New corn technology: scientists are all eyes and ears.

Corn and other crops contaminated with the fungus Aspergillus flavus give off a carcinogenic by-product called aflatoxin, which is blamed for high rates of liver cancer in Asia and Africa, where rice and corn are food staples. In the United States, aflatoxin's major threat is to farm animals, which can get sick or even die from consuming too much of the toxin. Scientists are working on ways to keep the deadly toxin out of the food supply. Two techniques under development identify aflatoxin-tainted corn by using infrared light to elicit telltale sounds and light from contaminated kernels. Other scientists hope to protect corn from A. flavus in the first place by designing genetically engineered aflatoxin-resistant grain species and by working with drugs such as oltipraz that reportedly detoxify aflatoxin already in the body.

A HYPOTHESIS has been presented implicating adrenal dehydroepiandrosterone (DHEA) and DHEA sulphate (DHEAS) in the aetiology of human breast cancer (Adams, 1977). The main metabolites of DHEA, produced by in vitro incubation with human mammary tumours, are 7oxygenated derivatives (7a-hydroxy DH-EA, 7P-hydroxy DHEA and 7-oxo DHEA) and 5-androstene-3g, 17/-diol (Li et al., 1978). The latter can cause translocation of the oestrogen receptor in both the uterus and mammary tumours Nicholson et al., 1978). Although it is difficult to be certain of the exact role played by the 7-hydroxylation system in mammary tissue, it has been suggested ) that it could function as an inactivation step to modify oestrogen receptor translocation by 33hydroxy-A5-androstenes, which are known to occur in relatively high concentrations in human mammary tumours (Maynard et al., 1977). Thus high levels of DHEAS in the blood could lead to high intracellular levels of DHEA and its metabolites formed in situ in mammary tissue.
Pathways of DHEA metabolism similar to those established in human mammary tumours also occur in 7,12-dimethylbenz-(a)anthracene (DMBA)-induced rat mammary tumours (Li et al., 1976). It was thus of interest to use this as a model to study the influence of an altered sexhormone milieu in tumour-bearing rats upon subsequent DHEA metabolism in the tumour. In parallel studies, the influence of an altered hormone milieu on oestrogen metabolism by these tumours has been investigated.
One other aspect of DHEA metabolism by human mammary tumours, i.e. formation of the sulphate ester, is of relevance when considered together with the sulphurylation of oestradiol-17/. Measurement of these conversions by the tumour in vitro has been shown to relate to the patient's subsequent prognosis (Dao & Libby, 1972). Since the enzymes responsible for the sulphurylation of DHEA and oestradiol-17,f are also present in DMBAinduced rat mammary tumours, the influence of similar hormone manipulations on steroid sulphurylation in the tumours has been included in the present, investigationl. MATERIALS  adenosine -3' -phosphate -5' -phosphosulphate ([35] PAPS) and unlabelled PAPS wvere prepared as described previously (Adams et al., 1974). DMBAindued tumnour8. Female Sprague-Dawrley rats (50 days old) wvere given a single oral dose of 20 mg DMBA in peanut oil. Animals wsere provided with rat chow, and water ad libitum and held in an animal house subjected to the normal daylight cycle. When turnours had reached an approximate size of 2 x 2 cm, oophorectomy was performed. Tumours which regressed after oophorectomy, as judged by measurement writh calipers daily for 10 days, wNere classed as hormone-dependent and provided the basis for this study. These rats wvere then divided into 3 groups: Group 1 did not receive any further treatment and acted as controls, Group 2 received daily s.c. injections of oestradiol-17f in normal saline (1.5 /.g/day) until regrowth of the tumour occurred. while Group 3 received oestradiol-17fl as in Group 2, but on the last 5 days before killing progesterone (2.5 mg) in propylene glycol wvas also injected s.c. (in this case regrowth for 20 days wNas allowed before progesterone was administered). All animals were killed at about the same time in the morning. Six tumours wrere subjected to the biochemical studies in each of the 3 groups, and sample sections were retained for histology. PAPS synthesis.-Tumour tissue (0 5 g) wa.s cut into small pieces and packed into the chamber of a stainless-steel cylinder block pre-chilled in dry ice and pounded with a close-fitting stainless-steel plunger. The resulting powder was then homogenized in 2 5 ml of 04154M KCI containing 0ImM dithiothreitol, using a Potter-Elvejhem homogenizer with Teflon head.
After centrifugation for 10 min at 6,000 g to remove cell debris, the supernatant was spun at 100.000 g for 1 h. An incubation mixture was set up which contained, in a total volume of 2 0 ml: 1 82 jumol ATP, 3 64 Ftmol MgCl2, 0 04 1.mol [35S] Na2SO4 (sp. act. 1500 d/min/pmol), 9 1imol Tris-HCl buffer (pH 7.4) and 0.1 ml of cytosol. Reaction was carried out for 30 min at 37°C and stopped by placing the tubes in a boiling water bath for 1 min. After centrifugation to remove any precipitated protein, a sample (10 ,tl) was applied as a 2cm streak to strips of Whatman No. 1 paper. Electrophoresis was carried out for 4 h in 0-05M sodium phosphate buffer, pH 6-0. A potential of 10 V/cm was applied and the procedure carried out in the cold

room. Radioactive [35S] PAPS and [35S]
adenosine-5'-phosphosulphate (APS) on paper electrophoretograms were detected by scanning with a Nuclear Chicago Actigraph III instrument, and identified by comparison with the relative mobilities of authentic compounds. The radioactive regions were cut out for liquid-scintillation counting. Activity of the PAPS-synthesizing system was then expressed as pmol/mg cytosol protein used in the incubation.
Oestrogen and steroid alcohol sulphotransferases. Sulphotransferase assays in tumours used preformed [35S] PAPS, and were carried out in incubation mixtures of the following composition: 0(02 [kmol [35S] PAPS (2 x 105 d/min), 3 p.mol MgC12, 8 Hmol Tris-HCl (pH 7.4), 7-5 nmol steroid in 5 ,ul of propylene glycol, and 0 1 ml of cytosol in a total volume of 0-15 ml. Incubation was continued for 1 h at 37°C. Controls contained propylene glycol alone. Steroid [35S] sulphate was assayed by extraction into ethyl acetate (Dao & Libby, 1972). Counts in the controls were subtracted. and the net counts expressed as pmol of steroid sulphate/mg protein/h. DHEA metabolism.-Incubation of finely minced tumour tissue (0 5 g) in 2 ml of modified Krebs-Ringer phosphate buffer (pH 7.4) wN-ith [4-14C] dehydroepiandrosterone (0 5 ,uCi) and an NADPH-generating system was carried out as described by Li et al. (1978). Products were extracted, separated, deter-mined and characterized as described in previous publications (Li et al., 1976(Li et al., , 1978. Oestrone metabolism.-Minced tumour tissue (0-5 g) was added to 5 ml Krebs-Ringer phosphate buffer (pH 7.4) fortified with 16 7mM glucose, 50mM sodium fumarate and 10mM nicotinamide. [4-14C] oestrone (015 ,tCi) in 0 05 ml ethanol was added and the mixture incubated at 37TC for 2 h with shaking in air. After addition of carriers (10 jug each of oestrone, oestradiol-17P and oestriol), the reaction was stopped by addition of 5 volumes of acetone. The mixture was homogenized and filtered and the residue washed with acetone. After removal of the acetone, the remaining aqueous phase was extracted with ether (3 x 15 ml) and, after drying (Na2SO4), the products were separated by thin-layer chromatography on silica gel using chloroform ethyl acetate (9:1, v/v). The [4-14C] oestradiol peak, revealed after scanning, was cut out and measured by liquid-scintillation counting. The identity of the oestradiol was confirmed by derivative formation and comparison with authentic compounds in various thin-layer chromatography systems.
Statistical evaluation.-Student's two-sided t test was applied. Tabulated results are expressed as means + s.e.
Values are means + s.e. when progesterone was included. However, owing to the variability between tumours, these results were not significant (Group 3 vs Group 1, P < 0 1, Table I). Tumours from oophorectomized animals (Group 1) generally showed degenerative regressional changes on histological examination. These changes occurred mostly in the form of pools of mucinous material deposited in the lobular units. In some tumours, the lobular units had lost their distinct architectural form and showed diffuse infiltration of the connective tissue. Tumours from animals receiving oestrogen (Group 2) were generally more cellular, the mucinous material had largely disappeared and infiltration of the connective tissue was common. Oestrogen + progesterone (Group 3) led to tumours with a higher degree of structural organization and an increased secretion of PAS+ material. The stroma of tumours from Groups I and 2 showed considerable variegation, and in some cases diffuse inflammatory response with the presence of mast cells.

Metabolism of [4-14C] DHEA and [4-14C] oestrone
Major products of the in vitro metabolism of DHEA by rat mammary tumours were previously shown to be 7acand 73hydroxy DHEA, 7-oxo DHEA and 5androstene-3/,17/3-diol (Li et al., 1976). Oestrogen administration to oophorectomized rats led to a drop in each of the 7-oxygenated products of DHEA. Although the total 7-oxygenated products oestradiol-17f, 4-83 + 1-27 23-6 + 4-39** 26-9 + 0 97*** were significantly reduced by some 37 % (P < 0 02) by oestrogen administration, reduced formation of the main component, 7a-hydroxy DHEA, was not significant (P <0 2). Progesterone tended to reverse the effects of oestrogen, as judged by formation of total 7-oxygenated products ( Table II). Formation of 5-androstene-3/,17/3-diol was unaltered by hormone administration. Incubation of [4-14C] oestrone (3 nmol) with 05g samples of tumour minces in Krebs-Ringer buffer fortified with glucose and nicotinamide, caused a 458% conversion to oestradiol-17/. Treatment of oophorectomized animals with oestrogen, or oestrogen + progesterone, caused a 5-fold increase in the rate of oestradiol formation from [4-14C] oestrone (Table II). DISCUSSION Levels of steroid sulphotransferases in human mammary tumours have been shown to relate to the overall prognosis of the patient, and to the response to hormone ablation (Dao & Libby, 1972). The ability of an individual tumour to generate sufficiently high concentrations of PAPS plays a dominant role in the determination of the degree of sulphurylation of steroids added to the incubations of the tumour extracts (Dao & Libby, 1972;Li et al., 1976;Adams & Chandra, 1977). From data of the type presented in the Figure, it was calculated that the concentration of PAPS in the incubations with  tumour cytosol preparations would only reach 1-2 tM. The Km for PAPS, using purified bovine placental oestrogen sulphotransferase, is 37 pM (Adams et al., 1974) and the corresponding values and DHEA sulphotransferase are 13 /M for enzyme from rat liver and 20 /tM for enzyme from bovine liver (Adams & McDonald, unpublished). Thus, assay of oestrogen and 3/hydroxysteroid sulphotransferases were carried out by addition of saturating concentrations of the cosubstrate PAPS. The specific increase in oestrogen sulphotransferase by progesterone (Table I), although not significant in the present experiments, is comparable to a similar influence of this hormone on the enzyme in the endometrium in the gilt (Pack & Brooks, 1974) and human (Buirchell & Hahnel, 1975) uterus. In these cases, the enzyme is induced by progesterone at the end of the proliferative phase of the oestrous and menstrual cycles, and appears to regulate the localized concentration of oestrogen by conversion to the water-soluble inactive conjugate. In the DMBA tumour progesterone may play a similar role, and would provide another example of a "modifying" effect of progesterone on oestrogen action. Although progesterone administration led to large increases in the activity of oestrogen sulphotransferase in some tumours, the effect was not consistent, and the results were not statistically significant. The reason for this is unknown. No apparent differences were noted on histological examination oftumour sections, and rates of tumour regression after oophorectomy were similar, and did not explain the variance in oestrogen sulphotransferase levels.
In mammary tumours from the rat (Li et al., 1976) and human (Couch et al., 1975;Li et al., 1978) 7-oxygenated products are the main metabolites formed from IDHEA in vitro. It has been established that these conversions are catalysed by enzymes, and are not due to chemical oxidation at the active methylene group at position 7 in 5-androstenes (Li et al., 1976). In an extensive series of experi-9 ments, no evidence was found that 7hydroxy DHEA synthesis represented an intermediate in a more elaborate biosynthetic pathway (Li & Adams, unpublished). However, the presence of the 7hydroxyl group in 3/-hydroxy-A5-steroids does convey certain properties which suggest that the 7-hydroxylase may play an important role in the cell. For example, introduction of the 7-hydroxyl group into 5-androstene-33,17/-diol markedly reduced this steroid's capacity to combine with the oestrogen receptor (Li et al., 1978). In addition, the ability of DHEA and 5-androstene-3/,17P-diol to inhibit oestrogen sulphotransferase was nullified by 7-hydroxylation .
Although oestrogen administration did significantly lower formation of 7-oxygenated derivatives of DHEA, the effect was not pronounced. In rat liver, the 7hydroxylase for DHEA has unusual properties compared to other steroid hydroxylases. It appears to fall more in the class of a "constitutive" enzyme, since it is present in the liver of sexually immature males and females and, in contrast to 16oz-hydroxylase, the presence of androgen at a critical period of neonatal life is not needed for expression of the enzyme in the liver (Tabei & Heinrichs, 1974). Both 7acand 16oz-hydroxylases for DHEA in adult male rat liver are sensitive to light/ dark modulations, and it has been suggested that the 7cx-hydroxylase is adrenaldependent, rather than both gonadand adrenal-dependent, as is the case of the 16a-hydroxylase (Schafer & Colas, 1975). The 17/-hydrogenation of DHEA to 5-androstene-3p,173-diol was unaffected by hormone administration (Table II) which appears to reflect the situation in rat liver; this transformation being unaffected by castration in either sex (Tabei & Heinrichs, 1974).
Mammary tumours whose growth was stimulated by oestrogen, or oestrogen + progesterone, showed a greater capacity to convert oestrone to oestradiol-17/ than regressing tumours (Table II). In vitro studies have shown that human mammary tumours cultured in the presence of either labelled oestrone or oestradiol-173 showed higher ratios of oestradiol-17,B to oestrone than did normal or benign tissues (Willcox & Thomas, 1972;Geier et al., 1975). The specific activities of oestradiol-178 dehydrogenase in subcellular fractions of breast-cancer tissues are lower than those from non-malignant tissue (Pollow et al., 1977). Thus, whilst a decrease in oestradiol-17/ dehydrogenase could explain the decrease in oestradiol-17/ metabolism in neoplastic mammary tissue, it is inconsistent with the observed higher conversions of oestrone to oestradiol-17,. King et al. (1965) made the interesting observation that faster-growing DMBAinduced rat mammary tumours converted oestrone to oestradiol at a greater rate than slower-growing tumours. One explanation for the in vitro results obtained with DMBA-induced tumours (Table II) and the human tumours referred to above, could be an altered redox state of pyrimidine nucleotides. There is well-documented evidence that oestradiol-17, induces glucose-6-phosphate dehydrogenase (GPDH) in both normal and neoplastic mammary tissue (Hilf et al., 1967;Ringler & Hilf, 1975). Furthermore, GPDH activity is higher in neoplastic tissue (Smith et al., 1966;Knox, 1967;Richards & Hilf, 1972) and neoplastic mammary tissue contains higher oestrogen-receptor concentrations than normal mammary tissue, thus suggesting a correlation between GPDH and oestrogen receptor (Daehnfeldt & Schulein, 1975). An increase in NADPH/NADP+, via raised GPDH levels, could then account for the increase in the rate of oestradiol-17, formation from oestrone in DMBA-induced tumours after oestrogen administration to oophorectomized animals.