Comparison of the in vitro conversion of estradiol-17β to estrone of normal and neoplastic human breast tissue

Molecular and Cellular Endocrinology, 0 Elsevier/North-Holland

6 (1977) 333-348 Scientific Publishers, Ltd.

COMPARISON OF THE IN VITRO CONVERSION OF ESTRADIOLI 7fl TO ESTRONE OF NORMAL AND NEOPLASTIC HUMAN BREAST TISSUE * Kunhard POLLOW, Elmar BOQUOI, Joachim BAUMANN, Manfred SCHMIDT-GOLLWITZER and Barbara POLLQW Ins&u t ftir Molekularbiologie und Biochemie, Universitiit Berlin, 1000 Berlin 33, G.F. R.

Frauenklinik

Charlottenburg

der Freien

Received 10 May 1976; accepted 20 August 1976

Specific activity of 17p-hydroxysteroid dehydrogenase (17&HSD) was measured in 48 tissue specimens of human female breast cancer and, in addition, 48 nonmalignant tissue specimens obtained in each case from the same cancer-bearing breast. In all cases the nonmalignant tissue showed greater conversion of estradiol-170 into estrone than the neoplastic tissues. In normal human breast tissue of premenopausal women specific enzyme activity depended on the phase of the menstrual cycle: the highest values of 17BHSD activity were found in the early secretory phase. To determine the intracellular distribution of the 17@-HSD,purified microsomes, mitochondria, peroxysomes, lysosomes, nuclei and cytosol fractions were prepared. The purity of each fraction was monitored by marker enzymes. It was found that the 17p-HSD was mainly located in mitochondria and microsomes. Furthermore it could be demonstrated that the microsomal enzyme was bound tightly to the membranes of the endoplasmic reticulum, while the mitochondrial 17p-HSD was mainly associated with the outer membranes’of the organelle. Kinetic parameters (Km-values, coenzyme requirements and maximal velocities) of a cytoplasmic, nuclear, mitochondrial and microsomal 17&HSD of normal and neoplastic human mammary tissue were compared. Maximal velocity was highest in enzyme preparations of normal mammary tissue obtained from premenopausal women in the early secretory phase. Kmvalues were nearly identical in normal and neoplastic mammary tissue preparations (approx. 1 X 10” M). NAD was more efficient than NADP as a cofactor. For the conversion of estradiol to estrone the optimum temperature was approximately 40°C and the optimum pH 9.5. For the reduction of estrone the optimum pH was 6.5. Sulphydryl groups were shown to be essential for catalysis. Keywords:

17pbydroxysteroid cancer.

dehydrogenase;

estrone;

estradiol-17P;

human breast

Since 1967 the activities of several selected enzymes in normal and neoplastic human breast tissues have been measured (Abraham and Bartley, 1974; McGuire et al., 1974). The studies were initiated to establish and identify characteristics of * Supported by Deutsche Forschungsgemeinschaft. 333

334

K. Pollow

et al.

human malignancies that were unique to neoplasia and to seek correlations of these parameters with the clinical course of the disease. Breast carcinoma was chosen because it is the most common m~ignant lesion in women today. Little information is available on the metabolism of estradiol in subcellular fractions of normal and neoplastic breast tissues (Geier et al., 1975) although it is well accepted that the mammary gland has the capacity to bind estradiol-170 specifically with high affinity due to the presence of protein macromolecules termed ‘estrogen receptors’ (McGuire et al., 1975). Thus, it was considered appropriate to study the 17~-hydroxysteroid dehydrogenase (17@-HSD) activity which catalyzes the conversion of estradiol-170 to estrone in various subcellular fractions of nonmalignant and neoplastic breast tissues. Furthermore, the question arose whether progesterone regulates the 17&HSD activity in rna~a~ tissue in the same way as in human endometrial tissue. In earlier publications (Pollow et al., 1975a--d, 1976a,b) it was pointed out that, in the human endometrium, the highest values of 17@-HSDactivity were found during the early secretory phase. The aim of the present investigation is therefore also to correlate the enzyme activity to the stage of the’menstrual cycle in tissue specimens of premenopausal women. MATERIALS AND METHODS Steroids

[4-14C]Estradiol-17@(spec. act. 56 mCi/mmol) and [4-‘4C]estrone (spec. act. 58 m~i/~ol) were purchased from the Radiochemical Centre, Amersham, England and checked for radiochemical purity by thin-layer chromatography on silica gel using benzene-methanol (19 : 1, v/v). The purity was 97-99% and did not change during use. All reference steroids were obtained from commercial sources. Chemicals NAD*, NADH, DNA (thymus), RNA (yeast), ADP, hexokinase, succinate, cyto-

chrome c (cow), ~ucose~-phosphate dehydrogenase, ~-oxo~utarate were obtained from Biochemica Boehringer, Mannheim; Triton X-100 was bought from Serva, Heidelberg; sucrose, CsCl, EDTA, triethanolamine were provided by Merck, Darmstadt; bovine serum albumin was obtained from Behring, Marburg; Lubrol WX was from I.C.I. Organics, Inc., Frankfurt; and Sephadex G-200, Ficoll were obtained from Pharmacia Fine Chemicals, Uppsala, Sweden. All chemicals were reagent grade. Analytical grade benzene, methanol, diethylether and chloroform (Merck) were distilled before use. Tissues

48 human breast cancer specimens were studied and, in addition, 48 nomalig nant tissue specimens obtained in each case from the same cancer-bearing breast

Conversion of estra&oI-I 70 to estrone

335

were analyzed (nonmalignant, defined as breast tissue free of abnormalities by light microscopic criteria). Tissue samples were obtained as mastectomy specimens and washed within minutes of removal in five volumes ice-cold 0.01 M Tris-HCl buffer (pH 7.4) containing 0.25 M sucrose. The tissues were dissected free of adipose and of necrotic tissue as completely as possible and divided for histopathologic~ examination and for 17@-HSDactivity assay. All p~menopaus~ rna~a~ tissue samples (3.5 cases) were correlated with respect to the menstrual cycle using histological examination of the endometrium according to the method of Noyes et al. (1950). Preparation of subcellular fractions

Tissue homogenization was performed at 4’C. The washed tissue was weighed and homogenized in 4 vol. of 0.01 M Tris-HCl buffer (PH 7.4) containing 1 mM EDTA, 12 mM mercaptoethanol, 0.25 M sucrose and 20% glycerol (v/v) (medium A) with a homogenizer fitted with a teflon pestle. The homogenate was filtered through four layers of cheese-cloth and fractionated as follows: flf nuclei Nuclei were sed~ented at 8509 for lO*min (rotor SS-34, Sorvall RC-2). The supernatant fluid was carefully decanted and retained while the sediment was suspended in medium A, layered on a cushion of 0.01 M Tris-HCl (pH 7.4) containing 2.4 M sucrose and 1 mM magnesium acetate and centrifuged at 60 OOOg for 60 min (rotor SW 25.2, Beckman ultracentrifuge L265B). The sediment was washed twice in medium A (washing 1 and 2 in table 5), suspended in 0.01 M Tris-HCl @H 7.4) contain~g 25 mM KCl, 2 mM CaCla, 24 mM thioglycerol, 0.25 M sucrose and 0.25% Triton X-100 (v/v) (medium B), layered on 2.2 M sucrose containing 25 mM KCl, 2 mM CaCla, and 24 mM thioglycerol (which did not contain Triton X-100) and sedimented at 60 OOOgfor 60 min. The resulting sediment was designated ‘purified nuclear fraction’. (2) ~itochond~,

perox~somes and lysosomes The 850 g supematant was centrifuged at 8000g for 20 min to sediment the mitochondria which were resuspended in medium A by hand homogenization and centrifuged at 9000 g for 15 min. This procedure was repeated once more. The sediment obtained after the final centrifugation was designated as ‘washed mitochondria’. Fractions enriched in mitochondria, ~roxysom~ and lysosomes were obtained by Fico~-latent ~entrifugation (Ficollgradient O--IS% w/v, in medium A, sucrose cushion 2.4 M; rotor SW 25.2, 7000 rev./min for 15 min) according to the method of Brown (1968). Further purification of these fractions was achieved by sucrose gradient centrifugation (20-60%, w/v; 30 000 rev./min for 4 h, rotor SW 65, Beckman). (3) Preparation of subm~tochond~a~ fractions

Preparation of submitochondri~ fractions was performed as described by Parsons et al. (1966), Yago et al. (1970) and Pollow et al. (1975a).

336

K. Pollow et al.

~4~Mic~s~~es

Microsomes were sedimented from the 9000g supernat~t at 105 000 g for 60 min (rotor 30, Beckman) and washed twice in medium A. Separation into rough and smooth membranes was performed according to the method of Dalhrer et al. (1966) by CsCl sucrose density gradient centrifugation. (5) Prolonged high speed centrifigation of cytosol The cytosol fraction (105 000 g supernatant) was diluted 1 : 4 with 0.01 M Tris-HCI, pH 7.4 containing 20% glycerol (v/v) and centrifuged for 30 h at 158 OOOg. The supernatant fluid was decanted and the pellet suspended in 0.01 M Tris-HCI containing 20% glycerol @Iv). Enzyme mmys 17$4XD activi@ For the oxidation reaction the standard reaction mixtures (total volume 4.1 ml) contained 0.01 M Tris-HCl buffer, pH 7.4,0.1 &i [4-r4C]estradiol-17fl (0.47 pg) plus 10 PM of unlabelled steroid (dissolved in 0.1 ml of propylene glycol, 2.4% v/v), 500 PM NAD+ and *enzyme preparations. The following amounts of enzyme (mg protein/O.5 ml) were added: 3.5 mg cytosol, 5 mg microsomes, 1.2 mg mitochondria, 0.7 mg peroxisomes, 0.4 mg lysosomes, 1.8 mg nuclei, 0.25-0.6 mg submitochondrial fractions. For the reduction reaction the conditions were 0.15 M phosphate buffer, pH 6.0, 0.1 MC1[4-r4C]estrone plus 10 FM of unlabelled steroid and 500 m NADH t Hf. The incubation temperature was 37°C. Reactions were started by addition of coenzymes and terminated (after 30 min of incubation) by addition of 5 ml ether/ chloroform (3 : 1, v/v). The extracts of the reaction mixtures (3 X 5 ml etherlchloroform) were pooled and evaporated under nitrogen and dissolved in 0.5 ml of benzene. An aliquot (50 ~1) was removed for liquid scintillation counting (Berthold BF 5000 liquid scintillation counter, Wildbad, GFR) in order to estimate the total amount of radioactive steroids present in the extract. The benzene extracts were dried down under nitrogen and the dry residues were transferred with 0.2 ml chloroform/methanol (1 : 1, v/v) to thin-layer plates (silica gel, 0.25 mm, Woelm, Eschwege, GFR). Thin-layer chromatography and identification of reaction products were performed as described previously (Pollow et al., 1975a,b). Marker enzymes

Glucosed-phosphate dehydrogenase was determined according to Biicher et al. (1964); succinate dehydrogenase according to Arrigoni and Singer (1962); urate oxidase according to Mahler et al. (1955); acid phosphatase according to Leighton et al. (1968); monoamine oxidase and adenylate kinase activities were assayed by the methods of Schnaitman and Greenawalt (1968); glutamate dehydrogenase activity was determined by the method of Bergrneyer (1962). Protein content was determined according to Lowry et al. (1951) with bovine

Conversion of estmdiol-i 7p to estrone

337

serum albumin as standard. DNA and RNA content were assayed according to Burton (1956) and Ceriotti (19.55) with DNA (thymus) and RNA (yeast) as standards.

RESULTS After incubation of 10 I.IM[4-‘4C]estradiol-17fl with homogenate as well as various subce~ular fractions of human nonrn~i~~t rn~rna~ tissue under standard conditions it is evident that most of the estradiol was transformed to estrone. No effort was made to identify trace amounts of other metabolites which occurred especially in incubations with microsomes and mitochondria. All breast cancer tissue samples produced the same estradiol-17fl metabolite found in normal mammary tissue preparations. Subcellulardistribution The distribution of 17fl-HSD and marker enzymes in the subcellular fractions of normal human mammary tissue is given in table 1. The 17/J-HSD activity was mainly associated with the mitochondrial and microsomal fractions. Cytosol and peroxysomal, lysosomal and purified nuclear fractions contained 10 to 20 times less activity. The degree of purity of subcellular fractions obtained by differential centrifugation was satisfactory. Although mitochondria could not be separated completely from lysosomes, peroxysomes and microsomes, the activity of extramitochondrial enzymes was relatively small. There was minor contamination of lysosomal fraction with mitochondria and peroxysomes; the purity of peroxysomes and microsomes was satisfacto~. In order to determine whether the 17&HSD activitjr in the cytosol was soluble or was bound to low-density particles, the supematant fraction obtained after sedimentation of the particulate fractions was centrifuged at 158 000 g for 30 h after 4fold dilution with 0.01 M Tris-HCl containing 20% glycerol (v/v) (table 2). No change in the specific activity of 17&HSD in the supernatant was observed. Fu~herlnore, NADPH-cytochrome~-reduct~e, which is a marker enzyme of the endoplasmic reticulum, could not be detected in the pellet. The pellet was also examined by electron microscopy. It consisted mainly of a fine granular material as described by Fawceff (1955) and by Palade and Siekevitz (1956) for prolonged highspeed sediments of liver cytosol. It is likely, therefore, that prolonged centrifugation leads to the sedimentation of enzyme aggregates of low density. Such aggregation has been described for the soluble 17/J-HSDof human placenta (Jarabak et al., 1966) and endometrium (Pollow et al., 197Sa). The procedure developed by Dallner et al. (1966) was used to obtain complete separation of smooth and rough microsomes. The smooth vesicles appear at the interface and the rough vesicles sediment through the 1.3 M sucrose layer to form a pellet.

Recovery

Homogenate Mitochondria Lysosomes Peroxysomes Microsomes Cytoso1 Nuclei

Subcelhdar fraction

1581 258 44 49 19.5 724 210

mg

Protein

93.6

100 16.3 2.8 3.1 12.3 45.8 13.3

Total (%) 100 85 -

248 n.m. n.m. n.m. n.m. 211 n.m. 85.0

S.A.

Total (%)

S.A.

287 227 30 19 n.m. ri.m. n.m.

SDH

G-6-P-DH

96.2

100 79.1 10.5 6.6 -

Total (%) 3058 158 2778 211 n.m. n.m. n.m.

S.A.

AP

102.9

100 5.2 90.8 6.9 -

Total (%) 213 18 23 178 n.m. n.m. n.m.

S.A.

uo

102.9

100 8.5 10.8 83.6 -

Total (%)

881 11 n.m. n.m. 808 n.m. n.m.

99.0

100 1.3 97.7 -

NADPH-cyticreductase S.A. Totai (%)

Total (%)

95.6

158.6 100 87.4 55.1 8.3 5.2 5.4 3.4 40.0 25.2 5.7 3.6 4.8 3.0

S-A.

1II@-HSD

Table 1 IntmceJlular distribution of marker enzymes and 17fi-HSD in secretory human mammary tissue. S.A. = specific enzyme activity (pmol/mg protein/ mm); n.m. = not measurable. G&P-DH, ghmose-6-phosphate dehydrogenase; SDH, succinate dehydrogenase; AP, acid phosphatase; UO, urate oxidase.

339

Conversion of estradiol-I7p to estrone

Table 2 Effect of long-time high-speed centrifugation (30 h, 158 000 g) on the activity of 178-HSD in the cytosol of secretory human mammary tissue after 4-fold dilution with 0.01 M Tris-HCl buffer, pH 7.4. Fraction

Total activity (pmol/min)

Total protein (mg)

Spec. act. 17p-HSD (pmol/mg protein/min)

cytoso1

578 140 417

101.4 26.4 77.2

5.7 5.3 5.4

Pellet (30 h, 158 OOOg) Sunematant (30 h, 158 000~)

From table 3 it can be seen that the specific activity of 17@HSD in rough and smooth microsomes was similar to that in crude double-washed microsomes. This indicates that microsomal 17fl-HSD is bound to the membranes of the endoplasmic reticulum. The purity of the nuclear fractions obtained by conventional differential centrifugation from human mammary homogenate was checked morphologically by phase-contrast microscopy, and chemically by means of RNA/DNA and protein/ DNA ratios (table 4). After three washings RNA/DNA ratios approximated 0.2 and protein/DNA ratios approximated 2. The nucleus-associated 17/3-HSD activity decreased during the purification procedure. In order to evaluate the contamination of mitochondria with the microsomal fraction NADPH-cytochrome-c-reductase, a microsomal marker enzyme was determined after several washings and compared with NADPH-cytochrome-c-reductase activity of microsomal fraction prepared from the same tissue (table 5). The specific activity of NADPH-cytochrome-creductase in double-washed mitochondrial suspension was reduced from an initial value of 11 pmol/mg proteinlmin to less than 1% of microsomal contamination in mitochondria washed four successive times.

Table 3 Specific activity of 17@HSD in crude and ‘fractionated’ microsomes of secretory human mammary tissue; double-washed crude microsomes were fractionated by the method of Dallner et al. (1966) (CsCl-sucrose gradient centrifugation) into smooth and rough particles. Fraction

Zone

Crude microsomes 0.25 M sucrose Smooth particles (interface 0.25/1.3 M sucrose) 1.3 M sucrose Rough particles (pellet in 1.3 M sucrose)

1 2 3 4

Spec. act. lllp-HSD (pmol/mg protein/min) 40 3.8 38.2 1.2 40.8

K. Pollow et al.

340 Table 4 Composition tion.

and 17@-HSDactivity of human mammary nuclei after different steps of purifica-

17pHSD activity RNA/DNA

Protein/DNA

(pmol/mg DNA/ min).

(pmol/mg protein/ min)

__ Pellet from washing 1

0.29 f 0.08

2.01 f 0.07

14.3

7.1

Pellet from washing 2

0.25 f 0.05

1.98 f 0.03

10.3

5.2

Purified nuclear fraction

0.23 f 0.03

1.95 f 0.02

9.4

4.8

n=6

n=6

n=6

n=6

The distribution of the marker enzymes and that of the 17fl-HSDactivity in purified subfractions of washed mitochondria of human mammary tissue is shown in table 6. These results suggest that the 17&HSD is associated with the outer membranes of the organelle rather than with the inner mitochondrial membrane. The soluble subfractions (matrix, intermembrane fraction) were not capable of converting estradiol to estrone. Fig. 1 shows that the specific activity of 17P-HSD in nonmalignant human mammary tissue of premenopausal women depends on the phase of the menstrual cycle. This is particularly evident in particulate fractions: the conversion of estradiol-17/3 to estrone was much higher in subcellular fractions obtained during the early secretory phase than during the proliferative phase. The conversion rates of estradioLl7P to estrone by various subcellular fractions

Table 5 Specific activity of NADPH-cytochromec-reductase and 17@-HSD(pmol/mg protein/min) human mammary mitochondria after several washings and in microsomes. Subcellular fraction

No. of washes

NADPHcytochrome-creductase

Mitochondria

2 3 4 2

11 n.m.

Microsomes

E’

Microsomal contamination (%) 1.4
17@-HSD

87.4 82.8 83.4 40

in

>

_t

-.-

14

-F

‘%=*_ _*--‘-

-.-.-.-

DAYS

Y .e

”

0. E

I

0

2-

4-

6-

-

2

phase

secretory

26

22

10

16

I I I I I I I

-L

-c

, , ,

proliferatwe

6

I , ,

._

.7

&

0

2

4

6

1

-I

2

6

10

fraction

pro!iferative

Nuclear

Micro~mes

_.-

4-._

phase

14

I

1

-*i.-

-- 7-- .-

I

-_

22

--

secretary

18

----._

_.__s”-

26

DAYS

Fig. 1. Dependence of specific 17P-HSD activity in mitochondria, microsomes, cytosol and nuclear fraction of normal human mammary tissue on the phase of the menstrual cycle; points are means of duplicates, horizontal bars indicate confidence limits of histological diagnosis.

,”

0

10

F f 0

9

-.-9-

-.=-

-.-

4-

s

-

-o-

-a-

60

80

1GQ

g

_o_-* -*.s_:-

Cvtosol

-A

-.-

-.-t-

-I-*--

-d-

I

_;-

H 8-

c 2

-a-

-o-‘?&

_e

Mitochondria

1

1

E = 10 -

00

100

E

342

K. Pollow et al.

Table 6 Specific activities (pmol/mg protein/min) of marker enzymes and 17@-HSDin purified subfractions of mitochondria from human mamma tissue. For details see under Materials and Methods. n.m. = not measurable. --___ SuccinateMonoamineAdenylateGlutamate17@-HSD DH oxidase kinase DH Washed mitochon~ia

227

113

158

Outer membranes

81

641

298

Inner membranes

682

n.m.

n.m.

211 148

n.m.

2113

n.m.

Intermembrane fraction

nm

‘ .

n.m.

1112

Matrix

n.m.

n.m.

nm.

987 n.m.

87.4 218 45.3

Fig. 2. Estradiol oxidation in various subcellular fractions of nonmaljgnant human breast tissue (solid bars) as compared with breast carcinomata (opened bars) obtained from the same organs. Incubation procedure see under Materials and Methods. Values presented as means f S.E.M. (vertical line in each bar); number of samples examined is shown at the base of each bar. early prol. f early proliferative phase (1st to 7th day of menstrual cycle); late prol. = late proliferative phase (8th to 13th day); early seer. = early secretory phase (14th to 21st day); late seer. = late secretory phase (22nd to 28th day).

Conversion of estradiol-I7p to estrone

343

Fig. 3. Estrone reduction in various subcellular fractions (solid bars) as compared with breast carcinoma (opened Incubation procedure see under Materials and Methods. (vertical line in each bar); number of sampfes examined breviations as in fii. 2.

of nonmalignant human breast tissue bars) obtained from the same organs. Values presented as means f S.E.M. is shown at the base of each bar. Ab-

of breast cancer tissue relative to the nonmalignant tissues from the same cancerbearing breast are summarized in fig. 2. In all cases the nonmalignant tissue showed greater conversion of estradiol-1713 into estrone than the neoplastic tissues. Using estrone as substrate the tendency was the same. 2

E2”EI

El+2

WI

542

1.5

Mit.

Mic.

wt.

Mic.

W.

Nut.

W

Nut.

Fig. 4. Specific activity of l?p-HSD in various subcellular fractions of nonm~ignant and neoplastic human breast tissue (obtained from the same organs) of postmenopaus~ women. Incubations procedures see under Materials and Methods. Mit.: mitochondrla; Mic.: microsomes; Cyt.: cytosol; Nut.: nuclei; stippled bars: nonmalignant human breast tissue; opened bars: neoplastic tissue.

K. Pollow

344

et al.

Fig. 3 shows that estrone was converted to estradiol-170 (under reducing conditions) more effectively by the nonmalignant mammary tissue than by cancerous tissues. Furthermore, nonmalignant and neoplastic mammary tissue samples from 13 postmenopausal patients were evaluated (fig. 4). In all tissue samples 17/I-HSD activity could be demonstrated. From fig. 4 it also becomes evident that all nonmalignant tissue samples incubated showed higher 17/3-HSD activity than did any specimen of breast cancer. Enzyme kinetics Fig. 5 shows that the optimal pH was approximately 9.5 for the oxidative reaction in 0.1 M glycine/NaOH buffer, and 6.5 for the reduction in 0.15 M phosphate buffer. The optimum temperature for the oxidation of estradiol-170 to estrone in various subcellular fractions was around 406C. The effects of metal ions, SH-group MITOCHONDRIA

I CYTOSOL

60

MICHOSOMES

6

NUCLEAR

5

6

FRACTION

7

8

9

10

Fig. 5. The effect of pH and the influence of buffer solution on the enzyme activity of the 17pHSD in various subcellular fractions of nonmalignant human mammary tissue. Standard assay conditions as described under Materials and Methods. (1 mg protein). The buffers used were as follows: 0.15 M phosphate buffer (pH 5.5~S.O), (0); 0.1 M glycine/NaOH buffer (pH 8.2-lO.O), (0).

345

Conversion of estradiol-I 7p to estrone

Table 1 Influence of metal ions, metal chelating agents and SH-group blocking agents upon 17~3-HSD activity of various subcellular fractions.of nonmalignant human mammary tissue. 17f3-HSDactivity (%) Concentration (M) None p-Chloromercuribenzoate Iodoacetamide EDTA Cysteine Zn2+ Mn2+

yOW3 1o-3 1o-3 1o-3 1o-5 1o-3 1o-5 1o-3 1o-3 1o-3

Cytosol

Mitochondria

Microsomes

Nuclear fraction

100 12 48 18 108 68 14 94 88 _

100 8 52 88 98 54 12 88 72 _

100 18 41 84 102 62 8 91 78 _

45

38

32

100 12 63 71 94 51 9 98 71 24

Table 8 Kinetic constants for 17!3-HSD in various subcellular fractions of nonmalignant human mammary tissue. The values represent means of 8 determinations each with a different preparation. Assay conditions were as described under Material and Methods. A = mammary tissue from the early proliferative phase; B = mammary tissue from the early secretory phase; C = tissue samples from human breast cancer obtained from premenopausal women in the early secretory phase (same patients summarized under B). Fixed substrate (10 ctM)

Km for estradiol (PM)

Vmax (pmol/mg protein/min)

A

B

C

A

NAD NADP

0.71 1.13

0.69 1.08

0.88 -

Microsomes

NAD NADP

0.61 1.23

0.57 1.18

Mitochondria

NAD NADP

0.81 1.02

Nuclear fraction

NAD NADP

1.14 1.53

Enzyme preparation

Cytoso1

B

C

3.8 1.7

12.3 3.8

-

0.73 -

21.7 8.3

108 24.3

14.7 -

0.8 1.05

1.03 -

33.8 14.7

132 25.3

11.3 -

1.06 1.48

1.18 -

2.3 0.8

8.7 2.1

1.8

0.7 -

blocking agents and EDTA of the 17P-HSD activity in various subcellular fractions are summarized in table 7. Sulphydryl blocking agents exerted a strong inhibitory effect. All metal ions also showed an inhibitory effect at a concentration of 10P3 M. The effect of the cstradiol-170 concentration on the 17&HSD activity correlated to the stage of the menstrual cycle was studied in cytosol as well as in mitochon-

346

K. Pollow et al.

drial, microsomal and nuclear fractions of normal and neoplastic mammary tissues. & and Max were calculated from double-reciprocal plots by the method of Lineweaver and Burk (1934) (table 8). Vmaxwas highest in enzyme preparations of normal mammary tissue specimens obtained from premenopausal women in the early secretory phase. &,-values were nearly identical in normal and neoplastic mammary tissue preparations (approx. 1 X 10m6M).

DISCUSSION Since estrogen receptors in the human mammary tissue have been demonstrated (McGuire et al., 1975; Jensen et al., 1971) the knowledge of estradiol metabolism by human mammary tissues is of crucial importance because there could be formation in this tissue of metabolites which could compete with estradiol for binding sites in the mammary receptor and because there could be steroidal abnormalities in association with adenocarcinoma of the breast. James et al. (1969), who infused [3H]estradiol-17fl into breast cancer patients, identified in the neoplastic tissue 3-50% of extractable steroids as [3H]estrone. Very recently, Geier et al. (1975) reported on the [4-‘4C]estradiol metabolism by nonmalignant and neoplastic human mammary tissue grown in organ culture. In all cases, estradiol-170 was converted to estrone, but the nonmalignant tissue samples (including tibroadenomas and cystic mastitis) showed greater conversion than the mammary cancer tissues. These results are in good agreement with ours: in all cases, the specific activity of 17@HSDwas found to be much higher in various subcellular fractions of nonmalignant mammary tissue than of breast cancer tissue samples. Contamination of the human mammary tissue samples with adipose tissue was found to be rare by microscopic examination. Moreover, Blean et al. (1974) demonstrated that the conversion of estradiol to estrone by human adipose tissue was very low. After the fractionating of mammary tissue by differential centrifugation the 17@ HSD was seen in the nuclear fraction, in the outer membranes of mitochondria, in the endoplasmic reticulum and in the soluble fraction. Similar results were obtained for the 17P-HSD from human endometrial and placental tissues (Pollow et al., 1974, 1975a). The purity of the various subcellular fractions was found to be satisfactory by measurement of marker erizymes, the greatest 17/3-HSDactivity being in mitochondria and microsomes. The intracellular distribution of the 17/J-HSD in neoplastic mammary tissue was similar to that in the corresponding fractions isolated from normal mammary tissue. The 17/3HSD of human mammary tissue shares many characteristics with the corresponding enzymes of placenta, liver and endometrium (Langer and Engel, 1958; Jarabak et al., 1966; Karavolas and Engel, 1966; Lehmann and Breuer, 1967; Ball and Breuer, 1970; Engel and Groman, 1974; Pollow et al.,, 1974, 1975a-d, 1976b).

Conversion of estradiol-17~ to estrone

341

Furthermore, we found the specific activity of 17/3-HSDin various subcellular fractions of nonmalignant mammary tissue, like the corresponding human endometrial enzyme (Pollow et al. 1975a-d, 1976a,b), to be dependent on the phase of the menstrual cycle. We were able to show that 17&HSD activity increases during the late proliferative phase with highest values, at ovulation. This increase in activity is mainly located in the microsomal and mitochondrial fractions. This result indicates that the mammary 170.HSD may play a physiologic role in the modulation of intracellular estradiol activity. It is difficult on the basis of the available data to speculate about the reasons for the changes in 170.HSD concentrations during the menstrual cycle. The intracellular 17&HSD levels parallel well with the serum estradiol concentrations during the menstrual cycle. However, the higher intracellular 170.HSD activities during the first ten days of the corpus luteum phase suggest an additional stimulating effect of progesterone (by way of a receptor mechanism?) upon the 17fl-HSD activity as in the human endometrial cell (Tseng and Gurpide, 1974; Pollow et al., 1975a,d). In this context our observation that in all tissue samples of mammary cancer the 170. HSD activity is low is of special interest. That means that the ‘defense mechanism’ against biologically highly active estradiol becomes less effective.

REFERENCES Abraham, S. and Bartley, J.C. (1974) In: Hormones and Cancer, Eds.: K.W. McKerns (Academic Press, Inc., New York - London) p. 29. Arrigord, 0. and Singer, T.P. (1962) Nature 193, 1256. Bergmeyer, H.U. (1962) Methoden der enzymatischen Analyse (Verlag Chemie, Weinheim) p. 752. Blean, G., Roberts, K.D. and Chapdelaine, A. (1974) J. Clin. Endocrinol. 39,236. Brown, D.H. (1968) Biochim. Biophys. Acta 162, 152. Bticher, T., Lutz, W. and Pelle, D. (1964) In: Handbuch der physiologisch- und pathologisch chemischen Analyse, Eds.: Hoppe-Seyler/Thierfelder (Springer Verlag, Berlin--GGttingenHeidelberg-New York, Vol. VI A) p. 292. Burton, K. (1956) Biochem. J. 62,3 15. Ceriotti, G. (1955) J. Biol. Chem. 214,59. Dallner, G., Siekevitz, P. and Palade, G.E. (1966) J. Cell Biol. 30,73. Fawceff, D.W. (1955) J. Natl. Cancer Inst. suppl. 5,147s. Geier, A., Horn, H., Levig, J.S., Lichtshtein, E. and Finkelstein, M. (1975) Eur. J. Cancer 11, 127. Gurpide, E. and Welch, M. (1969) J. Biol. Chem. 244,5159. James, F., Braunsberg, H., Irvine, W.T. and James, V.H.T. (1969) Steroids 15,669. Jarabak, J., Seeds, A.E. and Talalay, P. (1966) Biochemistry 5, 1269. Jensen, E.V., Block, G.E., Smith, S., Kyser, K. and DeSombre, E.R. (1971) Natl. Cancer Inst. Monogr. 34, 55. Leighton, F., Poole, B., Beaufay, A., Baudhuin, P., Coffey, J.W., Fowler, S. and de Duve, C. (1968) J. Cell Biol. 37,482. Lieberman, S., Katzenellenbogen, E.R., Schneider, R., Studer, P.E. and Dobriner, K. (1953) J. Biol. Chem. 205,87.

348

K. Pollow et al.

Lowry, O.H., Rosebrough, N.J., Parr, A.L. and Randall, R.J. (1951) .I. Biol. Chem. 193, 265. Mahler, H.R., Hiibscher, G. and Baum, H. (1955) J. Biol. Chem. 216,625. McGuire, W.L., Chamness, G.C., Costlow, M.E. and Shepherd, K.E. (1974) In: Ilormones and Cancer, Eds.: K.W. McKerns (Academic Press, Inc., New York-London) p. 76. McGuire, W.L., Carbone, P.P. and Vollmer, E.P. (1975) Estrogen Keccptors in Human Breast Cancer (Raven Press, New York). Noyes, R.W., Hertig, A.T. and Rock, J. (1950) Fertil. Steril. 1, 3. Palade, G.E. and Siekevitz, P. (1956) J. Biophys. Biochem. Cytol. 2, 171. Parsons, D.F., Williams, G.R. and Chance, B. (1966) Ann. N.Y. Acad. Sci. 137,643. Pollow, K., Sokolowski, G., Grunz, H. and Pollow, B. (1974) Hoppc-Seyler’s 2. Physiol. Chem. 355,501. Pollow, K., Liibbert, Il., Boquoi, I?., Kreuzcr, G., Jcskc, R. and Pollow, B. (1975a) Acta Fndocrinol. 79, 134. Pollow, K., Liibbert, H., Jeske, R. and Pollow, B. (1975b) Acta Endocrinol. 79, 146. Pollow, K., Liibbert, H. and Pollow, B. (1975~) Acta Endocrinol. 80, 355. Pollow, K., Boquoi, I<., Liibbert, H. and Pollow, B. (1975d) J. Endocrinol. 67, 13 1. Pollow, K., Liibbert, H. and Pollow, B. (1976a) J. Steroid Biochem. 7, 45. Pollow, K., Liibbert, l-1. and Pollow, B. (1976b) J. Steroid Biochem. in press. Schnaitman, C. and Greenawalt, J.W. (1968) J. Cell Biol. 38, 158. Tseng, L. and Gurpide, 1;. (1972) Am. J. Obstct. Gynecol. 114, 1002. Tseng, L. and Gurpide, E. (1974) Endocrinology 94,419. Yago, N., Kobayashi, S., Sekiyama, S., Kurokawa, H., Twai, Y., Suzuki, J. and Ichii, S. (1970) J. Biol. Chem. Tokyo 68, 775. Zaffaroni, A. and Burton, R.B. (1951) J. Biol. Chem. 193,749.