Plasma hormone levels and the incidence of carcinogen-induced mammary tumours in two strains of rat.

The incidence of mammary tumours developing after administration of the carcinogen DMBA (at 50 days of age) has been determined in 2 strains of Sprague-Dawley rat. Untreated animals of each strain were exsanguinated in dioestrus at a time corresponding to the early post-carcinogen stage (at 70 days of age) and the plasma concentrations of prolactin, oestradiol-17B and progesterone were measured by radioimmunoassay. In an inbred strain of rats, tumour-induction rate was 6-4% and plasma prolactin concentration was 2-5 x lower than that found in a random-bred strain with a tumour-induction rate of 41-6%. No difference was found between the 2 strains in the level of either ovarian hormone. It is concluded that the difference between these strains in mammary gland susceptibility to DMBA may be related to plasma prolactin concentration, but it is unlikely to be determined by the ovarian hormones.

that the difference between these strains in mammary gland susceptibility to DMBA may be related to plasma prolactin concentration, but it is unlikely to be determined by the ovarian hormones.
In the rodent, mammary tumours can be readily induced following the administration of 7,12-dimethyl benz(a)anthracene or " DMBA " (Huggins, Grand and Brillantes, 1961) and the sensitivity of the mammary gland to the carcinogen is dependent upon the prevailing endocrine status. In particular, prolactin (Meites, 1972) and ovarian hormones (Geyer et al., 1953;Huggins, Briziarelli and Sutton, 1959;Jabara and Harcourt, 1970) have been reported to alter the sensitivity of their target organ, the mammary gland, to tumorigenesis.
In a previous study from this laboratory (Boyns et al., 1973), it has been shown that in 2 different strains of Sprague-Dawley rat, the rate of tumour induction was much higher in the strain with the higher concentration of plasma prolactin at dioestrus. In view of uncertainty as to whether prolactin or an ovarian hormone has a dominant role in determining the rates of growth and induction of these tumours, we have repeated and extended this work to include assessment of the ovarian hormones, as well as of prolactin, in relation to tumour incidence.

MATERIALS AND METHODS
Animals.-The random-bred strain of rats (OLAC) was originally purchased from Oxford Laboratories. Fresh stock animals have recently been introduced into this strain from Oxford and this appears to have restored sensitivity to tumour induction, which was diminishing with time, to former high levels. The inbred strain of rats (ADRA) was originally purchased from the Animal Diseases Research Association, Edinburgh.
Tumour induction.-Two groups of rats (101 OLAC and 62 ADRA) were each treated by the intragastric administration of 30 mg DMBA, at 50-55 days of age. At this age, the inbred rats (ADRA) had a body weight significantly lower (by approximately 15%) than that of the random-bred rats (OLAC), and thus received a slightly higher dose of DMBA/g body wt. Subsequently, animals were palpated for tumours at regular intervals until approximately 6 months of age.
Collection of blood from animals in dioestrus.-Vaginal smears were taken daily from 13 OLAC rats and 12 ADRA animals, at approximately 70 days of age. After at least 2 oestrous cycles, rats were killed between 11.30 and 12.03 h, in dioestrus, by exsanguination through the abdominal aorta under ether anaesthesia, the blood being collected into a heparinized syringe. The OLAC rats all exhibited regular 4-day cycles, but the ADRA strain showed longer, irregular cycles as judged from the vaginal smears, with a tendency to show no obvious proestrous smear, and a prolongation of the oestrous phase. Accordingly, " dioestrus " was defined for all rats as the third day of sequence in which clear, characteristic smears were seen as follows: Day 1: oestrus, Day 2: metoestrus and Day 3: dioestrus (this therefore corresponded to the day before proestrus in the OLAC rats, but such was not necessarily the case in the ADRA animals). Blood was stored on ice and the plasma was separated within 30 min by centrifugation.
Determination of plasma hormone levels.-Plasma oestradiol-1 7B concentration was measured by radioimmunoassay, as previously described, using a single 3-0-ml portion of plasma (Hawkins et al., 1975). Assay sensitivity for this volume of plasma was 0-16 ng/100 ml plasma.
Plasma prolactin concentration was measured also by radioimmunoassay by the method described previously, with minor modifications (Hawkins et al., 1975). [1251] iodo-prolactin was prepared by the method of Redshaw and Lynch (1974) and the mean assay sensitivity using such preparations was 3-3 ng RP-1/ml (n = 22 assays, range 1 to 7-6 ng/ml). Inevitable deterioration of prolactin standard occurs on storage (NIAMD instructions) and a correction has been made for this using the quality control data, a procedure which results in values lowei than those previously reported by us.
Plasma progesterone concentration was measured using duplicate 100-1l portions of plasma by the method of Cameron and Scarisbrick (1973). Assay sensitivity (defined as the mass required to cause a 10% fall from the maximum found for the standard curve, corrected for losses on extraction) was 0-6 ng/ml plasma.

RESULTS
The rate of induction of mammary tumours was much greater in the randombred strain of rats than in the inbred ADRA strain (Table I). Although the tumours were not all examined histologically, in our experience the formation of fibromata generally occurs much later than 4 months after DMBA administration, and thus this difference almost certainly reflects the difference in induction of adenocarcinomata. % Incidence 41-6 6-4 All rats received 30 mg DMBA intragastrically at 50-55 days of age and were palpated at intervals until approximately 6 months of age. All palpable tumours > 0 5 cm in diameter were counted, irrespective of histology. Incidence was calculated as a percentage of all the rats treated. Examination of the plasma hormone concentration in these 2 strains ofrat (Table  II) revealed no significant difference between the animals in the levels of circulating ovarian hormones, but demonstrated the existence of a significant difference in the levels of circulating plasma prolactin.

DISCUSSION
In an earlier study (Boyns et al., 1973), it was found that in 3 strains of rat, the differences in incidence of tumour induction by DMBA were apparently associated with differences in circulating prolactin concentration measured at 50 days of age, the time of administration of the carcinogen. In the present work, the marked difference in tumour induction rate between the 2 strains of Sprague-Dawley rat has been confirmed, and measurement of the plasma hormone concentrations at 70-days of age shows that the difference in plasma prolactin levels between the strains is maintained from 50 to 70 days. This time would correspond to the early post-carcinogen period in rats treated with DMBA, and in view of the conclusion of Brown and Shellabarger (1974) that differences in tumour susceptibility to DMBA are determined after carcinogen-cell interaction, it may well be the time at which inter-strain differences are involved. In contrast to the difference between strains in plasma prolactin concentration at this stage, we found no difference in the plasma concentration of either of 2 major ovarian hormones. A similar, apparent association between increased incidence of tumour induction and increased plasma prolactin levels has been noted in rats fed on a high fat diet (Chan, Didato and Cohen, 1975) but in that study the ovarian hormones were not examined.
The effects of prolactin and ovarian hormones on the induction of mammary tumours by carcinogens are complex. The absence of either the pituitary gland (Foulds, 1975) or the ovary (Dao, 1962;Talwalker, Meites and Mizuno, 1964) has been reported to inhibit tumour induction.
Thus both prolactin (Meites, 1972) and oestrogen (Sinha and Dao, 1972) have been implicated in tumorigenesis, though tumours can be induced in ovariectomized rats treated by administration of prolactin and growth hormone (Talwalker et al., 1964). By contrast, high levels of either prolactin (Meites, 1972; Gala and Loginsky, 1973) or oestrogen (Nagasawa et al., 1974) may protect the mammary gland from carcinogenesis.
In the 2 strains of rat studied here it seems unlikely that the level of either of the 2 major ovarian hormones normally determines the susceptibility of the mammary gland to carcinogenesis.
It should be noted, however, that in the mouse, attempts to relate plasma, pituitary or urinary levels of prolactin to the incidence of spontaneous tumours in 2 different strains have led to the suggestion that, for prolactin, mammary turnover rather than plasma level may determine susceptibility to spontaneous mammary tumorigenesis (Sinha, Selby and Vanderlaan, 1974;.