- Email: [email protected]
Estrogen-dependent modulation of rat brain ascorbate levels and ischemia-induced ascorbate loss
Brain Research 803 Ž1998. 105–113
Research report
Estrogen-dependent modulation of rat brain ascorbate levels and ischemia-induced ascorbate loss June Kume-Kick ) , Margaret E. Rice Departments of Physiology and Neuroscience and Neurosurgery, New York UniÕersity Medical Center, 550 First AÕenue, New York, NY 10016, USA Accepted 9 June 1998
Abstract Brain ascorbate levels in young adult female rat are lower than those in males. Loss of ascorbate during ischemia is also less in females, suggesting lower oxidative stress. After ovariectomy, however, ischemia-induced loss equals that in males. In the present study, we determined ascorbate levels in maturing male and female rat brain to establish when the gender difference in content arises. We further investigated whether 17b-estradiol andror progesterone treatment modulate levels and ischemia-induced loss in ovariectomized females and compared these data with those from normal females in proestrus and estrus. Gender differences in brain ascorbate content were absent before puberty and persisted only in cortex in aging rats. Chronic estradiol treatment, whether alone or in combination with progesterone, prevented an ovariectomy-induced ascorbate increase in hippocampus and caused levels in cortex and cerebellum to fall below those of randomly sampled normal females. These same low levels were found during proestrus and estrus. Estradiol replacement after ovariectomy prevented enhanced ischemia-induced ascorbate loss in hippocampus, but not in cortex or cerebellum. Ischemia-induced losses in proestrus and estrus were similar to those in normal controls. Progesterone had little effect in any region. These data indicate that ascorbate content and redox balance in female brain are influenced postpubertally by estrogens in a region-selective manner. q 1998 Published by Elsevier Science B.V. All rights reserved. Keywords: GSH; Antioxidant; Progesterone; Ovariectomy; Oxidative stress; Estrous
1. Introduction Ascorbic acid Žascorbate. is a water-soluble, intracellular antioxidant that reaches total brain tissue levels in the millimolar range w21,29,32,40x and appears to be preferentially localized in neurons w41x. Ascorbate can scavenge reactive oxygen intermediates, including hydroxyl and peroxyl radicals, as well as superoxide anions w31,35,36x. Under normal conditions, ascorbate acts in concert with other cellular antioxidants, including antioxidant enzymes, to form an integrated defense against reactive oxygen species w7,23,44,49x. When the balance between the antioxidant network and prooxidant systems is altered, as in pathological conditions like cerebral ischemia w33x, oxidative stress occurs w33,44x. In adult rat brain, ascorbate levels are lower in females than males w3,15x, indicating an inherent gender difference in brain ascorbate regulation. Gonadectomy eliminates the Abbreviations: E 2 , 17b-estradiol; OVX, ovariectomy; FC, frontal cortex; HPC, hippocampus; CB, cerebellum ) Corresponding author. Fax: q1-212-689-0334; E-mail: [email protected]
gender difference in hippocampus and cerebellum, but not in frontal cortex w28x, suggesting that the factors which mediate the difference are regionally selective. Moreover, we have observed a gender-dependence in oxidative stress during ischemia, using ascorbate loss as a marker of redox imbalance w15,28x. Ischemia-induced loss is significantly lower in adult female brain compared to that in male. Importantly, this difference is eliminated after ovariectomy, when ascorbate loss in females increases to equal that in males in all regions tested w28x. Taken together, these findings support the hypothesis that sex hormones influence ascorbate levels and overall redox balance in female brain in a manner different from male brain. In the present study, we evaluated gender differences in brain ascorbate content in Long–Evans rats at different stages of sexual maturity: prepuberty wpostnatal day 15 ŽP15. and P25x, postpuberty Ž2 months., and in aging Ž1 year.. We also investigated whether chronic 17b-estradiol ŽE 2 . andror progesterone treatment affected ascorbate levels in adult females after ovariectomy. Furthermore, we tested the effects of these exogenously administered sex hormones on ischemia-induced ascorbate loss in ovariectomized females. Finally, we determined brain ascorbate
0006-8993r98r$19.00 q 1998 Published by Elsevier Science B.V. All rights reserved. PII: S 0 0 0 6 - 8 9 9 3 Ž 9 8 . 0 0 6 2 8 - 3
106
J. Kume-Kick, M.E. Rice r Brain Research 803 (1998) 105–113
levels and loss in normally cycling females during proestrus and estrus. Proestrus is an estrous state characterized, in part, by peaking serum estrogen levels, followed by estrus when there is a precipitous decline in estrogens w17x. Brain tissue levels of a second water-soluble antioxidant, glutathione ŽGSH., were also evaluated in these studies. In contrast to ascorbate, GSH levels in young adult rat brain are gender-independent and insensitive to gonadectomy w15,28x. Although GSH loss typically exceeds that of ascorbate under ischemic conditions w11,15,29x, loss of GSH is also gender-independent w15x. Consequently, evaluation of GSH levels in the present study served as an internal control against which to contrast gender-dependent ascorbate regulation.
tated. The scalp was then retracted and the head chilled on ice for approximately 1 min before the removal of the brain w15x. Tissue was prepared by dissecting 2–3 samples from frontal cortex, hippocampus, and cerebellum of each animal. Samples were quickly weighed in microcentrifuge vials, frozen on dry ice, then stored at y808C before processing. The model of ischemia used was decapitation ischemia w1,10,15,24,28x. In this paradigm, animals were anesthestized as described above, decapitated, and the head sealed in a water-tight container and maintained at 378C in a water bath for 1 h prior to tissue dissection. Samples were obtained for antioxidant determination, as described above. 2.2. OÕariectomy surgery
2. Experimental procedures 2.1. Animal handling Long–Evans hooded male and female rats were obtained from an in-house colony at the NYU Medical Center or from Charles River Laboratories ŽWilmington, MA.. All procedures for animal use were in strict accordance with the NIH Guide for the Care and Use of Laboratory Animals and were approved by the NYU Institutional Animal Care and Use Committee. Immature rats under 4 weeks of age were housed in individual litters with their dams. Adult rats were housed with 2–3 of the same sex per cage, maintained on a 12L:12D light cycle, and fed Purina rat chow and tap water ad libitum up to the time of experimentation. For studies of the development of the gender-dependence of brain ascorbate content, the number of rats per gender used in each group were: 5–6 at P15 and P25; eight at 2 months; and four at 1 year. In sex hormone manipulation experiments, five groups of adult ovariectomized females Ž3–4 months old. were used: 40 ovariectomy only ŽOVX., 18 ovariectomized animals given implants filled with cholesterol vehicle ŽOVX q CH., 36 ovariectomy plus E 2 ŽOVX q E 2 ., 18 ovariectomy plus progesterone ŽOVX q P., and lastly, 12 ovariectomy plus E 2 and progesterone ŽOVX q EP.. These ovariectomized animals were compared to 30 non-ovariectomized normal females, randomly sampled from a population of varying estrous states. In each group, half were used to determine basal ascorbate and the other half to determine ischemic loss. An additional group of 26 adult females Ž3 to 4 months old. were included in this study from which daily vaginal smears were obtained to confirm at least 2–3 normal consecutive cycles. Half of these animals were taken for determination of ascorbate levels and loss while in late estrus and the other half while in early to intermediate proestrus. To obtain samples for determination of ascorbate Žand GSH. content animals were anesthetized with 50 mg kgy1 sodium pentobarbital intraperitoneally Ži.p.. and decapi-
Adult female rats Ž3–4 months old. were ovariectomized in order to eliminate the primary sources of endogenous estrogen and progesterone. Briefly, ovaries were surgically removed by bilateral dorsal incision after ligation of fallopian tubes and surrounding fascia as described by Waynforth w51x. Prior to surgery, animals were anesthestized with 50 mg kgy1 sodium pentobarbital i.p. Rats were allowed to recover under post-operative care for 3 weeks prior to experiment. 2.3. Sex hormone replacement Rats receiving sex hormone replacement were implanted subcutaneously at the nape of the neck with hormone-filled silastic tubeŽs. immediately following ovariectomy. Like all OVX animals, implanted rats were allowed a 3-week recovery period after surgery. For E 2 implants Žone per animal., 1 cm lengths of silastic tubing Ž0.058 in. i.d, 0.077 in. o.d.. were packed with a mixture of 5% crystalline 17b-estradiol in cholesterol Žwrw. w30,48x. All tubes were sealed at both ends with medical grade adhesive and allowed to air-dry. For progesterone implants Žtwo per animal., 3 cm lengths of tubing were prepared and sealed with adhesive, using a method modified from the work of Attardi w4x. Animals given combined sex hormone treatments received one estradiol implant and two progesterone implants. Before implantation, tubes were tested overnight in saline containing 1% bovine serum albumin to verify that seals were intact. To ascertain sex steroid levels in hormonally manipulated rats, trunk blood was collected into heparinized tubes at time of decapitation. Plasma estradiol concentrations were assayed commercially ŽCornell Univ. Endocrinol. Labs., Ithaca, NY. by radioimmunoassay modified using a plasma extraction method as previously described w39x. Progesterone concentrations were also assayed commercially by either radioimmunoassay ŽCornell Univ. Endocrinol. Labs., Ithaca, NY. or chemiluminescent immulite assay ŽNYU Medical Center, NY, NY.. Vaginal smears taken from hormonally manipulated rats on the day of experiment were well-correlated with expected hormonal
J. Kume-Kick, M.E. Rice r Brain Research 803 (1998) 105–113
state w51x. Sex steroid levels were also sampled in normally cycling females to confirm the estrus or proestrus state indicated by vaginal smears. 2.4. Determination of brain ascorbate and GSH content Ascorbate and GSH content were assessed using reversed phase liquid chromatography with electrochemical detection, as previously described w40x. Tissue was sonicated in ice-cold, deoxygenated mobile phase, then centrifuged. Supernatant Ž10 ml. was injected directly into the HPLC system. Separation was on a 10 cm C 18 reversed phase column ŽHRA catecholamine column, ESA, Wiggens, MA., with a 7 mm guard column Ž1.5 cm, ODS, BAS.. For ascorbate and GSH detection, the detector was a gold amalgam electrode set at q0.15 V vs. AgrAgCl. For ascorbate only, a carbon electrode set at q0.70 V vs. AgrAgCl. The eluent was 10–20 mM monochloroacetic acid, pH 5.0, with 300 mg ly1 Myris-100 Žmyristyl dimethylbenzylammonium chloride., and 150 mg ly1 EDTA.
107
males; 3.77 " 0.05, females., or cerebellum Ž6.15 " 0.24, males; 6.30 " 0.13, females. Ž n s 14–15 samples per mean; not illustrated.. Nor were there differences at P25 in frontal cortex Ž3.48 " 0.06, males; 3.31 " 0.10, females., hippocampus Ž3.50 " 0.08, males; 3.51 " 0.06, females., or cerebellum Ž4.59 " 0.15, males; 4.66 " 0.20, females. Ž n s 15–18 samples per mean; Fig. 1.. At 2 months, however, as previously reported, ascorbate levels were significantly lower in females compared to males in all three regions. Taken together, these results suggest gender differences in basal ascorbate arise during puberty ŽFig. 1.. At 1 year of age, ascorbate content was still significantly lower in females than males in frontal cortex Ž2.82 " 0.04, males; 2.59 " 0.05, females; p - 0.001., but not in hippocampus Ž2.92 " 0.05, males; 2.79 " 0.05, females. or cerebellum Ž2.48 " 0.08, males; 2.47 " 0.05, females. Ž n s 12 samples per mean; Fig. 1.. The loss of gender difference in hippocampus reflected both a slight increase in female levels and decrease in males. In cerebellum, the loss of gender difference was from a significant decrease in males Ž p - 0.05., with no change in females.
2.5. Data analysis Data are expressed as mean " S.E.M. with units of mmol gy1 tissue wet weight for tissue ascorbate or GSH content. Gender differences in total tissue contents were evaluated using a t-test analysis. Comparisons of different hormone treatments or among different brain regions were statistically assessed with analysis of variance ŽANOVA.. Ascorbate loss was evaluated using two-way ANOVA, with basal vs. post-ischemic levels and hormone treatment considered as independent variables. Bonferroni–Dunns post hoc comparison test on least square means was used to assess statistical significance among different groups. Plasma neurosteroid levels were compared between appropriate controls and estrous states using t-test analysis. 2.6. Materials Silastic tubing and medical grade adhesive used to make silastic implants were obtained from Dow Corning Ž M idland, M I . . Ascorbic acid, GSH, EDTA, monochloroacetic acid, 17b-estradiol, progesterone, and cholesterol were obtained from Sigma ŽSt. Louis, MO.. The ion pairing agent used in HPLC analysis, Myris-100, was a gift of Jame Fine Chemicals ŽBound Brook, NJ..
3. Results 3.1. DeÕelopment of gender differences in brain ascorbate content At P15, there was no gender difference in ascorbate content Žmmol gy1 . in either frontal cortex Ž4.11 " 0.08, males; 4.08 " 0.07, females., hippocampus Ž3.75 " 0.05,
Fig. 1. Age-dependence of gender differences in rat brain ascorbate content. Gender differences were absent before puberty ŽP25., but were present shortly after puberty Ž2 months.. In aging animals Ž1 year., the difference is maintained in frontal cortex, but not in hippocampus or cerebellum. Data are mean"S.E.M.; ns12–24 samples per data point ŽFC, frontal cortex; HPC, hippocampus; CB, cerebellum.. Significance of differences between males and females at a given age indicated by: )) p- 0.01 and ))) p- 0.001.
108
J. Kume-Kick, M.E. Rice r Brain Research 803 (1998) 105–113
3.2. Sex hormone manipulation and brain ascorbate content Ascorbate content increased significantly following OVX in adult female hippocampus Ž p - 0.01., but not in cortex or cerebellum, when compared to the ascorbate contents of non-ovariectomized, randomly sampled females ŽFig. 2.. Data from ovariectomized rats treated with cholesterol vehicle ŽOVX q CH. were not significantly different from those of OVX only animals in any brain region tested. Notably, ascorbate levels in the hippocampus of OVX q CH animals were also significantly higher than non-ovariectomized controls Ž p - 0.05, Fig. 2.. In hippocampus, E 2 replacement after ovariectomy, whether alone or in combination with progesterone, returned ascorbate levels to those of non-ovariectomized controls. Ascorbate levels after E 2 treatment were significantly lower than those of OVX q CH animals ŽOVX q E 2 , p - 0.001; OVX q EP, p - 0.01.. Basal ascorbate levels in cortex and cerebellum were also decreased by E 2 treatment. Ascorbate content was significantly lower in the frontal cortex of OVX q E 2 Ž p - 0.001. and OVX q EP Ž p - 0.01. rats than in nonovariectomized controls that were randomly sampled throughout the estrous cycle Ž p - 0.001.. Cortical levels after E 2 treatment were also lower than in OVX q CH animals ŽOVX q E 2 , p - 0.001; OVX q EP, p - 0.05.. In cerebellum, E 2 treatment caused ascorbate levels to fall below those of non-ovariectomized controls ŽOVX q E 2 , p - 0.01; OVX q EP, p - 0.05.. Although these levels did not differ significantly from those in OVX q CH rats, they were lower than in OVX animals ŽOVX q E 2 , p - 0.001; OVX q EP, p - 0.01..
Progesterone replacement following ovariectomy ŽOVX q P. had much less effect on brain ascorbate levels than did E 2 . In frontal cortex, ascorbate in OVX q P animals did not differ significantly from either normal controls or OVX only ŽFig. 2.. Similarly, progesterone only partially reversed the ovariectomy-induced increase in hippocampus: OVX q P levels did not differ significantly from those in normal controls, however, neither did these differ from those in OVX only. Ascorbate content in all three brain regions in OVX q P animals was higher than in OVX q EP, although significantly only in hippocampus Ž p - 0.05. and cerebellum Ž p - 0.05. ŽFig. 2.. 3.3. Ischemia-induced ascorbate loss after sex hormone manipulation To assess the effect of E 2 or progesterone on ischemiainduced ascorbate loss, we determined tissue ascorbate contents after 1 h of decapitation ischemia at 378C and compared these to basal levels under each hormonal condition tested. Ascorbate loss in non-ovariectomized control females was significant only in cerebellum Ž p - 0.05. ŽFig. 3.. By contrast, ischemia-induced loss was significant in all regions tested after ovariectomy, whether OVX only Ž p 0.001. or OVX q CH Ž p - 0.001. ŽFig. 3.. In frontal cortex and cerebellum, loss remained significant after OVX even with exogenous replacement of estradiol ŽOVX q E 2 , p - 0.001., progesterone ŽOVX q P, p - 0.001. or both ŽOVX q EP, cortex, p - 0.01; cerebellum, p - 0.001. ŽFig. 3.. In hippocampus, however, ascorbate loss was decreased to that found in non-ovariectomized females in OVX q E 2 and OVX q EP animals, with post-ischemic levels that were not statistically different from basal levels in either hormone group ŽFig. 3.. Progesterone replacement alone had no effect on loss in hippocampus Žischemic vs. basal OVX q P, p - 0.01.. 3.4. Brain ascorbate leÕels in proestrus Õs. estrus adult females
Fig. 2. Effect of ovariectomy and sex hormone manipulation on brain ascorbate levels in female rats. Conditions were control Žnormal, randomly chosen females., ovariectomy ŽOVX., ovariectomy plus cholesterol vehicle ŽOVXqCH., ovariectomy plus 17b-estradiol ŽOVXqE 2 ., ovariectomy plus progesterone ŽOVXqP. or ovariectomy plus 17bestradiol and progesterone ŽOVXqEP.. Data are mean"S.E.M.; ns 26–74 samples per mean ŽFC, frontal cortex; HPC, hippocampus; CB, cerebellum.. Significance of differences from non-ovariectomized control data for a given region are indicated by: ) p- 0.05; )) p- 0.01; ))) p- 0.001.
Ascorbate levels in the frontal cortex of females in proestrus and estrus were significantly lower than the average levels found in randomly sampled, normally cycling animals Ž p - 0.001; Fig. 4A.. In cerebellum, ascorbate levels were also lower than those of average controls, although significantly only during estrus Ž p - 0.01.. These lower levels were similar to those in OVX animals with E 2 implants Žcompare Figs. 2 and 4A.. The ascorbate contents of cortex and cerebellum did not differ between proestrus and estrus. Hippocampal ascorbate levels in proestrus, by contrast, were the same as those in the average female population ŽFig. 4A.. Here too, however, proestrus levels were similar to those in OVX animals with E 2 implants.
J. Kume-Kick, M.E. Rice r Brain Research 803 (1998) 105–113
Fig. 3. Effect of ovariectomy and sex hormone manipulation on brain ascorbate loss in females rats after 1 h decapitation ischemia. Ascorbate loss is a marker for oxidative stress. Data are the difference between average basal Žsee Fig. 2. and post-ischemic levels for each condition; ns 22–74 samples for post-ischemic levels for each region and condition ŽFC, frontal cortex; HPC, hippocampus; CB, cerebellum.. Conditions were control, ovariectomy ŽOVX., ovariectomy plus cholesterol vehicle ŽOVXqCH., ovariectomy plus 17b-estradiol ŽOVXqE 2 ., ovariectomy plus progesterone ŽOVXqP. or ovariectomy plus 17b-estradiol and progesterone ŽOVXqEP.. Significance of difference between average basal and average post-ischemic ascorbate levels for each condition is indicated by: ) p- 0.05; )) p- 0.01; ))) p- 0.001.
Moreover, ascorbate content fell significantly Ž p - 0.05. between proestrus and estrus, such that hippocampal levels in estrus were significantly lower than in controls Ž p 0.01.. Ascorbate loss following 1 h decapitation ischemia was significant only in cerebellum during both proestrus Ž p 0.001. and estrus Ž p - 0.01.. This was the same pattern as in randomly chosen females w15x ŽFig. 4B..
109
Fig. 4. Brain ascorbate levels and ischemia-induced loss during estrus and proestrus compared to averaged data from randomly selected females. ŽA. Basal ascorbate content. Significance of differences between average control and proestrus or estrus data is indicated. Ascorbate levels in hippocampus ŽHPC. also differed significantly Ž p- 0.05. between estrus and proestrus, whereas those in frontal cortex ŽFC. or cerebellum ŽCB. did not. Data are mean"SEM Ž ns14–21 samples per mean for estrus and proestrus.. ŽB. Ischemia-induced ascorbate loss during proestrus and estrus. Data are the difference between average basal ŽA. and postischemic levels for each condition Ž ns14–21 samples per post-ischemic mean for estrus and proestrus.. Significance of differences between average basal and average post-ischemic ascorbate levels is indicated Ž) p- 0.05; )) p- 0.01; ))) p- 0.001..
3.5. Plasma sex hormone leÕels Plasma 17b-estradiol levels during proestrus were significantly greater Ž p - 0.05. than those during estrus, and
were comparable to levels measured in rats treated with E 2 implants ŽTable 1. w8x. Ovariectomy ŽOVX q CH. resulted in lower estradiol levels than those found in proestrus
Table 1 Plasma concentrations of progesterone and 17b-estradiol
Proestrus Žearly. Estrus Žlate. Ovariectomized Žno implant. ŽOVX. Ovariectomizedq cholesterol vehicle ŽOVX q CH. Ovariectomizedq E 2 treatment ŽOVX q E 2 . Ovariectomizedq progesterone treatment ŽOVX q P. Ovariectomizedq E 2 and progesterone treatment ŽOVX q EP. a
Progesterone Žng mly1 .
Ž n.
Estradiol Žpg mly1 .
Ž n.
8.7 " 1.8 6.2 " 0.7 0.8 " 0.2 a 3.0 " 0.5 b 5.3 " 1.6 10.0 " 1.0 c 17.6 " 1.3 d
Ž10. Ž6. Ž8. Ž23. Ž10. Ž19. Ž8.
e
Ž10. Ž6. Ž2. Ž22. Ž17.
- Proestrus Ž p - 0.01.; - estrus Ž p - 0.001.; - Proestrus Ž p - 0.01.; - estrus Ž p - 0.05.; ) OVX Ž p - 0.05.. c ) Estrus Ž p - 0.05.; ) OVX Ž p - 0.001.. d ) Proestrus Ž p - 0.05.; ) estrus Ž p - 0.001.; ) OVX q CH Ž p - 0.001.; ) OVX q P Ž p - 0.001.. e ) Estrus Ž p - 0.05.. f - Proestrus Ž p - 0.01.. g ) Estrus Ž p - 0.001.; ) OVX q CH Ž p - 0.001.. h ) Estrus Ž p - 0.01.; ) OVX q CH Ž p - 0.001.. NrA: Not available. Data are mean " SEM Ž n.. b
82 " 13 43 " 9 36 " 1.4 a 49 " 2.2 f 112 " 10 g NrA 117 " 16 h
Ž8.
110
J. Kume-Kick, M.E. Rice r Brain Research 803 (1998) 105–113
in ovariectomized females ŽFig. 5B., nor on ischemia-induced GSH loss. Lastly, there were no significant differences between GSH content or loss in estrus or proestrus females Žnot shown..
4. Discussion
Fig. 5. Gender-independent GSH levels in rat brain and lack of influence of ovarian sex hormone manipulation. ŽA. Brain GSH content Žmmol gy1 . was gender-independent in all regions at all ages tested. Data are mean"S.E.M.; ns12–24 samples per data point. ŽB. GSH levels in female brain were unaffected by ovariectomy ŽOVX., ovariectomy plus 17b-estradiol ŽOVXqE 2 ., or ovariectomy plus progesterone ŽOVXqP.. Data are mean"S.E.M.; ns 26–74 samples per mean ŽFC, frontal cortex; HPC, hippocampus; CB, cerebellum..
Ž p - 0.01., or in estradiol-treated, ovariectomized rats ŽOVX q CH: - OVX q E, p - 0.001; - OVX q EP, p - 0.001.. In contrast to estradiol, levels of progesterone in plasma during early Žto middle. proestrus were not significantly different from those during estrus ŽTable 1.. Levels of both hormones were consistent with data in the literature obtained using the same or similar assay methods w8,13,18x. Ovariectomy decreased plasma progesterone levels significantly below those in proestrus Ž p - 0.01. or estrus Žvs OVX, p - 0.001; vs. OVX q CH, p - 0.05.. Progesterone treatment, whether alone ŽOXV q P. or in combination with estradiol ŽOVX q EP., resulted in plasma concentrations that were comparable to or higher than those in intact animals in early proestrus or estrus ŽTable 1.. Plasma progesterone levels in OVX q P animals were somewhat lower than in OVX q EP Ž p - 0.001.. Concentrations reached with both implant protocols were below the peak progesterone levels that occur during late proestrus w45x. 3.6. Brain GSH leÕels and ischemia-induced loss No gender differences in cortical, hippocampal, or cerebellar GSH contents were found at any age examined ŽP15, P25, 2 months, or 1 year. ŽFig. 5A.. In addition, neither E 2 nor progesterone treatment had any effect on GSH content
The present report describes several important new findings about the relationship between brain antioxidant regulation and female sex hormones, especially estrogen. First, we have established that the gender difference in brain ascorbate content does not appear until puberty ŽFig. 1., when sex hormones have reached adult levels w37x. Puberty occurs at approximately P35 in female rats and P45 in males w37x. These data extend the findings of one earlier study of cortex only, in which an absence of gender difference in ascorbate levels was noted at birth ŽP1., with evidence of difference by P42 w3x. Second, the data demonstrate that changes in sex hormone levels in adulthood can have regionally selective effects on ascorbate regulation. As noted earlier, gonadectomy in young, postpubertal rats causes loss of the gender difference in hippocampus and cerebellum, but not in frontal cortex w28x. In hippocampus, this is the result of increased ascorbate levels in females, while in cerebellum it is the result of small, opposing changes in both genders. The present data demonstrate that such changes also occur in normal aging, with persistence of lower ascorbate levels in females compared to males in frontal cortex, but not in hippocampus or cerebellum ŽFig. 1.. By 1 year of age, the estrous cycle in female rats are becoming erratic and acyclic in nature w17,50x, reflecting altered circulating sex hormone levels. Although hormone levels were not determined in these older animals, the ascorbate data are consistent with changes in regulation in males as well as females. Similar to the effect of gonadectomy, loss of the gender difference in aging hippocampus was the result of increased ascorbate levels in females, as well as a slight decrease in males, whereas that in cerebellum was the result of a decrease in males only. Third, and most important, the data indicate that estrogen can modulate not only brain ascorbate levels, but also tissue redox status. This is most clearly demonstrated by the data from female hippocampus. Hippocampal ascorbate levels rise after ovariectomy w28x and in aging ŽFig. 1.. The OVX-induced increase could be reversed by chronic E 2 replacement ŽFig. 2.. Moreover, E 2 returned OVX-enhanced ascorbate loss during ischemia to that observed in non-ovariectomized controls ŽFig. 2.. This targeted regulation of hippocampal ascorbate levels and oxidative stress is consistent with previous reports that steroid hormones Že.g., E 2 . can induce plastic changes in normal female rat brain, including changes in apical dendritic spine density of CA1 pyramidal neurons w19,53x. The present data
J. Kume-Kick, M.E. Rice r Brain Research 803 (1998) 105–113
strengthen the concept that the sex hormone milieu of the mature hippocampus is important in modulating its neurochemistry and consequent physiology w52x. Finally, the region-selective response patterns to ovarian hormone manipulation reveal a distinct dissociation between effects on brain ascorbate regulation and on overall redox status. Most striking in this regard are the data from frontal cortex. Ascorbate levels in frontal cortex were unaltered by ovariectomy or aging. Nonetheless, chronic exposure to elevated levels of E 2 after ovariectomy caused cortical ascorbate levels to fall below those of nonovariectomized controls ŽFig. 2., as did the endogenous hormone increases that occurred during proestrus ŽFig. 4A.. Conversely, ischemia-induced ascorbate loss in frontal cortex increases after ovariectomy, reflecting enhanced oxidative stress w28x. This enhancement, however, was not affected by E 2 treatment ŽFig. 3.. A similar pattern of effects was seen in cerebellum, whereas these responses differed markedly from those in hippocampus, as already discussed. Several in vitro studies have indicated that estrogens exhibit direct antioxidant properties, particularly at high, non-physiological levels w5,22,34x. Two of the present findings argue most strongly against direct effects in vivo. First, the same levels of circulating E 2 that prevented enhanced ischemic-induced loss, i.e., increased oxidative stress, after ovariectomy in hippocampus had no effect on loss in frontal cortex or cerebellum. Second, although plasma E 2 levels fall by nearly 50% between proestrus and estrus, ischemia-induced ascorbate loss was identical in both states for cortex, hippocampus, and cerebellum. 4.1. Brain antioxidant modulation by oÕarian sex hormones The one uniform response to E 2 treatment in the three brain regions examined was a decrease in tissue ascorbate content, which indicates that estrogen may modulate ascorbate regulation. Moreover, the selective ascorbate increase in hippocampus, but not cortex or cerebellum, after ovariectomy suggests the existence of regionally distinct mechanisms of hormone dependence. One mechanism by which E 2 could cause a decrease in ascorbate levels is via inhibition of the ascorbate transporter, although the concentrations of estrogen found to be inhibitory in adrenal tissue in vitro w12x are higher than physiological levels. Progesterone can also inhibit ascorbate uptake in adrenal tissue w12x. The absence of effect of that steroid in the present studies may argue against transporter involvement, or possibly that the progesterone levels achieved with hormone implants were insufficient for this action. We have previously suggested that gender differences in ascorbate regulation are dependent on sex hormone status in females w28x. This hypothesis is further supported by the finding that brain ascorbate levels in proestrus and
111
estrus are generally lower than average female levels ŽFig. 4A.. The observation that the lowest brain ascorbate levels occurred following exposure to peak estradiol levels points to the tight homeostatic regulation of brain ascorbate content w46x. One exception to this tight control was the dramatic decrease in hipppocampal ascorbate content between proestrus and estrus ŽFig. 4A.. In hippocampus, the fall in ascorbate content following high levels of estrogens might have been further exacerbated by the dramatic decrease in CA1 pyramidal cell spine density that occurs between proestrus and estrus w19,53x. Other data from our laboratory indicate that brain ascorbate is preferentially localized in neurons w41x. The fall in hippocampal ascorbate content between proestrus and estrus would be consistent with the expected decrease in neuron volume. Brain antioxidant systems besides ascorbate are also gender-dependent in a region, age-, and sex hormone-dependent manner. In particular, Carrillo et al. w9x have found lower levels of manganese superoxide dismutase ŽMnSOD. activity in female compared to male rat brain in all regions tested, with regional variability in absolute levels. In males, Mn-SOD activity increases markedly with age, while female activity is much less affected. Mn-SOD activity in whole brain tissue is also modulated by sex hormones in females, with an increase after ovariectomy w38x. This increase in activity can be substantially reversed following either exogenous E 2 or progesterone treatment w38x. In addition, Taskiran et al. w47x recently reported that the activity of the antioxidant enzyme catalase is significantly higher in female cortex than male, but that there is no gender difference in other regions, including hippocampus and cerebellum. Taken with the present data, these results suggest that ovarian steroids may have a general, but region-selective role in regulation of brain antioxidant status. One exception to the list of gender-dependent antioxidant systems in brain is GSH. The present data indicate that GSH levels are gender-independent throughout maturation and aging and are not modulated by either E 2 or progesterone treatment. These data are consistent with our previous reports noting GSH to be gender-independent in young adult rat brain w15x and unaffected by gonadectomy w28x. Similarly, there is no apparent gender-dependence to the activity of the GSH-linked enzyme, GSH-peroxidase in rat brain w9x. One earlier report that described age-dependent changes in rat brain GSH levels w6x, noted a gender difference in 2 year-old animals, with lower whole brain levels in males compared to females. The difference between these data and the present results ŽFig. 4. might reflect either age Ž2 year vs. 1 year. or strain ŽWistar vs. Long–Evans. differences. It should be noted, however, that total tissue levels in the earlier report were lower than those presented elsewhere in the literature w43x. This suggests the possibility of GSH loss during processing or analysis, the effects of which could lead to apparent gender differences in content.
112
J. Kume-Kick, M.E. Rice r Brain Research 803 (1998) 105–113
4.2. Sex hormone-dependent modulation of oxidatiÕe stress The data reported here provide further support for the hypothesis that gender-dependent oxidative stress, mediated by ovarian sex hormones, may underlie the gender-dependence of neurodegenerative disorders linked to oxidative damage. Consistent with lower oxidative stress in female brain, damage following in vivo ischemia has been reported to be gender-dependent, with lower neuronal necrosis w22x and smaller ischemic lesion volumes w2,55x in females compared to males. Moreover, several recent clinical reports have described lower stroke mortality in women who have undergone sex hormone replacement therapy, compared to those who have not w14,16,20x. The mechanisms involved in ovarian sex hormone protection are not yet clear. The present results suggest E 2 treatment can decrease ischemia-induced oxidative stress, at least in hippocampus. Other studies investigating the effects of sex hormones on nerve trauma have suggested that E 2 , as well as other neurosteroids, can act as neuromodulators with protective effects including actions via hormone receptors w25–27,42,54x. The regionally selective effects of E 2 treatment in the present report point to the complexity of gender-dependent antioxidant regulation. The data indicate that the effect of a single sex hormone Že.g., E 2 . on a single antioxidant system Že.g., ascorbate., may differ from overall changes in redox balance, hence oxidative stress, in a specific brain region. Further investigation of underlying mechanisms of protection by ovarian sex hormones and the regional dependence of their actions should provide new insights into brain antioxidant regulation and its relationship to genderdependent neurological disorders. Acknowledgements This work was supported by NINDS grant NS-34115. The authors gratefully acknowledge Mort Levitz and Dr. Thomas Reimer for assistance in analyzing plasma steroid levels; Michelle Hsieh, Risa Koren, and Drs. Iolanda Russo-Menna and David Ferris for assistance in HPLC analysis; and Dr. Stephanie J. Cragg for critical discussion of the manuscript. References w1x K. Abe, S. Yoshida, B.D. Watson, R. Busto, K. Kogure, M.D. Ginsberg, a-Tocopherol and ubiquinones in rat brain subjected to decapitation ischemia, Brain Res. 273 Ž1983. 166–169. w2x N.J. Alkayed, I. Harukuni, A.S. Kimes, E.D. London, R.J. Traystman, P.D. Hurn, Gender-linked brain injury in experimental stroke, Stroke 29 Ž1998. 159–166. w3x J.H. Allison, M.A. Stewart, Myo-inositol and ascorbic acid in developing rat brain, J. Neurochem. 20 Ž1973. 1785–1788. w4x B. Attardi, Progesterone modulation of the luteinizing hormone surge: regulation of hypothalamic and pituitary progestin receptors, Endocrinology 115 Ž1984. 2113–2122.
w5x S. Ayres, M. Tang, R.M.T. Subbiah, Estradiol-17b as an antioxidant: some distinct features when compared with common fat-soluble antioxidants, J. Lab. Clin. Med. 128 Ž1996. 367–375. w6x E. Bien, G. Skorka, H. Rex, D. Kastner, Glutathione level in developing rat brain, Biogenic Amines 7 Ž1990. 275–281. w7x G.R. Buettner, The pecking order of free radicals and antioxidants: lipid peroxidation, a-tocopherol, and ascorbate, Arch. Biochem. Biophys. 300 Ž1993. 535–543. w8x R.L. Butcher, W.E. Collins, N.W. Fugo, Plasma concentration of LH, FSH, prolactin, progesterone and estradiol-17b throughout the 4-day estrous cycle of the rat, Endocrinology 94 Ž1974. 1704–1708. w9x M.C. Carrillo, S. Kanai, Y. Sato, K. Kitani, Age-related changes in antioxidant enzyme activities are region and organ, as well as sex, selective in the rat, Mech. Ageing Dev. 65 Ž1992. 187–198. w10x P.H. Chan, L. Chu, R.A. Fishman, Reduction of activities of superoxide dismutase but not glutathione peroxidase in rat brain regions following decapitation ischemia, Brain Res. 439 Ž1988. 388–390. w11x A.J. Cooper, W.A. Pulsinelli, T.E. Duffy, Glutathione and ascorbate during ischemia and postischemic reperfusion in rat brain, J. Neurochem. 35 Ž1980. 1242–1245. w12x A.F. DeNicola, M. Clayman, R.M. Johnstone, Hormonal control of ascorbic acid transport in rat adrenal glands, Endocrinology 82 Ž1968. 436–446. w13x W.S. Evans, B.J. Boykin, D.L. Kaiser, J.L.C. Borges, M.O. Thorner, Biphasic luteinizing hormone secretion in response to gonadotropinreleasing hormone during continuous perfusion of dispersed rat anterior pituitary cells: changes in total release and the phasic components during the estrous cycle, Endocrinology 112 Ž1983. 535–542. w14x M. Falkeborn, I. Persson, A. Terent, H.O. Adami, H. Lithell, R. Bergstrom, Hormone replacement therapy and the risk of stroke. Follow-up of a population-based cohort in Sweden, Arch. Intern. Med. 153 Ž1993. 1201–1209. w15x D.C. Ferris, J. Kume-Kick, I. Russo-Menna, M.E. Rice, Gender differences in cerebral ascorbate levels and ascorbate loss in ischemia, NeuroReport 6 Ž1995. 1485–1489. w16x F.F. Finucane, J.H. Madans, T.L. Bush, P.H. Wolf, J.C. Kleinman, Decreased risk of stroke among postmenopausal hormone users, Arch. Intern. Med. 153 Ž1993. 73–79. w17x M.E. Freeman, The ovarian cycle of the rat, in: E. Knobil, J.D. Neill ŽEds.., Physiology of Reproduction, Vol. 2, Raven Press, New York, 1988, pp. 1893–1928. w18x R.L. Goodman, A quantitative analysis of the physiological role of estradiol and progesterone in the control of tonic and surge secretion of luteinizing hormone in the rat, Endocrinology 102 Ž1978. 142– 150. w19x E. Gould, C.S. Woolley, M. Frankfurt, B.S. McEwen, Gonadal steroids regulate dendritic spine density in hippocampal pyramidal cells in adulthood, J. Neurosci. 10 Ž1990. 1286–1291. w20x F. Grodstein, M.J. Stampfer, G.A. Colditz, W.C. Willett, J.E. Manson, M. Joffe, B. Rosner, C. Fuchs, S.E. Hankinson, D.J. Hunter, C.H. Hennekens, F.E. Speizer, Postmenopausal hormone therapy and mortality, New Engl. J. Med. 336 Ž1997. 1769–1775. w21x R.A. Grunewald, Ascorbic acid in the brain, Brain Res. Rev. 18 Ž1993. 123–133. w22x E.D. Hall, K.E. Pazara, K.L. Linseman, Sex differences in postischemic neuronal necrosis in gerbils, J. Cereb. Blood Flow Metab. 11 Ž1991. 292–298. w23x B. Halliwell, How to characterize a biological antioxidant, Free Radic. Res. Commun. 9 Ž1990. 1–32. w24x S.C. Hong, Y. Goto, G. Lanzino, S. Soleau, N.F. Kassell, K.S. Lee, Neuroprotection with a calpain inhibitor in a model of focal cerebral ischemia, Stroke 25 Ž1994. 663–669. w25x K.J. Jones, Gonadal steroids as promoting factors in axonal regeneration, Brain Res. Bull. 30 Ž1993. 491–498. w26x K. Kujawa, K.J. Jones, Testosterone-induced acceleration of recov-
J. Kume-Kick, M.E. Rice r Brain Research 803 (1998) 105–113
w27x
w28x
w29x
w30x
w31x w32x w33x
w34x
w35x w36x
w37x
w38x
w39x
w40x
w41x
ery from facial paralysis in male hamsters: temporal requirements of hormone exposure, Physiol. Behav. 48 Ž1990. 765–768. J. Kume-Kick, F.L. Strand, Sex hormones affect muscle contractility and motor functional recovery following peroneal nerve crush, Exp. Neurol. 128 Ž1994. 115–123. J. Kume-Kick, D.C. Ferris, I. Russo-Menna, M.E. Rice, Enhanced oxidative stress in female rat brain after gonadectomy, Brain Res. 738 Ž1996. 8–14. P. Lyrer, H. Landolt, A. Kabersh, H. Langemann, H. Kaeser, Levels of low molecular weight scavengers in the rat brain during focal ischemia, Brain Res. 567 Ž1991. 317–320. M.Y. McGinnis, B. Parsons, T.C. Rainbow, L.C. Krey, B.S. McEwen, Temporal relationship between cell nuclear progestin receptor levels and sexual receptivity following intravenous progesterone administration, Brain Res. 218 Ž1981. 365–371. A. Meister, On the antioxidant effects of ascorbic acid and glutathione, Biochem. Pharmacol. 44 Ž1992. 1905–1915. K.H. Milby, A. Oke, R.N. Adams, Detailed mapping of ascorbate distribution in rat brain, Neurosci. Lett. 28 Ž1982. 15–20. L.E. Moore, R.J. Traystman, Role of oxygen free radicals and lipid peroxidation in cerebral reperfusion injury, Adv. Pharmacol. 31 Ž1994. 565–576. M. Nakano, K. Sugioka, I. Naito, S. Takekoshi, E. Niki, Novel and potent biological antioxidants on membrane phospholipid peroxidation: 2-hydroxy estrone and 2-hydroxy estradiol, Biochem. Biophys. Res. Commun. 142 Ž1987. 919–924. E. Niki, Vitamin C as an antioxidant, World Rev. Nutr. Dietet. 64 Ž1991. 1–30. M. Nishikimi, Oxidation of ascorbic acid with superoxide anion generated by the xanthine–xanthine oxidase system, Biochem. Biophys. Res. Commun. 63 Ž1975. 463–468. S.R. Ojeda, H.F. Urbanski, Puberty in the rat, in: E. Knobil, J.D. Neill ŽEds.., The Physiology of Reproduction, Vol. 2, Raven Press, New York, 1988, pp. 1699–1737. S. Pajovic, G. Nikezic, J.V. Martinovic, Effects of ovarian steroids on superoxide dismutase activity in the rat brain, Experientia 49 Ž1993. 73–75. T.J. Reimers, S.V. Lamb, S.A. Bartlett, R.A. Matamoros, R.B. Cowan, J.S. Engle, Effects of hemolysis and storage on quantification of hormones in blood samples from dogs, cattle and horses, Am. J. Vet. Res. 52 Ž1991. 1075–1079. M.E. Rice, E.J.K. Lee, Y. Choy, High levels of ascorbic acid, not glutathione, in the CNS of anoxia-tolerant reptiles contrasted with levels in anoxia-intolerant species, J. Neurochem. 64 Ž1995. 1790– 1799. M.E. Rice, I. Russo-Menna, Differential compartmentalization of
w42x
w43x
w44x w45x
w46x w47x
w48x
w49x
w50x
w51x w52x
w53x
w54x
w55x
113
brain ascorbate and glutathione between neurons and glia, Neuroscience 82 Ž1998. 1213–1223. R.L. Roof, R. Duvdevani, D.G. Stein, Gender influences outcome of brain injury: progesterone plays a protective role, Brain Res. 607 Ž1993. 333–336. B.R. Shivakuma, H.K. Anandatheerthavarada, V. Ravindranath, Free radical scavenging systems in developing rat brain, Int. J. Dev. Neurosci. 9 Ž1991. 181–185. H. Sies, Strategies of antioxidant defense, Eur. J. Biochem. 215 Ž1993. 213–219. M.S. Smith, M.E. Freeman, J.D. Neill, The control of progesterone secretion during the estrous cycle and early pseudopregnancy in the rat: prolactin, gonadotropin and steroid levels associated with rescue of the corpus luteum of pseudopregnancy, Endocrinology 96 Ž1975. 219–225. R. Spector, A.V. Lorenzo, Ascorbic acid homeostasis in the central nervous system, Am. J. Physiol. 225 Ž1973. 757–763. D. Taskiran, F.Z. Kutay, E. Sozmen, S. Pogun, Sex differences in nitriternitrate levels and antioxidant defense in rat brain, NeuroReport 8 Ž1997. 881–884. C.M. Vanin, N.J. MacLusky, M.D. Grynpas, R.F. Casper, The effect of three hormone replacement regimens on bone density in the aged ovariectomized rat, Fertility and Sterility 63 Ž1995. 643–651. G.T. Vatassery, W.E. Smith, H.T. Quach, Ascorbic acid, glutathione and synthetic antioxidants prevent the oxidation of vitamin E in platelets, Lipids 24 Ž1989. 1043–1047. F.S. vom Saal, C.E. Finch, Reproductive senescence: Phenomena and mechanisms in mammals and selected vertebrates, in: E. Knobil, J.D. Neill ŽEd.., The Physiology of Reproduction, Vol. 2, Raven Press, New York, 1988, pp. 2351–2413. H.G. Waynforth, Experimental and Surgical Technique in the Rat, Academic Press, London, 1988, pp. 160–163. C. Woolley, N. Weiland, B. McEwen, P. Schwartzkroin, Estradiol increases the sensitivity of hippocampal CA1 pyramidal cell to NMDA receptor-mediated synaptic input: correlation with dendritic spine density, J. Neurosci. 17 Ž1997. 1848–1859. C.S. Woolley, E. Gould, M. Frankfurt, B.S. McEwen, Naturally occurring fluctuation in dendritic spine density on adult hippocampal pyramidal neurons, J. Neurosci. 10 Ž1990. 4035–4039. W.H. Yu, Administration of testosterone attenuates neuronal loss following axotomy in the brain-stem motor nuclei of female rats, J. Neurosci. 9 Ž1989. 3908–3914. Y. Zhang, J. Shi, G. Rajakumar, A.L. Day, J.W. Simpkins, Effects of gender and estradiol treatment on focal brain ischemia, Brain Res. 784 Ž1998. 321–324.
Research report
Estrogen-dependent modulation of rat brain ascorbate levels and ischemia-induced ascorbate loss June Kume-Kick ) , Margaret E. Rice Departments of Physiology and Neuroscience and Neurosurgery, New York UniÕersity Medical Center, 550 First AÕenue, New York, NY 10016, USA Accepted 9 June 1998
Abstract Brain ascorbate levels in young adult female rat are lower than those in males. Loss of ascorbate during ischemia is also less in females, suggesting lower oxidative stress. After ovariectomy, however, ischemia-induced loss equals that in males. In the present study, we determined ascorbate levels in maturing male and female rat brain to establish when the gender difference in content arises. We further investigated whether 17b-estradiol andror progesterone treatment modulate levels and ischemia-induced loss in ovariectomized females and compared these data with those from normal females in proestrus and estrus. Gender differences in brain ascorbate content were absent before puberty and persisted only in cortex in aging rats. Chronic estradiol treatment, whether alone or in combination with progesterone, prevented an ovariectomy-induced ascorbate increase in hippocampus and caused levels in cortex and cerebellum to fall below those of randomly sampled normal females. These same low levels were found during proestrus and estrus. Estradiol replacement after ovariectomy prevented enhanced ischemia-induced ascorbate loss in hippocampus, but not in cortex or cerebellum. Ischemia-induced losses in proestrus and estrus were similar to those in normal controls. Progesterone had little effect in any region. These data indicate that ascorbate content and redox balance in female brain are influenced postpubertally by estrogens in a region-selective manner. q 1998 Published by Elsevier Science B.V. All rights reserved. Keywords: GSH; Antioxidant; Progesterone; Ovariectomy; Oxidative stress; Estrous
1. Introduction Ascorbic acid Žascorbate. is a water-soluble, intracellular antioxidant that reaches total brain tissue levels in the millimolar range w21,29,32,40x and appears to be preferentially localized in neurons w41x. Ascorbate can scavenge reactive oxygen intermediates, including hydroxyl and peroxyl radicals, as well as superoxide anions w31,35,36x. Under normal conditions, ascorbate acts in concert with other cellular antioxidants, including antioxidant enzymes, to form an integrated defense against reactive oxygen species w7,23,44,49x. When the balance between the antioxidant network and prooxidant systems is altered, as in pathological conditions like cerebral ischemia w33x, oxidative stress occurs w33,44x. In adult rat brain, ascorbate levels are lower in females than males w3,15x, indicating an inherent gender difference in brain ascorbate regulation. Gonadectomy eliminates the Abbreviations: E 2 , 17b-estradiol; OVX, ovariectomy; FC, frontal cortex; HPC, hippocampus; CB, cerebellum ) Corresponding author. Fax: q1-212-689-0334; E-mail: [email protected]
gender difference in hippocampus and cerebellum, but not in frontal cortex w28x, suggesting that the factors which mediate the difference are regionally selective. Moreover, we have observed a gender-dependence in oxidative stress during ischemia, using ascorbate loss as a marker of redox imbalance w15,28x. Ischemia-induced loss is significantly lower in adult female brain compared to that in male. Importantly, this difference is eliminated after ovariectomy, when ascorbate loss in females increases to equal that in males in all regions tested w28x. Taken together, these findings support the hypothesis that sex hormones influence ascorbate levels and overall redox balance in female brain in a manner different from male brain. In the present study, we evaluated gender differences in brain ascorbate content in Long–Evans rats at different stages of sexual maturity: prepuberty wpostnatal day 15 ŽP15. and P25x, postpuberty Ž2 months., and in aging Ž1 year.. We also investigated whether chronic 17b-estradiol ŽE 2 . andror progesterone treatment affected ascorbate levels in adult females after ovariectomy. Furthermore, we tested the effects of these exogenously administered sex hormones on ischemia-induced ascorbate loss in ovariectomized females. Finally, we determined brain ascorbate
0006-8993r98r$19.00 q 1998 Published by Elsevier Science B.V. All rights reserved. PII: S 0 0 0 6 - 8 9 9 3 Ž 9 8 . 0 0 6 2 8 - 3
106
J. Kume-Kick, M.E. Rice r Brain Research 803 (1998) 105–113
levels and loss in normally cycling females during proestrus and estrus. Proestrus is an estrous state characterized, in part, by peaking serum estrogen levels, followed by estrus when there is a precipitous decline in estrogens w17x. Brain tissue levels of a second water-soluble antioxidant, glutathione ŽGSH., were also evaluated in these studies. In contrast to ascorbate, GSH levels in young adult rat brain are gender-independent and insensitive to gonadectomy w15,28x. Although GSH loss typically exceeds that of ascorbate under ischemic conditions w11,15,29x, loss of GSH is also gender-independent w15x. Consequently, evaluation of GSH levels in the present study served as an internal control against which to contrast gender-dependent ascorbate regulation.
tated. The scalp was then retracted and the head chilled on ice for approximately 1 min before the removal of the brain w15x. Tissue was prepared by dissecting 2–3 samples from frontal cortex, hippocampus, and cerebellum of each animal. Samples were quickly weighed in microcentrifuge vials, frozen on dry ice, then stored at y808C before processing. The model of ischemia used was decapitation ischemia w1,10,15,24,28x. In this paradigm, animals were anesthestized as described above, decapitated, and the head sealed in a water-tight container and maintained at 378C in a water bath for 1 h prior to tissue dissection. Samples were obtained for antioxidant determination, as described above. 2.2. OÕariectomy surgery
2. Experimental procedures 2.1. Animal handling Long–Evans hooded male and female rats were obtained from an in-house colony at the NYU Medical Center or from Charles River Laboratories ŽWilmington, MA.. All procedures for animal use were in strict accordance with the NIH Guide for the Care and Use of Laboratory Animals and were approved by the NYU Institutional Animal Care and Use Committee. Immature rats under 4 weeks of age were housed in individual litters with their dams. Adult rats were housed with 2–3 of the same sex per cage, maintained on a 12L:12D light cycle, and fed Purina rat chow and tap water ad libitum up to the time of experimentation. For studies of the development of the gender-dependence of brain ascorbate content, the number of rats per gender used in each group were: 5–6 at P15 and P25; eight at 2 months; and four at 1 year. In sex hormone manipulation experiments, five groups of adult ovariectomized females Ž3–4 months old. were used: 40 ovariectomy only ŽOVX., 18 ovariectomized animals given implants filled with cholesterol vehicle ŽOVX q CH., 36 ovariectomy plus E 2 ŽOVX q E 2 ., 18 ovariectomy plus progesterone ŽOVX q P., and lastly, 12 ovariectomy plus E 2 and progesterone ŽOVX q EP.. These ovariectomized animals were compared to 30 non-ovariectomized normal females, randomly sampled from a population of varying estrous states. In each group, half were used to determine basal ascorbate and the other half to determine ischemic loss. An additional group of 26 adult females Ž3 to 4 months old. were included in this study from which daily vaginal smears were obtained to confirm at least 2–3 normal consecutive cycles. Half of these animals were taken for determination of ascorbate levels and loss while in late estrus and the other half while in early to intermediate proestrus. To obtain samples for determination of ascorbate Žand GSH. content animals were anesthetized with 50 mg kgy1 sodium pentobarbital intraperitoneally Ži.p.. and decapi-
Adult female rats Ž3–4 months old. were ovariectomized in order to eliminate the primary sources of endogenous estrogen and progesterone. Briefly, ovaries were surgically removed by bilateral dorsal incision after ligation of fallopian tubes and surrounding fascia as described by Waynforth w51x. Prior to surgery, animals were anesthestized with 50 mg kgy1 sodium pentobarbital i.p. Rats were allowed to recover under post-operative care for 3 weeks prior to experiment. 2.3. Sex hormone replacement Rats receiving sex hormone replacement were implanted subcutaneously at the nape of the neck with hormone-filled silastic tubeŽs. immediately following ovariectomy. Like all OVX animals, implanted rats were allowed a 3-week recovery period after surgery. For E 2 implants Žone per animal., 1 cm lengths of silastic tubing Ž0.058 in. i.d, 0.077 in. o.d.. were packed with a mixture of 5% crystalline 17b-estradiol in cholesterol Žwrw. w30,48x. All tubes were sealed at both ends with medical grade adhesive and allowed to air-dry. For progesterone implants Žtwo per animal., 3 cm lengths of tubing were prepared and sealed with adhesive, using a method modified from the work of Attardi w4x. Animals given combined sex hormone treatments received one estradiol implant and two progesterone implants. Before implantation, tubes were tested overnight in saline containing 1% bovine serum albumin to verify that seals were intact. To ascertain sex steroid levels in hormonally manipulated rats, trunk blood was collected into heparinized tubes at time of decapitation. Plasma estradiol concentrations were assayed commercially ŽCornell Univ. Endocrinol. Labs., Ithaca, NY. by radioimmunoassay modified using a plasma extraction method as previously described w39x. Progesterone concentrations were also assayed commercially by either radioimmunoassay ŽCornell Univ. Endocrinol. Labs., Ithaca, NY. or chemiluminescent immulite assay ŽNYU Medical Center, NY, NY.. Vaginal smears taken from hormonally manipulated rats on the day of experiment were well-correlated with expected hormonal
J. Kume-Kick, M.E. Rice r Brain Research 803 (1998) 105–113
state w51x. Sex steroid levels were also sampled in normally cycling females to confirm the estrus or proestrus state indicated by vaginal smears. 2.4. Determination of brain ascorbate and GSH content Ascorbate and GSH content were assessed using reversed phase liquid chromatography with electrochemical detection, as previously described w40x. Tissue was sonicated in ice-cold, deoxygenated mobile phase, then centrifuged. Supernatant Ž10 ml. was injected directly into the HPLC system. Separation was on a 10 cm C 18 reversed phase column ŽHRA catecholamine column, ESA, Wiggens, MA., with a 7 mm guard column Ž1.5 cm, ODS, BAS.. For ascorbate and GSH detection, the detector was a gold amalgam electrode set at q0.15 V vs. AgrAgCl. For ascorbate only, a carbon electrode set at q0.70 V vs. AgrAgCl. The eluent was 10–20 mM monochloroacetic acid, pH 5.0, with 300 mg ly1 Myris-100 Žmyristyl dimethylbenzylammonium chloride., and 150 mg ly1 EDTA.
107
males; 3.77 " 0.05, females., or cerebellum Ž6.15 " 0.24, males; 6.30 " 0.13, females. Ž n s 14–15 samples per mean; not illustrated.. Nor were there differences at P25 in frontal cortex Ž3.48 " 0.06, males; 3.31 " 0.10, females., hippocampus Ž3.50 " 0.08, males; 3.51 " 0.06, females., or cerebellum Ž4.59 " 0.15, males; 4.66 " 0.20, females. Ž n s 15–18 samples per mean; Fig. 1.. At 2 months, however, as previously reported, ascorbate levels were significantly lower in females compared to males in all three regions. Taken together, these results suggest gender differences in basal ascorbate arise during puberty ŽFig. 1.. At 1 year of age, ascorbate content was still significantly lower in females than males in frontal cortex Ž2.82 " 0.04, males; 2.59 " 0.05, females; p - 0.001., but not in hippocampus Ž2.92 " 0.05, males; 2.79 " 0.05, females. or cerebellum Ž2.48 " 0.08, males; 2.47 " 0.05, females. Ž n s 12 samples per mean; Fig. 1.. The loss of gender difference in hippocampus reflected both a slight increase in female levels and decrease in males. In cerebellum, the loss of gender difference was from a significant decrease in males Ž p - 0.05., with no change in females.
2.5. Data analysis Data are expressed as mean " S.E.M. with units of mmol gy1 tissue wet weight for tissue ascorbate or GSH content. Gender differences in total tissue contents were evaluated using a t-test analysis. Comparisons of different hormone treatments or among different brain regions were statistically assessed with analysis of variance ŽANOVA.. Ascorbate loss was evaluated using two-way ANOVA, with basal vs. post-ischemic levels and hormone treatment considered as independent variables. Bonferroni–Dunns post hoc comparison test on least square means was used to assess statistical significance among different groups. Plasma neurosteroid levels were compared between appropriate controls and estrous states using t-test analysis. 2.6. Materials Silastic tubing and medical grade adhesive used to make silastic implants were obtained from Dow Corning Ž M idland, M I . . Ascorbic acid, GSH, EDTA, monochloroacetic acid, 17b-estradiol, progesterone, and cholesterol were obtained from Sigma ŽSt. Louis, MO.. The ion pairing agent used in HPLC analysis, Myris-100, was a gift of Jame Fine Chemicals ŽBound Brook, NJ..
3. Results 3.1. DeÕelopment of gender differences in brain ascorbate content At P15, there was no gender difference in ascorbate content Žmmol gy1 . in either frontal cortex Ž4.11 " 0.08, males; 4.08 " 0.07, females., hippocampus Ž3.75 " 0.05,
Fig. 1. Age-dependence of gender differences in rat brain ascorbate content. Gender differences were absent before puberty ŽP25., but were present shortly after puberty Ž2 months.. In aging animals Ž1 year., the difference is maintained in frontal cortex, but not in hippocampus or cerebellum. Data are mean"S.E.M.; ns12–24 samples per data point ŽFC, frontal cortex; HPC, hippocampus; CB, cerebellum.. Significance of differences between males and females at a given age indicated by: )) p- 0.01 and ))) p- 0.001.
108
J. Kume-Kick, M.E. Rice r Brain Research 803 (1998) 105–113
3.2. Sex hormone manipulation and brain ascorbate content Ascorbate content increased significantly following OVX in adult female hippocampus Ž p - 0.01., but not in cortex or cerebellum, when compared to the ascorbate contents of non-ovariectomized, randomly sampled females ŽFig. 2.. Data from ovariectomized rats treated with cholesterol vehicle ŽOVX q CH. were not significantly different from those of OVX only animals in any brain region tested. Notably, ascorbate levels in the hippocampus of OVX q CH animals were also significantly higher than non-ovariectomized controls Ž p - 0.05, Fig. 2.. In hippocampus, E 2 replacement after ovariectomy, whether alone or in combination with progesterone, returned ascorbate levels to those of non-ovariectomized controls. Ascorbate levels after E 2 treatment were significantly lower than those of OVX q CH animals ŽOVX q E 2 , p - 0.001; OVX q EP, p - 0.01.. Basal ascorbate levels in cortex and cerebellum were also decreased by E 2 treatment. Ascorbate content was significantly lower in the frontal cortex of OVX q E 2 Ž p - 0.001. and OVX q EP Ž p - 0.01. rats than in nonovariectomized controls that were randomly sampled throughout the estrous cycle Ž p - 0.001.. Cortical levels after E 2 treatment were also lower than in OVX q CH animals ŽOVX q E 2 , p - 0.001; OVX q EP, p - 0.05.. In cerebellum, E 2 treatment caused ascorbate levels to fall below those of non-ovariectomized controls ŽOVX q E 2 , p - 0.01; OVX q EP, p - 0.05.. Although these levels did not differ significantly from those in OVX q CH rats, they were lower than in OVX animals ŽOVX q E 2 , p - 0.001; OVX q EP, p - 0.01..
Progesterone replacement following ovariectomy ŽOVX q P. had much less effect on brain ascorbate levels than did E 2 . In frontal cortex, ascorbate in OVX q P animals did not differ significantly from either normal controls or OVX only ŽFig. 2.. Similarly, progesterone only partially reversed the ovariectomy-induced increase in hippocampus: OVX q P levels did not differ significantly from those in normal controls, however, neither did these differ from those in OVX only. Ascorbate content in all three brain regions in OVX q P animals was higher than in OVX q EP, although significantly only in hippocampus Ž p - 0.05. and cerebellum Ž p - 0.05. ŽFig. 2.. 3.3. Ischemia-induced ascorbate loss after sex hormone manipulation To assess the effect of E 2 or progesterone on ischemiainduced ascorbate loss, we determined tissue ascorbate contents after 1 h of decapitation ischemia at 378C and compared these to basal levels under each hormonal condition tested. Ascorbate loss in non-ovariectomized control females was significant only in cerebellum Ž p - 0.05. ŽFig. 3.. By contrast, ischemia-induced loss was significant in all regions tested after ovariectomy, whether OVX only Ž p 0.001. or OVX q CH Ž p - 0.001. ŽFig. 3.. In frontal cortex and cerebellum, loss remained significant after OVX even with exogenous replacement of estradiol ŽOVX q E 2 , p - 0.001., progesterone ŽOVX q P, p - 0.001. or both ŽOVX q EP, cortex, p - 0.01; cerebellum, p - 0.001. ŽFig. 3.. In hippocampus, however, ascorbate loss was decreased to that found in non-ovariectomized females in OVX q E 2 and OVX q EP animals, with post-ischemic levels that were not statistically different from basal levels in either hormone group ŽFig. 3.. Progesterone replacement alone had no effect on loss in hippocampus Žischemic vs. basal OVX q P, p - 0.01.. 3.4. Brain ascorbate leÕels in proestrus Õs. estrus adult females
Fig. 2. Effect of ovariectomy and sex hormone manipulation on brain ascorbate levels in female rats. Conditions were control Žnormal, randomly chosen females., ovariectomy ŽOVX., ovariectomy plus cholesterol vehicle ŽOVXqCH., ovariectomy plus 17b-estradiol ŽOVXqE 2 ., ovariectomy plus progesterone ŽOVXqP. or ovariectomy plus 17bestradiol and progesterone ŽOVXqEP.. Data are mean"S.E.M.; ns 26–74 samples per mean ŽFC, frontal cortex; HPC, hippocampus; CB, cerebellum.. Significance of differences from non-ovariectomized control data for a given region are indicated by: ) p- 0.05; )) p- 0.01; ))) p- 0.001.
Ascorbate levels in the frontal cortex of females in proestrus and estrus were significantly lower than the average levels found in randomly sampled, normally cycling animals Ž p - 0.001; Fig. 4A.. In cerebellum, ascorbate levels were also lower than those of average controls, although significantly only during estrus Ž p - 0.01.. These lower levels were similar to those in OVX animals with E 2 implants Žcompare Figs. 2 and 4A.. The ascorbate contents of cortex and cerebellum did not differ between proestrus and estrus. Hippocampal ascorbate levels in proestrus, by contrast, were the same as those in the average female population ŽFig. 4A.. Here too, however, proestrus levels were similar to those in OVX animals with E 2 implants.
J. Kume-Kick, M.E. Rice r Brain Research 803 (1998) 105–113
Fig. 3. Effect of ovariectomy and sex hormone manipulation on brain ascorbate loss in females rats after 1 h decapitation ischemia. Ascorbate loss is a marker for oxidative stress. Data are the difference between average basal Žsee Fig. 2. and post-ischemic levels for each condition; ns 22–74 samples for post-ischemic levels for each region and condition ŽFC, frontal cortex; HPC, hippocampus; CB, cerebellum.. Conditions were control, ovariectomy ŽOVX., ovariectomy plus cholesterol vehicle ŽOVXqCH., ovariectomy plus 17b-estradiol ŽOVXqE 2 ., ovariectomy plus progesterone ŽOVXqP. or ovariectomy plus 17b-estradiol and progesterone ŽOVXqEP.. Significance of difference between average basal and average post-ischemic ascorbate levels for each condition is indicated by: ) p- 0.05; )) p- 0.01; ))) p- 0.001.
Moreover, ascorbate content fell significantly Ž p - 0.05. between proestrus and estrus, such that hippocampal levels in estrus were significantly lower than in controls Ž p 0.01.. Ascorbate loss following 1 h decapitation ischemia was significant only in cerebellum during both proestrus Ž p 0.001. and estrus Ž p - 0.01.. This was the same pattern as in randomly chosen females w15x ŽFig. 4B..
109
Fig. 4. Brain ascorbate levels and ischemia-induced loss during estrus and proestrus compared to averaged data from randomly selected females. ŽA. Basal ascorbate content. Significance of differences between average control and proestrus or estrus data is indicated. Ascorbate levels in hippocampus ŽHPC. also differed significantly Ž p- 0.05. between estrus and proestrus, whereas those in frontal cortex ŽFC. or cerebellum ŽCB. did not. Data are mean"SEM Ž ns14–21 samples per mean for estrus and proestrus.. ŽB. Ischemia-induced ascorbate loss during proestrus and estrus. Data are the difference between average basal ŽA. and postischemic levels for each condition Ž ns14–21 samples per post-ischemic mean for estrus and proestrus.. Significance of differences between average basal and average post-ischemic ascorbate levels is indicated Ž) p- 0.05; )) p- 0.01; ))) p- 0.001..
3.5. Plasma sex hormone leÕels Plasma 17b-estradiol levels during proestrus were significantly greater Ž p - 0.05. than those during estrus, and
were comparable to levels measured in rats treated with E 2 implants ŽTable 1. w8x. Ovariectomy ŽOVX q CH. resulted in lower estradiol levels than those found in proestrus
Table 1 Plasma concentrations of progesterone and 17b-estradiol
Proestrus Žearly. Estrus Žlate. Ovariectomized Žno implant. ŽOVX. Ovariectomizedq cholesterol vehicle ŽOVX q CH. Ovariectomizedq E 2 treatment ŽOVX q E 2 . Ovariectomizedq progesterone treatment ŽOVX q P. Ovariectomizedq E 2 and progesterone treatment ŽOVX q EP. a
Progesterone Žng mly1 .
Ž n.
Estradiol Žpg mly1 .
Ž n.
8.7 " 1.8 6.2 " 0.7 0.8 " 0.2 a 3.0 " 0.5 b 5.3 " 1.6 10.0 " 1.0 c 17.6 " 1.3 d
Ž10. Ž6. Ž8. Ž23. Ž10. Ž19. Ž8.
e
Ž10. Ž6. Ž2. Ž22. Ž17.
- Proestrus Ž p - 0.01.; - estrus Ž p - 0.001.; - Proestrus Ž p - 0.01.; - estrus Ž p - 0.05.; ) OVX Ž p - 0.05.. c ) Estrus Ž p - 0.05.; ) OVX Ž p - 0.001.. d ) Proestrus Ž p - 0.05.; ) estrus Ž p - 0.001.; ) OVX q CH Ž p - 0.001.; ) OVX q P Ž p - 0.001.. e ) Estrus Ž p - 0.05.. f - Proestrus Ž p - 0.01.. g ) Estrus Ž p - 0.001.; ) OVX q CH Ž p - 0.001.. h ) Estrus Ž p - 0.01.; ) OVX q CH Ž p - 0.001.. NrA: Not available. Data are mean " SEM Ž n.. b
82 " 13 43 " 9 36 " 1.4 a 49 " 2.2 f 112 " 10 g NrA 117 " 16 h
Ž8.
110
J. Kume-Kick, M.E. Rice r Brain Research 803 (1998) 105–113
in ovariectomized females ŽFig. 5B., nor on ischemia-induced GSH loss. Lastly, there were no significant differences between GSH content or loss in estrus or proestrus females Žnot shown..
4. Discussion
Fig. 5. Gender-independent GSH levels in rat brain and lack of influence of ovarian sex hormone manipulation. ŽA. Brain GSH content Žmmol gy1 . was gender-independent in all regions at all ages tested. Data are mean"S.E.M.; ns12–24 samples per data point. ŽB. GSH levels in female brain were unaffected by ovariectomy ŽOVX., ovariectomy plus 17b-estradiol ŽOVXqE 2 ., or ovariectomy plus progesterone ŽOVXqP.. Data are mean"S.E.M.; ns 26–74 samples per mean ŽFC, frontal cortex; HPC, hippocampus; CB, cerebellum..
Ž p - 0.01., or in estradiol-treated, ovariectomized rats ŽOVX q CH: - OVX q E, p - 0.001; - OVX q EP, p - 0.001.. In contrast to estradiol, levels of progesterone in plasma during early Žto middle. proestrus were not significantly different from those during estrus ŽTable 1.. Levels of both hormones were consistent with data in the literature obtained using the same or similar assay methods w8,13,18x. Ovariectomy decreased plasma progesterone levels significantly below those in proestrus Ž p - 0.01. or estrus Žvs OVX, p - 0.001; vs. OVX q CH, p - 0.05.. Progesterone treatment, whether alone ŽOXV q P. or in combination with estradiol ŽOVX q EP., resulted in plasma concentrations that were comparable to or higher than those in intact animals in early proestrus or estrus ŽTable 1.. Plasma progesterone levels in OVX q P animals were somewhat lower than in OVX q EP Ž p - 0.001.. Concentrations reached with both implant protocols were below the peak progesterone levels that occur during late proestrus w45x. 3.6. Brain GSH leÕels and ischemia-induced loss No gender differences in cortical, hippocampal, or cerebellar GSH contents were found at any age examined ŽP15, P25, 2 months, or 1 year. ŽFig. 5A.. In addition, neither E 2 nor progesterone treatment had any effect on GSH content
The present report describes several important new findings about the relationship between brain antioxidant regulation and female sex hormones, especially estrogen. First, we have established that the gender difference in brain ascorbate content does not appear until puberty ŽFig. 1., when sex hormones have reached adult levels w37x. Puberty occurs at approximately P35 in female rats and P45 in males w37x. These data extend the findings of one earlier study of cortex only, in which an absence of gender difference in ascorbate levels was noted at birth ŽP1., with evidence of difference by P42 w3x. Second, the data demonstrate that changes in sex hormone levels in adulthood can have regionally selective effects on ascorbate regulation. As noted earlier, gonadectomy in young, postpubertal rats causes loss of the gender difference in hippocampus and cerebellum, but not in frontal cortex w28x. In hippocampus, this is the result of increased ascorbate levels in females, while in cerebellum it is the result of small, opposing changes in both genders. The present data demonstrate that such changes also occur in normal aging, with persistence of lower ascorbate levels in females compared to males in frontal cortex, but not in hippocampus or cerebellum ŽFig. 1.. By 1 year of age, the estrous cycle in female rats are becoming erratic and acyclic in nature w17,50x, reflecting altered circulating sex hormone levels. Although hormone levels were not determined in these older animals, the ascorbate data are consistent with changes in regulation in males as well as females. Similar to the effect of gonadectomy, loss of the gender difference in aging hippocampus was the result of increased ascorbate levels in females, as well as a slight decrease in males, whereas that in cerebellum was the result of a decrease in males only. Third, and most important, the data indicate that estrogen can modulate not only brain ascorbate levels, but also tissue redox status. This is most clearly demonstrated by the data from female hippocampus. Hippocampal ascorbate levels rise after ovariectomy w28x and in aging ŽFig. 1.. The OVX-induced increase could be reversed by chronic E 2 replacement ŽFig. 2.. Moreover, E 2 returned OVX-enhanced ascorbate loss during ischemia to that observed in non-ovariectomized controls ŽFig. 2.. This targeted regulation of hippocampal ascorbate levels and oxidative stress is consistent with previous reports that steroid hormones Že.g., E 2 . can induce plastic changes in normal female rat brain, including changes in apical dendritic spine density of CA1 pyramidal neurons w19,53x. The present data
J. Kume-Kick, M.E. Rice r Brain Research 803 (1998) 105–113
strengthen the concept that the sex hormone milieu of the mature hippocampus is important in modulating its neurochemistry and consequent physiology w52x. Finally, the region-selective response patterns to ovarian hormone manipulation reveal a distinct dissociation between effects on brain ascorbate regulation and on overall redox status. Most striking in this regard are the data from frontal cortex. Ascorbate levels in frontal cortex were unaltered by ovariectomy or aging. Nonetheless, chronic exposure to elevated levels of E 2 after ovariectomy caused cortical ascorbate levels to fall below those of nonovariectomized controls ŽFig. 2., as did the endogenous hormone increases that occurred during proestrus ŽFig. 4A.. Conversely, ischemia-induced ascorbate loss in frontal cortex increases after ovariectomy, reflecting enhanced oxidative stress w28x. This enhancement, however, was not affected by E 2 treatment ŽFig. 3.. A similar pattern of effects was seen in cerebellum, whereas these responses differed markedly from those in hippocampus, as already discussed. Several in vitro studies have indicated that estrogens exhibit direct antioxidant properties, particularly at high, non-physiological levels w5,22,34x. Two of the present findings argue most strongly against direct effects in vivo. First, the same levels of circulating E 2 that prevented enhanced ischemic-induced loss, i.e., increased oxidative stress, after ovariectomy in hippocampus had no effect on loss in frontal cortex or cerebellum. Second, although plasma E 2 levels fall by nearly 50% between proestrus and estrus, ischemia-induced ascorbate loss was identical in both states for cortex, hippocampus, and cerebellum. 4.1. Brain antioxidant modulation by oÕarian sex hormones The one uniform response to E 2 treatment in the three brain regions examined was a decrease in tissue ascorbate content, which indicates that estrogen may modulate ascorbate regulation. Moreover, the selective ascorbate increase in hippocampus, but not cortex or cerebellum, after ovariectomy suggests the existence of regionally distinct mechanisms of hormone dependence. One mechanism by which E 2 could cause a decrease in ascorbate levels is via inhibition of the ascorbate transporter, although the concentrations of estrogen found to be inhibitory in adrenal tissue in vitro w12x are higher than physiological levels. Progesterone can also inhibit ascorbate uptake in adrenal tissue w12x. The absence of effect of that steroid in the present studies may argue against transporter involvement, or possibly that the progesterone levels achieved with hormone implants were insufficient for this action. We have previously suggested that gender differences in ascorbate regulation are dependent on sex hormone status in females w28x. This hypothesis is further supported by the finding that brain ascorbate levels in proestrus and
111
estrus are generally lower than average female levels ŽFig. 4A.. The observation that the lowest brain ascorbate levels occurred following exposure to peak estradiol levels points to the tight homeostatic regulation of brain ascorbate content w46x. One exception to this tight control was the dramatic decrease in hipppocampal ascorbate content between proestrus and estrus ŽFig. 4A.. In hippocampus, the fall in ascorbate content following high levels of estrogens might have been further exacerbated by the dramatic decrease in CA1 pyramidal cell spine density that occurs between proestrus and estrus w19,53x. Other data from our laboratory indicate that brain ascorbate is preferentially localized in neurons w41x. The fall in hippocampal ascorbate content between proestrus and estrus would be consistent with the expected decrease in neuron volume. Brain antioxidant systems besides ascorbate are also gender-dependent in a region, age-, and sex hormone-dependent manner. In particular, Carrillo et al. w9x have found lower levels of manganese superoxide dismutase ŽMnSOD. activity in female compared to male rat brain in all regions tested, with regional variability in absolute levels. In males, Mn-SOD activity increases markedly with age, while female activity is much less affected. Mn-SOD activity in whole brain tissue is also modulated by sex hormones in females, with an increase after ovariectomy w38x. This increase in activity can be substantially reversed following either exogenous E 2 or progesterone treatment w38x. In addition, Taskiran et al. w47x recently reported that the activity of the antioxidant enzyme catalase is significantly higher in female cortex than male, but that there is no gender difference in other regions, including hippocampus and cerebellum. Taken with the present data, these results suggest that ovarian steroids may have a general, but region-selective role in regulation of brain antioxidant status. One exception to the list of gender-dependent antioxidant systems in brain is GSH. The present data indicate that GSH levels are gender-independent throughout maturation and aging and are not modulated by either E 2 or progesterone treatment. These data are consistent with our previous reports noting GSH to be gender-independent in young adult rat brain w15x and unaffected by gonadectomy w28x. Similarly, there is no apparent gender-dependence to the activity of the GSH-linked enzyme, GSH-peroxidase in rat brain w9x. One earlier report that described age-dependent changes in rat brain GSH levels w6x, noted a gender difference in 2 year-old animals, with lower whole brain levels in males compared to females. The difference between these data and the present results ŽFig. 4. might reflect either age Ž2 year vs. 1 year. or strain ŽWistar vs. Long–Evans. differences. It should be noted, however, that total tissue levels in the earlier report were lower than those presented elsewhere in the literature w43x. This suggests the possibility of GSH loss during processing or analysis, the effects of which could lead to apparent gender differences in content.
112
J. Kume-Kick, M.E. Rice r Brain Research 803 (1998) 105–113
4.2. Sex hormone-dependent modulation of oxidatiÕe stress The data reported here provide further support for the hypothesis that gender-dependent oxidative stress, mediated by ovarian sex hormones, may underlie the gender-dependence of neurodegenerative disorders linked to oxidative damage. Consistent with lower oxidative stress in female brain, damage following in vivo ischemia has been reported to be gender-dependent, with lower neuronal necrosis w22x and smaller ischemic lesion volumes w2,55x in females compared to males. Moreover, several recent clinical reports have described lower stroke mortality in women who have undergone sex hormone replacement therapy, compared to those who have not w14,16,20x. The mechanisms involved in ovarian sex hormone protection are not yet clear. The present results suggest E 2 treatment can decrease ischemia-induced oxidative stress, at least in hippocampus. Other studies investigating the effects of sex hormones on nerve trauma have suggested that E 2 , as well as other neurosteroids, can act as neuromodulators with protective effects including actions via hormone receptors w25–27,42,54x. The regionally selective effects of E 2 treatment in the present report point to the complexity of gender-dependent antioxidant regulation. The data indicate that the effect of a single sex hormone Že.g., E 2 . on a single antioxidant system Že.g., ascorbate., may differ from overall changes in redox balance, hence oxidative stress, in a specific brain region. Further investigation of underlying mechanisms of protection by ovarian sex hormones and the regional dependence of their actions should provide new insights into brain antioxidant regulation and its relationship to genderdependent neurological disorders. Acknowledgements This work was supported by NINDS grant NS-34115. The authors gratefully acknowledge Mort Levitz and Dr. Thomas Reimer for assistance in analyzing plasma steroid levels; Michelle Hsieh, Risa Koren, and Drs. Iolanda Russo-Menna and David Ferris for assistance in HPLC analysis; and Dr. Stephanie J. Cragg for critical discussion of the manuscript. References w1x K. Abe, S. Yoshida, B.D. Watson, R. Busto, K. Kogure, M.D. Ginsberg, a-Tocopherol and ubiquinones in rat brain subjected to decapitation ischemia, Brain Res. 273 Ž1983. 166–169. w2x N.J. Alkayed, I. Harukuni, A.S. Kimes, E.D. London, R.J. Traystman, P.D. Hurn, Gender-linked brain injury in experimental stroke, Stroke 29 Ž1998. 159–166. w3x J.H. Allison, M.A. Stewart, Myo-inositol and ascorbic acid in developing rat brain, J. Neurochem. 20 Ž1973. 1785–1788. w4x B. Attardi, Progesterone modulation of the luteinizing hormone surge: regulation of hypothalamic and pituitary progestin receptors, Endocrinology 115 Ž1984. 2113–2122.
w5x S. Ayres, M. Tang, R.M.T. Subbiah, Estradiol-17b as an antioxidant: some distinct features when compared with common fat-soluble antioxidants, J. Lab. Clin. Med. 128 Ž1996. 367–375. w6x E. Bien, G. Skorka, H. Rex, D. Kastner, Glutathione level in developing rat brain, Biogenic Amines 7 Ž1990. 275–281. w7x G.R. Buettner, The pecking order of free radicals and antioxidants: lipid peroxidation, a-tocopherol, and ascorbate, Arch. Biochem. Biophys. 300 Ž1993. 535–543. w8x R.L. Butcher, W.E. Collins, N.W. Fugo, Plasma concentration of LH, FSH, prolactin, progesterone and estradiol-17b throughout the 4-day estrous cycle of the rat, Endocrinology 94 Ž1974. 1704–1708. w9x M.C. Carrillo, S. Kanai, Y. Sato, K. Kitani, Age-related changes in antioxidant enzyme activities are region and organ, as well as sex, selective in the rat, Mech. Ageing Dev. 65 Ž1992. 187–198. w10x P.H. Chan, L. Chu, R.A. Fishman, Reduction of activities of superoxide dismutase but not glutathione peroxidase in rat brain regions following decapitation ischemia, Brain Res. 439 Ž1988. 388–390. w11x A.J. Cooper, W.A. Pulsinelli, T.E. Duffy, Glutathione and ascorbate during ischemia and postischemic reperfusion in rat brain, J. Neurochem. 35 Ž1980. 1242–1245. w12x A.F. DeNicola, M. Clayman, R.M. Johnstone, Hormonal control of ascorbic acid transport in rat adrenal glands, Endocrinology 82 Ž1968. 436–446. w13x W.S. Evans, B.J. Boykin, D.L. Kaiser, J.L.C. Borges, M.O. Thorner, Biphasic luteinizing hormone secretion in response to gonadotropinreleasing hormone during continuous perfusion of dispersed rat anterior pituitary cells: changes in total release and the phasic components during the estrous cycle, Endocrinology 112 Ž1983. 535–542. w14x M. Falkeborn, I. Persson, A. Terent, H.O. Adami, H. Lithell, R. Bergstrom, Hormone replacement therapy and the risk of stroke. Follow-up of a population-based cohort in Sweden, Arch. Intern. Med. 153 Ž1993. 1201–1209. w15x D.C. Ferris, J. Kume-Kick, I. Russo-Menna, M.E. Rice, Gender differences in cerebral ascorbate levels and ascorbate loss in ischemia, NeuroReport 6 Ž1995. 1485–1489. w16x F.F. Finucane, J.H. Madans, T.L. Bush, P.H. Wolf, J.C. Kleinman, Decreased risk of stroke among postmenopausal hormone users, Arch. Intern. Med. 153 Ž1993. 73–79. w17x M.E. Freeman, The ovarian cycle of the rat, in: E. Knobil, J.D. Neill ŽEds.., Physiology of Reproduction, Vol. 2, Raven Press, New York, 1988, pp. 1893–1928. w18x R.L. Goodman, A quantitative analysis of the physiological role of estradiol and progesterone in the control of tonic and surge secretion of luteinizing hormone in the rat, Endocrinology 102 Ž1978. 142– 150. w19x E. Gould, C.S. Woolley, M. Frankfurt, B.S. McEwen, Gonadal steroids regulate dendritic spine density in hippocampal pyramidal cells in adulthood, J. Neurosci. 10 Ž1990. 1286–1291. w20x F. Grodstein, M.J. Stampfer, G.A. Colditz, W.C. Willett, J.E. Manson, M. Joffe, B. Rosner, C. Fuchs, S.E. Hankinson, D.J. Hunter, C.H. Hennekens, F.E. Speizer, Postmenopausal hormone therapy and mortality, New Engl. J. Med. 336 Ž1997. 1769–1775. w21x R.A. Grunewald, Ascorbic acid in the brain, Brain Res. Rev. 18 Ž1993. 123–133. w22x E.D. Hall, K.E. Pazara, K.L. Linseman, Sex differences in postischemic neuronal necrosis in gerbils, J. Cereb. Blood Flow Metab. 11 Ž1991. 292–298. w23x B. Halliwell, How to characterize a biological antioxidant, Free Radic. Res. Commun. 9 Ž1990. 1–32. w24x S.C. Hong, Y. Goto, G. Lanzino, S. Soleau, N.F. Kassell, K.S. Lee, Neuroprotection with a calpain inhibitor in a model of focal cerebral ischemia, Stroke 25 Ž1994. 663–669. w25x K.J. Jones, Gonadal steroids as promoting factors in axonal regeneration, Brain Res. Bull. 30 Ž1993. 491–498. w26x K. Kujawa, K.J. Jones, Testosterone-induced acceleration of recov-
J. Kume-Kick, M.E. Rice r Brain Research 803 (1998) 105–113
w27x
w28x
w29x
w30x
w31x w32x w33x
w34x
w35x w36x
w37x
w38x
w39x
w40x
w41x
ery from facial paralysis in male hamsters: temporal requirements of hormone exposure, Physiol. Behav. 48 Ž1990. 765–768. J. Kume-Kick, F.L. Strand, Sex hormones affect muscle contractility and motor functional recovery following peroneal nerve crush, Exp. Neurol. 128 Ž1994. 115–123. J. Kume-Kick, D.C. Ferris, I. Russo-Menna, M.E. Rice, Enhanced oxidative stress in female rat brain after gonadectomy, Brain Res. 738 Ž1996. 8–14. P. Lyrer, H. Landolt, A. Kabersh, H. Langemann, H. Kaeser, Levels of low molecular weight scavengers in the rat brain during focal ischemia, Brain Res. 567 Ž1991. 317–320. M.Y. McGinnis, B. Parsons, T.C. Rainbow, L.C. Krey, B.S. McEwen, Temporal relationship between cell nuclear progestin receptor levels and sexual receptivity following intravenous progesterone administration, Brain Res. 218 Ž1981. 365–371. A. Meister, On the antioxidant effects of ascorbic acid and glutathione, Biochem. Pharmacol. 44 Ž1992. 1905–1915. K.H. Milby, A. Oke, R.N. Adams, Detailed mapping of ascorbate distribution in rat brain, Neurosci. Lett. 28 Ž1982. 15–20. L.E. Moore, R.J. Traystman, Role of oxygen free radicals and lipid peroxidation in cerebral reperfusion injury, Adv. Pharmacol. 31 Ž1994. 565–576. M. Nakano, K. Sugioka, I. Naito, S. Takekoshi, E. Niki, Novel and potent biological antioxidants on membrane phospholipid peroxidation: 2-hydroxy estrone and 2-hydroxy estradiol, Biochem. Biophys. Res. Commun. 142 Ž1987. 919–924. E. Niki, Vitamin C as an antioxidant, World Rev. Nutr. Dietet. 64 Ž1991. 1–30. M. Nishikimi, Oxidation of ascorbic acid with superoxide anion generated by the xanthine–xanthine oxidase system, Biochem. Biophys. Res. Commun. 63 Ž1975. 463–468. S.R. Ojeda, H.F. Urbanski, Puberty in the rat, in: E. Knobil, J.D. Neill ŽEds.., The Physiology of Reproduction, Vol. 2, Raven Press, New York, 1988, pp. 1699–1737. S. Pajovic, G. Nikezic, J.V. Martinovic, Effects of ovarian steroids on superoxide dismutase activity in the rat brain, Experientia 49 Ž1993. 73–75. T.J. Reimers, S.V. Lamb, S.A. Bartlett, R.A. Matamoros, R.B. Cowan, J.S. Engle, Effects of hemolysis and storage on quantification of hormones in blood samples from dogs, cattle and horses, Am. J. Vet. Res. 52 Ž1991. 1075–1079. M.E. Rice, E.J.K. Lee, Y. Choy, High levels of ascorbic acid, not glutathione, in the CNS of anoxia-tolerant reptiles contrasted with levels in anoxia-intolerant species, J. Neurochem. 64 Ž1995. 1790– 1799. M.E. Rice, I. Russo-Menna, Differential compartmentalization of
w42x
w43x
w44x w45x
w46x w47x
w48x
w49x
w50x
w51x w52x
w53x
w54x
w55x
113
brain ascorbate and glutathione between neurons and glia, Neuroscience 82 Ž1998. 1213–1223. R.L. Roof, R. Duvdevani, D.G. Stein, Gender influences outcome of brain injury: progesterone plays a protective role, Brain Res. 607 Ž1993. 333–336. B.R. Shivakuma, H.K. Anandatheerthavarada, V. Ravindranath, Free radical scavenging systems in developing rat brain, Int. J. Dev. Neurosci. 9 Ž1991. 181–185. H. Sies, Strategies of antioxidant defense, Eur. J. Biochem. 215 Ž1993. 213–219. M.S. Smith, M.E. Freeman, J.D. Neill, The control of progesterone secretion during the estrous cycle and early pseudopregnancy in the rat: prolactin, gonadotropin and steroid levels associated with rescue of the corpus luteum of pseudopregnancy, Endocrinology 96 Ž1975. 219–225. R. Spector, A.V. Lorenzo, Ascorbic acid homeostasis in the central nervous system, Am. J. Physiol. 225 Ž1973. 757–763. D. Taskiran, F.Z. Kutay, E. Sozmen, S. Pogun, Sex differences in nitriternitrate levels and antioxidant defense in rat brain, NeuroReport 8 Ž1997. 881–884. C.M. Vanin, N.J. MacLusky, M.D. Grynpas, R.F. Casper, The effect of three hormone replacement regimens on bone density in the aged ovariectomized rat, Fertility and Sterility 63 Ž1995. 643–651. G.T. Vatassery, W.E. Smith, H.T. Quach, Ascorbic acid, glutathione and synthetic antioxidants prevent the oxidation of vitamin E in platelets, Lipids 24 Ž1989. 1043–1047. F.S. vom Saal, C.E. Finch, Reproductive senescence: Phenomena and mechanisms in mammals and selected vertebrates, in: E. Knobil, J.D. Neill ŽEd.., The Physiology of Reproduction, Vol. 2, Raven Press, New York, 1988, pp. 2351–2413. H.G. Waynforth, Experimental and Surgical Technique in the Rat, Academic Press, London, 1988, pp. 160–163. C. Woolley, N. Weiland, B. McEwen, P. Schwartzkroin, Estradiol increases the sensitivity of hippocampal CA1 pyramidal cell to NMDA receptor-mediated synaptic input: correlation with dendritic spine density, J. Neurosci. 17 Ž1997. 1848–1859. C.S. Woolley, E. Gould, M. Frankfurt, B.S. McEwen, Naturally occurring fluctuation in dendritic spine density on adult hippocampal pyramidal neurons, J. Neurosci. 10 Ž1990. 4035–4039. W.H. Yu, Administration of testosterone attenuates neuronal loss following axotomy in the brain-stem motor nuclei of female rats, J. Neurosci. 9 Ž1989. 3908–3914. Y. Zhang, J. Shi, G. Rajakumar, A.L. Day, J.W. Simpkins, Effects of gender and estradiol treatment on focal brain ischemia, Brain Res. 784 Ž1998. 321–324.