Pergamon Free Radical Biology & Medicine, Vol. 17, No. 6, pp. 493-499, 1994

Copyright © 1994 Elsevier Science Ltd Printed in the USA. All rights reserved

0891-5849/94 $6.00 + .00

0891-5849(94)E0073-R

Original Contribution

STEROIDOGENIC CAPACITY AND OXIDATIVE STRESS-RELATED PARAMETERS IN HUMAN LUTEAL CELL REGRESSION

MARGARITA VEGA, t '* TERESITA CASTILLO, ~ IVAN RETAMALES, § JORGE LAS HERAS, §

LUIGI DEVOTO,* and Lu1s A. VIDELA ~1

Departments of ~Cell Biology and Genetics, *Medicine, and IBiochemistry, *Research Institute for the Mother and Child, Faculty of Medicine, University of Chile, Chile; and ~Department of Pathology, San Borja-Arriarfin

Clinical Hospital, National Health Service, Chile

(Received 21 July 1993; Revised 16 December 1993; Re-revised 22 March 1994; Accepted 30 March 1994)

A b s t r a c t The steroidogenic capacity and oxidative stress-related parameters of the human corpus luteum (CL) at different stages of the luteal phase were studied under basal and human chorionic gonadotropin (hCG)-stimulated conditions. Mid CL exhibited the maximal steroidogenic capacity, together with lower levels of glutathione and higher thiobarbituric acid reactants content, macrophage count, and superoxide dismutase (SOD) activity, compared to the late CL. Addition of hCG to luteal cell cultures led to a preferential increase in progesterone synthesis in the late CL compared to the mid CL, without changes in the oxidative stress-related parameters, except for the increased SOD activity found in the late CL. It is concluded that an oxidative stress condition is established in the mid CL, coinciding with the maximal steroidogenic capacity and macrophage infiltration of the organ, which may be of relevance as one of the major mechanisms initiating CL involution in the human.

Keywords--Steroidogenesis, Oxidative stress, Glutathione, Superoxide dismutase, Lipid peroxidation, Macrophages, Human luteai cell regression, Free radicals

I N T R O D U C T I O N

Luteal regression or luteolysis is an example of cell function decrease, which is a normal and necessary event in the mammalian reproductive cycle. In the rat, corpus luteum (CL) involution has been related to an increased generation of reactive oxygen species (ROS), such a s 0 2 ° - and H202, by the intact ovary, t-3 One consequence of the production of free radicals in tis- sues is lipid peroxidation, 4 whose occurrence at the plasma membrane of the luteal cell may be involved in the observed gonadotropin receptor loss, diminished cyclic AMP formation, and, therefore, in the decrement of the steroidogenic ability of the CL during involu- tion. 5'6 Although the origin of 02"- and H202 in CL is unknown, their generation by luteal cells or by other cell types in the CL is possible. There is evidence that macrophages are present as normal components of the developing CL from several species, 7 cells that are characterized by their capacity for ROS formation upon stimulation, s In addition, CL macrophages pro-

Address correspondence to: Margarita Vega, Departamento de Biolog/a Celular y Genrtica, Facultad de Medicina, Universidad de Chile, Casilla 70061, Santiago-7, Chile.

493

duce different cytokines and noncytokines, 9'1° which have been suggested to modulate progesterone (P) bio- synthesis.ll'J2

Protection against ROS in cells is provided by en- zymes (SOD, catalase, and glutathione peroxidase), metabolites (GSH), or antioxidants vitamins. ~3 Both SOD and catalase are known to be present in the rat ovary, which, together with the vitamins E, C, and A, have been suggested to be endocrine-regulated] 3'~4 This suggestion is based on the findings that, during luteal functioning in the rat, luteinizing hormone (LH) seems to induce SOD, leading to H202 generation from 02"- to promote P synthesis by the peroxidase-ascor- bate system] 4 Furthermore, rat ovarian vitamin E and A levels are markedly increased during luteal regres- sion, an effect that is mimicked by the administration of either LH (vitamin E) or prostaglandin F2,~ (vitamin A), whereas a rapid and transient depletion of vitamin C is elicited by both treatments. 13 However, no infor- mation is available concerning the presence of antioxi- dant protective systems and their endocrine regulation in human CL, nor their relationship to human luteal regression. In this respect, LH and human chorionic gonadotropin (hCG) are known to stimulate human luteal P synthesis, an effect that was found to be age-

494 M. VEGA et aL

dependent.~5 In view of these considerations, this study evaluates the changes in oxidative stress-related pa- rameters of cells obtained from human CL at different stages of the luteal phase in relation to their capacity for P synthesis, assessed under basal and hCG-stimu- lated conditions, and to the macrophage count.

MATERIALS AND METHODS

Patients

CL were obtained from 28 normal women, 30 -40 years of age, requesting surgical sterilization at the San Borja-Arriar~in Hospital, National Health Service, Faculty of Medicine, University of Chile. The laparot- omy was scheduled in the mid-luteal phase (5 -9 days after LH peak) or in the late luteal phase (more than 9 days after LH peak). The study was approved by the Institutional Review Board, and all patients gave informed consent for removal of CL. After removal, CL were placed in sterile 0.15 M NaC1, washed with 0.15 M NaCI to eliminate blood, and a piece of each CL was placed in 10% (v/v) formalin for histologic examination.~5 The cycle date of each woman was con- firmed by CL and endometrial biopsies, ultrasonogra- phy, and by the respective plasma levels of LH (mid CL, 8.7 + 1.2 mlU/ml In = 10]; late CL, 10.2 _+ 0.9 [n = 10]), P (mid CL, 7.2 _+ 0.6 ng/ml [n = 10]; late CL, 4.5 ___ 0.8 [n = 10]), and total estrogens (mid CL, 122.6 + 11.3 pg/ml [n = 10]; late CL, 87.0 __+ 10.4 [n = 10]), determined in duplicate by previously described radioimmunoassays. 16

Cell dispersion and culture

Luteal cells were dispersed and cultured as de- scribed previously.16 Briefly, the luteal tissue was enzy- matically dissociated in Medium-199 pH 7.4, con- taining 26 mM NaHCO3, 0.1% (w/v) bovine serum albumin, 25 mM Hepes, 5 U of penicillin/ml, 370 U of collagenase/100 mg tissue, and 14 Kunitz U of DNase/100 mg tissue. Cells were counted in a hemocy- tometer, and their viability, assessed by the trypan blue exclusion method, was greater than 90% before and after the 24 h of culture. Immediately after plating, cultures were carried out at 37°C in the absence (basal conditions) or presence of hCG (10 IU/ml) (stimulated conditions) and were terminated after 24 h. The media were stored at -20°C until assayed for P and estradiol by radioimmunoassay, ~6 which were normalized to 10 6

viable cells.

and TBARS levels. GSH content was measured ac- cording to Tietze ~7 at 412 nm, in homogenates of luteal cells (105 cells/ml of 0.5 N HCIO4) obtained at different stages of the luteal phase, cultured for 24 h, and results were expressed as nmol/106 cells. SOD activity was estimated in cell homogenates (10 s cells/ml of 1.75 N tris buffer pH 7.4) by the inhibition of the rate of adrenochrome formation, in a reaction medium con- taining 1 mM epinephrine and 50 mM glycine buffer pH 10.2, at 480 nm. ~8 One SOD unit corresponds to the amount of enzyme producing 50% inhibition of adrenochrome formation, and results are expressed as U S O D / 1 0 6 cells. TBARS were determined according to Buege and Aust 19 at 535 nm, in luteal cells alone or luteal cells plus medium, after 24 h culture, and results are expressed as nmol/106 cells.

Immunohistochemical detection of macrophages

The determination of the number of macrophages in histological sections of CL of different ages was performed in 10 consecutive high power fields, in an area of 62,500 #m 2. The immunohistochemical identi- fication of macrophages was carried out by the avidin- biotin immunohistochemistry method, 2°'2~ detecting muramidase or lysozyme by a monoclonal antibody anti-human lysozyme raised in rabbits (Dako Corp., USA). The second antibody used was a biotinilated anti-rabbit immunoglobulin raised in pigs. The reaction was evidenced by avidin-biotin conjugated peroxidase using diaminobencidine as the chromogen.

All chemicals, culture media, and hormones used were obtained from Sigma Chemical Co. (St. Louis, MO), except for the lymphocyte separation medium (Litton Bionetics, Kensington, MD) and collagenase (Worthington Biochemical Corp., Freehold, NJ). The data are presented as means _ SE for the indicated number of separate studies (n). The statistical signifi- cance between mean values was assessed either by Student's t-test for unpaired results or by a covariance study for missing observations using the Lineal Gen- eral method and the nonparametric procedure of Mann-Whytney. Wilcoxon and Kruskall-Wallis methods were employed for comparison of the number of macrophages between the different stages of the luteal phase. 22

Oxidative stress-related parameters

Oxidative stress-related parameters included the de- termination of the total content of GSH, SOD activity,

RESULTS

Basal P synthesis in mid luteal cells was found to be 2.3-fold greater than that in late luteal cells (p <

Oxidative stress and human luteolysis 495

A) 3 0 0

• ~- 2 5 0

c u 150 0 ~- 0

® ~ 1 0 0

O ~

a_ 5 0

M I D C L LATE

ITI

r---i Basal

~//,~ h C G

CL

B ) MID CL LATE CL

tO

I l J e- t'~ tIE

- i "o O O ~

W e-

Fig. I. Basal and fiCG-stimulated progesterone (P) synthesis (A) and estradiol (E2) synthesis (B) by dispersed mid and late human luteal cells cultured for 24 fi. Values shown represent the means _+ S E for 10 mid and 10 late corpora lutea (CL), cultured in duplicate. Significance of the changes in P synthesis between mid and late CL under basal conditions, **p < 0.01, and under hCG-stimulated conditions, *p < 0.05. Significance of the changes in E2 synthesis between mid and late CL under basal and hCG-stimulated conditions, in cells from both luteal phases, *p < 0.05.

0.01; Fig. IA), whereas that of estradiol was enhanced by 53% upon involution (mid CL, 4500 _+ 1200 pg/ 10 6 cells/24 h In = 10]; late CL, 6900 +__ 1700 In = 10]; p < 0.05; Fig. 1B). Addition o fhCG to the culture medium led to a significant enhancement in P synthe- sis, both in late luteal cell (1.8-fold increase; p < 0.05) and in mid luteal cell (1.6-fold increase; p < 0.05) cultures (Fig. 1A).

Under basal conditions, the content of GSH in mid luteal ce l l s w a s 0.71 _ 0 .20 nmol /106 ce l l s (n = 16),

w h e r e a s r e g r e s s i n g lu teal ce l l s e x h i b i t e d va lues that

are 5 .4 - fo ld h i g h e r (p < 0 .001; Fig . 2A) . T h e p r e s e n c e

o f h C G in the cu l tu re m e d i a d id no t s ign i f i can t ly a l ter

the content of GSH in cells obtained from both mid and late luteal phases (Fig. 2A). Data presented in Fig. 2B indicate that the activity of SOD in the late luteal cells is 41% lower (p < 0.05) than that observed in mid luteal cells, under basal conditions. In addition, hCG significantly enhanced SOD activity (86%; p < 0.05) in regressing luteal cells, without modifying that of cells from the mid luteal phase (Fig. 2B). Fig. 2C shows the level of TBARS in cells from both luteal phases. As can be observed, the content of TBARS was 35% lower in late luteal cells compared to that in cells from mid CL (p < 0.05) under basal conditions. The level of TBARS in mid and late luteal cells was not modified by hCG (Fig. 2C), whereas that observed

A) (4)

, ~ O,J • ,,-- f J e -

-i- -6 0") E (.9 ,_

6

2

- MID CL LATE C L

mJ

I I B a s a l

~,-/~"~ h C G

B)

. m I D

~ o

o o g

120

8 0

4 0

MID

:I_" C L LATE C L

c) 6

c u 4 o

if} - -

m --6 2

110 , -

MID C L LATE CL

Fig. 2. Glutathione (GSH) content (A), superoxide dismutase (SOD) activity (B), and thiobarbituric acid reactants (TBARS) levels (C) in mid and late human luteal cells cultured for 24 h, under basal and hCG-stimulated conditions. Values shown represent the means +_ SE for 16 mid and l0 late corpora lutea (CL), cultured in duplicate. Significance studies: (A) Mid CL vs. late CL under basal and hCG- stimulated conditions, *p < 0.001 ; effect ofhCG vs. basal conditions both in mid and late CL, not significant (NS). (B) Mid CL vs. late CL under basal conditions, **p < 0.05; effect of hCG vs. basal conditions: NS in mid CL and **p < 0.05 in late CL. (C) Mid CL vs. late CL under basal conditions, **p < 0.05, and NS under hCG- stimulated conditions; effect of hCG vs. basal conditions: NS in mid and late CL.

496 M. VEGA et al.

5

e "~ 4 i

m

m

I

Basal Z,5 ~uM estradiol

Fig. 3. Thiobarbituric acid reactants (TBARS) levels in mid human luteal cells cultured for 24 h in the absence (basal conditions) or presence of 2.5 #M estradiol. Values shown correspond to the means + S E for 6 mid corpora lutea per experimental condition. *p < 0.05, compared to basal conditions.

after the addition of 2.5 #M estradiol to mid luteal cells cultured for 24 h was diminished by 40% (Fig. 3).

The assessment of the number of macrophages pres- ent in CL throughout the luteal phase is shown in Fig. 4. The highest number of macrophages per unit area was observed in mid CL, with significantly lower counts being found in the early and late CL (p < 0.001). Macrophages were localized at random in the CL, independently of the stage of the luteal phase. In fact, they were observed on the cortical region of the CL, as well as toward the central part of the gland (Fig. 5). Furthermore, macrophages establish a relationship with other cel l types o f the CL, including active luteal

0

( D

0

. u

C

~0 Q)

0 r"

0

U 0

E

o

..Q

E Z

4 0

3 0

2 0

IO

L

. , . . . , . . , .

. . . . , . . . .

. . . . . . . . . . .

. . . . . . , . - . .

~ . . . . . . . . . , . ~ , . . . . . . . . . . . . . . . . , . . . . . . .

~ , . . . . . . . . . . .

. : . : . : . : . 5 : , . . . . . . . . . . . .

. . . . . . . . . . .

. . , . . . . . . . . .

. . . , . . . . , . , .

. . . . , . . . . . . . . : , : . : . : . : . ; , : . . . . . . . . . . . . . . . . . . , . . , . , . : . : . : . : . : . ; . ; . . . . . . . , , . . . . . . . , . , . , . . . . . . . . . . . . . , . . . . . . . . . . , . , . . . .

Ear ly Mid Late

Stages of luteol phase

Fig. 4. Mean number of macrophages in human corpus luteum (CL) of different ages. Values shown represent the means _+ S E for 8 early CL, 7 mid CL, and 7 late CL. Values for the early CL correspond to previous unpublished data, obtained immediately after ovulation (1 - 4 days after LH peak). *p < O.001, compared to both early and late CL.

ceils (Fig. 5A), involuted luteal cel ls (Figs. 5A, 5B), and endothelial cells (Fig. 5B).

DISCUSSION

Data presented in this work show that the steroido- genic capacity of the CL, as well as its GSH levels, SOD activity, and TBARS content, are subject to major changes during the human luteal phase, and in response to gonadotropin action. Furthermore, macrophages are present in the human CL during its lifespan, at a num- ber that is dependent on the age of the tissue, probably playing a role in the regulation of the endocrine func- tion of the gland.

The maximal steroidogenic capacity of the CL is achieved in the intermediate stage of the luteal phase, together with a lower level of GSH and a higher T B A R S content, compared to the late CL. Although an increased level of TBARS is indicative of an en- hancement in cellular lipid peroxidation, 4 these prod- ucts could also derive from independent processes, such as prostaglandin biosynthesis. 23 In this respect, studies in primates 5 and in human females 24 have shown that higher rates of prostaglandin synthesis oc- curred in the late CL stage, at which lower TBARS levels were observed (Fig. 2C), compared to the mid CL. Furthermore, E2, which results in the production of luteal prostaglandin F2~ during the mid to late luteal phase, 5 led to a diminution in CL T B A R S levels when added in vitro (Fig. 3). These observations, and the dependency of prostaglandin synthesis upon lipid per- oxidation for activation,' suggest that the former path- way is likely to account for a minor proportion of the enhanced TBARS levels observed in the mid CL. However, further studies are needed to clarify this view. L o w GSH levels in the mid CL could be the result of an enhanced utilization of the tripeptide by different cellular mechanisms. In fact, GSH could be used for reduction o f hydroperoxide intermediates, de- rived from both the steroidogenic 2s and lipid peroxida- tion 4 pathways, or H2Oz, through the action of glutathi- one peroxidase. Alternatively, GSH could be utilized for a direct interception of oxygen free radicals arising either from the cellular metabolism of the luteal cells 25 or from activated macrophagesfl whose number is sig- nificantly higher in the mid CL compared to that in the early and late luteal phases. Thus, a greater pro- oxidant activity seems to be achieved in the luteal cells from the mid CL, despite the higher S O D activity observed in these cel ls in relation to that found in the regressing CL, leading to an oxidative stress condition with the respective lipid peroxidative response.

The decrement in the basal capacity for P produc- tion observed in cells obtained from late CL is indica- tive o f natural luteal regression. At this stage of the

Oxidative stress and human luteolysis 497

A

Fig. 5. Histological sections of human corpus luteum showing the interactions of (A) a macrophage (big arrow) with an active luteal cell (single arrow) and with an involuted cell ("K"-cell) (double arrow). (B) Interaction of a macrophage (big arrow) with endothelial cells (single arrow) and with an involuted cell ("K"-cell; double arrow). Note the difference in nuclear morphology of both ceils. (Avidin-biotin immunohistochemical method, anti-human lysozyme, 400x).

498 M. VEGA et al.

luteal phase, the pro-oxidant act ivi ty of the luteal cells is dras t ical ly d iminished compared to the mid CL, as ev idenced by the significant reduction in the l ipid per- oxidat ive index, TBARS. This f inding correlates with the marked enhancement in the content o f cel lular G S H and with the d iminut ion in the number of macrophages , despi te the lower SOD activi ty found. In this respect, the significant direct correlat ion between P synthesis and SOD activi ty observed during luteolysis (r = 0.78; p < 0.05) suggests a par t ic ipat ion of ROS in the ste- ro idogenic abi l i ty of human luteal cells, as previously demonst ra ted for the rat CL. t4 Fur thermore , two addi-

tional factors could play a role in reducing the pro- oxidant capaci ty in the regress ing CL, namely, (a) the increased basal product ion of E2 over that of the mid CL, phenolic steroid known to exhibit an ant ioxidant act ivi ty greater than that of vi tamin E 26 and that sig-

nificantly decreased the T B A R S content of mid luteal cells after culture for 24 h; and (b) the repor ted age- related decrease in the ratio of unsaturated/saturated fatty acids of the p lasma membrane of nonluteal cells, z7 which may lead to substrate def ic iency for the peroxi- dat ing action o f ROS. In this situation, high levels of G S H could result f rom a decreased demand for hydroperox ide metabol i sm associa ted with the d imin- ished s teroidogenic capaci ty, a s t imulat ion of the bio- synthesis o f G S H determined by the low level of thiol 2s

prevai l ing in the mid luteal phase, or both. In agreement with this work in human CL, recent studies in the rat have shown higher levels of G S H in luteal regression. ~ 3

hCG was found to preferent ia l ly increase P synthe- sis in late luteal cel ls compared to those obtained from the mid CL when added to the culture medium. Fur- thermore, hCG did not produce major changes in the measured oxidat ive s tress-related parameters of the mid and late CL, except for the marked increase in the act ivi ty of SOD observed in the regressing luteal cells. These observat ions suggest that hCG has an important luteotropic role in the regress ing CL, with the concomi- tant increase in ant ioxidant protect ion against luteo- lytic mediators . The mechan i sm by which hCG en- hances SOD act ivi ty remains to be elucidated.

In conclusion, an oxida t ive stress condit ion is estab- l ished in the human mid CL, which is c losely related to the prevai l ing cel lular G S H status rather than to the SOD activity, and that coincides with the maximal s teroidogenic capaci ty and macrophage infiltration of the organ. This pro-oxidant condi t ion may have an important role as one of the mechanisms init iat ing CL involution, by inducing structural al terat ions of the p lasma membrane of the luteal cells, able to disrupt the regula tory processes control l ing P biosynthesis .

Acknowledgements--This work was supported by FONDECYT

(1272/91). The authors thank the National Hormone and Pituitary Program (NIADDK), Baltimore, MD, for providing the reagents for LH assay.

REFERENCES

1. Carlson, J. C.; Wu, X. M.; Sawada, M. Oxygen radicals and the control of ovarian corpus luteum function. Free Radic. Biol. Med. 14778-84; 1993.

2. Sawada, M.; Carlson, J. C. Rapid plasma membrane changes in superoxide radical formation, fluidity, and phospholipase A2 activity in the corpus luteum of the rat during induction of luteolysis. Endocrinology 12872992-2998; 1991.

3. Riley, J. C. M.; Behrman, H. R. In vivo generation of hydrogen peroxide in the rat corpus luteum during luteolysis. Endocrinol- ogy 12871749-1753; 1991.

4. Kappus, H. Lipid peroxidation: Mechanisms, analysis, enzymol- ogy and biological relevance. In: Sies, H., ed. Oxidative stress. London: Academic Press, 1985:273-310.

5. Aulette, F. J.; Flint, P. F. Mechanisms controlling corpus luteum function in sheep, cows, nonhuman primates, and women, spe- cially in relation to the time of luteolysis. Endocr. Rev. 9:88- 105; 1988.

6. Wang, H. Z.; Wu, Y. F.; Lu, S. H.; Sun, Z. D.; Shen, W. X.; Zhon, W.; Gong, Y. T. The hormonal control of human luteal cells. In: Hasseltine, F. P.; Finallay, J. K., eds. Growth Jhctor infertility regulation. New York: Cambridge University Press; 1991:99-113.

7. Bagavandoss, P.; Kunkel, S. L.; Wiggins, R. C.; Keyes, P. L. Tumor necrosis factor a (TNF-o0 production and localization of macrophages and T lymphocytes in the rabbit corpus luteum. Endocrinology 122:1185-1187; 1988.

8. Nathan, C. F. Secretion of oxygen intermediates: Role in effector functions of activated macrophages. Fed Proc. 417 2206-2211 : 1982.

9. Weiss, S. J.; Lobuglio, A. F. Biology of disease: Phagocyte- generated oxygen metabolites and cellular injury. Lab. Invest. 47:5-18; 1982.

10. Kirsch, T. M.; Vogel, R. L.; Frickinger, G. L. Macrophages: A source of luteotropic cybernins. Endocrinology 11371910-1912: 1983.

11. Behrman, H. R.; Preston, S. L. Luteolytic actions of peroxide in rat ovarian cells. Endocrinology 124:2895-2900; 1989.

12. Agrawal, P.; Laloraya, M. M. Induction of peroxidase in corpora lutea of rat ovary by lutropin. Biochem. J. 1667205-208; 1977.

13. Aten, R. F,; Duarte, K. M.; Behrman, H. R. Regulation of ovar- ian antioxidant vitamins, reduced glutathione, and lipid peroxi- dation by luteinizing hormone and prostaglandin F2,. Biol. Re- prod. 467401-407; 1992.

14. Laloraya, M.; Kumar, G. P.; Lalotaya, M. M. Changes in the levels of superoxide anion radical and superoxide dismutase during the estrous cycle of Rattus norvegicas and induction of superoxide dismutase in rat ovary by lutropin. Biochem. Bio- phys. Res. Commun. 1877146-153; 1988.

15. Vega, M.; Devoto, L; Navarro, V.; Castro, O.; Kohen, P. In vitro net progesterone production by human corpora lutea: Effect of human chorionic gonadotropin, dibutyryl adenosine-3',5'- monophosphate, cholera toxin and forskolin. Z Clin. Endocrinol. Metab. 65:747-752; 1987.

16. Devoto, L.; Vega, M.; Navarro, V.; Sir, T.; Alba, F.; Castro, O. Regulation of steroid hormone synthesis by human corpora lu- tea: Failure of follicle-stimulating hormone to support steroido- genesis in vivo and in vitro. Fertil. Steril. 517628-633; 1989.

17. Tietze, F. Enzymic method for quantitative determination of nanogram amounts of total and oxidized glutathione: Applica- tion to mammalian blood and other tissues. Anal. Biochem. 27:502-522; 1969.

18. Misra, H. P.; Fridovich, 1. The role of superoxide anion in the autoxidation of epinephrine and a simple assay for superoxide dismutase. Z Biol. Chem. 24773170-3175; 1971.

19. Buege, J. A.; Aust, S. D. Microsomal lipid peroxidation. Meth- ods Enzymol. 527302-310; 1978.

Oxidative stress and human luteolysis 499

20. Pinkus, G. S.; Said, J. W. Profile of intracytoplasmic lysozyme in normal tissues, myeloproliferative disorders, hairy cell leuke- mia, and other pathologic processes: An immunoperoxidase study of paraffin sections and smears. Am. J. Pathol. 89:351- 366; 1977.

21. Keskaw, S.; Chung, P.; Milon, G.; Gordon, S. Lysozyme is an inducible marker of macrophage activation in murine tissues as demonstrated by in situ hybridization. J. Exp. Med. 174:1049- 1058; 1991.

22. Matthews, D. E.; Farewell, V. T. Estadfstica mddica: Aplicaci6n e interpretaci6n. Barcelona: Salvat; 1990.

23. Cadenas, E.; Sies, H.; Nastainczyk, W.; Ullrich, V. Singlet oxy- gen formation by low-level chemiluminescence during enzy- matic reduction of prostaglandin Gz to H2. Hoppe-Seyler's Z Physiol. Chem. 364:519-528; 1983.

24. Vijayakumar, R.; Walters, W. A. W. Human luteal tissue prosta- glandin, 17/3-estradiol, and progesterone in relation to the growth and senescence of the corpus luteum. Fertil. Steril. 39:298-303; 1983.

25. Little, C. Steroid hydroperoxides as substrates for glutathione peroxidase. Biochim. Biophys. Acta 284:375-381; 1972.

26. Nakano, M.; Sugioka, K.; Naito, I.; Takekoshi, S.; Niki, E. Novel and potent biological antioxidants on membrane phospho- lipid peroxidation: 2-hydroxy estrone and 2-hydroxy estradiol. Biochem. Biophys. Res. Commun. 142:919-924; 1987.

27. Cand, F.; Verdetti, J. Superoxide dismutase, glutathione peroxi- dase, catalase and lipid peroxidation in the major organs of the aging rat. Free Radic. Biol. &led. 7:59-63; 1989.

28. Richman, P. G.; Meister, A. Regulation of y-glutamylcysteine synthetase by non-allosteric feedback inhibition by glutathione. J. Biol. Chem. 250:1422-1426; 1975.

ABBREVIATIONS

CL--corpus luteum (corpora lutea) E2--estradiol GSH--reduced glutathione hCG--human chorionic gonadotropin H20/--hydrogen peroxide LH--luteinizing hormone O2"---superoxide radical P - - progesterone ROS-reactive oxygen species SOD-superoxide dismutase TBARS--thiobarbituric acid reactants