Subcellular Localization of LH Releasing Activity in the Rat Hypothalamus * CAROL A. TABER AND HARRY J. KARAVOLASf Department of Physiological Chemistry, Endocrinology-Reproductive Physiology Program, and Waisman Center on Mental Retardation, University of Wisconsin, Madison, Wisconsin 53706 synaptosomes, or mitochondria, caused an increase in LH release and total LH while pituitary LH remained unchanged. This fraction appears to be predominantly composed of electron dense vesicles of varying sizes. Also of interest is another fraction which decreased LH release and was again not associated with myelin, synaptosomes, or mitochondria. The decreased release was associated with an increased pituitary LH while total LH remained unchanged. Of the six subcellular fractions obtained from the "crude mitochondrial pellet" only these two caused significant changes in LH release. Two other neural tissues, the cerebellum and cerebral cortex, were similarly processed. None of these subcellular fractions significantly altered LH release and/or synthesis. (Endocrinology 96: 446, 1975)
ABSTRACT. Separation of participate fractions associated with LH releasing activity has been effected by differential centrifugation followed by sucrose density gradient centrifugation. The fractions were monitored for LH releasing activity by an in vitro incubation with rat pituitaries and for subcellular organelles by electron microscopy. LH released into the incubation medium and contained in the pituitaries was measured by radioimmunoassay (RIA). After differential centrifugation only the "crude mitochondrial fraction" caused an increase of LH release accompanied by a depletion of pituitary LH and a rise in total LH. Further fractionation of this pellet on a discontinuous sucrose gradient resulted in three opaque bands and three clear areas which were removed as six fractions. Only one fraction, not associated with myelin,
T
HAT the hypothalamus contains a releasing factor or hormone for LH (LRF) has been well established (1-6). Moreover, its chemical structure has been identified and the polypeptide LRF synthesized (7-9). However, the subcellular site(s) of storage and biosynthesis of LRF have not been completely elucidated. In 1970, Ishii (10), using differential centrifugation and sucrose density centrifugation techniques, localized the LH releasing activity of equine median eminence to a subcellular particulate fraction containing electron dense vesicles. In the same year, Clementi et al. (11) using similar techniques localized the FSH and GH releasing activities of the male rat median eminence to a subcellular particulate fraction containing electron dense vesicles and dopamine.
Received May 17, 1974. * This investigation was supported by U.S.P.H.S. Grant No. 5-T01-HD-00104-08, 1R01-HD-05414-02 and 2-P01-HD-03352-06 from NICHD and a Ford Foundation Grant No. 630-0505A. f NICHD Research Career Development Awardee No. l-K4-HD-70,006-02.
The assays used for FRF and GRF activities were in vivo pituitary depletion methods, while Ishii (10) used ovarian ascorbic acid depletion as a method of determining LRF activity. Earlier work done in this laboratory (12) utilized differential centrifugation, Nucleopore filtration, and Sepharose 2B gel filtration to localize a subcellular particulate fraction which caused increased synthesis of ovulating hormone1 (OH) in an in vitro pituitary organ culture. The hypothalamic stimulator of LH synthesis was associated with the 1-15,000 x g particulates and was not retained by a 0.5 (x (diameter pore size) Nucleopore filter. Sepharose 2B gel filtration showed two zones of OH synthesizing activity separated by a distinct area which caused inhibition of OH synthesis. Similar data were obtained using either the ovulating hormone assay or the ovarian ascorbic acid depletion assay. 1
The term ovulating hormone (OH) is used to represent the gonadotropin(s) necessary for the occurrence of ovulation (14).
446
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 20 November 2015. at 03:32 For personal use only. No other uses without permission. . All rights reserved.
SUBCELLULAR LRF LOCALIZATION IN RAT MBH
Johansson et al. (13) in 1973 using differential centrifugation followed by Ficoll gradient centrifugation localized the FSH, LH, GH, and TSH releasing activities of procine hypothalami to three subcellular particulate fractions which contained mitochondria. The releasing activities could be further separated into distinct activities for FSH and LH, GH, and TSH by partition chromatography on Sephadex G-25. The present study was done to localize the subcellular site(s) of the LH releasing activity in the rat medial basal hypothalamus (MBH) and to identify by electron microscopy the subcellular organelles and/or par-. ticulates associated with LH releasing activity. Methods used for detecting the subcellular localization were differential centrifugation and sucrose density gradient centrifugation rather than Sepharose 2B gel filtration, and an LH specific radioimmunoassay rather than the ovulating hormone assay used previously (12). These and similar subcellular localization experiments should be helpful in elucidating the scheme of molecular events in the feedback control of LRF release and/or synthesis. Materials and Methods Tissue preparation. Thirty nonestrous, 70-dayold female rats, obtained from Holtzman Co., Madison, Wisconsin, were used as donors of medial basal hypothalami (MBH) for each of three experiments. Nonestrous rats were used since previous workers (14,15) found the lowest LH releasing activity and stimulation of OH synthesis during estrus. The rats were sacrificed between 7:00 and 9:00 AM CST and the medial basal hypothalami removed as previously reported (14). The tissues were then weighed and homogenized in a 0.32 M sucrose solution (10% w/v) at 4 C by hand using fifteen up and down movements of a Thomas glass homogenizer with a teflon pestle. Subcellular fractionation (Fig. 1). Three replicate subcellular fractionations were done. The procedure of Gray and Whittaker (16) was used to obtain the crude nuclei and three primary fractions (P2 = crude mitochondrial fraction, P3 = microsomal fraction, and S = supernatant).
447
The crude nuclear pellet was separated into two fractions, which we also term primary fractions, by the procedure of L0vtrup-Rein and McEwen (17) (Pj = cellular debris fraction and P-1A = nuclear fraction). The nuclear fraction was collected as a clear pellet at the bottom of the tube while the cellular debris fraction formed a pink beige band at the interface of the 0.32M and 2.39M sucrose. Further fractionation of P2, the crude mitochondrial fraction, on a discontinuous sucrose gradient (16) resulted in three opaque and three clear zones which were separated into six fractions and referred to as secondary fractions P 2 -l, P2-2, P2-3, P2-4, P2-5, and P2-6. The secondary fractions were collected by means of a Luer-lok syringe with a bent needle and diluted to 0.32M sucrose. They were concentrated by centrifugation at 100,000 x g for 1 hr and then resuspended in 1.0 ml of 0.32M sucrose prior to assay, as were all other fractions. Portions of each of the primary and secondary fractions were used for electron microscopy and determination of LH content. The remainder of each fraction was analyzed for LH releasing activity by the assays described below, after disrupting the particulates by grinding with a low clearance McShan homogenizer at 4 C for 30 sec. Failure to grind the subcellular particulates, especially fraction P2-5, consistently yielded low LH releasing activity. In vitro assays for LH releasing activity. The procedure developed by Mittler and Meites (6) was used for the in vitro assay of LH releasing activity with the exception that female Holtzman rats, 26 days of age, were used as pituitary donors rather than adult male Sprague-Dawley rats. Kragt and Dahlgren (18,19) showed that pituitary content of FSH and LH was 4-5 times higher in rats at 20 days of age than at or after puberty. For each incubation, opposite halves of four anterior pituitaries were placed in two separate 10 ml Erlenmeyer flasks containing 1.5 ml of medium 199 (Microbiological Associates, Bethesda, Maryland). One flask was designated the test flask; the other flask containing the four opposing halves was designated the control flask. Preincubations were carried out in a Dubnoff metabolic shaker (60 cpm) at 37 C in an atmosphere of 95% O 2 -5% CO2 for 30 min. At the end of this time the medium 199 was removed from each flask. In the case of the test flasks
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 20 November 2015. at 03:32 For personal use only. No other uses without permission. . All rights reserved.
448
TABER AND KARAVOLAS
Endo • 1975 Vol 96 • No 2
Subcellular fractionation of the rat medial basal hypothalamus tissue homogenate (10% w/v) spin at lOOOxg for 10 min. wash, resuspend, and recentrifuge 2X
I crude nuclear fraction
combined supernatants spin at 17,000xg for 60 min.
resuspend in 1.0 ml 0.32 M sucrose, layer on 2.5 ml 0.32 M sucrose mixed with 13.5 ml 2.39 M sucrose; spin at 53,500xg 45 min. P-1A purified nuclei
P-l cell debris
supernatant
P-2 crude mitochondria
spin at 100,000xg for 90 min.
resuspend in 1.0 ml 0.32 M sucrose; layer on a discontinuous gradient of 6.S ml 0.8 M sucrose and 6.4 ml 1.2 M sucrose; spin at 100,000xg for 120 min.
FIG. 1. Subcellular fractionation scheme. For details, see Materials and Methods section.
P-3 microsomes S-supernatant
P2-l
0.32 M sucrose
P,-?
white band between 0.32 and 0.8 M sucrose
P,-3
0.8 M sucrose
P--4
brown band between 0.8 and 1.2 M sucrose
P,-5
1.2 M sucrose
P.-6
pale beige pellet below 1.2 M sucrose
the original medium was replaced with 1.0 ml of fresh medium 199 and 0.5 ml of the homogenized subcellular fraction. The corresponding control flasks received 1.0 ml medium 199 and 0.5 ml 0.32M sucrose. Each set of incubations for the three subcellular fractionations also included a pair of incubation flasks which were both control flasks. Incubation was carried out for 6 hr and terminated by placing the flasks on ice. The medium was removed and quick-frozen in small aliquots for subsequent radioimmunoassay (RIA) of LH. For the determination of LH content in the pituitaries after incubation, the four halves were suspended in 1.5 ml of fresh medium 199, and homogenized well with a McShan all glass tissue grinder. The homogenate was centrifuged at 20,000 x g for 30 min yielding a supernatant containing the pituitary LH (W. H. McShan and K. W. Thompson, unpublished). The supernatant fluid was then quick-frozen in small aliquots for subsequent RIA of LH.
The stimulation of LH release in vitro by subcellular fractions P2 and P2-5 (appropriately diluted with either 0.32M sucrose or medium 199) was linear over a range of 0.1 ml to 1 ml. The Mittler and Meites assay procedure (6), as modified above, also produced a proportional LH release with 5 ng and 10 ng doses of synthetic LRF (Spectrum Medical Industries, Los Angeles, California) added in 0.5 ml of 0.32M sucrose and 1 ml medium 199. Radioimmunoassay ofLH. The LH in the pituitary cultures was measured by a double antibody RIA developed by Beamer et al. (20). LH antibodies, LH for iodination, and standard reference preparations for LH, FSH, and TSH were obtained from the National Institute for Arthritis and Metabolic Diseases. The second antibody (goat anti-rabbit gamma globulin) was purchased from Antibodies Inc., Davis, California. Radioiodine 131 was purchased from Union Carbide, Tuxedo, New York, and used for
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 20 November 2015. at 03:32 For personal use only. No other uses without permission. . All rights reserved.
SUBCELLULAR LRF LOCALIZATION IN RAT MBH iodination of LH according to the method of Greenwood (21). The validity of the RIA was ascertained according to criteria established by Vivian and Labella (22). The procedure of Beamer et al. (20) utilizes a sodium barbital buffer system, pH 8.6, for preparation of standards, extracts and subsequent dilutions. Before RIA of the experimental samples, we demonstrated that standard and pituitary LH preparations in medium 199 diluted with the barbital buffer exhibited dose response curves which were parallel to curves given by these preparations in the barbital buffer system of Beamer et al. (20). The linear range of the standard curve (percent bound vs. log dose) extended from 5 ng to 320 ng. The cross reactivity of the standard reference preparations of FSH and TSH in the LH assay were tested and found to be less than 3%. Each assay was assessed by a method of statistical control developed by Rodbard et al. (23). The cumulative within assay variance was 2.1% and the overall between assay variance was 2.3%. 95% confidence limits were computed for each standard curve (24) and extreme outliers were discarded (22). All test and control pituitary and medium preparations for each of the three sets of incubations were assayed in triplicate at two dilutions in three separate radioimmunoassays. The LH values of the incubation media and pituitaries were computed as ng/mg pituitary. The means (±SEM) of the 18 values for each test and control sample were calculated and the LH releasing activity in each fraction was expressed as a percent of the control. The percent-of-control values for each subcellular fraction from the three experiments were averaged and the standard error of the mean determined. A two-way ANOVA was done (24) with the experiment as one variable and the subcellular fraction as the second variable. Orthogonal comparisons were made between groups to test for significant difference (24) and all groups were compared to the double control group. Electron microscopy. A portion of each subcellular fraction obtained in Fig. 1 was prepared for electron microscopy according to the Method of Hayat (25). Glutaraldehyde and osmium tetroxide were used for fixation and postfixation. The pellets were embedded in Epon-Araldite (Electron Microscopy Sciences, Fort Washington,
449
Pennsylvania) in such a way that sections would contain a cross section of the pellet from top to bottom. The blocks were sectioned on a Reichert thermal advance microtome and collected on Formvar coated copper grids. Some of the sections were stained with uranyl acetate and lead citrate for better contrast. Sections were viewed with a Hitachi HU12 electron microscope at magnifications from 2-30,000. Results LH releasing activity of the primary subcellular fractions. The five primary fractions were assayed for their ability to increase LH release from cultured anterior pituitaries. P2, the crude mitochondrial fraction, was the only fraction which caused a highly significant increase in LH release (Fig. 2a). This increased release was also accompanied by a highly significant depletion of pituitary LH (Fig. 2b) and a significant rise in the total LH in the incubation (i.e., LH in medium plus pituitaries) (Fig. 2c). None of the other four fractions caused a significant change in LH release into the medium or in pituitary LH content. LH releasing activity of the secondary subcellular fractions. The six secondary fractions obtained by sucrose density centrifugation of P2 were also assayed. Fraction P2-5 caused a highly significant increase in LH release into the medium (Fig. 2d) and a highly significant rise in total LH (Fig. 2f). However, this was not accompanied by a depletion of pituitary LH (Fig. 2e). Of interest are the results found with fraction P 2 -l, indicating a significant decrease in LH release into the medium (Fig. 2a). This was accompanied by a significant increase in pituitary LH content (Fig. 2e) but no change in total LH content (Fig. 2f) compared to control. There were no significant differences between the other fractions. LH releasing activity of other neural tissues. Two other neural tissues, the cerebellum
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 20 November 2015. at 03:32 For personal use only. No other uses without permission. . All rights reserved.
TABER AND KARAVOLAS
450
d) LH released into medium
a) LH released into medium
300
c
control
200
n
e) pituitary LH
b) pituitary LH
s?
n
rT
3 200 •o u «o to
1
1—=1
S 100 K •1 5
**
f) total LH
c) total LH
200 i
*|
i
,
100 P1
P1A P2
P3
S
LI
hi
P2-1 P2-2 P2-3 P2-4 P2-5 P2-6 C
MBH PRIMARY
MBH
SECONDARY
SUBCELLULAR FRACTIONS
SUBCELLULAR
FRACTIONS
En do i 1975 Vo! 96 < No 2
below the sensitivity limits of our RIA. The LH content of the various MBH subcellular fractions ranged from 3-24 ng per 0.5 ml. Thus, the LH present in the 0.5 ml aliquot of the MBH subcellular fractions added to the incubations, represents less than .015% of the LH measured in the incubation medium. Electron microscopy. The fractions were monitored by electron microscopy and the primary fractions, consistent with expected results (16,17), contained cellular debris, nuclei, mitochondria, microsomes, and free ribosomes respectively. P2, the crude mitochondrial fraction, also contained myelin fragments, synaptosomes, and scattered electron dense vesicles. After further centrifugation of this fraction on a discontinuous sucrose density gradient, the particulates were somewhat separated. Fraction P2-2 contained mainly myelin fragments. Fraction P2-3 consisted of myelin
FIG. 2. LH releasing activity of the subcellular fractions of the MBH. The designations on the abscissa, except for C (double control group), refer to the different subcellular fractions derived from the scheme shown in Fig. 1. Since ANOVA indicated no significant difference between the three experiments, the data were combined and the results shown represent the mean percent of control values (see text for details). Standard errors of the mean (SEM) are not shown on the bar graphs since in no case did the SEM exceed 0.5%. The shaded areas on the bar graphs for the double control groups (C) indicate the mean differences in LH release or content between the two control flasks. **Highly significant p < 0.01. *Significant p < 0.05.
and the cerebral cortex, were examined to determine the tissue specificity of the LH releasing activity. These tissues were fractionated and processed as described above for the MBH. None of the fractions significantly altered pituitary LH content or release. LH content of the subcellular fractions. The subcellular fractions of each of the three tissues were examined for endogenous LH content. The LH values of the fractions from the cerebellum and cerebral cortex were
FIG. 3. A representative electron micrograph of fraction P2-5 derived from the crude mitochondria! fraction (P2). As shown, electron dense vesicles of at least two Size ranges (A and B) are the predominant subcellular organelle. No mitochondria, synaptosomes, or myelin were detected in any of the EM sections of this fraction, (x 19,000).
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 20 November 2015. at 03:32 For personal use only. No other uses without permission. . All rights reserved.
SUBCELLULAR LRF LOCALIZATION IN RAT MBH fragments and synaptosomes. Synaptosomes of varying structural integrity dominated fraction P2-4 while P2-5 contained predominantly electron dense vesicles of various sizes (Fig. 3). Fraction P2-6 was composed of mitochondria with rough endoplasmic reticulum. Fraction P2-l did not yield a firm enough specimen to section. Negative staining of the fraction with 1% phosphotungstic acid showed small filamentous structures which could be membrane fragments. Discussion In agreement with our earlier work (12), the crude mitochondrial fraction obtained by differential centrifugation (P2) contained LH releasing activity (Fig. 2a). When monitored by electron microscopy, this fraction contained myelin, synaptosomes, mitochondria, and electron dense vesicles. These particulates were partially separated by further centrifugation on a discontinuous sucrose density gradient. Fraction P2-5 from this procedure showed an increased ability to release LH (Fig. 2d). This fraction contained predominantly electron dense vesicles (Fig. 3) in accordance with our previous work (12) and that of Ishii (10) and Clementi (11). Since the fraction contains electron dense vesicles of various sizes, we would anticipate that some would be associated with other releasing factors and/or biogenic amines. We have not yet assayed the fractions for these compounds. As shown in Fig. 2f, the total LH in the incubation as well as the LH released into the culture medium was increased. This suggests the possibility that fraction P2-5 which stimulates LH release from the pituitary may also increase pituitary LH synthesis (26,27). Unlike the results with fraction P2, no significant depletion of pituitary LH was observed with fraction P2-5. At this time we have no explanation for this difference. The failure of the cerebral cortical and cerebellar fractions to stimulate LH release from cultured pituitaries shows that the stimulation found with the MBH is not a general property of nervous tissue. Hence,
451
we assume that the release of LH from the MBH subcellular preparations was probably due wholly or in part to specific hypothalamic LH releasing factors such as the decapeptide, LRF, and not to general cellular constituents such as cAMP, prostaglandins, etc. The concentrations of cAMP (28) and prostaglandins (29) in these three brain tissues have been reported to be of the same order of magnitude (for each compound). Presumably the above and other cellular constituents common to these tissues, would have manifested themselves equally in their ability to cause release of LH whether in cerebellar, cerebral cortical, or hypothalamic tissues. Furthermore, the brain subcellular distribution profiles of cAMP (30,31) and prostaglandins (32) are different from our observed MBH subcellular localization of LH releasing activity. The fractionation of P2 by a sucrose density gradient also resulted in a fraction (P2-l) collected in 0.32M sucrose which significantly inhibited LH release from cultured anterior pituitaries. This agrees with our earlier, work (12) although the previous method used for separating the particulates in P2 was Sepharose 2B gel filtration. The particulate nature of this fraction is not known although the presence of small membrane fragments was detected by negative staining. The presence of both releasing and inhibitory activity in the crude mitochondrial fraction, P2, is consistent with the fact that fraction P2-5 had greater stimulatory activity than fraction P2. It would be premature at this juncture to deduce the existence of a hypothalamic LH release-inhibitory substance from these preliminary data. Several alternative explanations could be proffered on the basis of these observations. Until further investigations reveal the nature and significance of this inhibitory activity, suffice it to say that it represents an interesting finding in our investigation of the subcellular localization of LH releasing activity which warrants further investigation. The subcellular localization of LH releasing activity ("LRF") to a particulate fraction
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 20 November 2015. at 03:32 For personal use only. No other uses without permission. . All rights reserved.
TABER AND KARAVOLAS
452
(P2-5) containing electron dense vesicles may indicate that this and other hypophysiotropins are contained in storage vesicles or granules analogous to the storage granules of the anterior and posterior pituitary hormones. Other investigators (10,11) have found similar results.2 Further studies of these particulates is needed to resolve whether: 1) the LH releasing activity is contained in or adsorbed onto the electron dense vesicles which constitute the predominant subcellular organelle in P2-5; 2) separate populations (e.g., size) of the electron dense vesicles contain different hypophysiotropins and/or biogenic amines; and 3) different areas or cell types contain specific hormone or biogenic amine-containing electron dense vesicles. At any rate, the localization of this activity to a particulate fraction rich in electron dense vesicles separate from synaptosomes, nuclei, mitochondria, cell membranes, etc., should facilitate investigation into the sequential scheme of molecular events in the control (feedback) of LRF release and/or synthesis and possibly serve as a model for the subcellular localization of other hypophysiotropins.
8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23.
References 1. Arimura, A., A. V. Schally, T. Saito, E. Mulder, and C. Y. Bowers, Endocrinology 80: 515, 1967. 2. Clemens, J., and C. J. Shaar, Proc Int Soc Psychoneuroendocrinology, Brooklyn 1970, p. 356. 3. Crighton, D. B., H. P. G. Schneider, and S. M. McCann, Endocrinology 87: 323, 1970. 4. Jackson, G. L., K. R. Stevens, and A. V. Nalbandov, Endocrinology 83: 234, 1968. 5. McCann, S. M., S. Taleisnik, and H. M. Friedman, Proc Soc Exp Biol Med 104: 432, 1960. 6. Mittler, J. C , and J. Meites, Endocrinology 88: 500, 1966. 7. Matsuo, H., Y. Baba, R. M. G. Nair, A. Arimura,
24.
25.
26. 27. 28. 29. 30. 31.
2
Also, since the submission of this paper, Pelletier et al. (33) using an immunohistochemical technique at the electron microscopic level, have observed that LRF is contained in the secretory granules of some nerve endings in the palisade layer of the median eminence.
32. 33.
Endo 1975 Vol 96 . No 2
and A. V. Schally, Biochem Biophys Res Commun 43: 1334, 1971. Baba, Y., H. Matsuo, and A. V. Schally, Biochem Biophys Res Commun 44: 459, 1971. Arimura, A., H. Matsuo, Y. Baba, L. Debeljuk, J.. Sandow, and A. V. Schally,Endocrinology 90: 163, 1972. Ishii, S., Endocrinology 86: 207, 1970. Clementi, F., B. Ceccarelli, M. Ceratie, L. Demonte, M. Felici, M. Motta, and A. Pecile,/ Endocrinol 48: 205, 1970. Karavolas, H. J., R. K. Meyer, K. J. Deighton, and S. Adrouny, Endocrinology 88: 969, 1971. Johansson, K. N. G., B. L. Currie, K. Folkers and C. Y. Bowers, Biochem Biophys Res Commun 52: 967, 1973. Deighton, K. J., and R. K. Meyer, Endocrinology 84: 778, 1969. Ramirez, S., V. Domingo, and' C. H. Sawyer, Endocrinology 76: 282, 1965. Gray, E. G., and V. P. Whittaker, Anatomy 96: 79, 1962. L^vtrup-Rein, H., and B. S. McEwen, J Cell Biol 30: 405, 1966. Kragt, C. L., and J. Dahlgren, Neuroendocrinology 9: 30, 1972. Spona, J., and O. Luger, FEBS Lett 32: 52, 1973. Beamer, W. G., S. M. Murr, and I. I. Geschwind, Endocrinology 90: 823, 1972. Greenwood, F. C , W. M. Hunter, and T. S. Glover, Biochem J 89: 114, 1963. Vivian, S. R., and F. S. Labella,/ Clin Endocrinol Metab 33: 225, 1971. Rodbard, D., P. L. Rayford, J. A. Cooper, and G. T. Ross,/ Clin Endocrinol Metab 28: 1442, 1968. Snedecor, G. W., and W. G. Cochran, Statistical Methods, ed. 6, The Iowa State University Press, Ames, Iowa, 1967, p. 267. Hayat, M. A., Principles and Techniques of Electron Microscopy, Van Nostrand Reinhold Co., New York, 1970, p. 35. Redding, T. W., A. V. Schally, A. Arimura, and H. Matsuo, Endocrinology 90: 764, 1972. Serra, G. B., and A. R. Midgely, Endocrinology 91: 962, 1972. Schmidt, M. J., D. E. Schmidt, and G. A. Robison Science 173: 1142, 1971. von Euler, U. S., and R. Eliasson, Prostaglandins, Academic Press, New York, 1967. Lust, W. D., and N. D. Goldberg, Pharmacologist 12: 290, 1970. Goldberg, N. D , W. D. Lust, R. F. O'Dea, S. Wei, and A. G. O'Toole. In Greengard, P., and E. Costa, Role of Cyclic AMP in Cell Function, Raven Press, New York, 1970, p. 67. Kataoka, K., P. W. Ramwell, and S. Jessup, Science 157: 1187, 1967. Pelletier, G., F. Labrie, R. Purrani, A. Arimura, and A. V. Schally, Endocrinology 95: 314, 1974.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 20 November 2015. at 03:32 For personal use only. No other uses without permission. . All rights reserved.
ABSTRACT. Separation of participate fractions associated with LH releasing activity has been effected by differential centrifugation followed by sucrose density gradient centrifugation. The fractions were monitored for LH releasing activity by an in vitro incubation with rat pituitaries and for subcellular organelles by electron microscopy. LH released into the incubation medium and contained in the pituitaries was measured by radioimmunoassay (RIA). After differential centrifugation only the "crude mitochondrial fraction" caused an increase of LH release accompanied by a depletion of pituitary LH and a rise in total LH. Further fractionation of this pellet on a discontinuous sucrose gradient resulted in three opaque bands and three clear areas which were removed as six fractions. Only one fraction, not associated with myelin,
T
HAT the hypothalamus contains a releasing factor or hormone for LH (LRF) has been well established (1-6). Moreover, its chemical structure has been identified and the polypeptide LRF synthesized (7-9). However, the subcellular site(s) of storage and biosynthesis of LRF have not been completely elucidated. In 1970, Ishii (10), using differential centrifugation and sucrose density centrifugation techniques, localized the LH releasing activity of equine median eminence to a subcellular particulate fraction containing electron dense vesicles. In the same year, Clementi et al. (11) using similar techniques localized the FSH and GH releasing activities of the male rat median eminence to a subcellular particulate fraction containing electron dense vesicles and dopamine.
Received May 17, 1974. * This investigation was supported by U.S.P.H.S. Grant No. 5-T01-HD-00104-08, 1R01-HD-05414-02 and 2-P01-HD-03352-06 from NICHD and a Ford Foundation Grant No. 630-0505A. f NICHD Research Career Development Awardee No. l-K4-HD-70,006-02.
The assays used for FRF and GRF activities were in vivo pituitary depletion methods, while Ishii (10) used ovarian ascorbic acid depletion as a method of determining LRF activity. Earlier work done in this laboratory (12) utilized differential centrifugation, Nucleopore filtration, and Sepharose 2B gel filtration to localize a subcellular particulate fraction which caused increased synthesis of ovulating hormone1 (OH) in an in vitro pituitary organ culture. The hypothalamic stimulator of LH synthesis was associated with the 1-15,000 x g particulates and was not retained by a 0.5 (x (diameter pore size) Nucleopore filter. Sepharose 2B gel filtration showed two zones of OH synthesizing activity separated by a distinct area which caused inhibition of OH synthesis. Similar data were obtained using either the ovulating hormone assay or the ovarian ascorbic acid depletion assay. 1
The term ovulating hormone (OH) is used to represent the gonadotropin(s) necessary for the occurrence of ovulation (14).
446
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 20 November 2015. at 03:32 For personal use only. No other uses without permission. . All rights reserved.
SUBCELLULAR LRF LOCALIZATION IN RAT MBH
Johansson et al. (13) in 1973 using differential centrifugation followed by Ficoll gradient centrifugation localized the FSH, LH, GH, and TSH releasing activities of procine hypothalami to three subcellular particulate fractions which contained mitochondria. The releasing activities could be further separated into distinct activities for FSH and LH, GH, and TSH by partition chromatography on Sephadex G-25. The present study was done to localize the subcellular site(s) of the LH releasing activity in the rat medial basal hypothalamus (MBH) and to identify by electron microscopy the subcellular organelles and/or par-. ticulates associated with LH releasing activity. Methods used for detecting the subcellular localization were differential centrifugation and sucrose density gradient centrifugation rather than Sepharose 2B gel filtration, and an LH specific radioimmunoassay rather than the ovulating hormone assay used previously (12). These and similar subcellular localization experiments should be helpful in elucidating the scheme of molecular events in the feedback control of LRF release and/or synthesis. Materials and Methods Tissue preparation. Thirty nonestrous, 70-dayold female rats, obtained from Holtzman Co., Madison, Wisconsin, were used as donors of medial basal hypothalami (MBH) for each of three experiments. Nonestrous rats were used since previous workers (14,15) found the lowest LH releasing activity and stimulation of OH synthesis during estrus. The rats were sacrificed between 7:00 and 9:00 AM CST and the medial basal hypothalami removed as previously reported (14). The tissues were then weighed and homogenized in a 0.32 M sucrose solution (10% w/v) at 4 C by hand using fifteen up and down movements of a Thomas glass homogenizer with a teflon pestle. Subcellular fractionation (Fig. 1). Three replicate subcellular fractionations were done. The procedure of Gray and Whittaker (16) was used to obtain the crude nuclei and three primary fractions (P2 = crude mitochondrial fraction, P3 = microsomal fraction, and S = supernatant).
447
The crude nuclear pellet was separated into two fractions, which we also term primary fractions, by the procedure of L0vtrup-Rein and McEwen (17) (Pj = cellular debris fraction and P-1A = nuclear fraction). The nuclear fraction was collected as a clear pellet at the bottom of the tube while the cellular debris fraction formed a pink beige band at the interface of the 0.32M and 2.39M sucrose. Further fractionation of P2, the crude mitochondrial fraction, on a discontinuous sucrose gradient (16) resulted in three opaque and three clear zones which were separated into six fractions and referred to as secondary fractions P 2 -l, P2-2, P2-3, P2-4, P2-5, and P2-6. The secondary fractions were collected by means of a Luer-lok syringe with a bent needle and diluted to 0.32M sucrose. They were concentrated by centrifugation at 100,000 x g for 1 hr and then resuspended in 1.0 ml of 0.32M sucrose prior to assay, as were all other fractions. Portions of each of the primary and secondary fractions were used for electron microscopy and determination of LH content. The remainder of each fraction was analyzed for LH releasing activity by the assays described below, after disrupting the particulates by grinding with a low clearance McShan homogenizer at 4 C for 30 sec. Failure to grind the subcellular particulates, especially fraction P2-5, consistently yielded low LH releasing activity. In vitro assays for LH releasing activity. The procedure developed by Mittler and Meites (6) was used for the in vitro assay of LH releasing activity with the exception that female Holtzman rats, 26 days of age, were used as pituitary donors rather than adult male Sprague-Dawley rats. Kragt and Dahlgren (18,19) showed that pituitary content of FSH and LH was 4-5 times higher in rats at 20 days of age than at or after puberty. For each incubation, opposite halves of four anterior pituitaries were placed in two separate 10 ml Erlenmeyer flasks containing 1.5 ml of medium 199 (Microbiological Associates, Bethesda, Maryland). One flask was designated the test flask; the other flask containing the four opposing halves was designated the control flask. Preincubations were carried out in a Dubnoff metabolic shaker (60 cpm) at 37 C in an atmosphere of 95% O 2 -5% CO2 for 30 min. At the end of this time the medium 199 was removed from each flask. In the case of the test flasks
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 20 November 2015. at 03:32 For personal use only. No other uses without permission. . All rights reserved.
448
TABER AND KARAVOLAS
Endo • 1975 Vol 96 • No 2
Subcellular fractionation of the rat medial basal hypothalamus tissue homogenate (10% w/v) spin at lOOOxg for 10 min. wash, resuspend, and recentrifuge 2X
I crude nuclear fraction
combined supernatants spin at 17,000xg for 60 min.
resuspend in 1.0 ml 0.32 M sucrose, layer on 2.5 ml 0.32 M sucrose mixed with 13.5 ml 2.39 M sucrose; spin at 53,500xg 45 min. P-1A purified nuclei
P-l cell debris
supernatant
P-2 crude mitochondria
spin at 100,000xg for 90 min.
resuspend in 1.0 ml 0.32 M sucrose; layer on a discontinuous gradient of 6.S ml 0.8 M sucrose and 6.4 ml 1.2 M sucrose; spin at 100,000xg for 120 min.
FIG. 1. Subcellular fractionation scheme. For details, see Materials and Methods section.
P-3 microsomes S-supernatant
P2-l
0.32 M sucrose
P,-?
white band between 0.32 and 0.8 M sucrose
P,-3
0.8 M sucrose
P--4
brown band between 0.8 and 1.2 M sucrose
P,-5
1.2 M sucrose
P.-6
pale beige pellet below 1.2 M sucrose
the original medium was replaced with 1.0 ml of fresh medium 199 and 0.5 ml of the homogenized subcellular fraction. The corresponding control flasks received 1.0 ml medium 199 and 0.5 ml 0.32M sucrose. Each set of incubations for the three subcellular fractionations also included a pair of incubation flasks which were both control flasks. Incubation was carried out for 6 hr and terminated by placing the flasks on ice. The medium was removed and quick-frozen in small aliquots for subsequent radioimmunoassay (RIA) of LH. For the determination of LH content in the pituitaries after incubation, the four halves were suspended in 1.5 ml of fresh medium 199, and homogenized well with a McShan all glass tissue grinder. The homogenate was centrifuged at 20,000 x g for 30 min yielding a supernatant containing the pituitary LH (W. H. McShan and K. W. Thompson, unpublished). The supernatant fluid was then quick-frozen in small aliquots for subsequent RIA of LH.
The stimulation of LH release in vitro by subcellular fractions P2 and P2-5 (appropriately diluted with either 0.32M sucrose or medium 199) was linear over a range of 0.1 ml to 1 ml. The Mittler and Meites assay procedure (6), as modified above, also produced a proportional LH release with 5 ng and 10 ng doses of synthetic LRF (Spectrum Medical Industries, Los Angeles, California) added in 0.5 ml of 0.32M sucrose and 1 ml medium 199. Radioimmunoassay ofLH. The LH in the pituitary cultures was measured by a double antibody RIA developed by Beamer et al. (20). LH antibodies, LH for iodination, and standard reference preparations for LH, FSH, and TSH were obtained from the National Institute for Arthritis and Metabolic Diseases. The second antibody (goat anti-rabbit gamma globulin) was purchased from Antibodies Inc., Davis, California. Radioiodine 131 was purchased from Union Carbide, Tuxedo, New York, and used for
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 20 November 2015. at 03:32 For personal use only. No other uses without permission. . All rights reserved.
SUBCELLULAR LRF LOCALIZATION IN RAT MBH iodination of LH according to the method of Greenwood (21). The validity of the RIA was ascertained according to criteria established by Vivian and Labella (22). The procedure of Beamer et al. (20) utilizes a sodium barbital buffer system, pH 8.6, for preparation of standards, extracts and subsequent dilutions. Before RIA of the experimental samples, we demonstrated that standard and pituitary LH preparations in medium 199 diluted with the barbital buffer exhibited dose response curves which were parallel to curves given by these preparations in the barbital buffer system of Beamer et al. (20). The linear range of the standard curve (percent bound vs. log dose) extended from 5 ng to 320 ng. The cross reactivity of the standard reference preparations of FSH and TSH in the LH assay were tested and found to be less than 3%. Each assay was assessed by a method of statistical control developed by Rodbard et al. (23). The cumulative within assay variance was 2.1% and the overall between assay variance was 2.3%. 95% confidence limits were computed for each standard curve (24) and extreme outliers were discarded (22). All test and control pituitary and medium preparations for each of the three sets of incubations were assayed in triplicate at two dilutions in three separate radioimmunoassays. The LH values of the incubation media and pituitaries were computed as ng/mg pituitary. The means (±SEM) of the 18 values for each test and control sample were calculated and the LH releasing activity in each fraction was expressed as a percent of the control. The percent-of-control values for each subcellular fraction from the three experiments were averaged and the standard error of the mean determined. A two-way ANOVA was done (24) with the experiment as one variable and the subcellular fraction as the second variable. Orthogonal comparisons were made between groups to test for significant difference (24) and all groups were compared to the double control group. Electron microscopy. A portion of each subcellular fraction obtained in Fig. 1 was prepared for electron microscopy according to the Method of Hayat (25). Glutaraldehyde and osmium tetroxide were used for fixation and postfixation. The pellets were embedded in Epon-Araldite (Electron Microscopy Sciences, Fort Washington,
449
Pennsylvania) in such a way that sections would contain a cross section of the pellet from top to bottom. The blocks were sectioned on a Reichert thermal advance microtome and collected on Formvar coated copper grids. Some of the sections were stained with uranyl acetate and lead citrate for better contrast. Sections were viewed with a Hitachi HU12 electron microscope at magnifications from 2-30,000. Results LH releasing activity of the primary subcellular fractions. The five primary fractions were assayed for their ability to increase LH release from cultured anterior pituitaries. P2, the crude mitochondrial fraction, was the only fraction which caused a highly significant increase in LH release (Fig. 2a). This increased release was also accompanied by a highly significant depletion of pituitary LH (Fig. 2b) and a significant rise in the total LH in the incubation (i.e., LH in medium plus pituitaries) (Fig. 2c). None of the other four fractions caused a significant change in LH release into the medium or in pituitary LH content. LH releasing activity of the secondary subcellular fractions. The six secondary fractions obtained by sucrose density centrifugation of P2 were also assayed. Fraction P2-5 caused a highly significant increase in LH release into the medium (Fig. 2d) and a highly significant rise in total LH (Fig. 2f). However, this was not accompanied by a depletion of pituitary LH (Fig. 2e). Of interest are the results found with fraction P 2 -l, indicating a significant decrease in LH release into the medium (Fig. 2a). This was accompanied by a significant increase in pituitary LH content (Fig. 2e) but no change in total LH content (Fig. 2f) compared to control. There were no significant differences between the other fractions. LH releasing activity of other neural tissues. Two other neural tissues, the cerebellum
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 20 November 2015. at 03:32 For personal use only. No other uses without permission. . All rights reserved.
TABER AND KARAVOLAS
450
d) LH released into medium
a) LH released into medium
300
c
control
200
n
e) pituitary LH
b) pituitary LH
s?
n
rT
3 200 •o u «o to
1
1—=1
S 100 K •1 5
**
f) total LH
c) total LH
200 i
*|
i
,
100 P1
P1A P2
P3
S
LI
hi
P2-1 P2-2 P2-3 P2-4 P2-5 P2-6 C
MBH PRIMARY
MBH
SECONDARY
SUBCELLULAR FRACTIONS
SUBCELLULAR
FRACTIONS
En do i 1975 Vo! 96 < No 2
below the sensitivity limits of our RIA. The LH content of the various MBH subcellular fractions ranged from 3-24 ng per 0.5 ml. Thus, the LH present in the 0.5 ml aliquot of the MBH subcellular fractions added to the incubations, represents less than .015% of the LH measured in the incubation medium. Electron microscopy. The fractions were monitored by electron microscopy and the primary fractions, consistent with expected results (16,17), contained cellular debris, nuclei, mitochondria, microsomes, and free ribosomes respectively. P2, the crude mitochondrial fraction, also contained myelin fragments, synaptosomes, and scattered electron dense vesicles. After further centrifugation of this fraction on a discontinuous sucrose density gradient, the particulates were somewhat separated. Fraction P2-2 contained mainly myelin fragments. Fraction P2-3 consisted of myelin
FIG. 2. LH releasing activity of the subcellular fractions of the MBH. The designations on the abscissa, except for C (double control group), refer to the different subcellular fractions derived from the scheme shown in Fig. 1. Since ANOVA indicated no significant difference between the three experiments, the data were combined and the results shown represent the mean percent of control values (see text for details). Standard errors of the mean (SEM) are not shown on the bar graphs since in no case did the SEM exceed 0.5%. The shaded areas on the bar graphs for the double control groups (C) indicate the mean differences in LH release or content between the two control flasks. **Highly significant p < 0.01. *Significant p < 0.05.
and the cerebral cortex, were examined to determine the tissue specificity of the LH releasing activity. These tissues were fractionated and processed as described above for the MBH. None of the fractions significantly altered pituitary LH content or release. LH content of the subcellular fractions. The subcellular fractions of each of the three tissues were examined for endogenous LH content. The LH values of the fractions from the cerebellum and cerebral cortex were
FIG. 3. A representative electron micrograph of fraction P2-5 derived from the crude mitochondria! fraction (P2). As shown, electron dense vesicles of at least two Size ranges (A and B) are the predominant subcellular organelle. No mitochondria, synaptosomes, or myelin were detected in any of the EM sections of this fraction, (x 19,000).
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 20 November 2015. at 03:32 For personal use only. No other uses without permission. . All rights reserved.
SUBCELLULAR LRF LOCALIZATION IN RAT MBH fragments and synaptosomes. Synaptosomes of varying structural integrity dominated fraction P2-4 while P2-5 contained predominantly electron dense vesicles of various sizes (Fig. 3). Fraction P2-6 was composed of mitochondria with rough endoplasmic reticulum. Fraction P2-l did not yield a firm enough specimen to section. Negative staining of the fraction with 1% phosphotungstic acid showed small filamentous structures which could be membrane fragments. Discussion In agreement with our earlier work (12), the crude mitochondrial fraction obtained by differential centrifugation (P2) contained LH releasing activity (Fig. 2a). When monitored by electron microscopy, this fraction contained myelin, synaptosomes, mitochondria, and electron dense vesicles. These particulates were partially separated by further centrifugation on a discontinuous sucrose density gradient. Fraction P2-5 from this procedure showed an increased ability to release LH (Fig. 2d). This fraction contained predominantly electron dense vesicles (Fig. 3) in accordance with our previous work (12) and that of Ishii (10) and Clementi (11). Since the fraction contains electron dense vesicles of various sizes, we would anticipate that some would be associated with other releasing factors and/or biogenic amines. We have not yet assayed the fractions for these compounds. As shown in Fig. 2f, the total LH in the incubation as well as the LH released into the culture medium was increased. This suggests the possibility that fraction P2-5 which stimulates LH release from the pituitary may also increase pituitary LH synthesis (26,27). Unlike the results with fraction P2, no significant depletion of pituitary LH was observed with fraction P2-5. At this time we have no explanation for this difference. The failure of the cerebral cortical and cerebellar fractions to stimulate LH release from cultured pituitaries shows that the stimulation found with the MBH is not a general property of nervous tissue. Hence,
451
we assume that the release of LH from the MBH subcellular preparations was probably due wholly or in part to specific hypothalamic LH releasing factors such as the decapeptide, LRF, and not to general cellular constituents such as cAMP, prostaglandins, etc. The concentrations of cAMP (28) and prostaglandins (29) in these three brain tissues have been reported to be of the same order of magnitude (for each compound). Presumably the above and other cellular constituents common to these tissues, would have manifested themselves equally in their ability to cause release of LH whether in cerebellar, cerebral cortical, or hypothalamic tissues. Furthermore, the brain subcellular distribution profiles of cAMP (30,31) and prostaglandins (32) are different from our observed MBH subcellular localization of LH releasing activity. The fractionation of P2 by a sucrose density gradient also resulted in a fraction (P2-l) collected in 0.32M sucrose which significantly inhibited LH release from cultured anterior pituitaries. This agrees with our earlier, work (12) although the previous method used for separating the particulates in P2 was Sepharose 2B gel filtration. The particulate nature of this fraction is not known although the presence of small membrane fragments was detected by negative staining. The presence of both releasing and inhibitory activity in the crude mitochondrial fraction, P2, is consistent with the fact that fraction P2-5 had greater stimulatory activity than fraction P2. It would be premature at this juncture to deduce the existence of a hypothalamic LH release-inhibitory substance from these preliminary data. Several alternative explanations could be proffered on the basis of these observations. Until further investigations reveal the nature and significance of this inhibitory activity, suffice it to say that it represents an interesting finding in our investigation of the subcellular localization of LH releasing activity which warrants further investigation. The subcellular localization of LH releasing activity ("LRF") to a particulate fraction
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 20 November 2015. at 03:32 For personal use only. No other uses without permission. . All rights reserved.
TABER AND KARAVOLAS
452
(P2-5) containing electron dense vesicles may indicate that this and other hypophysiotropins are contained in storage vesicles or granules analogous to the storage granules of the anterior and posterior pituitary hormones. Other investigators (10,11) have found similar results.2 Further studies of these particulates is needed to resolve whether: 1) the LH releasing activity is contained in or adsorbed onto the electron dense vesicles which constitute the predominant subcellular organelle in P2-5; 2) separate populations (e.g., size) of the electron dense vesicles contain different hypophysiotropins and/or biogenic amines; and 3) different areas or cell types contain specific hormone or biogenic amine-containing electron dense vesicles. At any rate, the localization of this activity to a particulate fraction rich in electron dense vesicles separate from synaptosomes, nuclei, mitochondria, cell membranes, etc., should facilitate investigation into the sequential scheme of molecular events in the control (feedback) of LRF release and/or synthesis and possibly serve as a model for the subcellular localization of other hypophysiotropins.
8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23.
References 1. Arimura, A., A. V. Schally, T. Saito, E. Mulder, and C. Y. Bowers, Endocrinology 80: 515, 1967. 2. Clemens, J., and C. J. Shaar, Proc Int Soc Psychoneuroendocrinology, Brooklyn 1970, p. 356. 3. Crighton, D. B., H. P. G. Schneider, and S. M. McCann, Endocrinology 87: 323, 1970. 4. Jackson, G. L., K. R. Stevens, and A. V. Nalbandov, Endocrinology 83: 234, 1968. 5. McCann, S. M., S. Taleisnik, and H. M. Friedman, Proc Soc Exp Biol Med 104: 432, 1960. 6. Mittler, J. C , and J. Meites, Endocrinology 88: 500, 1966. 7. Matsuo, H., Y. Baba, R. M. G. Nair, A. Arimura,
24.
25.
26. 27. 28. 29. 30. 31.
2
Also, since the submission of this paper, Pelletier et al. (33) using an immunohistochemical technique at the electron microscopic level, have observed that LRF is contained in the secretory granules of some nerve endings in the palisade layer of the median eminence.
32. 33.
Endo 1975 Vol 96 . No 2
and A. V. Schally, Biochem Biophys Res Commun 43: 1334, 1971. Baba, Y., H. Matsuo, and A. V. Schally, Biochem Biophys Res Commun 44: 459, 1971. Arimura, A., H. Matsuo, Y. Baba, L. Debeljuk, J.. Sandow, and A. V. Schally,Endocrinology 90: 163, 1972. Ishii, S., Endocrinology 86: 207, 1970. Clementi, F., B. Ceccarelli, M. Ceratie, L. Demonte, M. Felici, M. Motta, and A. Pecile,/ Endocrinol 48: 205, 1970. Karavolas, H. J., R. K. Meyer, K. J. Deighton, and S. Adrouny, Endocrinology 88: 969, 1971. Johansson, K. N. G., B. L. Currie, K. Folkers and C. Y. Bowers, Biochem Biophys Res Commun 52: 967, 1973. Deighton, K. J., and R. K. Meyer, Endocrinology 84: 778, 1969. Ramirez, S., V. Domingo, and' C. H. Sawyer, Endocrinology 76: 282, 1965. Gray, E. G., and V. P. Whittaker, Anatomy 96: 79, 1962. L^vtrup-Rein, H., and B. S. McEwen, J Cell Biol 30: 405, 1966. Kragt, C. L., and J. Dahlgren, Neuroendocrinology 9: 30, 1972. Spona, J., and O. Luger, FEBS Lett 32: 52, 1973. Beamer, W. G., S. M. Murr, and I. I. Geschwind, Endocrinology 90: 823, 1972. Greenwood, F. C , W. M. Hunter, and T. S. Glover, Biochem J 89: 114, 1963. Vivian, S. R., and F. S. Labella,/ Clin Endocrinol Metab 33: 225, 1971. Rodbard, D., P. L. Rayford, J. A. Cooper, and G. T. Ross,/ Clin Endocrinol Metab 28: 1442, 1968. Snedecor, G. W., and W. G. Cochran, Statistical Methods, ed. 6, The Iowa State University Press, Ames, Iowa, 1967, p. 267. Hayat, M. A., Principles and Techniques of Electron Microscopy, Van Nostrand Reinhold Co., New York, 1970, p. 35. Redding, T. W., A. V. Schally, A. Arimura, and H. Matsuo, Endocrinology 90: 764, 1972. Serra, G. B., and A. R. Midgely, Endocrinology 91: 962, 1972. Schmidt, M. J., D. E. Schmidt, and G. A. Robison Science 173: 1142, 1971. von Euler, U. S., and R. Eliasson, Prostaglandins, Academic Press, New York, 1967. Lust, W. D., and N. D. Goldberg, Pharmacologist 12: 290, 1970. Goldberg, N. D , W. D. Lust, R. F. O'Dea, S. Wei, and A. G. O'Toole. In Greengard, P., and E. Costa, Role of Cyclic AMP in Cell Function, Raven Press, New York, 1970, p. 67. Kataoka, K., P. W. Ramwell, and S. Jessup, Science 157: 1187, 1967. Pelletier, G., F. Labrie, R. Purrani, A. Arimura, and A. V. Schally, Endocrinology 95: 314, 1974.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 20 November 2015. at 03:32 For personal use only. No other uses without permission. . All rights reserved.
