Pluripotent stem cell lines, embryonic stem (ES) cells areundifferentiated cell lines that are capable of forming virtu-ally any cell type in the body.2) Mouse ES cells have beenuseful and versatile cells to study developmental biology, andare indispensable for creation of knockout mice to analyzefunction of specific protein.3) ES cell lines derived from notonly mouse but also human have been established and theirwide developmental potential and unlimited life span in cul-ture make them extremely interesting and important for basicand applied research.4) They can undergo differentiation invivo and in vitro, allowing them to be used for the studybasic processes in developmental biology, and be used as arenewable cell source for cell engineering, transplantationand regenerative medicine.

Maintenance of ES cells in the undifferentiated state wasperformed by the culture with feeder cells. There, isolatedinner cell mass-derived cells were seeded to gelatin-coateddishes containing a confluent layer of mitomycin C-inacti-vated STO fibroblasts.5,6) Because STO cells are easily cul-tured for a long period, they have been then often used asfeeder cells to culture mouse ES cells. This indicated thatfeeder cells such as STO fibroblasts and mouse embryo fi-broblasts (MEF) provided some factors to promote self-re-newal and suppress differentiation. Thereafter, the cytokineleukemia inhibitory factor (LIF) was found to be one of im-portant factors for maintaining the pluripotency of mouse EScells.7,8) LIF activates signal transducer and activator of tran-scription 3 (STAT3) which has been shown to play an essen-tial role in maintaining self-renewal, while it also induces ac-tivation of a number of signaling protein including extracel-lular signal-regulated kinase (ERKs) which appears to pro-mote differentiation.9—12) Therefore, it is considered that thebalance between STAT3 and ERKs signaling is critical fordetermining the fate of undifferentiated ES cells.13) In con-trast, maintenance of human ES cells still required a mono-layer of feeder cells, since LIF has no effect on preventingdifferentiation of human ES cells. Mouse ES cells have re-cently been cultured in the presence of LIF and fetal bovineserum (FBS) or serum replacement (SR) on gelatin-coateddishes in feeder cell-free system. However, FBS is put undersuspicion for its unknown factor to promote differentiation

and SR often decreases the proliferation. Under those situa-tions, we thought that the re-examination of feeder cell func-tion for maintaining undifferentiated ES cells was necessaryfor establishment of the proper ES cell culture system.

We found here that conditioned medium (CM) derivedfrom STO cells enhanced proliferation of undifferentiated EScells through the promotion of ES cell attachment on gelatin-coated dishes. The result gave useful information for the roleof feeder cells in the maintenance of undifferentiated EScells.

MATERIALS AND METHODS

Cell Cultures Mouse 129/sv ES cell lines (passage 15,Dainippon Pharmaceutical Corporation, Osaka, Japan) wereroutinely cultured on tissue culture plates (Falcon) coatedwith 0.1% (v/v) gelatin (Dainippon Pharmaceutical Corpora-tion) in KnockoutTM Dulbecco’s modified Eagle’s medium(DMEM) (Invitrogen, Scottland) in the presence of 15%(v/v) KnockoutTM serum replacement (Invitrogen), 0.1 mM b-mercaptoethanol, 2 mM L-glutamine, 0.1 mM non-essentialamino acids, 100 units/ml penicillin, 0.1 mg/ml streptomycin,and 1000 units/ml LIF (Chemicon) in 5% CO2 incubator at37 °C. Cells were trypsinized and replated every 3 d.

Mouse fibroblast STO cells (ATCC, Rockville, MD,U.S.A.) were cultured in DMEM in the presence of 15% fetalbovine serum (FBS) (Trace), 5 mg/ml D-glucose, 1.5 mg/mlsodium bicarbonate, 100 units/ml penicillin, 0.1 mg/ml strep-tomycin) in 5% CO2 incubator at 37 °C. Cells were treatedwith 10 mg/ml mitomycin C (Wako, Osaka) for 4 h,trypsinized, and replated on tissue culture dishes coated with0.1% (v/v) gelatin to utilize as feeder cells.

CM from STO Cells STO cells were maintained in cul-ture flasks. The medium was changed to a culture mediumfor ES cells (KnockoutTM DMEM in the presence of 15%(v/v) KnockoutTM SR, 0.1 mM b-mercaptoethanol, 2 mM L-glutamine, 0.1 mM non-essential amino acids, 100 units/mlpenicillin, 0.1 mg/ml streptomycin, and 1000 units/ml LIF).STO cells were cultured for 3 d, and the cultured mediumwas centrifuged at 1000 rpm for 5 min. The supernatant wasused as CM after the filtration by a 0.45 mm mesh filter (Mil-

August 2006 1747

© 2006 Pharmaceutical Society of Japan∗ To whom correspondence should be addressed. e-mail: mamoru@dpc.agu.ac.jp

Biol. Pharm. Bull. 29(8) 1747—1750 (2006)Notes

Conditioned Medium from Feeder STO Cells Increases the Attachment ofMouse Embryonic Stem Cells

Kenichiro AMANO, Tadahide FURUNO,1) and Mamoru NAKANISHI*,1)

Graduate School of Pharmaceutical Sciences, Nagoya City University; Tanabe-dori, Mizuho-ku, Nagoya 467–8603, Japan.Received January 24, 2006; accepted May 30, 2006; published online June 2, 2006

Mouse embryonic stem (ES) cell lines, which were established by culturing on feeder cells, have usually beencultured without feeder cells in the presence of leukemia inhibitory factor. However, proliferating rate of ES cellsin the condition is often lower than that with feeder cells. Here, we found that conditioned medium (CM) fromfeeder cells (STO cells) increased the number of undifferentiated cells in a culture dish by promoting attachmentof ES cells. The attached cells were increased in 4 h after replating ES cells in the presence of CM from STOcells, and they formed flat colonies composed of undifferentiated cells. This culture system with CM from feedercells is useful in preparing a large number of well-defined ES cells. In addition, from present experiments wefound that the dish double-coated with gelatin and CM was extremely useful for culturing ES cells.

Key words embryonic stem cell; feeder STO cell; conditioned medium

lipore).In coating culture dishes with CM, the medium containing

30% CM was added to non-coated or gelatin-coated dishes.After they were incubated overnight, the medium were re-moved. They were washed PBS, and then used for the experi-ments.

Cell Counts ES cells were replated on gelatin-, CM-, orgelatin plus CM-coated dishes (1.0�106 cells/dish). Afterculturing in SR-, CM-, or FBS-containing medium for appro-priate period, we counted the number of ES cells attached todishes. Cells were washed with PBS to remove floating cells,and then attached cells were collected by trypsinization.They were counted with Burker-Turk cell-counting plate.

SDS-PAGE To extract proteins existing in gelatin- andgelatin plus CM-coated dishes, we scratched the bottom ofdishes by rubber policeman in lysis buffer (25 mM Tris–HCl,0.15 M NaCl, 5 mM MgCl2, 1 mM EDTA, 10% glycerol, 1 mM

dithiothreitol, 1mM sodium vanadate, 0.5% Nonidet P-40,1 mM phenylmethylsulfonyl fluoride, 4 mg/ml leupeptin (pH7.5)). Then, we added acetone, kept them at �20 °C for 1 h,and centrifuged them at 23500�g for 5 min. After removingsupernatants, the resulting precipitations were solubilized bytreatment with Laemmli buffer at 100 °C for 5 min, separatedby electrophoresis in 10% SDS-polyacrylamide gel andstained with 0.25% Coomassie Brilliant Blue R250 (Wako).

Data Presentation Data were presented as mean�S.E.M. In statistical analysis, data were analyzed by unpairedStudent’s t-test, 2-tailed. p�0.05 was accepted as a level ofstatistically significant difference.

RESULTS

To study the effects of CM from STO cells on proliferationof ES cells, we measured the number of ES cells after cultur-ing in the medium for ES cells containing CM from STOcells on tissue culture dishes coated with gelatin in the pres-ence of LIF. Some floating ES cells differentiated, however,more than half of ES cells (ca. 60%) attached on gelatin-coated dishes. The attached cells formed colonies whosemorphology was characteristic of undifferentiated ES cellcolonies; that is, the boundary of cells was apparently ob-scure in flat colonies.14) We checked that the ES cell in suchcolonies expressed the stem cell markers alkaline phos-phatase, which were lost on removal of LIF when the EScells differentiated into a variety of cell types.15) The numberof undifferentiated ES cells attached to gelatin-coated dishesdid not increase in the medium containing 10% CM but in-creased in the medium containing more than 20% CM asshown in Fig. 1. This suggested that some soluble factors inCM accelerated the proliferation of ES cells. We used the 30 % CM-containing medium in the following experiments.

Here, we compared the number of ES cells attached to gel-atin-coated dishes among the culture media containing 15%SR, 15% FBS, and 30% CM from STO cells. Figure 2 showsthe time-courses of attached ES cells in three differentmedia. The number of ES cells cultured in CM-containingmedium for 24 h was apparently higher than those in SR- andFBS-containing media as shown in Fig. 2. When ES cellswere cultured for a long-time (2 or 3 d) in FBS-containingmedium, the number of the cells increased to the values inCM-containing medium. The values in SR-containing

medium were much lower than those in FBS- or CM-contain-ing medium (data not shown).

As described above, the attached ES cells formed flatcolonies where the boundary of cells was apparently obscurein SR-containing medium in the presence of LIF, as shown inFig. 3a. Removal of LIF for 3 d induced the differentiation ofES cells and the morphological changes of colonies to bethick (Fig. 3b). ES cells in FBS-containing medium alsoshowed some thick colonies as like as the case of LIF with-draw after 3-d culture as shown in Fig. 3c. On the contrary,the ES cell colonies in CM-containing medium were normalshape (Fig. 3d).

Next, we studied the effects of dishes pre-coated withgelation and/or CM on the attachment of ES cells. We madethis experiment using three kinds of culture dishes, whichwere gelatin-coated, CM-coated and gelatin plus CM-coatedones. The number of attached ES cells to gelatin-coateddishes for 2-h culture was much higher in CM-containingmedium (grey column in left) than in SR-containing medium(white column in left) in Fig. 4a. Further, the number of at-tachment of ES cells increased in the dishes pre-coated withCM, whether the medium contained SR or CM. In addition,the number in the dishes pre-coated with both gelatin andCM also more increased in comparison with that in the CM-coated dishes. To detect CM-containing factors which in-duced the promotion of ES cell attachment, we analyzed theproteins on the bottom of dishes coated with gelatin and gel-atin plus CM. Two significant bands of proteins at ca. 7 kDaand ca. 34 kDa were observed by SDS-PAGE in gelatin plusCM-coated dishes, but not in gelatin-coated ones, as shown

1748 Vol. 29, No. 8

Fig. 1. Effect of CM from STO Cells on ES Cell Proliferation

The numbers of ES cells attached to gelatin-coated dishes were counted at 3 d afterreplating 1.0�106 cells in the medium containing CM from STO cells. Triplicate cellcounts were performed in each experiment, and the experiment was repeated threetimes. ∗∗ p�0.01 compared with the medium in the absence of CM from STO cells.

Fig. 2. Effect of CM from STO Cells on Attachment of ES Cells

The numbers of ES cells attached to gelatin-coated dishes which were replated at1.0�106 cells/dish and were cultured in the SR (15%)-containing medium (�), FBS(15%)-containing medium (�), and CM (30%)-containing medium (�) were counted.Triplicate cell counts were performed in each experiment, and the experiment was re-peated three times. ∗ p�0.05 and ∗∗ p�0.01 compared with the SR- and FBS-contain-ing media.

in Fig. 4b.

DISCUSSION

A monolayer of inactivated feeder cells was required forestablishment of mouse ES cell lines, because feeder cellsare considered to produce some crucial factors either to pro-mote self-renewal or suppress differentiation. After LIF wasidentified as an important factor for the activity, mouse EScells have been usually maintained in gelatin-coated dishes in

the presence of LIF without feeder cells in the medium con-taining FBS or SR.16) However, the supplement of FBSshould be completely eliminated from the in vitro culture ofES cells because the composition of serum is not well de-fined. Serum contains uncharacterized growth and differenti-ation factors, and has variations among their lots. The growthrate of ES cells was actually high in the medium containingFBS, but they formed thick colonies which were likely to bedifferentiating. At present, there are a number of commer-cially available chemically defined synthetic serum substi-tutes such as KnockoutTM SR.17) SR is completely devoid ofany undefined growth factors or differentiation-promotingfactors. In this experiment, ES cells formed flat colonies, andproliferated in undifferentiated states in the medium contain-ing SR, but the growth rate was not so high. All of these re-sults indicated that the CM from STO cells seems to be use-ful in preparing a large number of well-defined ES cells.

We focused here on the role of feeder cells in the prolifera-tion and maintenance of undifferentiated ES cells, and suc-ceeded to find an important role of feeder cells to mouse EScells. The CM from STO cells showed the promotion of at-tachment to culture dishes in ES cell. The procedure of EScell culture using CM from feeder cells described in the pres-ent paper has two major advantages: first; ES cells in thisculture system are well-defined because there is no anxietyabout the direct contamination of feeder cells, and second;they are prepared at low passage number because a largenumber of ES cells adhere to culture dishes. The adhesion toextracellular matrix of ES cells is necessary for the mainte-nance of undifferentiated state. In fact, floating ES cells areknown to be led to apoptotic cell death or differentiationthrough the formation of embryonic body.18,19) The presentresults clearly indicated that the dish double-coated with gel-atin and CM was extremely useful for culturing ES cells. Wedetected two kinds of proteins from the gelatin plus CM-coated dishes. Especially, a strong band was observed at ca.34 kDa, suggesting that this protein might be entactin fromthe previous paper analyzing CM from MEF and STO cellsby two-dimensional electrophoresis mass spectrometry.20)

The identification of this protein is needed by other experi-ments, but the present results suggested that some compo-nents of soluble proteins derived from STO cells seem toplay a significant role in the attachment of ES cells.

Actually, we can obtain more undifferentiated ES cells inroutine culture using CM from STO cells which was able tostock for at least 6 months at �20 °C. The data presentedhere show that the ES cell culture system utilizing CM fromfeeder cells enables us to prepare a large number of well-de-fined ES cells in the feeder cell-free and serum-free condi-tion.

REFERENCES AND NOTES

1) Present address: School of Pharmacy, Aichi-Gakuin University,Kusumoto-cho, Chikusa-ku Nagoya 464–8650, Japan.

2) Wobus A. M., Boheler K. R., Physiol. Rev., 85, 635—678 (2005).3) Robertson G. R., Bradley A., Kuehn M., Evans M., Nature (London),

323, 445—448 (1986).4) Thompson J. A., Itskovitz-Eldor J., Shapiro S. S., Waknitz M. A.,

Swiergiel J. J., Marshall V. S., Jones J. M., Science, 282, 1145—1147(1998).

5) Evans M. J., Kaufman M. H., Nature (London), 292, 154—156 (1981).6) Martin G. R., Proc. Natl. Acad. Sci. U.S.A., 78, 7634—7638 (1981).

August 2006 1749

Fig. 3. Differential Interference Contrast Images of ES Cell Colonies

(a) ES cell colonies cultured in the SR (15%)-containing medium in the presence ofLIF. (b) ES cell colonies cultured in the SR (15%)-containing medium in the absence ofLIF for 3 d. (c) ES cell colonies cultured in FBS (15%)-containing medium in the pres-ence of LIF for 3 d. (d) ES cell colonies cultured in CM (30%)-containing medium inthe presence of LIF for 3 d. Bars �100 mm.

Fig. 4. Promotion of Attachment of ES Cells to Culture Dishes Coatedwith CM from STO Cells

(a) The numbers of ES cells attached to culture dishes coated with gelatin, CM, orgelatin plus CM were counted at 2 h after replating 1.0�106 cells in the SR (15%)-con-taining medium (white bars) and CM (30%)-containing medium (grey bars) werecounted. Triplicate cell counts were performed in each experiment, and the experimentwas repeated three times. ∗∗ p�0.01 compared with the SR-containing medium in thegelatin-coated dish. # p�0.05 compared with the CM-containing medium in the gelatin-coated dish. (b) Proteins extracted from gelatin-coated dishes (lane 1) and the gelatinplus CM-coated dish (lane 2) were analyzed by SDS-PAGE. Equal amounts of protein(10 mg/lane) were electrophoresed.

7) 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., Gough N. M.,Nature (London), 336, 684—687 (1988).

8) Smith A. G., Heath J. K., Donaldson D. D., Wong G. G., Moreau J.,Stahl M., Rogers D., Nature (London), 336, 688—690 (1988).

9) Matsuda T., Nakamura T., Nakao K., Arai T., Katsuki M., Heike T.,Yokota T., EMBO J., 18, 4261—4269 (1999).

10) Niwa H., Burdon T., Chambers I., Nichols J., Smith A., Genes Dev.,12, 2048—2060 (1998).

11) Burdon T., Stracey C., Chambers I., Nichols J., Smith A., Dev. Biol.,210, 30—43 (1999).

12) Furuno T., Hirashima N., Onizawa S., Sagiya N., Nakanishi M., J. Im-munol., 166, 4416—4421 (2001).

13) Burdon T., Smith A., Savatier P., Trends Cell Biol., 12, 432—438(2002).

14) Smith A., “Embryonic Stem Cells,” Section II, Chap. 10, ed. by Mar-shak D. R., Gardner R. L., Gottlieb D., Cold Spring Harbor LaboratoryPress, New York, 2001, pp. 205—230.

15) Pease S., Braghetta P., Gearing D., Grail D., Williams R. L., Dev. Biol.141, 344—352 (1990).

16) Heng B. C., Cao T., Haider H. K., Wang D. Z., Sim E. K., Ng S. C.,Cell Tissue Res., 315, 291—303 (2004).

17) Goldsborough M. D., Tilkins M. L., Price P. J., Lobo-Alfonso J., Mor-rison J. R., Stevens M. E., Focus, 20, 8—12 (1998).

18) Martin M. J., Muotri A., Gage F., Varki A., Nat. Med., 11, 228—232(2005).

19) Doetschman T. C., Eistetter H., Katz M., Schmidt W., Kemler R., J.Embryol. Exp. Morphol., 87, 27—45 (1985).

20) Lim J. W., Bodnar A., Proteomics, 2, 1187—1203 (2002).

1750 Vol. 29, No. 8