Structural changes in nuclear chromatin in rat pituitary after chronic stress of low intensity.код для вставкиСкачать
THE ANATOMICAL RECORD 220:125-131(1988) Structural Changes in Nuclear Chromatin in Rat Pituitary After Chronic Stress of Low Intensity DYMITR KOMITOWSKI, SHUHEI MUTO, fiRGEN WEISS, BERTOLD S C H M I n , AND GEORGE T. TAYLOR Institute for Experimental Pathology, German Cancer Research Center, 0-6900Heidelberg, FRG (D.K., S. M., B.S.); Center for Molecular Biology, University of Heidelberg, 0-6900 Heidelberg, FRG (J.W.); Laboratory for Psychobiology, University of Missouri-St. Louis, ‘ :‘1:) St. Louis, MO 63121 (G. 1 ABSTRACT Acute, intense sources of “psychogenic” stress clearly modify the structure and function of the hypophysis, and there are concomitant changes in many peripheral physiological systems. Less dramatic sources of stress yield more equivocal results. An experiment is reported in which nuclear morphology of adenohypophyseal cells from 49 male rats exposed to a chronic, low-intensity stressor was examined both by conventional histological and computer-assisted-image-processing methods. The hypothesis tested was that an unequivocal pattern of morphological changes in the nucleus and nuclear chromatin would be revealed by image processing. Rats were killed after living for a year in a relatively low-stress environment, “crowded” in groups of five animals per cage. The control condition was a minimal stress environment of two rats per cage. Results suggested few signs of pathology from peripheral measures of hypophyseal activity, and direct light microscopic examination of the gland revealed no differences between the two groups. Analysis of computer-enhanced images of the pars distalis nuclei from the adenohypophysis, on the other hand, generated findings that were statistically and biologically significant. Nuclear size increased in the stress condition, the number of chromatin and area occupied by the particles increased, and the position of chromatin shifted toward the periphery of the nucleus. Perhaps more important, optical density analysis indicated that chromatin was less tightly packed in the experimental animals. Implications are that chronic, low-intensity stress modulates nuclear structural changes from a dormant to an active state that portend changes in the peripheral systems influenced by the hypophysis. Recent progress in the study of the pituitary has confirmed functions that were suggested five decades ago. Moreover, there is now a clear picture of such features as cellular localization and functional consequences of structural changes in the pituitary. A key question that remains unanswered is the interrelation among different cell types (Kovacs, 1984; Tixier-Vidal et al., 1984). The anterior lobe of the gland contains distinctive cell types that influence a notable array of physiological systems. Somatotropic and thyrotropic cells influence metabolism, gonadotropic and lactotropic cells affect fertility, and adrenocorticotropic cells influence various homeostatic processes (Martin, 1985). Stress can influence all these peripheral systems (McGrady and Chakraborty, 1983; Munck et al., 1984; Pollard, 19841, and there is general agreement that the stress effects are mediated by the hypophysis (Hennessy and Levine, 1979). The possibility that everyday sources of “psychological” stress may influence the genesis and onset of pathological states has been entertained for years, but there is little empirical support (Locke et al., 1984). Experi0 1988 ALAN R. LISS, INC. mental findings with animal models, most commonly rodents, indicate that acute, intense stressors can induce pathological changes in liver (Hidalgo et al., 19861, testes (Collu et al., 1979), digestive tract (Glavin, 19851, and circulatory system (Fokkema and Koolhaas, 1985). Direct histological analysis of acutely stressed animals is unusual, but Moriarty et al. (1975) reported cytoloplasmic changes in parenchymal cells of the intermediate lobe of the rat hypophysis. The most reliable of the acute stressors, in comparison with a nonstressed control animal, are immobilization by restraint (Quirce and Maickel, 1981) and electric footshock (Friedman and Iwai, 1976). Subtle manipulations, on the other hand, have yielded more equivocal results. For example, both predictable and unpredictable footshock (Maier et al., Received April 13, 1987; accepted August 11, 1987. Address reprint requests to Prof. Dr. Dymitr Komitowski, Institute for Experimental Pathology, German Cancer Research Center, Im Neuenheimer Feld 280, D-6900 Heidelberg, FRG. 126 D. KOMITOWSKI ET AL. Fig. 1. A general microscopic view (a) of hypophysis from a male rat framed indicates the region of measurements made by image analysis. showing the three major anatomical subdivisions: pars distalis (PD), H.E. x 125. Higher-magnification photomicrographs of pars distalis pars intermedia (PI), and pars nervosa (PN). The section of PD that is from a representative (b) control and stressed (c) rat. H.E. ~ 5 0 0 . 1986)similarly increased the release of glucocorticosteroids and ACTH, and reliable dose-response changes in physiology with different stress intensities have proved difficult to demonstrate (Murison et al., 1986; Natelson et al., 1987). One problem may be that the acute stressors that have been employed bear questionable relation to the evolutionary history of the experimental animal (Brain and Benton, 1979; Domanski et al., 1986). More “ecologically valid” sources of stress may yield more reliable data. In a n initial attempt to establish a more valid stressor, we reported recently on the effects of manipulating aggressiveness of male rats as a source of chronic social stress (Taylor et al., 1987). The experimental paradigm employed was notable mostly for its complexity. A simpler chronic social stressor would be to house different numbers of animals in similar-size cages (Brayton and Brain, 1984; Gamallo et al., 1986). The apparent low intensity of “overcrowding,” however, provokes few reliable changes in pituitary or peripheral structures 127 STRESS AND CHROMATIN STRUCTURE Fig. l(c). (Benton et al., 1979; Shalicky et al., 1984). Rather than being a disadvantage for the present experiment, however, the unreliability of manipulating housing conditions was a mechanism by which to test the effectiveness of computer-assisted image processing. Using subvisual quantitative parameters, the system can describe very subtle morphological changes (Curcio and Sloan, 1986; Panno and Nair, 1986). The basic function of the adenohypophysis is its hormonal activity, and function is related to structure of both nucleus and nuclear chromatin (Alberts et al., 1983; Thrall et al., 1978).The hypothesis tested was that there would be systematic changes in the morphology of pars distalis cell nuclei, as revealed by image processing, in the hypophysis of male rats housed chronically in a low-stress environment. cessing. Hypophyses prepared for routine light microscopy were fixed immediately in buffered formaldehyde (pH 7.4). The tissue was imbedded in paraffin blocks, cut at 2 pm (Autocut 1140, Reichert-Jung), and stained with haematoxylin and eosin. In order to differentiate cell types, from every four Feulgen-stained sections, one section was stained by the Glenner and Lillie (1957) procedure and a second section was stained by PAS (Pearse, 1968). For image analysis, serial paraffin sections from hypophyses were prepared as described above for light microscopy and were stained by Feulgen's method (Pearse, 1968). In the Feulgen reaction the fluorescent staining displays DNA in nuclear chromatin (Ross and Reith, 1985). MATERIALS AND METHODS Animals and Maintenance The system for image analysis includes a light microscope (Axiomat, Zeiss), plumbicon scanner, Quantimet 720 analyzer (Cambridge Instruments Inc., Dortmund), and color display. Computers were VAX 11/750,PDP 11/ 34, and LSI 11/23 models (Digital Equipment Corp., Maynard). There are several discrete steps used by the computer to analyze tissue. Details are presented in more technical publications by Komitowski and associates (Komitowski and Zinser, 1985; Komitowski et al., 19831, but a brief description of the process employed for the present research follows; the major steps are presented in the second figure. Initially, an image of the Feulgen-stained section of the nuclei from a section of adenohypophyseal tissue was generated by microscopic illumination. The images were digitized in a matrix of 768 x 512 picture elements, or pixels. The optical density of each pixel was evaluated by the comDuters relative to "may" levels remesentin; 64 different levels of light intensity (Fig. 2a.b). Each &el corresDonds to 36 n& of the obi&t. agd , the images 'were segmLntd (Fig. 2c,d) by a polir ckrdinate method (Zinser and Komitowski, 1983).Parameters The animals were male Wistar rats man: Wistaroutbred stock) that were killed and examined at 14 months of age, 1 year after being housed in groups of two (n=24) or five (n=25) rats per cage. Cages were clear plastic with inside dimensions of 37.7 cm x 21.5 cm x 15 cm (total available space = 12,158 cm3).Room conditions were standardized with 22 f2"C, 55 f 5 % humidity, and 12-hour light: 12-hour darkness that were controlled automatically. Water and food (Ssniff, Soest) were freely available. At the end of a year in the environment, all animals were killed by carbon dioxide asphyxiation. Tissue Preparation At necromv the animals were weighed. adrenal glands and hypophises were excised, andU wet weight; were obtained. Testes were removed, weighed, and prepared for sperm counts by using a method described earlier (Taylor et al., 1983). Hypophyses from each group were prepared for conventional light microscopy or image pro- Image Processing , ~ 128 D. KOMITOWSKI ET AL. Fig. 2. Representative nuclei from control (a,c, and e) or stressed (b, d, and f) rats demonstrating the steps involved in computer-assisted image analysis (see text). ~9,000. of nuclear contents were described by using a n algorithm that enables the localization of condensed chromatin regions within the matrix based on locally adaptive thresholding. These were defined as regions with high optical density or chromatin particles. Each chromatin particle could be identified and the number of particles per fixed area determined; evaluation of parameters such a s area occupied by chromatin and position of the particle within the nucleus was also possible (Fig. 2e,f). The final step is optical density analysis, or the amount of light passing through the chromatin that provides a n accurate measure of the relative degree of chromatin condensation or packing (Yaniv and Cereghini, 1986). A total of 889 and 1,266 nuclei were analyzed in this manner from the control and experimental groups, re- STRESS AND CHROMATIN STRUCTURE 129 Fig. 2(e,f). Quantitative Image Analysis spectively. As described previously (Komitowski et al., 1986), a n area of visual field was selected a t random from tissue sections, and all nuclei within that visual field were analyzed. We did not distinguish among cell types for the analysis because the cells are functional interdependent (Thrall et al., 19781, and our interests are in the structural properties that are common to all pars distalis cells (Allen et al., 1977).Also, we examined only the nucleus and its chromatin because structural changes at that level determine transcriptional processes that modulate hormone synthesis and morphological changes (Alberts et al., 1983) in the gland and, subsequently, in peripheral physiological systems. RESULTS Light Microscopy Excised glands and adenohypophyseal tissue were prepared for conventional histological analyses. Two pathologists in the histodiagnostic division of the German Cancer Research Center independently examined tissue sections without knowledge of which group the animals belonged to. Results were described in the same manner in which other tissues are described during routine histological service to researchers in the center. There were infrequent pathological changes in peripheral tissues including pneumonia, endocardia1 fibrosis, and nephrosis, but these findings were not related to differences in housing conditions. Morphology of individual hypophyseal cells, including integrity of the nuclear envelope, membranes, cytoplasm, and nucleus, appeared quite normal in both sets of animals (Fig. 1). Percentages of acidophilic, basophilic, and chromophobic cells relative to the total cell population did not differ among the rats nor from those values reported in the literature (Nakane, 1973). Analyses of nuclei and nuclear chromatin were the primary interest. The parameters examined in each nucleus were size of nucleus, number of particles and area occupied by chromatin, position of particles relative to the periphery of the nucleus, and, finally, the degree of packing of chromatin fibers. Area occupied by chromatin was calculated in pm2 of nucleus to allow comparisons of nuclei of different sizes. The capacity of chromatin packing to organize with higher-order foldings of DNA (Nicolini, 1983) allows number, area, and density of chromatin to be calculated independently of each other. The data were combined for the animals in each group and appear in Table 1. A series of t-tests at 5% significance level were used to analyze those data, and the results revealed statistically reliable values for each of the five parameters. Animals from the experimental group had adenohypophyseal nuclei that were larger than those from the control group. Experimental males also had higher numbers of chromatin, and the area occupied by chromatin particles was greater. The position of chromatin within the nucleus was more peripherally located, and chromatin particles were more loosely packed in experimental males. Multivariate analysis of variance indicates that nuclei from the two groups were clearly from different statistical populations, F(5, 2149) = 289.91, P< .01. Finally, representative nuclei were selected from adenohypophyseal tissue obtained from a rat from each group, and the image-analysis process is shown in Figure 2. Figure 2a,b shows the initial image of a typical nucleus projected by computer. Figure 2c,d demonstrates the intial image of the nucleus with its detected contour. Figure 2e,f is the computer-generated digitized image in which one can note clearly different areas occupied by densely compacted chromatin. 130 D. KOMITOWSKI ET AL TABLE 1. Computer-assisted quantitative image analysis findings from nuclei and nuclear chromatin from adenohypophysisof rats maintained under chronic low-intensity stress Mean Control Measure Nuclear area (pm') Number of particles Area occupied by particles (pm') Optical density of particle (per pm') Percentage of particles in periphery SEM t-value Change (%) 33.54 13.52 13.19 +31.4 0.2' + 0.03' + 16.7 + 16.7 26.2 & 0.4 17.8 & 0.2l 20.57 -32.1' 60.6 rt 0.3 61.8 f 0.3' 3.04 group 22.3 f 0.2 25.1 & 0.2 4.02 k 0.04 Stress group 29.3 29.3 4.69 0.1' +2.0 'Significant difference from the control group ( P < .01). 'Decrease in this parameter value indicates more light is passing through the nucleus. Peripheral Measures Of secondary interest was a general measure of each of the broad functions of adenohypophyseal cells. Specifically, body weight was used as a gross measure of disruptions in metabolism, and a t-test on those data indicated that there were no differences between the groups, mean = 541 f SEM = 12 gm, for controls and 522 f 11 for experimental animals. Similarly, there were no differences in a crude measure of adrenal-pituitary interactions; wet weights in mgI100 gm body weight of adrenal and pituitary glands were 10.5f0.6 and 2.2k0.2, respectively, for the controls and 10.7f0.4 and 2.350.2, respectively, for the experimental animals. The measure of fertility was number of sperm; again the groups did not differ significantly. The values of control males and experimental males in x106 per gm testicular tissue were 82.8k8.6 and 70.6 f3.3, respectively. DISCUSSION Data from the present experiment suggest that adenopophyses of male rats undergo structural modification during exposure to stress. There were two contributions from the present study to the literature on stress-pathology relations. First, a computer-assisted examination of sectioned pars distalis tissue revealed systematic quantitative changes that were undetected with conventional analyses. Second, a chronic social stressor of relatively low intensity, living in crowded conditions for a year, induced structural modifications that were reliable and common to the varied cell types in the gland. Previous findings from animals maintained under mildly stressful conditions have proved difficult to replicate (Brain and Benton, 1979). The apparent conclusion is that mild stress provokes no systematic pathological changes or that the results reflect individual differences to stress Nogel, 1985). If only typical markers of the effects of stress on peripheral systems were used, the present data would have presented a similar picture. Body weights, wet weights of glands, and sperm counts did not differ under the two housing conditions. Conventional histological analyses with light microscopy were no more helpful. Characterization of adenohypophyseal tissue suggested that there were no reliable differences between the two ~ O U D Sof rats. Quantitative image processing on the z h e r h a n d , sug- gested clearly that the stress provoked systematic changes in adenohypophyseal nuclei. There were five parameters of the nuclei that were analyzed with computerenhanced image processing, and differences between the groups were statistically and biologically significant. Size of the nucleus increased by approximately 30% under stress. Number of chromatin particles increased, as well a s the area within the nucleus occupied by chromatin. Location of chromatin changed with a small, but statistically significant, shift toward the periphery of the nucleus. Finally, there was a loosening in the packing of chromatin. These latter data may prove to be the most important finding and may demonstrate the usefulness of image processing as a powerful supplemental tool for special problems in histology (Panno and Nair, 1986). Structural change in the hypophysis has profound implications on the integrity of varied peripheral physiological systems. There were no differences, at least none detected by traditional measures, between the peripheral systems of the two groups. The chromatin data, however, suggest that the pituitary was undergoing structural changes that later would be clearly revealed in both central and peripheral systems. There is evidence from postmortem analyses of the hypophysis that many people harbor small "silent" adenomas without exhibiting detectable clinical symptoms (Hausson et al., 1982;Krieger, 1984).Although we found no obvious tumors, there were changes in the nucleus that suggest that cells from rats under stress were in a state of activation. There are data in the literature indicating that transcriptionally active nuclei are characterized by increased size of the nuclear envelope (Franke et al., 1981), and nuclei of stressed animals were hypertrophic. Stressed animals also exhibited increased chromatin, which suggests higher DNA content (Wolley et al., 1982) and activation of the nucleus (Haag et al., 1977; Nicolini et al., 1982). Although movement of chromatin toward the periphery of the nucleus in crowded rats is of uncertain biological significance, the capacity of optical density analysis to demonstrate a loosening of tightly bound chromatin is of special interest. Only a small fraction of the DNA in nuclei of a cell is transcriDtionallv active (Darnell. 1982). Although some of thesg genes &e likely permanently' inactivGed dur- STRESS AND CHROMATIN STRUCTURE ing the developmental process, the remaining inactive genes must await signals that convert inactive chromatin into its active counterpart (Yaniv and Cereghini, 1986).The suggestion that psychosocial stress can act as one of these signals remains a n attractive hypothesis without empirical support (Grossarth-Maticek et al., 1985).The present data offer some support for that idea. A prominent feature that distinguishes active from inactive chromatin is the packing of the particles. Active chromatin is more loosely packed (Nicolini, 1983). The present data suggest that low-intensity chronic social stress can activate formerly inactive chromatin. LITERATURE CITED Alberts, B., D. Bray, J. Lewis, M. Raff, D. Roberts, and J.D. Watson 1983 Molecular Biology of the Cell. Garland Pub. Inc., New York. Allen, M.B., G.H. Greeley, A. Costoff, V.B. Mahesh, and K.Y. Little 1977 Experimental models for study of pituitary function. In: The Pituitary. M.B. Allen and V.B. Mahesh, eds. Academic Press, New York, pp. 235-259. Benton, D., P.F. Brain, and J.F. Goldsmith 1979 Effects of prior housing on endocrine responses to differential caging in male TO-strain mice. Physiol. Psychol., 7:89-92. Brain, P.F., and D. Benton 1979 The interpretation of physiological correlates of differential housing in laboratory rats. Life Sci., 24t99116. Brayton, A.R., and P.E. Brain 1974 Studies on the effects of differential housing on some measures of disease resistance in male and female laboratory mice. J. Endocrinol., 61:XIVIII. Collu, R., Y. Tache, and J.R. Ducharme 1979 Hormonal modifications induced by chronic stress in rats. J. Steroid Biochem., 11:989-1000. Curcio, C.A., and K.R. Sloan 1986 Computer-monitored morphometry using video-mixed microscopic images and computer graphics. Anat. Rec., 214:329-337. Darnell, J.E. 1982 Variety in the level of gene control in eukaryotic cells. Nature, 297:365-371. Domanski, E., F. Przekop, E. Wolinska-Witort, K. Mateusiak, L. Chomicka, and S. Garwacki 1986 Differential behavioural and hormone responses to two different stressors (footshocking and immobilization) in sheep. Exp. Clin. Endocrinol., 88:165-172. Fokkema, D.S., and J.M. Koolhaas 1985 Acute and conditioned blood pressure changes in relation to social and psychosocial stimuli in rats. Physiol. Behav., 34:33-38. Franke, W.W., U. Scheer, G. Krohne, and E.-D. Jarasch 1981 The nuclear envelope and the architecture of nuclear periphery. J. Cell Biol. 91:39-50. Friedman, R. and J. Iwai 1976 Genetic predisposition and stress-induced hypertension. Science, 193:161-162. Gamallo, A., A. Villanua, G. Trancho, and A. Fraile 1986 Stress adaptation and adrenal activity in isolated and crowded rats. Physiol. Behav., 36:217-221. Glavin, G.B. 1985 Effects of morphine and naloxone on restraint-stress ulcers in rats. Pharmacology, 3157-60. Glenner, G.G., and R.D. Lillie 1957 A rhodocyan technique for staining the anterior pituitary. Stain Technol., 32:187-190. Grossarth-Maticek, R., J. Bastiaan, and D.T. Kanazir 1985 Psychosocia1 factors as strong predictors of mortality from cancer, ischemic heart disease and stroke-the Yugoslav prospective study. J. Psychosom. Res., 29t167-176. Haag, D., F. Schlieter, V. Ehemann, and K. Goerttler 1977 Cytological and cytophometric studies on DMBA-induced changes of the conjunctival epithelium in Syrian golden hamsters. Z. Krebsforsch., 89:201-214. Hausson, J., C. Charpin, P. Jaquet, J.C. Lissitzky, F. Grisoli, and M. Toga 1982 Corticolipotropin immunoreactivity in silent chromophobe adenomas. Arch. Pathol. Lab. Med., 106:25-30. Hennessy, J.W., and S. Levine 1979 Stress arousal and the pituitaryadrenal system. A psychoendocrine hypothesis. In: Progress in Psychobiology and Physiological Psychology. J.M. Sprague and A.N. Epstein, eds. Academic Press, New York, Volume 8, pp. 134179. Hidalgo, J., A. Armario, R. Flos, and J.S. Garvey 1986 Restraint stress induced changes in rat liver and serum metallothionein and in Zn metabolism. Experientia, 42: 1006-1010. Komitowski, D., P. Schmezer, B. Schmitt, and S. Muto 1986 Image analysis of hepatocyte nuclei in assessing Di(2-ethylhexy1)phthalate effects eluding detection by conventional microscopy. Toxicology, 41:ll-19. Komitowski, D., and G. Zinser 1985 Quantitative description of chromatin structure during neoplasia by the method of image process- 131 ing. Anal. Quant. Cytol. Histol., 7:178-182. Komitowski, D., G. Zinser, and C. Stute 1983 Digital picture analysis as an integral part of the information system for experimental oncopathology of the German Cancer Research Center. Methods Inf. Med.. 22:69-74. Kovacs, K. 1984 Light and electron microscopic pathology of pituitary tumors: Immunohistochemistry. In: Secretory Tumors of the Pituitary Gland. P.M. Black, ed. Raven Press, New York, pp. 365-375. Krieger, D.T. 1984 Pituitary ACTH hyperfunction: Physiopathology and clinical aspects. In: Pituitary Hyperfunction: Physiopathology and Clinical Aspects. F. Camanni and E.E. Muller, eds. Raven Press, New York, pp. 221-234. Locke, S.E., L. Kraus, J. Leserman, M.W. Hurst, J.S. Heisel, and R.M. Williams 1984 Life change stress, psychiatric symptoms and natural killer cell activity. Psychosom. Med., 46:441-453. Maier, S.T., S.M. Ryan, C.M. Barksdale, and N.H. Kalin 1986 Stressor controllability and the pituitary-adrenal system. Behav. Neurosci., 1OOt669-674. Martin, C.R. 1985 Endocrine Physiology. Oxford University Press, New York. McGrady, A.V., and J. Chakraborty 1983 Effects of stress on the reproductive system of male rats and mice. Arch. Androl., 10:95-101. Moriarty, G.C., N.S. Halmi, and M. Moriarty 1975 The effect of stress on rat pituitary. Endocrinology, 961426-1436. Munck, A,, P.M. Guyre, and N.J. Holbrook 1984 Physiological functions of glucocorticoids in stress and their relation to pharmacological actions. Endocr. Rev., 5:25-44. Murison, R., J.B. Overmieir, and E.J. Skoglund 1986 Serial stressors: Prior exposure to a stressor modulates its later effectiveness on gastric ulceration and corticosteroid release. Behav. Neural Biol., 45: 185-195. Nakane, P.K. 1973 Distribution of gonadotropic cell in the anterior pituitary gland of the rat. In: The Regulation of Mammalian Reproduction. S.J. Segal, R. Crozier, P.A. Corfman, and P.G. Condliffe, eds. Charles C. Thomas Publishers, Springfield, IL,pp. 79-88. Natelson, B.H., D. Creighton, R. McCarty, W.N. Tapp, D. Pitman, and J. Ottenweller 1987 Adrenal hormonal indices of stress in laboratory rats. Physiol. Behav., 39:117-125. Nicolini, C. 1983 Chromatin structure: From nuclei to genes. Anticancer Res.. 2:63-86. Nicolini, C., G. Brambilla, E. Patrone, S. Ridella, and A. Chiabrera 1982 Nuclei, chromatin and water: Alternations and mechanisms for chemically induced carcinogenesis. In: Chemical Carcinogenesis. C. Nicolini, ed. Plenum Press, New York, pp. 381-434. Panno, J.P., and K.K. Nair 1986 Effects of increased lifespan on chromatin condensation in the adult male housefly. Mech. Ageing Dev., 35:31-38. Pearse, A.G.E. 1968 Histochemistry. Churchill Livingstone Publishers, Edinburgh. Pollard, I. 1984 Effects of stress administered during pregnancy on reproductive capacity and subsequent development of the offspring of rats: Prolonged effects on the litters of a second pregnancy. J. Endocrinol., 100:301-306. Quirce, C.M., and R.P. Maickel 1981Alterations of biochemical parameters by acute and repetitive stress situations in mice. Psychoneuroendocrinology, 6:91-97. Ross, M.H., and E.J. Reith 1985 Histology. Harper and Row, New York. Skalicky, M., H. Bubna-Littitz, and G. Hofecker 1984 The influence of persistant crowding on the age changes of behavioural parameters and survival characteristics of rats. Mech. Ageing Dev., 28:325336. Taylor, G.T., J. Weiss, and D. Komitowski 1983 Reproductive physiology and penile papillae morphology of rats after sexual experience. J. Endocrinol., 98.551-563. Taylor, G.T., J. Weiss, and R. Rupich 1987 Male rat behavior, endocrinology and reproductive physiology in a mixed-sex, socially stressful colony. Physiol. Behav., 39:429-433. Thrall, C.L., R.A. Webster, and T.C. Spelsberg 1978 Steroid receptor interaction with chromatin. In: The Cell Nucleus. H. Busch, ed. Academic Press, New York, pp. 461-529. Tixier-Vidal, A., C. Tougard, and A. Morin 1984 Cellular events involved in transport, storage and release of pituitary hormones. In Pituitary Hyperfunction: Physiopathology and Clinical Aspects. F. Camanni and E.E. Muller, eds. Raven Press, New York, pp. 71-83. Vogel, W.H. 1985 Coping, stressors and health consequences. Neuropsychobiology, 13:129-135. Wolley, R.C., K. Schreiber, L.G. Koss, M. Karas, and A. Sherman 1982 DNA distribution in human colonic carcinomas and its relation to clinical behavior. J. Natl. Cancer Inst., 69:15-22. Yaniv, M., and S. Cereghini 1986 Structure of transcriptionally active chromatin. CRC Crit. Rev. Biochem., 21:l-26. Zinzer, G., and D. Komitowski 1983 Segmentation of cell nuclei in tissue sections analysis. J. Histochem. Cytochem., 31:94-100.