[methods in molecular biology] epidermal cells volume 585 || establishment of spontaneously...

Chapter 5

Establishment of Spontaneously Immortalized KeratinocyteLines from Wild-Type and Mutant Mice

Julia Reichelt and Ingo Haase

Abstract

A considerable number of transgenic or knockout mice in which epidermal keratinocytes have beentargeted die shortly after birth due to barrier defects. In this case, recovery and cultivation of keratinocytesfrom these animals provide an opportunity for in vitro studies. Working with isolated keratinocytes is alsointeresting for certain experiments which cannot be performed in live animals.

Primary human keratinocytes can be kept in culture for a variable number of passages and then senesce.Immortalization can be achieved by transduction with constructs encoding viral genes. Murine keratino-cytes can be kept in culture as primary cells. Naturally the numbers of cells obtained by direct isolation frommouse epidermis is restricted and sometimes not sufficient for certain biochemical analyses. To overcomethis restriction some permanent murine keratinocyte lines have been generated by transfection with SV40Tor HPV E6E7 genes. This is, however, not suitable if established or hypothetical biochemical links existbetween these genes and the pathways or processes to be analysed in the respective experiment.

We describe an easy and reproducible method of establishing permanent keratinocyte lines from sponta-neously immortalized primary murine keratinocytes. This method employs co-cultivation of keratinocytes with3T3-J2 fibroblast feeder cells for several passages during which immortalization occurs. The resulting kerati-nocyte lines do not only grow infinitely but, in many cases, individual lines from the same genetic backgroundalso exhibit similar growth characteristics, hence they are especially valuable for comparative studies.

Key words: Mouse keratinocytes, Immortalization, 3T3-J2, Fibroblast feeder cells.

1. Introduction

The methods currently used to isolate and cultivate murine kera-tinocytes for experimental purposes can have major disadvantages.

Although primary keratinocytes are probably closest to the invivo situation, their isolation and cultivation in sufficient numbersrequires large amounts of neonatal mice bred just for this purpose.

K. Turksen (ed.), Epidermal Cells, Methods in Molecular Biology 585,DOI 10.1007/978-1-60761-380-0_5, ª Humana Press, a part of Springer Science+Business Media, LLC 2005, 2010

59

Furthermore, these cultures are not sustainable. Primary keratino-cytes tend to differentiate rapidly in culture and cannot be propa-gated for long in order to yield a sufficient pool of comparable cellsfor large-scale experiments. Furthermore, the fact that primaryepidermal cultures may harbour other epidermal cell types, e.g.melanocytes, is often disregarded.

Mouse keratinocytes have been immortalized in the past byusing transfection with recombinant retroviruses encoding HPV-16 E6 and E7 genes (1) or by isolating cells from transgenic micecarrying an inducible SV40 T antigen gene (2) (ImmortoMouse,Charles River Laboratories). However, these immortalized kerati-nocytes harbour the risk of displaying unexpected or unwantedmodifying effects caused by the transforming gene or the viralvector (3, 4).

Here, we describe a simple method to establish spontaneouslyimmortalized keratinocyte lines from wild-type or from geneticallymodified mice. Several of the keratinocyte lines generated in thisway have by now been serially cultivated for more than 250passages with no obvious changes in morphological or growthcharacteristics.

The protocol starts with the isolation of keratinocytesfrom neonatal mouse skin and their co-cultivation with 3T3fibroblast feeder cells for four to eight passages (1–3 months).From these isolated cells multiple keratinocyte clones emergewhich proliferate and continue growing as immortalized kera-tinocyte lines without feeder cells after the initial four to eightpassages.

We compared several wild-type keratinocyte lines obtainedfrom BABL/c mice with each other and with keratin 10(K10)-deficient keratinocyte lines isolated from K10 knockoutmice (5–8). K10 expression is restricted to differentiatedkeratinocytes present in the suprabasal epidermis or in differ-entiating keratinocytes grown in high-calcium medium. Inlow-calcium medium K10 is not expressed and thereforeK10-deficient keratinocytes should behave like wild-typecells. We analysed several lines of wild-type and K10–/– kera-tinocytes and found that all tested lines showed the samegrowth and migration characteristics (see Fig. 5.1). Impor-tantly, the spontaneous immortalization did not impair theintrinsic differentiation potential of the keratinocytes. Switch-ing confluent cultures of spontaneously immortalized kerati-nocytes from low- to high-calcium medium inducesdifferentiation and stratification.

The presented method has been used to establish epidermal aswell as oesophageal keratinocyte lines from distinct wild-typemouse strains (e.g. BALB/c and C57Bl/6) as well as from geneti-cally altered mouse strains (e.g. K10–/–, K10T, K14-Cre/Ikk2fl/fl

(9) and N17Rac1 transgenic mice (10)).

60 Reichelt and Haase

The major advantages of this method for producing sponta-neously immortalized mouse keratinocyte lines over primary kera-tinocytes are absence of other contaminating epidermal cells(e.g. melanocytes, dendritic T lymphocytes), the indefinite prolif-eration capacity and reproducibility of experiments due to stabilityof the keratinocyte characteristics. In contrast to keratinocytes

a

cell

nu

mb

er

days after seeding

5 10 151000

10 000

100 000

1 000 000

20

x

wt, 27

wt, 30

K10-/-, 4

K10-/-, 6

b

12 h0 h 6 h0 h 6 h 12 h

time after wounding in hrs

wo

un

d s

ize

in m

m wt, 27

wt, 30

0.4

0.8

1.2

3 6 9 120

time after wounding in hrs

wo

un

d s

ize

in m

m

0.4

0.8

1.2

3 6 9 120

K10-/-, 6

K10-/-, 8

Fig. 5.1. Characterization of wild-type and keratin 10-deficient murine keratinocyte lines. Characterization was initiated atpassage 20. (a) For the generation of a growth curve 2 � 103 keratinocytes/cm2 were seeded onto collagen I-coated35 mm dishes. Cell numbers of duplicates were determined at 1, 2, 3, 4, 7, 11, 16 and 21 days after seeding by countingtrypsinized cells in a counting chamber. Plating efficiency was low due to sparse seeding but comparable in the testedkeratinocyte lines (28%). Growth rates were determined in the exponential growth phase and saturation density was givenby the maximum cell number obtained at the end of the growth curve (2.2 � 105 keratinocytes/cm2). The growth curvesof distinct wild-type (wt) and K10–/– keratinocyte lines showed no differences. The keratinocyte clones were numberedconsecutively when established and these numbers are given in the legend. (b) Scratch assays were used in order todetermine the migration potential of keratinocytes from different lines. Keratinocytes were seeded onto collagen I-coated35 mm dishes and grown to confluency. Prior to the analysis of migration, cells were treated with mitomycin C (seeSection 3.1.2) to inhibit cell proliferation. The monolayers were scratched in the middle of the dish with a yellowmicropipette tip to yield a gap of about 1.2 mm width. The closure of the gap was followed by measuring the distance ofthe wound edges every hour over a period of 15 hours, during which the cells were kept at 32�C and 5% CO2 in anincubation chamber on the microscope. Migration rate was about 0.1 mm/hour in wt and keratin 10–/– keratinocyte lines(two lines per genotype were tested). Figure (a) by Monika Loeher and (b) by Heike Stachelscheid.

Establishment of Spontaneously Immortalized Keratinocyte Lines 61

immortalized through transduction with viral genes the sponta-neously immortalized keratinocyte lines presented here show nounwanted effects due to the presence of foreign protein or DNA.

2. Materials

2.1. Feeder Cell

Cultivation1. 3T3-J2 fibroblasts feeder cells were a generous gift of Fiona

Watt, Cambridge, UK (11).

2. Dulbecco’s modified Eagle’s medium (DMEM, Gibco/BRL) supplemented with 10% foetal calf serum (FCS) and100 U/ml penicillin and 100 mg/ml streptomycin (suppliedas 100� stock solution, Invitrogen).

3. Mitomycin C (Applichem, Germany). Prepare 100� stocksolution (0.4 mg/ml) in PBS and store aliquots at –20�C.Add 10 ml/ml DMEM + 10% FCS.

4. 0.05% Trypsin and 0.02% ethylenediamine tetraacetic acid(EDTA, Gibco/BRL).

2.2. Recovery of

Neonatal Mouse Skin

1. Betaisodona, 10% iodine (Mundipharma), which is equi-valent to povidone-iodine, PVP-I.

2. Dispase (Gibco/BRL), prepare 5 U/ml or 10 mg/ml solu-tion in PBS and sterile filter.

3. 35 mm plastic dishes or six-well plates (Nunc/Nunclon, Fal-con or TPP).

4. Sterile instruments (forceps, scissors), PBS (phosphate buf-fered saline) and 70% ethanol.

2.3. Culture of

Keratinocytes

1. Sterile instruments (forceps, scalpels), PBS and 70% ethanol.

2. 35 mm plastic dishes or six-well plates (Nunc/Nunclon,Falcon or TPP).

3. Collagen I from rat tail (Becton Dickinson), 50 mg/ml0.02 M acetic acid. The diluted solution is stored at 4�C andcan be used several times for coating.

4. FCS Gold (PAA), chelex treated to remove Ca2+: 20 g chelex100 (Bio-Rad)/500 ml FCS Gold is left to rotate O/N at4�C, then pre-cleared by paper filtration, sterile filtered andstored in 50 ml aliquots at –20�C.

5. FAD medium: DMEM/HAM’s F12 3.5:1.1, low-calcium(0.05 mM Ca2+) (custom made by Biochrom, Berlin,Germany).


6. FAD medium is supplemented to yield ‘‘complete FAD med-ium’’ before use with 10% chelex-treated FCS gold, 0.18 mMadenine, 0.5 mg/ml hydrocortisone, 5 mg/ml insulin, 10–10

M cholera toxin (all from Sigma), 10 ng/ml EGF (Invitro-gen), 2 mM glutamine, 1 mM pyruvate, 100 U/ml penicillinand 100 mg/ml streptomycin (all supplied as 100� stocksolutions, Invitrogen).

Supplement stock solutions are prepared as follows eitherunder sterile conditions or sterile filtered and stored inaliquots at –20�C unless otherwise indicated:a) Adenine (250� ): 45 mM in 50 mM HCl. Add 2 ml stock

solution/460 ml medium.

b) Hydrocortisone (2000� ): 1 mg/ml in ethanol. Dilute 1:4with FAD medium. Add 1 ml stock solution/460 ml FADmedium.

c) Insulin (1000� ): 5 mg/ml in 5 mM HCl. Add 0.5 mlstock solution/460 ml FAD medium.

d) EGF (1000� ): 10 mg/ml in FAD medium. Add 0.5 mlstock solution/460 ml FAD medium.

e) Cholera toxin (10–5 M): 1 mg/1.18 ml in sterile water,stored at 4�C. Add 5 ml stock solution/460 ml FADmedium.

7. 0.05% Trypsin and 0.02% ethylenediamine tetraacetic acid(EDTA, Gibco/BRL).

8. 0.02% EDTA in PBS.

9. Keratinocyte freezing solution: 90% chelex-treated FCSGold, 10% dimethyl sulphoxide (DMSO, cell culture grade,Applichem, Germany); aliquots are stored at –20�C.

10. CO2 incubator at 32�C.

3. Methods

3.1. Feeder Cell

CultivationFeeder cell expansion must be well timed with mouse breeding tomake sure enough feeder cells are available on the day after birth ofpups and thereafter. For a litter of six pups, one just confluent 10 cmPetri dish should be available on the day after mouse skin recovery tobe plated at a density of 1.5–3 � 104 cells/cm2 (see Note 1).

The 3T3-J2 subclone of Swiss mouse 3T3 fibroblasts has beenwidely used to support human keratinocyte cultivation (12, 13) aswell as other epithelial cells (14–16) and has also proven useful inour hands for the establishment of permanent murine keratinocytelines.


1. 3T3-J2 fibroblasts are routinely grown in DMEM + 10% FCS.They are split twice weekly when just confluent at no morethan 1:3 (see Fig. 5.2b and Note 2).

2. In order to use 3T3-J2 feeder cells in co-culture with keratino-cytes, culture a confluent Petri dish for 2–3 hours with DMEM +10% FCS containing mitomycin C at 37�C (see Note 3).

b

e

a

dc

f

Fig. 5.2. Morphology of murine keratinocytes and 3T3-J2 fibroblast feeder cells.(a) Confluent keratinocyte monolayers show the typical cobblestone-like pattern inlow-calcium FAD medium. (b) The photo shows confluent 3T3-J2 fibroblast. At thiscell density, feeder cells should be routinely split or used for mitomycin C treatment andco-culture with keratinocytes. (c) A colony of keratinocytes (encircled) emerging 12 daysafter co-cultivation of primary keratinocytes with feeder cells. (d) 3T3 feeder cells 3 daysafter mitomycin C treatment. (e) Typical morphology of spontaneously immortalizedmurine keratinocytes about 12 hours after splitting. A group of cells in the centre showthe typical cobblestone morphology of keratinocytes growing at higher density.(f) Splitting at too high dilutions often causes differentiation of keratinocytes. Differen-tiated keratinocytes may appear enlarged (arrowheads) or appear spindle shaped(arrows) causing an irregular pattern at confluency. Bars, 100 mm. Images were takenby Dr. Anne Vollmers.


3. Before adding trypsin solution, wash cells thoroughly twicewith PBS to remove mitomycin C.

4. Wash the fibroblasts once with trypsin/EDTA solution, addfresh trypsin/EDTA and incubate the cells for about 2 min at37�C. Resuspend the detached cells subsequently in kerati-nocyte medium (complete FAD medium) and centrifuge for5 min at 220g. Suspend the cell pellet in fresh complete FADmedium by pipetting three to four times and seed the fibro-blasts then directly onto collagen I-coated dishes at a densityof 1.5–3 � 104/cm2. When first seeding primary keratino-cytes and for the first passage using the higher density offeeder cells (approximately 3 � 104 cells/cm2) isrecommended.

5. The feeder cells may be seeded either prior to or together withthe keratinocytes.

3.2. Recovery of

Neonatal Mouse Skin

1. Decapitate neonatal mice and place them on ice for 1 hour(see Note 4).

2. From this point, a sterile laminar flow cabinet as well as sterilesolutions and instruments are required.

3. Use one mouse after the other and keep the remainingcorpses on ice.

4. Disinfect mouse corpses with Betaisodona (1:1 diluted withPBS) for 1 min (e.g. in a Petri dish), rinse two times briefly inPBS, place for 1 min in 70% ethanol and finally rinse briefly inPBS.

5. Remove extremities and cut the skin dorsally from head totail. Peel the skin off in one piece using forceps and the dullparts of scissors.

6. Spread the skin on a 35 mm plastic dish with the epidermisside facing upwards.

7. Add 2 ml of dispase solution carefully from the side so that theskin floats.

8. Seal the dish with parafilm and incubate O/N (or at least for 7hours) at 4�C (see Note 5).

3.3. Culture of

Keratinocytes

If keratinocytes from transgenic mice and wild-type control miceor keratinocytes from distinct wild-type mouse strains are to begrown at the same time, cross-contaminations must be carefullyavoided at any stage. Established keratinocyte lines should beregularly checked, e.g. by PCR genotyping, to ensure theirpurity.

1. Before starting, treat feeder cells with mitomycin C for 2–3hours (see Section 3.1).


2. Cell culture dishes (35 mm) have to be coated with collagenI solution for 1 hour at room temperature and then washedcarefully two times with PBS before cells are plated (see Note 6).

3. After incubation with dispase (see Section 3.2.7), remove theepidermis from the dermis with forceps (see Note 7).

4. Mince the epidermis in a Petri dish by moving two scalpelsagainst each other in opposite directions for 1–3 min to yield amush. The cells and tissue pieces must not dry out.

5. Suspend the minced epidermis of each mouse in 1.5 ml com-plete FAD medium and agitate for 30 min at 1000 rpm in amicrocentrifuge tube.

6. Seed keratinocytes and epidermal fragments from each indi-vidual mouse separately on collagen I-coated dishes togetherwith 3T3-J2 feeder cells. A subconfluent 10 cm dish of mito-mycin C-treated feeder cells is sufficient for six to twelve35 mm dishes of keratinocytes.

7. Grow the keratinocytes at 32�C and 5% CO2.

8. Primary keratinocytes can be seen on the following day, butmany of them will differentiate and die.

9. Change medium twice or three times weekly and add feedercells at least at every medium change and if required also in-between medium changes. In low-calcium medium, themitomycin C-treated feeder cells usually die after 3-4 days(see Fig. 5.2d) and therefore care must be taken to ensurethe presence of sufficient feeder cells.

10. Usually, after about 1–3 weeks keratinocyte clones withthe typical cobblestone-like appearance will be observed(see Fig. 5.2c and Note 8). Epidermal cell cultures ofdark coated mouse strains may appear increasingly blackduring the first passages due to melanocyte growth(see Note 9).

11. At the beginning, split keratinocytes at 1:1. At later passages(4–5), splitting at 1:2 or 1:3 is possible. The day after split-ting, the sparsely seeded keratinocytes do not show the typicalcobblestone pattern (see Fig. 5.2e) which only develops whenthe cultures grow and become more confluent.

12. For splitting, wash keratinocytes with PBS, incubate withEDTA solution for 5 min at RT and finally incubate for8–12 min with trypsin/EDTA solution at 37�C until theycome off easily with or often without gently knocking on theside of the flask.

13. Add complete FAD medium and resuspend the cells by pipet-ting up and down three to five times. After centrifugation for5 min at 220g resuspend the pellet in complete FAD mediumand seed the cells onto collagen I-coated dishes.


14. The keratinocytes have to be grown with feeder cells for atleast 4–8 passages before trying to omit the feeder cells (seeNote 10).

15. Keratinocytes may be frozen at a density of 1–1.5 � 106

cells/ml in keratinocyte freezing solution.

16. After thawing, keratinocytes should be seeded densely (about105 cells/cm2) (see Note 11).

17. The spontaneously immortalized mouse keratinocytes may bedifferentiated by addition of calcium to the complete FADmedium at a final concentration of 1.2 mM (high-calciummedium). Confluent cultures stratify when grown in high-calcium medium.

4. Notes

1. 3T3-J2 fibroblasts should be split routinely or used for co-culture with keratinocytes as soon as they get confluent(see Fig. 5.2b).

2. 3T3-J2 fibroblasts are subject to normal ageing. Make sure tonote passage numbers and to freeze enough aliquots at lowpassage. Cultures of 3T3-J2 fibroblasts should not be usedany longer if the typical phenotype changes and cells acquire aspindle cell phenotype as this can be a sign of transformation.

3. About 5 ml of mitomycin C-containing medium is sufficientfor a 10 cm Petri dish.

4. Neonates on days 0 or 1 postpartum yield the highest numberof isolated keratinocytes. According to our experience, effi-ciency of isolation worsens dramatically when mice after day 3postpartum are used. Keratinocytes from unborn mice at E18or E19 can be isolated using the same protocol.

5. It is best to use the litter immediately after birth which is oftenin the morning. Then incubate with dispase for 7 hours dur-ing the day and go on with the protocol on the same day. Ifyou start the recovery of neonatal mouse skin in the after-noon, terminate the digestion with dispase early the nextmorning.

6. The washed dishes can be left with traces of PBS at roomtemperature for a few hours until the keratinocytes are readyto be seeded. Alternatively, pre-coated dishes (e.g. Biocoatsix-well plates, Beckton Dickinson) may be used.

7. The epidermis comes off in one piece as a thin white layer,whereas the residual tissue appears pink, shiny and slimy.


8. Depending on the mouse strain used as well as the nature ofthe mutation and age variations it might take up to 2 monthsbefore keratinocyte colonies become visible.

9. During the first eight or more passages, the cultures stillcontain melanocytes and possibly other resident epidermalcell types. These cells cannot always be easily distinguishedwithout specific staining in cultures of white mouse strains,e.g. BALB/c, but melanocytes are clearly recognizable asblack cells in cultures from coloured mouse strains, e.g.C57Bl/6 mice. The melanocytes are gradually depleted withincreased passaging.

10. If high amounts of large, differentiated or irregularly shapedkeratinocytes become a problem in an established mousekeratinocyte line (compare Fig. 5.2f and a), the addition offeeder cells for 1–2 passages and/or seeding at a higherdensity (less that 1:1) may be helpful.

11. In order to enhance plating efficiency after thawing, kerati-nocytes may optionally be seeded together with feeder cellsirrespective of their passage number. One subconfluent toconfluent 10 cm dish of 3T3-J2 feeder cells (0.8–1.6 �106) is sufficient for twelve 35 mm dishes. The feeder cellsare lost at the first passage after thawing and may subse-quently be omitted.

Acknowledgements

We thank Semra Frimpong for excellent technical assistance andDr. Penny Lovat and Dr. Anne Vollmers for critically reading themanuscript. This work was supported by the Deutsche For-schungsgemeinschaft (DFG).

References

1. Woodworth, C.D., Gaiotti, D., Michael, E.,Hansen, L., and Nees, M. (2000) Targeteddisruption of the epidermal growth factorreceptor inhibits development of papillomasand carcinomas from human papilloma-virus-immortalized keratinocytes. CancerRes 60, 4397–402.

2. Jat, P.S., Noble, M.D., Ataliotis, P., Tanaka,Y., Yannoutsos, N., Larsen, L., and Kioussis,D. (1991) Direct derivation of conditionallyimmortal cell lines from an H-2 Kb-tsA58transgenic mouse. Proc Natl Acad Sci USA88, 5096–100.

3. DiPersio, C.M., Shao, M., Di Costanzo, L.,Kreidberg, J.A., and Hynes, R.O. (2000)Mouse keratinocytes immortalized withlarge T antigen acquire alpha3beta1 integ-rin-dependent secretion of MMP-9/gelati-nase B. J Cell Sci 113, 2909–21.

4. Lamar, J.M., Iyer, V., and DiPersio, C.M.(2008) Integrin alpha3beta1 potentiatesTGFbeta-mediated induction of MMP-9 inimmortalized keratinocytes. J Invest Derma-tol 128, 575–86.

5. Reichelt, J., Breiden, B., Sandhoff, K., andMagin, T.M. (2004) Loss of keratin 10 is


accompanied by increased sebocyte prolif-eration and differentiation. Eur J Cell Biol83, 747–59.

6. Reichelt, J., Bussow, H., Grund, C., andMagin, T.M. (2001) Formation of a normalepidermis supported by increased stability ofkeratins 5 and 14 in keratin 10 null mice.Mol Biol Cell 12, 1557–68.

7. Reichelt, J., Furstenberger, G., and Magin,T.M. (2004) Loss of keratin 10 leads tomitogen-activated protein kinase (MAPK)activation, increased keratinocyte turnover,and decreased tumor formation in mice. JInvest Dermatol 123, 973–81.

8. Reichelt, J., and Magin, T.M. (2002) Hyper-proliferation, induction of c-Myc and 14-3-3sigma, but no cell fragility in keratin-10-null mice. J Cell Sci 115, 2639–50.

9. Pasparakis, M., Courtois, G., Hafner, M.,Schmidt-Supprian, M., Nenci, A., Toksoy,A., Krampert, M., Goebeler, M., Gillitzer,R., Israel, A., Krieg, T., Rajewsky, K., andHaase, I. (2002) TNF-mediated inflamma-tory skin disease in mice with epidermis-spe-cific deletion of IKK2. Nature 417, 861–6.

10. Tscharntke, M., Pofahl, R., Chrostek-Grashoff, A., Smyth, N., Niessen, C., Nie-mann, C., Hartwig, B., Herzog, V., Klein,H.W., Krieg, T., Brakebusch, C., and Haase,I. (2007) Impaired epidermal wound heal-ing in vivo upon inhibition or deletion ofRac1. J Cell Sci 120, 1480–90.

11. Watt, F.M. (1998) Cultivation of humanepidermal keratinocytes with a 3T3 feederlayer. In: Celis, J.E., ed. Cell Biology: ALaboratory Handbook, Vol. 1. New York,Academic Press, pp. 113–8.

12. Todaro, G.J., and Green, H. (1963) Quan-titative studies of the growth of mouseembryo cells in culture and their develop-ment into established lines. J Cell Biol 17,299–313.

13. Rheinwald, J.G. (1989) Methods for clonalgrowth and serial cultivation of normalhuman epidermal keratinocytes andmesothelial cells. In: Baserga, R., ed. CellGrowth and Division. A Practical Approach.IRL Press, Oxford, pp. 81–94.

14. Sugimachi, K., Sosef, M.N., Baust, J.M.,Fowler, A., Tompkins, R.G., and Toner,M. (2004) Long-term function of cryopre-served rat hepatocytes in a coculture system.Cell Transplant 13, 187–95.

15. Kobayashi, C., Kagami, H., Kito, K., Ishi-kawa, K., Ebisawa, K., Ueda, M., and Ter-asaki, H. (2005) Selective and efficientculturing of retinal pigment epithelial cellsusing a feeder layer. Cytotherapy 7, 427–37.

16. Honda, M.J., Shimodaira, T., Ogaeri, T.,Shinohara, Y., Hata, K., and Ueda, M.(2006) A novel culture system for porcineodontogenic epithelial cells using a feederlayer. Arch Oral Biol 51, 282–90.


[methods in molecular biology] epidermal cells volume 585 || establishment of spontaneously...

Documents