MICROSCOPY RESEARCH AND TECHNIQUE 34:133-138 (1996) Sexual Dimorphism in the Harderian Gland of the Syrian Hamster Is Controlled and Maintained by Hormones, Despite Seasonal Fluctuations in Hormone Levels: Functional Implications GERALD R. BUZZELL Department of Anatomy and Cell Biology, University of Alberta, Edmonton, Alberta T6G 2H7,Canada KEY WORDS Sexual dimorphism, Harderian gland, Syrian hamster, Histology, Porphyrin, Lipid droplets, Photoperiods ABSTRACT The Harderian gland of the Syrian hamster (Mesocricetus aurutus) is unusual amongst rodents in the degree of dimorphism present. Other types of hamsters have Harderian glands which are apparently identical in male and female animals. Laboratory populations of Syrian hamsters are derived from very limited genetic stock, which makes one concerned lest they not be representative of wild populations; however, until wild stocks of M.uuratus become available, we should assume that insights derived from studies of dimorphism in Syrian hamsters represent important considerations for the life of these animals. Two dimorphic features are the histology and the porphyrin content of the Harderian glands. About 95% of the lipid droplets in female glands are small (type l),whereas only about 65%of those in males is type 1,with the other 35% being type 2 (large droplets). Five weeks of castration of males led to an increase in type 1 droplets to 90%. On the other hand, 2 weeks treatment of females with testosterone led to a reduction in type 1droplets to about 82%. Short day photoperiods led to a large increase in type 2 droplets in both males and females (to 52% in males, 35% in females after 8 weeks). These results suggest that the lipid contained in type 2 droplets is important to hamsters of both sexes during the winter. Porphyrin concentrations are 100-1,000 times higher in females than males, and this is largely controlled by testosterone as orchidectomy leads to increased male levels and testosterone treatment leads to reduced female levels. However, a number of treatments which also lead to reduced testosterone levels do not lead to increased porphyrins and may, in fact, prevent the rise which would normally follow orchidectomy. One of these antiporphyrinogenic treatments is exposure to short day photoperiods. Thus, the sexual differences in porphyrin levels in Syrian hamsters are maintained, despite seasonal fluctuations in hormone levels. This suggests that this dimorphism is important for the fundion of the gland. o 1996 Wiley-Liss, Inc. INTRODUCTION It is now over 40 years since Christensen and Dam (1953)and Wooley and Worley (1954) reported that the Harderian glands of Syrian golden hamsters exhibit marked sexual differences. Since that time, and particularly since the classic paper of Hoffman (19711, Mesocricetus auratus has been a favourite experimental animal for researchers investigating Harderian glands. We hope that understanding the hormonal control of dimorphic features will allow us to understand the significance (i-e.,the function) of Harderian glands in the overall biology of the hamster and that these studies will contribute insights into problems with wide biological importance. SEXUAL DIMORPHISM IN SYRIAN HAMSTER HARDERIAN GLANDS The Extent of the Phenomenon A list of characteristics known to differ between the sexes in Syrian hamsters is given in Table 1. It is evi- 0 1996 WILEY-LISS, INC. dent from this list that male and female Harderian glands are quite different in Syrian hamsters. However, Harderian glands are not the only features of Syrian hamsters which differ in males and females. For instance, females have high levels of a serum protein, called “female protein” (homologous to human C-reactive protein and serum amyloid P component) which is present in only very low concentrations in males (Coe, 1977). Like dimorphism in the Harderian gland, the significance of serum female protein is unknown. Is the Syrian Hamster Harderian Gland Unique? I have recently had the opportunity to examine Harderian glands of several species of hamsters histologically. Male-female differences, if they exist a t all, Received January 1, 1995; accepted in revised form March 16, 1996. Address reprint requeste to Gerald R. Buzzell, Department of Anatomy and Cell Biology, Universib of Alberta, Edmonton, Alberta T6G 2H7,Canada. 134 G.R. BUZZEU TABLE I . Semially dimorphic features of Syrian (golden) hamster Harderian glands Details Reference Feature d Larger in Size._. _ Darkin P Color Hoffman, 1971 6:type I & II cells Histology 4: type I cells Bucana and Nadakavukaren, 1972 6:polytubular complexes Ultrastructure 4: membranous organelles Murawski et al., 1991 6: 3 types of alkyl-diacylglycerol Lipids Q : 2 types of alkyl-diacylglycerol Lin and Nadakavukaren, 1981 6 : shorter chains Fatty acids Buzzell et al., 1991 Extra bands in 6 on SDS-PAGE Proteins Puig-Doming0 et al., 1988 4:more than d Somatostatin Hoffman, 1971 4:100-1,OOO times more than d Porphyrins Thompson et al., 1984 Mostly higher activity in P Enzymes of porphyrin synthesis Hoffman et al., 1985 Higher in P Melatonin Menendez-Pelaezet al., 1988 NAT higher activity in c3 Enzymes of melatonin synthesis HIOMT higher activity in P Hoffman and Jones, 1981 4:higher Na, Mn, Ca Metals 6:higher Fe, Mo McBlain et al., 1994 Higher,,B in intact P Androgen receptor Payne et al., 1982 4:35 times more than S Mast cells Pangerl et al., 1989 More in P B-receDtors are not obvious in Djungarian (Phodopus sungorus), Armenian (Cricetulus migrutorius), or Chinese (C. gresius) hamsters. It may or may not be significant that Armenian hamsters also show smaller sex differences in serum female protein levels than do Syrian hamsters (Coe and Ross, 1990). This brings up an important question: Is Mesocricetus uurutus unique, and, if so, can we learn any general truths from its study? brandti) which we examined recently also showed gender differences. These doubts notwithstanding, applying Occam’s razor to this problem dictates that we take what we see at face value and assume that the dimorphism we see in the laboratory evolved in the wild. Let us, therefore, look at the control of some of these dimorphic features in more depth. I will discuss the histology and the porphyrin content. The Syrian Hamster The Syrian hamster was first described by the brothers Alexander and Patrick Russell in 1797,from their observations in Aleppo, a city in northern Syria (then part of the Ottoman Empire). It was recognized as a separate species and named by George Waterhouse, curator of the Zoological Society of London, in 1839. The Syrian hamsters we u0e today are descendants of three surviving hamster siblings, captured near Aleppo by the zoologist Israel Aharoni of the Hebrew University of Jerusalem in 1930.From these three animals are descended all of the hamsters in laboratories and pet stores today. “here have been only two more recent captures of wild Syrian hamsters, but I do not believe that these are represented in the genotypes of generally available hamsters (Murphy, 1985). HARDERIAN GLAND HISTOLOGY The Harderian glands are tubular glands, consisting of secretory units with lipid-secreting cells arranged around a central lumen. Nuclei are generally spherical and basally located. Cytoplasm is often scanty, as most of the cell is occupied by secretory lipid droplets. Female Harderian glands contain cells whose lipid droplets are typically small. It has been the practice in the past to refer to these as type I cells (e.g., Hoffman, 1971). Ldpez et al. (19931,however, recently pointed out that these cells differ ultrastructurally from type I cells of male hamsters; they used the more noncommittal term “female secretory cell” to describe them. Most of the secretory cells in the female gland are of this type. In male glands, cells whose light-microscopical appearance is identical to those of females are abundant. These are the type I cells. In addition, another type of cell has large secretory lipid droplets and generally more abundant cytoplasm. These are type I1 cells. We have looked in more detail at the secretory lipid droplets in hamster Harderian glands by a morphometric approach. We considered small lipid droplets to be type 1 and large lipid droplets to be type 2. We then examined histologic sections under an ocular graticule and counted (in several fields) the numbers of times the graticule lines intersected over type 1 and type 2 droplets. From these numbers, we calculated the proportion of each lipid droplet type in the section and, by extension, in the gland. Is the Syrian Hamster in Captivity Similar to That in the Wild? Thus, despite its popularity in laboratories and pet stores, all the Syrian hamsters we use are descended from a very small gene pool. Thus, it is possible that the sexual differences we see are due to a founder effect which has bred true. I fear that this doubt will be with us until wild hamsters from the Aleppo region are sampled and compared with captive populations, and I am enough of a realist to doubt that this will happen soon. However, it may be relevant that Harderian glands from the closely related Turkish hamsters (Mesocricetus DIMORPHISM IN HAMSTER HARDERIAN GLAND 1 .o Z 0 F 11L 0 1 .o 135 z Z 0 - k0.5 [L 0 11L rx 0 a 0.5 0 DL a a 0.0 0.0 - 9 0 9 TES LD SD Fig. 1. Effects of castration or short day photoperiods on proportions of type 1(clear bars) and type 2 (cross-hatched bars) lipid droplets in the Harderian glands of male Syrian hamsters. LD, long days (14:lO);SD, short days (8:16). Fig. 2. Effects of testosterone or short day photoperiods on proportions of type 1(clear bars) and type 2 (cross-hatched bars) lipid droplets in the Harderian glands of female Syrian hamsters. LD, long days (14:lO);SD, short days (8:16);TES, given testosterone pellets. The dimorphism in lipid droplets is thus apparent quantitatively. Male glands typically have about 65% type 1 lipid droplets, whereas those of females have about 95% type 1droplets. Androgen is a major control of this. Thus, castration of male hamsters leads to an increase in type 1droplets to 90% after 5 weeks (Fig. 1).On the other hand, beeswax pellets containing testosterone implanted into females for 2 weeks led to a significant reduction in type 1droplets to about 82% (Fig. 2). These observations, while important, are based on experimental procedures which do not occur in nature. Androgen levels and levels of other hormones do, however, fluctuate naturally in hamsters, mediated by naturally occurring decreased day length in the autumn. This is the basis of seasonal reproduction. Syrian hamsters interpret less than 12.5 h of light per day as a short day; serum gonadotrophin, prolactin, thyroid hormone, and sex steroid levels decrease. In males, the testes and accessory sex glands atrophy due to low testosterone levels. Females become anestrous, their ovaries hypertrophy, and their uteri atrophy. These changes reverse spontaneously in the spring. The phenomenon of seasonal reproduction is controlled by the pineal gland, as pinealectomy prevents these changes from occurring (Reiter, 1981). Exposure of male hamsters to short days reduces testosterone levels; however, it does not mimic castration. In fact, the opposite is true. Type I1 cells predominate and morphometric studies confirm that type 1 lipid droplets decreased from 64% in long days to 48% in short days (Fig. 1). Female hamsters react to short days in a similar manner to males. Eight weeks in short days lead to a conspicuous increase in type I1 cells. The percentage of type 1 lipid droplets decreased from 94% to 65% (Fig. 2). To summarize, Harderian lipid droplets in both males and females change their characteristics with the approach of winter, suggesting a function in adaptation to cold weather. This change may be, for instance, in the viscosity of the lipid and suggests that the theory of Thiessen (1992) that Harderian secretions are groomed into the pelage in winter to insulate the animal may be correct. We are studying Harderian lipids in relation to the photoperiod to try to clarify this matter. PORPHYRIN IN HARDERIAN GLANDS High porphyrin concentrations are a feature of most rodent Harderian glands, and the significance of this is unknown. The sex difference in Harderian porphyrin concentrations in Syrian hamsters is truly remarkable. In our experience, male porphyrin levels are in the 4-10 ng/mg tissue range, whereas those of females are in the 1-4 pg/mg range. In other words, female Harderian glands have 100-1,000 times higher porphyrin levels than those of males. This is reflected in the external appearance of the glands (male glands are pale; those of females are dark) and in histology, as female glands show deposits of precipitated porphyrin. Thus, dimorphism is seen by a huge preponderance of porphyrins in female glands over those of males. This is largely controlled by androgens. Thus, castrating males leads to a very rapid and very striking rise in porphyrin concentrations. Within 1 week, there is a tenfold rise in porphyrin concentrations, and within 4-6 weeks porphyrin concentrations near those of fe- G.R. BUZZELL L 0 1 2 3 WEEKS CASTRATED Fig. 3. Effects of castration of up to 3 weeks duration on porphyrin concentrations in the Harderian glands of male Syrian hamsters. Data from Buzzell et al. (1991). males are achieved (Fig. 3) (Buzzell et al., 1991). On the other hand, treating a female hamster with testosterone leads to a decrease in porphyrin concentrations (though not so precipitous nor so striking as the rise seen in males) (Buzzell et al., 1992). Thus, low testosterone levels lead to high porphyrin levels, whereas high testosterone levels have the opposite effect. However, not all procedures which lower testosterone levels lead to elevated porphyrin concentrations, and these procedures may prevent most of the porphyrinogenesis resulting from castration. Procedures with this antiporphyrinogenic effect include hypophysectomy, blinding, and short day photoperiods (Fig. 4) (Buzzell et al., 1994; Hoffman, 1971; Hoffman et al., 1990). These observations suggested that some hormone whose levels are altered by short photoperiod (but not by castration) is necessary for low testosterone levels to induce porphyrinogenesis. Prolactin is such a hormone. Prolactin secretion is unusual amongst pituitary hormones in that its release is chronically inhibited by hormones of the hypothalamus. Dopamine is the major prolactin release inhibiting hormone. Bromocriptine is an ergot alkyloid which is a dopamine agonist; it is used experimentally and clinically to suppress prolactin secretion. We therefore treated hamsters with bromocriptine to determine whether lowering prolactin levels also prevents the porphyrinogenic effects of castration. We found our hypothesis confirmed. Bromocriptine had no effect on Harderian porphyrin levels in intact males. In castrated males, however, bromocriptine prevented most of the rise which follows castration in its absence (Buzzell et al., 1989). Further experiments strengthened this relationship. Removal of the pituitary had a similar effect to short days or bromocriptine: no significant rise in porphyrins, whatever the gonadal state. However, removal of the pituitary and its replacement in the sella turcica or its insertion under the kidney capsule, procedures - - LD LO SD SD INTACT CAS INTACT CAS Fig.4. Effects of castration and photoperiod on porphyrin concentrations in the Harderian glands of male Syrian hamsters.CAS,castrated; LD,long days (1410);SD, short days (816).Data from Buzzell et al. (1994). which restore prolactin secretion while keeping levels of other pituitary hormones low, led to full or partial restoration of Harderian porphyrin levels to those seen after castration (Fig. 5) (Buzzell et al., 1992). Thus, the evidence is good that the control of Harderian porphyrins in male hamsters involves the interaction of at least two hormones: testosterone and prolactin. In addition, we have evidence that thyroid hormones are also involved but in a way which we have not yet sorted out. Both hypothyroid and hyperthyroid conditions prevent the porphyrinogenic effects of castration (Hoffman et al., 1989, 1990). Female porphyrin levels are high and are difficult to alter experimentally. Testosterone treatment, hypophysectomy, hypothyroidism, or hyperthyroidism leads to a drop in female hamster porphyrin levels (Buzzell et al., 1992; Hoffman et al., 1989, 1990; Marrufo et al., 1989). However, none of these treatments leads to levels as low as those of males. Other endocrine paradigms, some of which prevent the rise in porphyrin levels following castration in males (e.g., short days, bromocriptine), have no effect on female porphyrin levels (Buzzell et al., 1989, 1994; Hoffman et al., 1990; Menendez-Pelaez et al., 1992; Rodriguez-Colunga et al., 1991; Spike et al., 1985, 1990). FUNCTIONAL IMPLICATIONS The Harderian gland of Mesocricetus uurutus is unusual in the degree of sexual dimorphism displayed. Striking amongst dimorphic features are porphyrin content and lipid droplet types. Both are under androgenic control; however, they respond to seasonal fluctuations in androgen levels in different ways. Porphyrin levels are kept low in males and high in females, despite seasonal fluctuations in reproductive DIMORPHISM IN HAMSTER HARDERIAN GLAND 0.0 J 1 137 A., Pangerl, B., Vaughan, M.K., and Reiter, R.J. (1989) Bromocriptine prevents the castration-induced rise in porphyrin concentration in the Harderian glands of the male Syrian hamster, Mesocricetus auratus. J. Exp. Zool., 249:172-176. Buzzell, G.R., Menendez-Pelaez,A., Chlumecky, V., and Reiter, R.J. (1991) Gender differences and time course of castration-induced changes in porphyrins, indoles, and proteins in the Harderian glands of the Syrian hamster. Can. J. Physiol. Pharmacol., 69: 1814-1818. Buzzell, G.R., Hoffman, R.A., Vaughan, M.K., and Reiter, R.J. (1992) Hypophysectomy prevents the castration-induced increase in porphyrin concentrations in the Harderian glands of the male golden hamster: A possible role for prolactin. J. Endocrinol., 133:29-35. Buzzell, G.R., Menendez-Pelaez,A., Hoffman, R.A., Rodriguez, C., and Antolin, 1. (1994) The androgenic control of porphyrin in the Harderian glands of the male Syrian hamster is modulated by the photoperiod, which suggests that the sexual differences in porphyrin concentrations in this gland are important functionally. Anat. Rec., 24052-58. v L HPX rAP1 L A Fig. 5. Effects of hypophysectomy (HPX) and removal of the anterior pituitary and its reinsertion into the sella turcica ([API) on porphyrin concentrations in the Harderian glands of castrated male Syrian hamsters. Data from Buzzell et al. (1992). hormone levels. We do not yet know what the function of this porphyrin is. However, the fact that nature has designed a complex homeostatic control system implies that the marked sexual differences in the concentrations of Harderian porphyrin are critical to the functions of both the Harderian glands and their porphyrin in this species. Lipid droplet types, while also being controlled by androgens, are not consistent year round. Thus, both male and female glands have increased amounts of type 2 lipid in the short days of autumn and winter, suggesting that this feature is important to the animal at this time of the year. Again, we do not know the function of either Harderian porphyrin or lipid. Nor do we know how universally applicable these results are. However, I feel that knowledge of the Harderian gland in one species is an important step towards a more general understanding. I hope that this is a contribution toward that goal. ACKNOWLEDGMENTS I am especially indebted to Roger Hoffman and Russel Reiter, my friends and collaborators. Studies in my laboratory are funded by the Canadian Natural Sciences and Engineering Research Council. Several of the studies described here were done in collaboration with the late Armando Menhdez-Pellez. His recent death was a shock t o all of us. The field of Harderian gland research has lost a major player; I have lost a good friend and colleague. REFERENCES Bucana, C.D., and Nadakavukaren, M. (1972) Fine structure of the hamster Harderian gland. Z. Zellforsch., 120178-187. Buzzell, G.R., Menendez-Pelaez,A., Porkka-Heiskanen, T., Pangerl, Christensen, F., and Dam, H. (1953) A sexual dimorphism of the Harderian glands in hamsters. Acta Physiol. Scand., 27:333-336. Coe. J.E. (1977) A sex-limited serum Drotein of Svrian hamsters: Definition of female protein and regula'tion by tesksterone. Roc. Natl. Acad. Sci. U.S.A., 74730-733. Coe, J.E., and Ross, M.J. (1990) Armenian hamster female protein: a pentraxin under complex regulation. Am. J. Physiol., 259R341RXA9 _ _ I I. Hoffman. R.A. (1971) Influence of some endocrine elands. hormones and blinding on the histology of the Harderian-glands of golden hamsters. Amer. J . Anat., 132:463-478. Hoffman, R.A., and Jones, J.W. (1981) Concentrationsof metals in the Harderian glands of male and female golden hamsters. Comp. Biochem. Physiol. 69A153-156. Hoffman, R.A., Johnson, L.B., and Reiter, R.J. (1985) Harderian glands of golden hamsters: Temporal and sexual differences in immunoreactive melatonin. J. Pineal Res., 2:161-168. Hoffman, R.A., Wertz, P., and Habeeb, P. (1989) Harderian glands of golden hamsters: Morphological and biochemical responses to thyroid hormones. J. Comp. Physiol. B, 159:293-299. Hoffman, R.A., Habeeb, P., and Buzzell, G.R. (1990) Further studies on the regulation of the Harderian glands of golden hamsters by the thyroid gland. J. Comp. Physiol. B, 160:269-275. Lin, W., and Nadakavukaren, M.K. (1981) Harderian gland lipids of male and female golden hamsters. Comp. Biochem. Physiol. 70B: 627-630. Upez, J.M., Tolivia, J., Alvarez-Uria, M., Payne, A.P., McGadey, J., and Moore, M.R. (1993) An electron microscopic study of the Harderian gland of the Syrian hamster with particular reference to the process of formation and discharge of the secretory vacuoles. Anat. Rec.,235:342-352. Marmfo, B., Menendez-Pelaez,A., Buzzell, G.R., Gonzalez-Brito, A., and Reiter, R.J. (1989) 5a-dihydrotestosterone administration converts indolamine metabolism and porphyrin content of the female Syrian hamster Harderian gland to the male type. Roc.Soc. Expl. Biol. Med., 192:192-195. McBlain, W.A., Hoffman, R.A., and Buzzell, G.R. (1994) Androgen receptor in the Harderian glands of the golden hamster: Characterization and the effects of androgen deprivation, the pituitary, and gender. J. Exp. Zool., 268442-451. Menendez-Pelaez,A., Reiter, R.J., Guerrero, J.M., Puig-Domingo,M., and Howes, K.A. (1988) Sexual dimorphism in N-acetyltransferase activity, hydroxyindole-0-methyltransferaseactivity, and melatonin content in the Harderian gland of Syrian hamsters: Changes following gonadectomy. Proc. Soc. Exp. Biol. Med., 187287-291. Menendez-Pelaez, A., Rodriguez-Colunga, M.J., Rodriguez, C., Tolivia, D., and Dominguez, P. (1992) Effects of human chorionic gonadotropin and progesterone administration on porphyrin biosynthesis and histology of the Harderian glands in male and female Syrian hamsters. Biol. Reprod., 47:307-315. Murawski, U., Platte, C., and Scotti, A. (1991) A triple methylbranched alkoxylipid as the main component of Harderian gland lipids in the golden hamster (Mesocricetus aumtus). Endocrinolo@a, 38:275-278. Murphy, M.R. (1985) History of the capture and domestication of the Syrian golden hamster (Mesocricetus auratus Waterhouse).In: The Hamster. Reproduction and Behavior. H.I. Siegel, ed. Plenum Press, New York, pp. 3-20. Pangerl, A,, Pangerl, B., Buzzell, G.R., Jones, D.J., and Reiter, R.J. 138 G.R. BUZZELL (1989)Characterization of 6-adrenoceptors in the Syrian hamster Harderian gland Sexual differences and effects of either castration or superior cervical ganglioneetomy.J. Neurosci. Res., 22:456-460. Payne, A.P., McGadey, J., Johnston, H.S., Moore, M.R., and Thomp son, G.G. (1982)Mast cells in the hamster Harderian gland Sex differences, hormonal control and relationship to porphyrin. J. Anat., 135:451-461. Puig-Domingo, M., Guerrero, J.M., Reiter, R.J., Peinado, M.A., Menendez-Pelaez, A., Santana, C., and Webb, S.M.(1988)Androgenic control of immunoreadive somatostatin in the Harderian gland of the Syrian hamster. J. Reprod. Fertil., 82:753-759. Reiter, R.J. (1981)The pineal and its hormones in the control of reproduction in mammals. Endocrinol. Rev., 1:109-131. Rodriguez-Colunga,M.J., Fernandez, C., Antolin, I., Rodriguez, C., Tolivia, D., and Menendez-Pelaez,A. (1991)Chronic administration of melatonin induces changes in porphyrins and in the histology of male and female hamster Harderian gland Interrelation with the gonadal status. J. Pineal Res., 11:42-48. Spike, R.C., Johnston, H.S., McGadey, J., Moore, M.R., Thompson, G.G., and Payne, A.P. (1985)Quantitative studies on the effects of hormones on structure and porphyrin biosynthesis in the Harderian gland of the female golden hamster. 1.The effects of ovariectomy and androgen administration. J. Anat., 14259-72. Spike, R.C., Payne, A.P., Thompson, G.G., and Moore, M.R. (1990) High-performance liquid chromatographic analyses of porphyrins in hamster Harderian glands. Biochim. Biophys. Acta, 10341-3. Thiessen, D. (1992)The function of the Harderian gland in the Mongolian gerbil, Meriones unguicuhtus. In. Harderian Glands. Porphyrin Metabolism, Behavioral and Endocrine Effects. S.M. Webb, R.A. Hoffman, M.L.Puig-Domingo, and R.J. Reiter, eds. SpringerVerlag, Berlin, pp. 127-140. Thompson, G.G., Hordovatzi, X., Moore, M.R., McGadey, J., and Payne, A.P. (1984)Sex differences in haem biosynthesis and porphyrin content in the Harderian gland of the golden hamster. Int. J. Biochem., 16:849-852. Wooley, G.W., and Worley, J. (1954)Sexual dimorphism in the Harderian gland of the hamster (Cricetusaumtuss)Anat. Rec.,118416417.