embriologia, Universita di Padova, Via Falloppio 50, I-35100 Padova, Italy. a Present address: Immunex Corporation, 51 University Street, Seattle, Washington 98101, USA. ’ To whom correspondence should be addressed at the Ludwig Institute for Cancer Research, PO Royal Melbourne Hospital, Victoria 3050, Australia. ’ Abbreviations used: ES, embryonic stem; LIF, leukemia inhibitory factor; BRL, buffalo rat liver; DIA. differentiation-inhibiting activity. 0012-1606/90 $3.00 Copyright All rights
0 1990 by Academic Press, Inc. of reproduction in any form reserved.
344
Isolation of ES Cell Lines in LIF
PEASE ET AL.
FIG. 1. Isolation and culture of MBL ES cell lines. Blastocysts from 129&v He mice were placed in ES cell culture media (see Methods). Within 6 days the inner cell mass form a distinct clump of cells on the adherent trophectoderm cells (a). Phase contrast microscopy of typical MBL ES cell colonies at passage 2 (b). When the ES cells were prevented from adhering in the absence of LIF they formed embryoid bodies (c). Once attached to the dish differentiated cells could be cultured from the embryoid bodies (d).
tory factor (LIF) shared a number of features in common with DIA (Williams et ah, 198813; Smith et aZ., 1988; Gough and Williams, 1989). Using ES cell lines previously isolated on feeders we were able to demonstrate that purified recombinant LIF prevents the difTABLE ISOLATION
OF 129/Sv CONTAINING
1
HE ES CELL LINES IN MEDIA RECOMBINANT LIF
Number of blastocysts Method of isolation
Isolated
Attached
Number of ES cell lines isolated
Explanted Immunosurgery
9 7
9 5
4 2
Note. Blastocysts were either placed directly into ES cell culture media or treated first by immunosurgery as described under Materials and Methods. The number of blastocysts that attached to the cell culture dish and the number of cells with ES morphology that survived the first passage are indicated.
ferentiation of ES cells in a dose-dependent manner (Williams et aZ., 198813; Gearing et aZ., 1989). Importantly, ES cells maintained in LIF retained the ability to form germ-line chimeric mice indicating that LIF is sufficient to maintain the developmental potential of ES cell lines previously isolated and maintained on feeders (Williams et aa, 1988b). However the ability of LIF to replace feeder cells during the isolation of ES cell lines has not been tested. In this report we demonstrate that LIF can substitute for feeder cells in the isolation of ES cell lines and that cell lines isolated under these conditions can subsequently form germ-line chimeras. MATERIALS
Embryo
Isolation
AND METHODS
and Manipulation
Mice were maintained on a 12-hr light/l2-hr dark cycle with a dark cycle midpoint at 1:00 AM. Blastocysts were collected by flushing the uterine horns 4 days after natural mating had taken place (as judged by the pres-
346
DEVELOPMENTALBIOLOGY VOLUME141,199O TABLE
2
KARYOTYPEOFTHEMBL
CELLLINES Chromosome
Cell line
Karyotype
MBL-1 MBLQ MBL-4 MBL-5 Note.
Passage
XY XY x0 XY ES cell chromosome
Modal number (percentage)
number 8 14 11 11
spreads
were
40 40 39 40
prepared
and examined
as described
ence of a vaginal plug on the morning of Day 1). Embryos used to isolate ES cells were obtained by mating inbred 129&v He mice. Host embryos for injecting with ES cells were from inbred C57BLN mice. Pseudopregnant foster mothers were obtained by mating vasectomized outbred ICR mice with (C57BL/6J X SJL) Fl mice and the manipulated embryos transferred to these mice 3 days after mating as described in Hogan et al. (1986). Embryo and ES Cell Culture Embryos were isolated and cultured in M2 and Ml6 culture medium, respectively. ES cell isolation and cul-
05
Maximum
and
minimum
values
for,
z 100 z E 90 $ .-; so E z 5 m
." E E G 0.
(89) (48) (69) (89) under
counts
38
39
40
41
42
1 0 0 0
3 1 20 3
32 14 9 25
0 8 1 2
0 5 0 0
Materials
and Methods.
ture was carried out in ES cell culture medium consisting of Dulbecco’s or Glasgow’s modified Eagles medium supplemented with 15% fetal calf serum, 0.1 mM@-mercaptoethanol and 1000 units/ml (10 rig/ml) recombinant LIF. All tissue culture dishes were preincubated with 0.1% gelatin in phosphate-buffered saline for a minimum of 30 min at room temperature. The fetal calf serum (Sebak GmbH, D-8359 Aidenbach, FRG) and recombinant LIF were tested to ensure that they could support the clonal growth of established ES cell lines. Recombinant human LIF was prepared by expression as a fusion protein with glutathione S-transferase in Escher&a coli, purified on a glutathione-agarose affinity matrix and biologically active LIF released by cleavage with thrombin (Gearing et al, 1989). For chromosome and karyotype analysis metaphase spreads were prepared essentially as described by Robertson (1987) except that the concentration of Colcemid was increased to 0.1 pg/ml.
70
Establishment
60
Blastocysts from 129/Sv He mice were either directly explanted into ES cell culture media or first pretreated by immunosurgery. Immunosurgery was carried out as described in Solter and Knowles (1975). Briefly, blastocysts were cultured for 1 day in Ml6 to allow hatching. The embryos were then treated with rabbit anti-mouse antibodies (gift from S. Hunter) for 20 min; the blastocysts were washed with Ml6 and then treated with guinea pig complement to lyse the trophectoderm. The inner cell mass cells were incubated for a further 24 hr in ES cell media and again treated by immunosurgery.
0 1000
250
60 Units
15 of LIF per
4
0
ml
FIG. 2. Differentiation of MBL ES cell lines. MBL-1, MBL-2, and MBL-5 ES cell lines previously isolated and maintained in culture media containing 1000 units/ml of recombinant LIF for l&21, and 17 passages, respectively, were plated out as single cells (250 cells per 35-mm dish). After 24 hr the media was changed to vary the concentration of recombinant LIF from 0 to 1000 units/ml. After 7 days the majority of the colonies examined consisted of distinct stem or differentiated cell populations (see Fig. 3). The few colonies consisting of mixed cell populations were characterized according to the relative proportions of each cell type.
of ES Cell Lines
Analysis of Stem and Bferentiated
Cell Markers
To test the ability of the ES cells to differentiate in vitro, the cells were plated out and incubated in decreasing concentrations of recombinant LIF. On Day 1 ES cells were plated out onto 35-mm dishes (250 cells per dish) in 1000 units/ml of recombinant LIF. The following day the media was changed to media containing de-
PEASE ET AL.
Isolation of ES Cell Lines in LIF
347
colony morphology and staining for alkaline phosphatase. ES cell colonies incubated with or without 1000 units/ml recombinant LIF were further characterized by immunofluorescence carried out as described in Rudnicki and McBurney (1987) using the ECMA-7 and TROMA-1 monoclonal antibodies (Kemler, 1980; Brulet et cd, 1980). Production
of Chimeric
Mice
Chimeric mice were produced as follows. A single-cell suspension of ES cells was prepared by collection of individual colonies in phosphate-buffered saline, treatment for 5 min in 0.5 mM EGTA, and incubation in trypsinEDTA solution (with 1% chick serum) for a further 5 min at 4°C. Between 15 and 20 ES cells (in Dulbecco’s modified Eagles medium with 10% fetal calf serum, 3000 units/ml DNase 1 [Sigma] buffered in 20 mMHepes [pH 81) were injected into each blastocyst and subsequently transferred into pseudopregnant foster mothers. Chimeric mice were identified by coat pigmentation. RESULTS
Isolation
FIG. 3. Morphology of MBL ES cells in the presence and absence of recombinant LIF. MBL-1~18 ES cells were transferred to media containing 1000 units/ml recombinant LIF (a) or to normal culture media (b) as described in Fig. 2. After 7 days the cultures were stained to detect alkaline phosphatase which is expressed by the stem cells but not their differentiated derivatives (see Methods). Differentiated cells were then stained using hemotoxylin. Compact stem cell colonies could also be distinguished from diffuse differentiated colonies.
creasing concentrations of LIF. Duplicate dishes were prepared for each of the ES cell lines tested. On Day 7 the cells were fixed and the number of colonies containing stem cells estimated by assaying for alkaline phosphatase (Sigma diagnostic kit No. 86, Sigma, St. Louis, MO) which is expressed in the ES cells but not their differentiated derivatives. Differentiated ES cells were counterstained with hemotoxylin. ES cell colonies were classified as stem or differentiated colonies according to
of Embryonic
Stem Cells
Blastocysts were obtained from naturally mated 1291 Sv He mice and ES cell lines isolated by two different methods. In the first method blastocysts were placed in ES cell culture media containing 1000 units/ml (10 ng/ ml) recombinant LIF. Within 48 hr the trophectoderm had attached to the tissue culture dish. After a further 4 days the inner cell mass had formed a distinct clump of cells on the trophectoderm cells (Fig. la). These clumps were picked using a glass capillary, trypsinized to dissociate the clump, and plated into lo-mm diameter tissue culture dishes. In this experiment nine inner cell masses were transferred and this stage was designated as passage 0. Within 6 to 13 days, four of the wells contained 5-8 distinct ES cell colonies (Table 1). One of the ES cell lines was lost after one passage whilst the remaining three ES cell lines, designated MBL-1, MBL-2, and MBL-3, were cultured for further analysis (Fig. lb). The other technique used to isolate ES cells was immunosurgery. Using this method the trophectoderm of hatched blastocysts is destroyed using anti-mouse antibodies and complement. The inner cell mass is then maintained in culture for 24 hr and again treated with anti-mouse antibodies and complement to remove the outer primitive endoderm cells and leave the pluripotent epiblast cells. The following day the remaining cells were then dispersed following trypsinization and plated into ES cell media on lo-mm dishes. After maintaining the dispersed cells for a further 19 to 22 days, stem cell
348
DEVELOPMENTAL BIOLOGY
VOLUME 141,199O
TROMA-1
ECMA-7
LIF
a
No LIF
d FIG. 4. Immunofluorescence of MBL-1 ES cells. MBL-1 ES cells were transferred to media containing 1000 units/ml recombinant LIF (a, b) or normal culture media (c, d) as described in Fig. 2. After 7 days the colonies were fixed and immunofluorescence carried out on the two colony types using the ECMA-7 and TROMA-1 monoclonal antibodies (see Methods).
colonies were identified in two wells (Table 1). These were expanded for analysis and designated MBL-4 and MBL-5 at passage 1. fiflerentiation
of MBL ES Cells in Vitro
In the presence of LIF the ES cells remained as normal stem cells which were cultured for over 25 passages (over 75 days in culture) and displayed no visible changes in their growth characteristics. To investigate the ability of the MBL ES cells to differentiate, MBL-1, MBL-2, and MBL-5 cells were cultured in ES cell medium without LIF in bacterial petri dishes to prevent attachment to the dish. Within 11 days the cells formed simple embryoid bodies and after 26 days, complex cystic embryoid bodies had formed (Fig. lc). When transferred to a normal tissue culture dish the embryoid bodies attached to the surface of the dish and several differentiated cell types became apparent including cardiac and skeletal muscle and fibroblastic cells (Fig. Id).
To investigate the consequence of removal of LIF from the cell culture medium ES cells were plated out as single cells in media containing varying concentrations of LIF. After 7 days, stem cell colonies were identified as compact colonies of small cells which expressed alkaline phosphatase, which is expressed by ES cells but not their differentiated derivatives (Fig. 2). Decreasing the concentration of LIF in the cell culture medium resulted in an increase in the number of cell colonies containing large, flat differentiated cells (Fig. 3). In the absence of LIF over 80% of the colonies were differentiated as assessed by cell morphology and lack of the alkaline phosphatase activity. To confirm that the MBL ES cells displayed normal stem cell markers, immunofluorescence was carried out using the ECMA-7 and TROMA-1 antibodies which recognize a stem-cell-specific surface-antigen and the keratin-like filaments present in differentiated ES cells, respectively (Kemler, 1980; Brulet et al, 1980). Stem
PEASE ET AL.
analysis of MBL-5 FIG. 5. Karyotype chromosome complement.
Iso!.atim of ES Cell Lines in LIF
ES cells. Karyotype analysis of the MBL-5
cells maintained in 1000 units/ml LIF retained their stem cell phenotype as judged by their ability to bind to ECMA-‘7 but not TROMA-1 (Figs. 4a and 4b). Cells cultured in the absence of LIF for 7 days bound TROMA-1 but failed to bind ECMA-7 suggesting that these cells were differentiated (Figs. 4c and 4d). Characterization
of MBL ES Cell Lines
It is generally accepted that only ES cells with normal karyotypes are able to contribute to the germ line of
ES cell line at passage 11 showing a normal XY euploid
chimeric mice. Furthermore it has been demonstrated that XY ES cell lines are more stable than XX ES cell lines in culture (Robertson et cd, 1983). To identify XY ES cell lines DNA was prepared from each of the five MBL cell lines and hybridized with a radioactive labeled probe of Y353B DNA fragment, which specifically recognizes repetitive sequences present on the Y chromosome. Three of the cell lines (MBL-1, MBL-2, and MBL-5) were positive for the Y chromosome (data not shown).
350
DEVELOPMENTAL BIOLOGY TABLE 3a GENERATION OF CHIMERIC MICE WITH MBL-1 ES CELLS
Passage Blastocysts number transferred 14 15 16
17 Total
Mice born
Chimeras
Male chimeras
19 7
9 7
5
3
2
5 20
3 13
4 3 4
2 2
51
32(63%)
16(50%)
9(56%)
Estimated % of ES cell contribution 90,70,60,30,30 40, 40, 20, 20 80, 25, 20 90, 40, 30, 10
Chromosome counts of the ES cells revealed that the three ES cell lines containing the Y chromosome (MBL1, MBL-2, and MBL-5) all had a modal number of 40. Indeed 89% of the MBL-1 and MBL-5 cells had the correct modal number (Table 2). Karyotype analysis by G-banding revealed that the MBL-1 and MBL-5 cells with the modal number of chromosomes had a normal chromosome content with no visible rearrangements (Fig. 5). The MBL-2 ES cells had a high proportion of cells with more than 40 chromosomes; this was not due to one rearrangement but rather to a number of separate mutations including duplication of the Y chromosome, trisomy of chromosome 8 and trisomy of chromosome ‘7. The larger number of mutations in the MBL-2 cell line may reflect the high passage number, passage 14, when the karyotype analysis was carried out. To examine the karyotype of ES cells lacking the Y chromosome the MBL-4 ES cell line was karyotyped. The MBL4 cells were predominantly X0 with a modal chromosome number of 39 (Table 2). The X0 status of the
TABLE 3b GENERATION OF CHIMERIC MICE WITH MBL-5 ES CELLS Passage Blastocysts number transferred
Mice born
Chimeras
Male chimeras
13 17
14 63
1 22
1 8
1 6
Total
77
23 (29%)
9 (39%)
7 (78%)
Estimated % of ES cell contribution 50 4 x 100,60 50, 40, 10
Note. MBL-1 and MBL-5 ES cells isolated and cultured in ES cell media containing 1000 units/ml (10 rig/ml) recombinant LIF, for the indicated passage number, were injected into C57BL/6J blastocysts which were then transferred into pseudopregnant foster mothers. Only foster mothers which became pregnant were counted in estimating the percentage of progeny born. Chimeric pups were identified by the presence of agouti coat pigmentation from ES-derived cells as the cells derived from the C5’7BL/6J blastocysts produce black hairs. The ES cell contribution in individual chimeras was estimated from the proportion of the coat with agouti pigmentation.
VOLUME 141.1890 TABLE 4a GERM-LINE TRANSMISSION OF MBL-1 ES-DERIVED CELLS
Passage number of MBL-1 cells
Estimated % of ES cell contribution
Number of offspring Total
Black
Agouti
14 14 14 15 17
60 30 30 40 40
31 20 21 29 18
7 20 21 29 10
24 0 0 0 8
MBL-4 cells may suggest that this ES cell line was originally XX but had lost one copy of the X chromosome as the partial or complete deletion of one copy of the X chromosome in ES cells which were originally typed as XX has been reported previously (Robertson et aZ., 1983).
Production
of Germ-Line
Chimeras
The most crucial test for the developmental potential of ES cells is their ability to form viable germ-line chimeras. The MBL-1 and MBL-5 ES cell line were selected to make chimeras due to the high number of karyotypitally normal cells in these independently isolated ES cell lines. The MBL-1 and MBL-5 ES cells were injected into preimplantation embryos isolated from C57BL/6J mice at the blastocyst stage. C57BL/6J mice were selected as embryo donors as previous work had demonstrated that these mice were the best strain for the efficient formation of germ-line chimeras when using 129/ Sv derived ES cell lines (Williams et al, 1988b; Pease and Williams, 1990). Since the ES cells were isolated from 129&v He mice which carry the dominant agouti allele and C57BL/6J mice are homozygous recessive at
TABLE 4b GERM-LINE TRANSMISSION OF MBL-5 ES-DERIVED CELLS Passage number of MBL-5 cells
Estimated % of ES cell contribution
Number of offspring Total
Black
13 17 17 17
50 100 10 60
7 11 12 10
6 0 4 10
Agouti 1 10 6 0
Note. To assess the contribution of MBL ES cells, cultured for the indicated passage number, to the germ cells of male MBL-l.C57BL/6J and MBL-5.C57BL/6J chimeras they were mated to C57BL/6J females. The pups born were then examined for agouti coat color which indicates that the germ cells were derived from MBL-1 or MBL-5 ES cells.
PEASEETAL.
351
Isolation of ES Cell Lines in LIF
the same locus it is possible to phenotypically identify chimeric offspring by the presence of agouti coat coloration. Of the offspring analyzed, 50 and 39% of the mice were MBL-1 and MBL-5 chimeras, respectively (Tables 3a and 3b). In individual mice chimerism was variable; with some chimeras agouti coat contribution was as low as 10% whilst in others all of the coat appeared to be agouti and so derived from ES cells. Although all of the chimeras appeared normal some of the mice, including two with a 90% contribution from MBL-l-derived cells, were lost before they could be mated. These losses resulted primarily from infantile diarrhea, a condition to which 129 mice appear to be particularly susceptible. All the remaining male chimaeras were mated to C57BL/6J mice. Two of the five MBL-1 chimeras, which had an ES cell contribution of between 30 and 60%, contained ESderived germ cells as assessed by the presence of agouti pups (Table 4a). Furthermore of the four male MBL-5 chimeras mated to date three had agouti progeny, again indicating that the ES cells had contributed to the germ cells in these mice (Table 4b). DISCUSSION
In this report we have demonstrated that karyotypitally normal ES cell lines can be isolated in the absence of feeder cells if the cell culture media is supplemented with recombinant LIF. These ES cells retain their ability to differentiate in vitro in the absence of LIF. Most importantly the ability of both the MBL-1 and MBL-5 ES cells to form germ-line chimeras indicates that they are developmentally pluripotent. These results clearly eliminate the possibility that feeder cells contribute to the isolation of ES cells either by the production of factors other than LIF or by other more esoteric mechanisms such as the interchange of genetic material between the two cell types. Our ability to isolate ES cells following immunosurgery in the absence of feeders suggests that cell-cell interaction between other embryonic cell types is not critical in the isolation of new ES cell lines. The XY MBL-5 ES cells which were isolated by immunosurgery retained their complete developmental potential as judged by the ability of these cells to form germ-line chimeras. One reason behind our attempts to isolate ES cell lines in the presence of LIF was that under certain circumstances feeder cells produce suboptimal quantities of LIF. We have demonstrated that the amount of LIF present in culture medium conditioned by feeder cells can be as low as 500 units/ml (5 rig/ml) which is below the optimal concentration (1000 units/ml; 10 rig/ml) necessary to culture some ES cells (Williams et ah, 1988b). Moreover, following the addition of fresh medium to
feeder cells, it is likely that the level of LIF in the medium will initially be even lower until the medium has been “conditioned” by the feeder cells. From the studies described here it appears that ES cell lines can be isolated in the presence of LIF at frequencies comparable to that achieved to using feeder cells (Evans and Kaufman, 1981; Martin, 1981; Doetschman et a& 1985). Furthermore the developmental potential of the MBL-1 and MBL-5 ES cells in viva is similar to ES cell lines isolated with feeders (Bradley et al., 1984; Gossler et al., 1986; Williams et aZ., 198813). The ability to isolate and maintain ES cells in the absence of feeders and conditioned media will greatly facilitate studies of ES cells. It should now be possible to carry out detailed studies on ES cell cultures to identify other factors and gene products that influence the control of proliferation and differentiation in these early embryonic cells. We thank Erwin Wagner and Thomas Graf for supporting this project; Rolf Kemler for the ECMA-7 and TROMA-1 antibodies; Susan Hunter for the rabbit anti-mouse antibodies; and David Stoddart and Elaine Major for maintaining the mouse colonies. We are grateful to Phil Wilcox and Natalie Danford for assistance with the chromosome and karyotype analysis and our colleagues at the Ludwig Institute for Cancer Research and Walter and Eliza Hall Institute for advice and critical reading of the manuscript. Work at the Walter and Eliza Hall Institute was supported by the National Health and Medical Research Council (Canberra), the Anti-Cancer Council of Victoria, and AMRAD Corporation (Melbourne). REFERENCES BRADLEY, A., EVANS, M., KAUFMAN, M. H., and ROBERTSON,E. (1984). Formation of germ-line chimaeras from embryo-derived teratocarcinema cell lines. Nature (Lvndm) 309,255-256. BRULET, P., BABINET, C., KEMLER, R., and JACOB, F. (1980). Monoclonal antibodies against trophectoderm-specific markers during mouse blastocyst formation. Proc. Natl. Acad. Sci. USA 77.4113-4117. CAPECCHI, M. R. (1989). Altering the genome by homologous reeombination. Science 244,1288-1292. DOETSCHMAN, T. C., EISTETTER, H., KATZ, M., SCHMIDT, W., and KEMLER, R. (1985). The in vitro development of blastocyst derived embryonic stem cell lines: Formation of visceral yolk sac, blood islands, and myocardium. J. Endnyol. Exp. Morphol. 87,27-45. EVANS, M. J., and KAUFMAN, M. H. (1981). Establishment in culture of pluripotential cells from mouse embryos. Nature (London) 292,154156. EISTETTER, H. R. (1989). Pluripotent embryonal stem cell lines can be established from disaggregated mouse morulae. Dev. Growth mer. 31,275-282. GEARING, D., NICOLA, N. A., METCALF, D., FOOTE, S., GOUGH, N. M., and WILLIAMS, R. L. (1989). Production of leukemia inhibitory factor (LIF) in Escher&&a coli and its use in embryonic stem (ES) cell culture. Biotechnology 7,1157-1161. GOUGH, N. M., and WILLIAMS, R. L. (1989). The pleiotropic actions of Leukaemia Inhibitory Factor. Cancer Cells 1,77-80. GOSSLER, A., DOETSCHMAN, T., KORN, R., SERFLING, E., and KEMLER, R. (1986). Transgenesis by means of blastocyst-derived embryonic stem cell lines. Proc. Natl Acad Sci. USA 83,9065-9069. HANDYSIDE, A. H., O’NEILL, G. T., JONES, M., and HOOPER, M. L.
352
DEVELOPMENTAL BIOLOGY
(1989). Use of BRL-conditioned medium in combination with feeder layers to isolate a diploid embryonal stem cell line. Roux’s Arch. Dev. Biol. 198.48-55. HOGAN, B., COSTANTINI, F., and LACY, E. (1986). “Manipulating the Mouse Embryo, A Laboratory Manual.” Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. KEYLER, R. (1980). In “Progress in Developmental Biology Band 26” (H. W. Sayer, Ed.), p. 175 Fischer, Stuttgart, 1980. MARTIN, G. R. (1981). Isolation of a pluripotent cell line from early mouse embryos cultured in medium conditioned by teratocarcinoma stem cells. Proc Natl. Acad Sci. USA 78,7634-7638. PEASE, S., and WILLIAMS, R. L. (1990). Formation of germ-line chimaeras from embryonic stem cells maintained with recombinant myeloid leukaemia inhibitory factor. Exp. Cell Res. in press. ROBERTSON,E. J. (1987). Embryo-derived stem cell lines. In “Teratocarcinomas and Embryonic Stem Cells, A Practical Approach” (E. J. Robertson, Ed.), pp. 71-112. IRL Press, Oxford. ROBERTSON,E. J., EVANS, M. J., and KAUFMAN, M. H. (1983). X-chromosome instability in pluripotential stem cell lines derived from parthenogenetic embryos. J. Ew&yol. Exp. Mmyhol. 74,297-309. RUDNICKI, M. A., and MCBURNEY, M. W. (198’7). Cell culture methods and induction of differentiation of embryonal carcinoma cell lines. In “Teratocarcinomas and Embryonic Stem Cells, A Practical Approach” (E. J. Robertson, Ed.), pp. 19-49. IRL, Oxford.
VOLUME 141,199O
SCHWARTZBERG,P. L., GOFF, S. P., and ROBERTSON,E. J. (1989). Germline transmission of a c-abl mutation produced by targeted gene disruption in ES cells. Science 246,799-803. SMITH, A. G., and HOOPER, M. L. (1987). Buffalo rat liver cells produce a diffusible activity which inhibits the differentiation of murine embryonal carcinoma and embryonic stem cells. Den Biol 121,1-9. SMITH, A. G., HEATH, J. K., DONALDSON, D. D., WONG, G. G., MOREAU, J., STAHL, M., and ROGERS, D. (1988). Inhibition of pluripotential embryonic stem cell differentiation by purified polypeptides. Nature (London) 336,688-690. SOLTER, D. and KNOWLES, B. B. (1975). Immunosurgery of mouse blastocysts. Proc. Natl. Acad. Sci. USA 72,5099-5102. WILLIAMS, R. L., COURTNEIDGE, S. A., and WAGNER, E. F. (1988a). Embryonic lethalities and endothelial tumours in chimaeric mice expressing polyoma virus middle T oncogene. Cell 52,121-131. WILLIAMS, R. L., HILTON, D. J., PEASE, S., WILLSON, T. A., STEWART, C. L., GEARING, D. P., WAGNER, E. F., METCALF, D., NICOLA, N. A., and GOUGH, N. M. (198813). Myeloid leukaemia inhibitory factor maintains the developmental potential of embryonic stem cells. Nature (London) 336,684-687. ZIJLSTRA, M., LI, E., SAJJADI, F., SUBRAMANI, S., and JAENISCH, R. (1989). Germ-line transmission of a disrupted B,-microglobulin gene produced by homologous recombination in embryonic stem cells. Nature (London) 342,435-438.
