DOMESTIC ANIMAL ENDOCRINOLOGY Vol. 5(2):141-147, 1988

INFLUENCE OF EPIDERMAL GROWTH FACTOR ON GROWTH OF BOVINE MAMMARY TISSUE IN ATHYMIC NUDE MICE*

Lewis G. Sheffield and In Suh Yuh

Dairy Science Department University of Wisconsin, Madison, Wl 53706

Received July 16, 1987

ABSTRACT

Bovine mammary tissue obtained from midpregnant Holstein heifers by surgical biopsy was transplanted subcutaneously to ovariectomized athymic nude mice (n =" 5 heifers). After 3 weeks recovery, mice were either sham operated or sialoadenectomized (sub- mandibular salivary glands removed). After an additional week, mice were injected with saline or 17[~-estradiol + progesterone (llJg + 1 mg/day) for 2 days. In addition, half of the sialoadenectomized mice were injected with epidermal growth factor (5 ~tg/ day). Grafted tissue was removed and rate of deoxyribonucleic acid (DNA) synthesis estimated by incorporation of 3H thymidine. Est_radiol + progesterone increased the incorporation of 3H thymidine from 77 + 20 dpm/Ix8 DNA to 472 + 53 dpm/ltg DNA. In sialoadenectomized mice, DNA synthesis was increased from 88 + 16 dpm/l~g DNA (saline treated) to 360 + 29 dpm/lag DNA (estradiol + progesterone treated). In staloadenectomized mice treated with epidermal growth factor, DNA synthesis in es- tradiol + progesterone treated mice was 529 + 36 dpm/tt8 DNA, compared to 112 + 30 dpm/l~g DNA in sialoadenectomtzed mice treated with epidermal growth factor. These data indicate that sialoadenectomy of athymic nude mice decreased the ability of transplanted bovine mammary tissue to increase DNA synthesis in response to estradlol and progesterone. This inhibition was removed by epidermal growth factor treatment. These data suggest a physiological role of epidermal growth factor in regulating de- velopment and hormone responsiveness of bovine mammary tissue.

INTRODUCTION

Epidermal growth factor (EGF) is a single chain polypeptide of 53 amino acids with a molecular weight of 6045 Da (1,2). It is found in high concen- trations in the submandibular salivary glands of mice (18), but has been identified in other tissues as well (4-6). In vitro, EGF has a variety of biological actions, including increasing growth of many tissues (1,2). That EGF increases growth of rodent mammary tissue in v i tro is well documented (7,8). However, a mammogenic role of EGF in v ivo has only recently been demonstrated (9). Okamoto and Oka (9) reported that sialoadenectomy of female mice reduced subsequent survival of offspring, probably due to decreased mammary gland size and milk production. The decreased offspring survival was alleviated by EGF injections. Subsequently, Sheffield and Welsch (10) demonstrated that sialoadenectomy reduced the ability of mouse mammary glands to respond to ovarian steroids in vtvo. Responsiveness of mouse mammary tissue to ovarian steroids could be restored by subcutaneous transplants of salivary tissue. Thus, there is mounting evidence that the salivary gland has an endocrine role in regulating mammary gland development, at least in the mouse. The effect of the salivary gland on mammary gland development may be mediated by EGF (9). However, all experiments to date have been performed using the mouse mammary gland as a model. The objective of this study was to determine if

Copyright © 1988 by DOMENDO, INC. 141 0739-7240/88/$3.00

142 SHEFFIELD AND YUH

EGF has a mammogenic effect on bovine mammary tissue in vivo. Since EGF is not available in quantities sufficient to inject into cows, this study utilized bovine mammary tissue transplanted subcutaneously to athymic nude mice as a model system.

MATERIALS AND METHODS

Female athymic nude mice (Harland Sprague Dawley, Madison, WI) were housed under aseptic conditions in a small animal isolater (CCI Equipment, Marietta, OH). All food, water, cages and bedding were sterilized prior to use. Aseptic technique was used when working with mice or tissue to be transplanted to mice.

Bovine mammary tissue was obtained from mid pregnant (4-6 months) Holstein heifers by surgical biopsy (11) (n----5 heifers). Mammary tissue was cut into slices .3 - .5 mm thick with a Stadie Riggs tissue slicer, cut into pieces approximately 3 - 5 mm 2 and placed subcutaneously in 5 week old ovariec- tomized athymic nude mice via a dorsal incision (12,13).

After three weeks recovery, mice were either sham operated or sialoade- nectomized (submandibular salivary glands removed, other salivary glands left in place). After an additional week, mice were injected daily for 2 days with saline or 17~-estradiol + progesterone (1 [tg/day + 1 mg/day). In addition, half of the sialoadenectomized mice were injected with EGF (5 ~tg/day) (Sigma Chemical Co., St. Louis, MO)(one mouse per heifer per treatment). Mice with tissue from a given heifer were housed together (in the same cage), injected together, killed together and tissue processed together.

Twenty four hours after the last injection, mice were anesthetized with pentobarbital, blood collected from the vena cava and grafted bovine mammary tissue removed. Death of the mouse was insured by cervical dislocation. Because of possible circadian rhythms in serum EGF concentration (15), mice were maintained on a 12 hr light: 12 hr dark photoperiod. Mice were killed within about 1 hr of each other, beginning about 1 hr before the beginning of the light phase of the photoperiod (i.e., while serum EGF concentrations were high). Mouse serum was frozen for later analysis of EGF by radioimmunoassay (15). Assay buffer used for EGF determination was O.O5M phosphate buffered (pH 7.4) saline (0.9% NaCI) containing 1% bovine serum albumin (radioim- munoassay grade, Sigma Chemical Co., St. Louis, MO) and O. 1% sodium azide. Each assay tube contained 50 ~tl of standard EGF (receptor grade, Sigma) or mouse serum (undiluted or diluted 1:4 with assay buffer), 100 lxI12q-EGF (Amersham Co., Arlington Heights, IL) (25,000 cpm/tube diluted in assay buffer) and 100 }11 rabbit anti-EGF (Calbiochem, LaJolla, CA) (diluted 1:100,000 in assay buffer containing 4% normal rabbit serum). Each sample was assayed in duplicate. The assay tubes were incubated 48 hr at 5C. Sheep anti rabbit serum (Sigma) (diluted 1:7 in assay buffer) was added to each tube (100 ~tl) and the assay tubes incubated an additional 24 hr. Tubes were then centrifuged (4,O00 x g for 15 min), the pellet washed in cold assay buffer and the tubes (representing bound EGF) counted in a gamma ray spectrometer. This procedure resulted in an assay detection limit of approximately 10 pg/tube (0.2 ng/ml undiluted serum). Within the limits of the assay, 1:4 dilution of samples did not affect results. Intra- and inter-assay coefficients of variation were 10% and 13%, respectively. Addition of known quantities of purified (receptor grade) EGF to pooled mouse serum gave the expected increase in EGF (within the limits of the assay).

EPIDERMAL GROWTH FACTOR AND MAMMARY GROWTH 143

One or two pieces of transplanted bovine mammary tissue were fixed in 10% buffered formalin and used for histological examination. Remaining bovine tissue was placed in Eagle's Minimum Essential Medium supplemented with non-essential amino acids (Sigma) containing 2 ~tCi/ml methyl 3H-thymidine (78.8 Ci/mMole)(New England Nuclear, Boston, MA). Tissue was incubated for 4 hr at 37C in an atmosphere of 5% CO2/95% air. After incubation, tissue was frozen on dry ice, homogenized in .9% NaCI and proteins and nucleic acids precipitated with 10% trichloroacetic acid (TCA). Excess 3H was removed by washing twice with cold 5% TCA. Lipids were removed by sequential washes with methanol:chloroform (2:1, vol:vol), ethanol and ethyl ether. Deoxyri- bonucleic acid (DNA) was extracted by heating with 5% perchloric acid (70C, 30 rain). An aliquot of the extract was neutralized and counted for 3H by liquid scintillation and the remainder used to determine DNA by the diphen- ylamine reaction (16). DNA synthesis rate was defined as dpm 3H/lig DNA.

Data were analyzed by analysis of variance as a randomized complete block design, with heifers considered random blocking factors. Treatment standard errors were corrected for block effects as described by Anscomb and Tukey (17). Planned contrasts (18) were used to compare responsiveness of DNA synthesis to estradiol and progesterone in bovine mammary tissue in sham operated mice, sialoadenectomized mice or sialoadenectomized mice treated with EGF. For sham operation, sialoadenectomy and sialoadenectomy + EGF, mean responsiveness of bovine mammary tissue DNA synthesis to estradiol and progesterone was estimated by subtracting DNA synthesis of tissue from saline treated mice from that of estradiol + progesterone treated mice. Comparisons of hormone responsiveness were made between sham operated and sialoade- nectomized mice, between sialoadenectomized and sialoadenectomized-EGF treated mice and between sham operated and sialoadenectomized-EGF treated mice. Unless otherwise stated, comparisons were two sided with P set at .05 (18,19).

RESULTS

Transplanted bovine mammary tissue retained essentially normal morphology (Figure 1). Histological morphology was not altered by sialoadenectomy, EGF or estradiol + progesterone. Mouse body weights (Figure 2A) were not affected

Fi B. 1. Histological section (hemotoxylin and eosln stained) of bovine mammary tissue maintained su~ 'utaneously in an athymlc nude mouse for 4 weeks. (x160).

144

_o uJ

>- ¢3 o en

A

3° 1

20

10

MOUSE BODY WEIGHTS

• SALINE [ ] E + P

SHEFFIELD AND YUH

SHAM S IAL . SIAL. + EGF

A . J

z Lk o LU

r r LU CO

B

2.5

2.0

1.5

1.0

0.5

0.0

MOUSE SERUM EGF

• SAUNE [ ] E + P

SHAM S IAL . SIAL.+EGF

Fig. 2. Body weights (A) and serum epidermal growth factor (EGF) (B) of athymic nude mice sham operated (SHAM) or sialoadenectomized (SIAL) and injected for 2 days with saline, estradiol + progesterone (E + P) and/or EGF. (n = 5 per group.)

by sialoadenectomy, hormone or EGF treatment (P>.05). Sialoadenectomy significantly reduced serum EGF concentrations (to undetectable concentrations (e.g.<0.2 ng/ml) (Figure 2B), while serum EGF was detectable in sham operated mice. Estradiol + progesterone treatment for 2 days did not significantly affect serum EGF concentration. Serum EGF was very low or undetectable (<0.2 ng/ ml; considered 0 for purposes of statistical analysis) in sialoadenectomized EGF treated mice, possibly because of the 24 hr lag between the last EGF injection and blood collection.

In sham-operated, saline-treated mice, the DNA synthesis rate of transplanted bovine mammary tissue was 77 _+ 20 dpm/gg DNA. This was increased (P<.O5) to 472 _+ 53 dpm/gg DNA by estradiol + progesterone, for a hormone response of 395 dpm/~g DNA (Figure 3). In sialoadenectomized mice, the hormone responsiveness was 272 dpm/~tg DNA (88 _+ 16 dpm/gg DNA in saline treated, 360 -+ 29 dpm/~tg DNA in estradiol + progesterone treated). The response to estradiol + progesterone in sialoadenectomized mice was significantly greater than zero (P<.05) and also significantly less than the response in sham operated mice (P<.O5). In sialoadenectomized mice treated with EGF, DNA synthesis of bovine mammary tissue in saline treated mice (112 _+ 30 dpm/ ~tg DNA) was not different than in tissue maintained in sham operated or

EPIDERMAL GROWTH FACTOR AND MAMMARY GROWTH 145

Z a

E a .

'13

600'

500'

400

300

200

• SAUNE

100

0 S H A M S I A L . SIAL. + EGF

Fig. 3. Incorporation of 3H thymidine into deoxyribonucleic acid (DNA) of bovine mammary tissue maintained in athymic nude mice that were sham operated (SHAM) or sialoadenectomized (SIAL) and injected for 2 days with saline, estradiol + progesterone (E + P) and/or epidermal growth factor (EGF). Mean and standard error of tissue from 5 heifers.

sialoadenectomized mice (P>.05). However, in sialoadenectomized mice treated with EGF, estradiol and progesterone, DNA synthesis was 529 +- 36 dpm/pg DNA, for a response of 417 dpm/pg DNA. This response was significantly greater than sialoadenectomized mice not treated with EGF (P<.05) but not different than sham operated mice (P>.05), indicating that EGF could alleviate the decreased hormone responsiveness induced by sialoadenectomy.

DISCUSSION

Results of histological analysis indicated that bovine mammary tissue was readily maintained in athymic nude mice. These results were consistent with previous studies (12-14). In addition to maintaining normal morphology, transplanted bovine mammary tissue retains the ability to respond to a variety of mammogenic and lactogenic hormones (12-14). Sheffield and Welsch (12,13) observed a significant increase DNA synthesis in bovine mammary tissue main- rained subcutaneously in athymic nude mice treated with estradiol, proges- terone, cholera toxin, growth hormone and/or prolactin. Welsch et al (14) observed that estradiol + progesterone + growth hormone + prolactin induced DNA synthesis rates in transplanted bovine mammary tissue approximately equal to that of mid pregnant cows. In this study, bovine mammary tissue also retained the ability to elicit a substantial (approximately 4 fold) increase in DNA synthesis in response to short term (2 day) treatment with estradiol + progesterone. However, due to the short duration of the treatments, differences in morphology or epithelial tissue area were not apparent in histological sections.

Results of this study suggest that the mouse submandibular salivary gland may secrete an endocrine factor important for normal growth of xenografted bovine mammary tissue. In particular, this factor would appear to be required for the t r~splanted bovine mammary tissue to respond fully to other mam- mogens (i.e., estl~diol and progesterone). EGF would appear to be a likely candidate for the salivary gland factor regulating mammary gland hormone responsiveness. EGF injections were found to restore hormone responsiveness in this study. Okamoto and Oka (9) found that EGF injections alleviated the sialoadenectomy-induced decrease in offspring survival, presumably by in- creasing mammary gland size and milk production. In another study, Sheffield

146 SHEFFIELD AND YUH

and Welsch (1 O) observed that EGF restored responsiveness of mouse mammary tissue to estradiol and progesterone.

EGF receptors have been identified in rodent mammary tissue (20) and the intramammary concentration of EGF receptors increases during early pregnancy (21). In addition, Vonderhaar (22) found that estradiol and progesterone injections substantially increased EGF receptor content of mouse mammary tissue. Recently, Spitzer and Grosse (23) identified EGF receptors in bovine mammary tissue. They observed molecular weights of 160 kDa, 145 kDa and 115 kDa for EGF binding proteins. In addition, they observed a dissociation constant of 1-3 nM, similar to that of mouse mammary tissue. However, the amount of EGF bound by bovine mammary membranes was about an order of magnitude greater than what has previously been reported for mouse mammary gland membranes. The study by Spritz and Grosse (23) also indicated that mammary gland membranes from pregnant cows had approximately twice the EGF binding capacity of those from lactating cows. They suggested that this indicated a greater EGF binding in rapidly proliferating bovine mammary tissue. However, they did not examine tissue from non-pregnant, non-lactating animals, so this conclusion should be considered tentative.

Serum EGF concentrations have been shown to increase during pregnancy in mice and humans (15,24) and were decreased by sialoadenectomy in this and previous studies (9,15). EGF is known to increase rodent mammary epithelial growth in vi tro (7,8). This would suggest that EGF might act directly on the mammary gland to increase growth. However, EGF may also act at extramammary sites to alter product ion of other mammogenic hormones. For example, EGF has recently been shown to augment growth hormone secretion by pituitary fragments in vitro (25). Thus, while EGF appears capable of affecting the mammary gland directly, it may also have extramammary actions affecting mammary development.

All of these factors suggest a mammogenic role for EGF. Although the mammogenic properties of EGF have been best described in rodents, the results of this study indicate that the mammogenic actions of EGF may not be restricted to rodents but appear to apply to the bovine as well, at least when tissue is transplanted to athymic nude mice. Whether the results are indicative of tissue in the intact cow is not known at this time.

ACKNOWLEDGEMENTS AND FOOTNOTES

'Supported by University of Wisconsin College of Agricultural and Life Sciences and Hatch Grant 3108.

Address correspondence and reprint requests to: Dr. L.G. Sheffield, Dairy Science Department, 266 Animal Science Building, University of Wisconsin, Madison, WI 53706.

REFERENCES

1. Carpenter G, Cohen S. Epidermal growth factor. Annu Rev Biochem 48:193-216, 1979.

2. Cohen S, Savage CR, Jr. Recent studies on the chemistry and biology of epidermal growth factor. Recent Prog Hormone Res 30:551-574, 1974.

3. Turldngton RW, Males JL, Cohen S. Synthesis and storage of epithelial-epidermal growth factor in submandibular gland. Cancer Res 31:252-256, 1971.

4. Fallon JH, Seroogy KB, Loughlin SE, Morrison RS, Bradshaw RA, Knauer DK, Cun- ningham DD. Epidermal growth factor immunoactivity in the central nervous system: Location and development. Science 224:1107-1109, 1984.

5. Oka Y, Orth DN. Human plasma epidermal growth factor is associated with blood platelets. J Clin Invest 72:249-259, 1983.

EPIDERMAL GROWTH FACTOR AND MAMMARY GROWTH 147

6. Shikata H, Utsumi N, Hiramatsu M, Minami N, Nemoto N, Shidata T. Immunological localization of nerve growth factor and epidermal growth factor in guinea pig prostate gland. Histochemistry 80:411-413, 1984.

7. Taketani Y, Oka T. Epidermal growth factor stimulates cell proliferation and inhibits functional differentiation of mouse mammary epithelial cells in culture. Endocri- nology 113:871-877, 1983.

8. Yang J, Guzman R, Richards J, Imagawa W, McCormick K, Nandi S. Growth factor- and cyclic nucleotide-induced proliferation of normal and malignant mammary epithelial cells in primary culture. Endocrinology 107:35-41, 1980.

9. Okamoto S, Oka T. Evidence for physiological function of epidermal growth factor: Pregestational sialoadenectomy of mice decreases milk production and increases offspring mortality during lactation period, proc Natl Acad Sci USA 81:6059-6063, 1985.

10. Sheffield LG, Welsch CW. Influence of submandibular salivary glands on hormone responsiveness of mouse mammary glands. Proc Soc Exp Biol Med, 186:368-377, 1987.

11. Byatt JC, Bremel RD. Lactogenic effect of bovine placental lactogen on pregnant rabbit but not pregnant heifer mammary gland explants. J Dairy Sci 69:2066-2071, 1986.

12. Sheffield LG, Welsch CW. Transplantation of bovine mammary tissue to athymic nude mice: Growth subcutaneously and in mammary gland-free fat pads. J Dairy Sci 69:1141-1147, 1986.

13. Sheffield LG, Welsch CW. Cyclic nucleotide concentration and protein kinase activities of bovine mammary tissue maintained in athymic nude mice: Effects of mammogenic and lactogenic hormones. J Dairy Sci, 71:75-83, 1988.

14. Welsch CW, McManus MJ, DeI-Ioog JV, Goodman GT, Tucker HA. Hormone-induced growth and lactogenesis of grafts of bovine mammary gland maintained in the athymic "nude" mouse. Cancer Res 39:2046-2050, 1979.

15. Kurachi H, Oka T. Changes in epidermal growth factor concentrations of subman- dibular gland, plasma and urine of normal and sialoadenectomized female mice during various reproductive stages. J Endocrinol 106:197-202, 1985.

16. Burton KA. A study of the conditions and mechanisms of the diphenylamine reaction for the colorimetric estimation of ~leoxyribonucleic acid. Biochem J 62:315-323, 1956.

17. Anscomb FJ, Tukey JW. The examination and analysis of residuals. Technometrics 5:141-160, 1963.

18. Gill JL. Design and Analysis of Experiments in the Animal and Medical Sciences. Vol. 1. Iowa State University Press, Ames, 1978.

19. Gill JL. Design and Analysis of Experiments in the Animal and Medical Sciences. Vol. 2. Iowa State University Press, Ames, 1978.

20. Taketani Y, Oka T. Biological action of epidermal growth factor and its functional receptors in normal mammary epithelial cells. Proc Nati Acad Sci USA 80:2647- 2650, 1983.

21. Edery M, Pang K, Larson L, Colsi T, Nandi S. Epidermal growth factor receptor levels in mouse mammary glands in various physiological states. Endocrinology 117:405-411, 1985.

22. Vonderhaar BK. Hormones and growth factors in mammary gland development. In: Control of Cell Growth and Proliferation, Veneziale CM (ed). Van Norstand-Rein- hold, New York, p. 11-33, 1985.

23. Spitzer E, Grosse R. EGF receptors on plasma membranes purified from bovine mammary gland of lactating and pregnant animals. Biochem Internatl 14:581-588, 1987.

24. Ances IG. Serum concentration of epidermal growth factor in human pregnancy. Am J Obstet Gynecol 115:357-362, 1973.

25. Ikeda H, Mitsuhashi T, Kubota K, Kuzuya N, Uchimura H. Epidermal growth factor stimulates growth hormone secretion from superfused rat adenohypophyseal frag- ments. Endocrinology 115:556-558, 1984.