IN VITRO Volume13, No. 5, 1977 All rights reserved9

THE EFFECTS OF VARIOUS MAMMALIAN SERA ON ATTACHMENT EFFICIENCY AND THYMIDINE INCORPORATION IN PRIMARY CULTURES OF MOUSE MAMMARY EPITHELIAL CELLS MARTIN K. FELDMAN' ANDDONALD L. WONG Bruce Lyon Memorial Research Laboratory, Children's Hospital Medical Center,

5lst and GroveStreets, Oakland, California 94609

SUMMARY

Mammary epithelial cells from 16- to 17-day pregnant BALB/c mice were cultured in various mammalian sera to determine the kind of serum which stimulates optimal attachment efficiency and thymidine incorporation. Of those sera tested, horse, bovine, lamb, goat and fetal bovine provided the highest attachment efficiency, whereas rat, mouse and human gave the lowest. Rabbit serum stimulated the highest thymidine incorporation into TCA-insoluble material with goat and rat providing the lowest. These results suggest that sera which provide the highest attachment efficiency for primary cultures are not the best stimulants of DNA synthesis and show that an inverse relationship exists between cell attachment and thymidine incorporation for any given type of mammalian serum.

Key words: serum; attachment efficiency; D N A synthesis; primary cultures; thymidine. INTRODUCTION

supplemented with Gentamycin (50 /~g per ml, Schering Corp., Kenilworth, N.J.) and the appropriate serum supplements. This medium was selected above others for its demonstrated ability to (a) support growth (9k (b) stimulate milk protein synthesis (11), and (c) maintain an epithelioid morphology in our cultures during the course of these studies (unpublished observations). Sera selected included the following: bovine (Grand Island, Santa Clara, Calif.), fetal bovine (Grand Island), goat (Antibodies, Inc., Davis, Calif.), horse (Grand Island), human (Microbiological Associates, Los Angeles, Calif. ), lamb (Grand Islandk rabbit (Antibodies, Inc., Grand Island and Microbiological Associates), and rat (Microbiological Associates). The same lots of each serum were used throughout the experiments. Attachment efficiency. Plating attachment was MATERIALS AND METHODS determined as follows: Cells initially were plated Cell culture techniques. Mammary glands from at a density of 7 x 104 cells per cm 2 and incubated 16- to 17-day pregnant BALB/c mice (Cancer Reat 37~ in an atmosphere of 95% air:5% C02. search Laboratory, Berkeley, Calif.j were col- The media contained 20% of the various sera to lected, minced and dissociated with collagenase be tested. Forty-eight hr after the addition of the (Type I, Worthington Biochemical Corp., Freecells to each well (2-cm2 surface area, Flow Labhold, N.J.) as previously reported (9). Cells ini- oratories, Inglewood, Calif.), the medium was retially were plated in a BHK-21 medium (10) moved, and the surface of each well was washed with buffer (phosphate-buffered saline, pH 7.2, 1To whom requests for reprints should be sent. 0.01 M) in order to remove the unattached cells. 275 Most primary monolayer cultures of mammalian cells require the presence of serum or serumlike proteins. Enhancement of cellular attachment to the substratum and the stimulation of growthrelated events are among the most important functions attributed to serum {1-8). Since relatively small numbers of mouse mammary epithelial cells (MMEC) are normally obtained for cultivation from virgin, mid-pregnant or lactating mice, we wished to develop conditions which would support optimal levels of both attachment and growth. Therefore, the purpose of this study was to discover which type of mammalian serum stimulates optimal attachment efficiency and which type produces the highest levels of thymidine incorporation.

276

FELDMAN AND WONG

The attached cells were trypsinized from the surface and counted electronically using the Cytograph, Model 6300A (Bio/Physics Corp., Mahopac, N.Y.). All reported analyses were performed on the same day. Results represent the average of four determinations and are reported as the percentage of cells which adhere to the surface of each well as compared to the initial number seeded. Thymidine incorporation. The incorporation of thymidine was determined in another series of cultures using standard radiolabeling techniques. First, cells were plated in medium containing 20% fetal bovine serum (FBS) for the initial 3 days of cultivation. This level was reduced to 5% FBS on the 3rd day, and to 5% of the selected serum on the 4th day of cultivation. The control consisted of cultures similarly treated but exposed to medium free of serum on the final day of culture. After 20 hr, all cultures were exposed to 3Hthymidine (45 Ci per mmol, lot T R I C 418, Amersham/Searle, Arlington Heights, Ill.) added to the present serum-containing medium at a concentration of 1 ~Ci per ml of medium. Two hr later, the cells were incubated in fresh unlabeled medium for 10 min, washed twice with cold 5% trichloroacetic acid (TCA, Mallinckrodt, Inc., St. Louis, Mo.) and placed in the cold for the next 6 days with fresh TCA to remove any unincorporated thymidine. The cells then were washed again with distilled water and dissolved in 98% formic acid. Aliquots were assayed for radioactivity. The results are expressed as counts per minute of 3Hthymidine incorporated into TCA-insoluble material during the 2-hr incubation period. Labeling index. M M E C cultures were grown and treated as indicated in the incorporation studies presented above. Cultures were labeled with 3H-thymidine (same lot as above) for 2 hr, washed twice with cold 5% TCA over a 2-day period, rinsed in distilled water and dried. Plates were dipped in NTB3 emulsion (Eastman Kodak, Rochester, N.Y.), dried and stored in the dark for 4 days at 4~ After development (Kodak D-19 developer), the autoradiographs were examined and the percentage of labeled to unlabeled cells was determined in randomly selected fields.

R ESU LTS

Highest attachment efficiency resulted when cells were seeded in the presence of horse serum (Table 1). Succeedingly lower levels were obtained in the presence of bovine, lamb, goat and

TABLE 1 ATTACHMENT EFFICIENCY OF VARIOUS SERA WHEN APPLIED TO MOUSE MAMMARYE P1THELIAL C ELLS Sera

AttachmentEfficiency(% }a

1. 2. 3. 4. 5. 6. 7.

Horse {Grand Island) 13.34 _+1.60 Bovine (Grand Island) 12.23 _+0.33 Lamb {Grand Island) 10.95 _+0.87 Goat (Antibodies, Inc. ) 9.76 _+0.99 Fetal Bovine (Grand Island) 8.87 _+0.45 Porcine (Grand Island) 5.34 _+0.19 Rabbit 1 (Grand Island) 5.24 -+ 1.12 Rabbit 2 (Antibodies, Inc. ) 2.13 _+0.19 Rabbit 3 (Microbiological) 2.73 _+0.40 8. Rat(Microbiological) 1.45_+0.19 9. Mouse (Microbiological) 1.21 _+0.25 10. Human (Microbiological) 1.21 _+0.10 11. None 1.34-+ 0.02 a 1.4 x l0 s cells plated per well; _+S.E. of mean; N = 4. fetal bovine sera. The lowest values occurred in the presence of human, mouse or rat sera. The highest rate of thymidine incorporation was obtained when cells were exposed to rabbit serum (RbS) (Table 2). Media containing human, mouse, horse and porcine sera also stimulated relatively high levels of thymidine incorporation. The lowest incorporation occurred in the presence of either rat or goat sera. In order to establish a possible relationship between these two events, plating attachment and thymidine incorporation were plotted for each tested serum (Fig. 1). With the exception of rat serum, which failed to stimulate either plating attachment or thymidine incorporation, all other sera fell into one of two categories: (a) those which TABLE 2 INCORPORATION OF TRITIATED THYMIDINE INTO TCA-INSOLUBLE MATERIAL IN MAMMARYEPITHELIAL CELLS EXPOSED TO VARIOUSSERA Sera

1. Rabbit 1 (Grand Island) Rabbit 2 (Antibodies, Inc. ) Rabbit 3 (Microbiological) 2. Human (Microbiological) 3. Mouse(Microbiological) 4. Horse (Grand Island) 5. Porcine (Grand Island) 6. Fetal Bovine (Grand Island) 7. Bovine (Grand Island) 8. Lamb (Grand Island) 9. Goat (Antibodies, Inc.) 10. Rat (Microbiological) 11. None a _+S.E. of mean; N :- 4.

~H-ThymidineIncorp. lepta/wellja

23,014 _+1,209 38,622 -+ 1,560 23,799 -+2,373 12,076 -+ 939 11,566_+ 1,176 9,535 _+ 484 9,477_+ 663 6,784 _+ 399 4,870 _+ 281 4,678 _+ 276 2,335_+ 430 1,843 _+ 56 878_+ 51

277

SERUM EFFECTS ON MAMMARY CELLS TABLE 3 NS = NO Se,=m R~S= Rat Serum GS = Goot Se,um LS = i omb Serum BS 9 8ov,~ Serum FBS= Fetol BOVI~ Serum HS : HOrSe Serum P5 = Porcine Serum MS = Mou~ Secure ;.4us= Human Serum RbS= ~ab.,r Ser.m

I HS BS 9 LS eGS

E

LABELING INDEX IN

Sera

RbSo =Rts Mse

J 5

i io

EPITHELIAL

%

1. Rabbit I (Grand Island) Rabbit 2 (Antibodies, Inc. ) Rabbit 3 (Microbiological) 2. Fetal Bovine (Grand Island) 3. Rat (Microbiological) 4. None

ePS

N;

MOUSEM A M M A R Y

C E L L S E X P O S E D TO S E L E C T E D M A M M A L I A N S E R A Labeled Cells

1.62 2.24 1.51 0.61 0.12 0.00

,H~s / i

i eo

3 H - T h y m i d i n e incorporation (cpm ~ IO~/well)

FIG. 1. Attachment efficiencyversus thymidine incorporation in various mammalian sera. N = 4 for each point. permitted high plating attachment and low thymidine incorporation, or (b) those which allowed for low plating attachment and high thymidine incorporation. In no instance did a single type of serum encourage high plating and high thymidine incorporation at the same time. It would appear, therefore, that an inverse relationship exists between these two events for any given type of serum. In order to further assess this relationship and to obtain some indication of the mechanism involved, we selected a representative serum from each of our resultant spectra for continued testing. FBS was chosen to encourage high plating efficiency and low thymidine incorporation, whereas RbS was selected for its ability to stimulate low plating efficiency and high thymidine incorporation, To confirm that the high rates of thymidine incorporation represented actual increases in the number of cells entering the cell cycle rather than simply alterations in the uptake of 3H-thymidine, autoradiographs of cultures exposed to RbS and FBS were prepared. The results indicated the media containing any of the available lots of RbS produced significantly higher labeling indices than media containing FBS or no serum at all (Table 3L Furthermore, the relative percentages obtained for each group reflected the relative rate of incorporation of 3H-thymidine into TCA-insoluble material. A time-course study of thymidine incorporation revealed that cells exposed to either RbS- or FBScontaining media produced almost identical patterns of DNA synthesis (Fig. 2L Following an initial, and similar, burst of synthesis, the cells began a second round of synthesis 12 to 32 hr after addition of the test media. It was during this see-

ond peak of DNA synthesis, which occurred 20 to 22 hr after adding the selected serum-containing medium, that RbS stimulated markedly higher levels of thymidine incorporation than did FBS. The similarity of the two curves, however, suggests that a quantitative, rather than a qualitative, difference exists between the two types of sera and their ability to stimulate the incorporation of the nuclcoside. By comparing the levels of thymidine incorporation in cells exposed to media containing RbS or FBS over a wide range of concentrations, the 2o

~ ~ o

J5

" o

\

g

",

Jo

|

~ 5 , -r

0

I 0

[ S

I

IS I

I

;

I 4

I 32

I

I 40

I

I 4S

Time (hrs) FIG. 2. The effect of fetal bovine and rabbit sera on thymidine incorporation in primary cultures of murine mammary epithelial cells. O, 5% fetal bovine serum; e, 5% rabbit serum. +_S.E. of mean reported; N = 4 for each point.

278

FELDMAN AND WONG

80

to high plating attachment characteristic of cells exposed to medium containing only FBS.

7O

DISCUSSION ~'- 60 o x

~ 50 .T,. o ~ 4o y/.

30 "o E

~_

////

2o

o

2.5

5.0

7.5

I0

20

Fetal bovine serum (%)

o

25

5.0

,.5

,o

~o"

Robbit serum (%)

FIG. 3. The effect of various concentrations of fetal bovine and rabbit sera on thymidine incorporation in primary cultures of murine mammary epithelial cells. _+ S.E. of mean reported; N = 4 for each point. curves generated were similar (Fig. 3L The difference between the two sera once again seems quantitative. For example, 5% RbS provided equivalent levels of thymidine incorporation as 20% FBS. By mixing RbS with FBS, M M E C responded with intermediate levels of thymidine incorporation, but never attained significantly higher levels of plating attachment than that obtained in RbSsupplemented medium alone (Table 4L Thus the presence of RbS in the medium was antagonistic TABLE 4 EFFECT OF VARIOUS COMBINATIONS OF FETAL BOVINE AND RABBIT SERA ON ATTACHMENT AND THYMIDINE INCORPORATION IN PRIMARY CULTURES OF MOUSE MAMMARY EPITHELIAL CELLS Serum Supplements Type

Fetal bovine Rabbit Fetal bovine plus Rabbit Fetal bovine plus Rabbit

Amount

Attachment Efficiency (%)

2.5% 6.30-+0.38 5.0% 10.67 -+0.67 2.5% 1.27-+0.18 5.0% 2.70_+0.24 2.5% 2.71 -+0.23 2.5% 5.0% 2.10+0.20 5.0%

~H-Thymidine Incorp. (cpm/wen)

13,510_+ 971 15,758 _+ 900 13,744+ 631 23,753_+1,110 17,979-+2,078 20,445+2,001

The results obtained in this study suggest that optimal plating attachment and thymidine incorporation cannot be obtained through supplementation of medium with a single type of serum. Since an inverse relationship exists between these two biologic events for any given serum, one type of serum should be selected at the time of seeding and another once the cells have attached successfully to the surface of the culture vessel. Mixing sera which possess abilities to produce high attachment efficiencies (i.e. FBS) with those which stimulate high thymidine incorporation (i.e. RbS) does not produce an additive response. In the case of MMEC, maximum numbers of cells can be obtained in the shortest period of time by seeding the cells in the presence of 20% horse, bovine, lamb, goat, or fetal bovine sera and then changing to a medium containing 5% RbS 48 to 72 hr later. This regimen already has been adopted successfully in our laboratory for routine preparations of cultures. Although it is unlikely that this particular combination of sera would provide optimal results with mammary epithelial cells from other species, or even with other murine cell types, the need to select two different serum supplements to optimize attachment and thymidine incorporation may be necessary for different cell systems as well. Since some sera are less expensive than others and the production of more cultures with fewer numbers of cells would significantly lower quantifies of tissue necessary, a marked saving in cost could be realized by the investigator. Then too, large numbers of cells often are not available or require tedious procedures to secure the amounts of tissue necessary to produce the final cultures. Thus we suspect that savings in significant amounts of time and money can be realized by a careful analysis of one's system with regard to attachment and cell growth requirements.

REFERENCES 1. Dulbecco, R. 1970. Behavior of tissue culture cells infected with polyoma virus. Proc. Natl. Acad. Sci. U.S.A. 67: 1214-1220. 2. Todaro, G. J., Y. Matsuya, S. Bloom, A. Robbins, and H. Green. 1967. Stimulation of RNA synthesis and cell division in resting cells by a factor present in serum. In: V. Defendi, and M. Stoker

SERUM EFFECTS ON MAMMARY CELLS

3.

4.

5.

6.

(Eds.j, Growth Regulating Substances for Animal Cells in Culture. Wistar Institute Press, Philadelphia. pp. 87-98. Todaro, G. J., G. K. Lazar, and H. Green. 1965. The initiation of cell division in a contact-inhibited mammalian cell line. J. Cell Comp. Physiol. 66: 325-333. Wiebel, F., and R. Baserga. 1969. Early alteration in amino acid pools and protein synthesis of diploid fibroblasts stimulated to synthesize DNA by addition of serum. J. Cell Physiol. 191-202. Clarke, G. D., M. P. G. Stoker, A. Ludlow, and M. Thornton. Requirement of serum for DNA synthesis in BHK-21 cells: effects of density, suspension and virus transformation. Nature 227: 798-801. Holley, R. W. 1974. Serum factors in growth control. In: B. Clarkson, and R. Basera (Eds.}, Control of Proli[eration in Animal Cells. Cold Spring Harbor, N.Y., Vol. 1. pp. 13-18.

279

7. Temin, H. M., R. W. Pierson, and N. C. Dulak. 1972. The role of serum in the control of multiplication of avian and mammalian cells in culture. In: G. H. Rothblatt, and V. J. Cristofalo (Eds.D, Growth, Nutrition and Metabolism of Cells in Culture. Vol. 1. Academic Press, pp. 49-81. 8. Gospodarowicz, D., and J . S . Moran. 1976. Growth factors in mammalian cell culture. Annu. Rev. Biochem. 45: 531-558. 9. Foster, R. C., and M. K. Feldman. 1975. Technique for mammary epithelial cell isolation and cultivation. Procedure 46131. TCA Manual 1: 27-30. 10. MacPherson, I., and M. Stoker. 1962. Polyoma transformation of hamster cell clones - - an investigation of genetic factors affecting cell competence. Virology 16: 147-151. 11. Feldman, M. K. 1974. A direct radioimmunoassay for mouse casein. Comp. Biochem. Physiol. 49: 127-135.

T h e authors wish to thank Drs. Roy C. Foster and S. A b r a h a m for their thoughtful comments and encouragement during the course of the research, and Ms. Jolyce Hardesty for preparation of the final manuscript. This research was supported in part by G r a n t No. C A 15764 from the National Cancer Institute, National Institutes of Health, D H E W .