THE ANATOMICAL RECORD 204333-339 (1982) Morphometric Analysis of Testicular Leydig Cells in Normal Adu It Mice H. MORI, D. SHIMIZU, R. FUKUNISHI, AND A. KENT CHRISTENSEN Department of Patholosy and Centml Research Laboomtory,Ehime UniversiCy School of Medicine, Shigenobu, Ehime, Japan (H.M.,D.S., RE.) and Department of Anatomy and Cell Biology, The University of Michigan Medical School, Ann Arbor, Michigan 48109 (A.K.C.) ABSTRACT Stereological analysis was carried out on Leydig cells in perfusion-fixed testes of normal adult mice. In a decapsulated testis, the seminiferous tubules occupy 89.3% and the interstitial tissue makes up 10.7% of the volume of the testis parenchyma. The Leydig cells comprise 3.8% of testicular volume. There are 24.9 million Leydig cells per cm3 (or gm)of tissue. An average Leydig cell has a volume of 1,533 pm3 and a surface area of 1150 pm2. The smooth endoplasmic reticulum (SER) is the most prominent organelle in the Leydig cells, and has a membrane surface area of 2,428cm2 per cm3 of fresh testis tissue, which is 8.5 times the surface area of the plasma membrane and constitutes 56.9% of the total membranes in Leydig cells. Mitochondria occupy 10.1% of the Leydig cell volume or 11.4% of cytoplasmic volume. The inner mitochondrial membrane (including tubular or vesicular cristae) provides a surface area of about 2855 km2/cell and is 2.26 times that of the outer membrane. There are approximately 712 cm2 of inner membranes per cm3 tissue. Mouse Leydig cells have numerous lipid droplets, which average 147 per cell and occupy 5.1% of the cell volume. Leydig cells of the mammalian testis secrete testosterone, necessary to maintain spermatogenesis in the seminiferous tubule and to regulate male target tissue throughout the body. The cytoplasm of Leydig cells characteristically contains an abundant smooth endoplasmic reticulum, known to be the site of many enzymes of testosterone synthesis. Other important steroidogenicenzymes are found on the inner mitochondria1membrane. Lipid droplets contain cholesterol esters that may act as substrates for androgen biosynthesis. For a review of Leydig cell structure and function, see Christensen (1975). In correlating structure with function in Leydig cells, it is important to have accurate quantitative information on cell numbers and on the volumes and membrane surface areas of the various organelles, especially those involved in testosterone synthesis. Modern methods of morphometry (reviewed by Weibel and Bolender, 1973)allow such information to be gained from light and electron micrographs. We have previously published rather extensive morphometric studies on Leydig cells in the testes of normal adult rats (Mori and Chris0003-276X/82/2044-0333$02.50 0 1982 ALAN R. LISS, INC. tensen, 1980),guinea pigs (Mori et al., 19801, and elderly men (Mori et al., 1982).In the present study we extend these investigations to the Leydig cells of normal adult mice. Although the Leydig cells of several species of mammals have been the subject of morphometric investigation (see Discussion, and Christensen and Peacock, 19801,this to our knowledge is the first such study in the mouse. MATERIALS AND METHODS Five young adult male mice (CD-1strain, 78 days old, from Charles River Breeding Labs, Wilmington, MA) were used after one week of acclimation. They were maintained on usual lab chow and water ad libitum. The right testis was fixed by perfusion through the thoracic aorta with 3% glutaraldehyde buffered with 0.1 M s-collidine,pH 7.4, for 15 minutes at room temperature under ether anesthesia. After perfusion, the testes were cut into six slices, perpendicular to the long axis of the testis, and were washed in buffer overnight or longer. Alternating slices were utiReceived July 6, 1982; Accepted Aug 27, 1982. 334 H. MORI ET AL. lized for light and electron microscopy, making three slices for each. Subsequent procedures to prepare the samples for stereology were essentially the same as described previously (Mori and Christensen, 1980;Christensen and Peacock, 1980). A detailed account of stereological theory and practice used in this study can be found in review articles (e.g., Weibel and Bolender, 1973) and in one of our previous papers (Mori and Christensen, 1980)."he volume of the interstitial tissue of the testis and the number and volume of Leyig cells were measured at the light-microscope level, while the number, volume and surface area of Leydig cell organelles were analyzed a t the electron-microscopelevel. Volume densities were determined by pointcounting and were expressed as the volume of a particular structure per unit volume within a specified reference space. The stereological analysis of mouse Leydig cells required four sampling stages, since the cellular components exhibited a broad range of size and frequency (Table 1). In stages I and 11, numerical (Nv) and volume (Vv) densities were estimated on light-microscope sections by viewing the specimen through an eyepiece grid containing a square lattice of 441 points in an area equal to 1 cm2.Numerical densities were calculated by means of the Floderus (1944)equation. In stages 111and IV,numerical and volume densities of Leydig cell organelles were estimated, using a square, double-lattice test sheet (1:4/ 108:432),with an area equal to 442 cm2,placed over the electron-microscope prints. For the measurement of surface density (Sv), a coherent multipurpose grid was used, containing 90 test points and 45 test lines of 2.42 cm, providing a total intersection length of 109 cm. Stereological analysis produces intrinsic systematic errors which yield an overestimation or an underestimation of the values, de- Fig. 1. Low-power light micrograph from a mourn testis fixed by perfusion with s-collidine-bufferedglutaraldehyde and embedded in glycol methacrylate. Tissue relationships are well preserved, providing favorable material for morphometry. x 80. Fig. 2. Mouse Leydig cells seen in this light micrograph exhibit a densely stained cytoplasm. Most of the cells contain numerous lipid droplets.Arrows indicate mamphages. X 1,050. 335 MORPHOMETRY OF TESTICULAR LEYDIG CELLS TABLE 1. Sampling stages of stereology on mouse Leydig cells Stage I Stage I1 Stage I11 Stage IV Electron microscopy x 10,250 SV of Leydig cells N v and Vv of organelles 60 micrographs Electron microscopy ~83,200 Sv and Vv of organelles Level Magnification Density* and components determined Light microscopy x 200 VV of interstitial tissue Light microscopy x 400 Nv and Vv of Leydig cells Size per animal 40-60 fields 100 fields 60 micrographs 'Nv, Vv, Sv. Number, volume and surface area per unit volume, respectively pending on the conditions. These errors have been corrected according to methods outlined in detail elsewhere (Mori and Christensen, 1980; Weibel and Paumgartner, 1978). RESULTS TABLE 2. Basic data on mice examined Animal No. 1 2 Body weight (g) 37 33 Testis weight Testis volume (g) km3) after perfusion fixation 0.130 0.128 0.125 0.123 The perfusion-fixed testicular tissue of the 3 36 0.128 0.123 -_ present study is well preserved and would thus 4 0.119 33 0.124 0.122 0.127 34 appear to be favorable for stereological anal- 5 ysis. Tissue relationships are well maintained 0.127 0.122 34.6 at the light-microscope level (Figs. 1,2), with Mean +0.001 10.001 k0.8 fS.E. little artifactual expansion of intercellular Fixed testis volume was estimated by dividing the testis weight space. Organelles viewed in Leydig cells at the by specific gravity of 1.042 k 0.0008 (N = 8), whereas the electron microscope level (Figs. 3,4) show no specific gravity for fresh testis was 1.055 f 0.0009 (N= 4). apparent artifactual changes, and their membranes are clearly visible. The fine structure was essentially as has been described for mouse Leydig cells by Christensen and Fawcett (1966). The testes used in this study weighed 0.127 The smooth endoplasmic reticulum (SER), 2 0.001 gm (mean SEMI and were estimated to have a volume of 0.122 f 0.001 cm3 an important site of steroidogenic enzymes (Christensen, 19751, has a volume of 103 pm3 (Table 2). Our data, listed in Table 3, are expressed in in an average Leydig cell, constituting 6.7%of terms of three stereological parameters, based the total cellular volume. These values are on fresh testis (without the capsule): Numer- measurements of the organelle itself, namely, ical density (N = the number of structures per the volumes of the membrane and the contents, cm3 of fresh testis), volume density (V = the and do not refer merely to the volume of genvolume of a structure per cm3 of fresh testis), eral areas of the cytoplasm in which SER preand surface density (S = the surface area of a dominates. The surface area of the SER is 2428 structure per cm3of fresh testis). Since one cm3 cm2/cm3tissue. This amounts to a surface area of fresh testis weighs 1.055 gm,the value per of 9736 p,m2 for the SER of a n average Leydig cm3 is essentially the same as the value per cell, which is 56.9%of the total membrane surface area in the cell and is 8.5 times greater gm of fresh testis. In a decapsulated testis, the testicular tissue than that of the plasma membrane. Mitochondria are also sites of enzymes imconsists of 89.3% seminiferous tubules and 10.7%interstitial tissue. The Leydig cells oc- portant in testosterone biosynthesis in the cupy 3.8%of total volume in the decapsulated Leydig cell (Christensen, 1975). The average testis. One cubic centimeter of mouse testis mitochondrion is 0.43 pm in diameter and 2.72 contains 24.9 f 1.28 million Leydig cells, on pm long. There are 605 mitochondria in a n the average, which means that a 35-gm mouse average Leydig cell, comprising a volume of would have about 6 million Leydig cells in both 156 pm3 per cell, which is 10.1% of the cell volume or 11.4%of the cytoplasm. The surface testes. An average Leydig cell has a volume of 1533 area of the outer mitochondria1 membrane is pm3and a surface area of 1150 pm2.Its nucleus 315 crn2/cm3tissue or 1262 cLm2/cell,while that has a volume of 161 pm3, constituting 10.5% of the inner membranes (including the tubular cristae) is 712 cm2/cm3tissue or 2855 pm2/cell. of the cell volume. ~ * 336 H. MORI ET AL 337 MORPHOMETRY OF TESTICULAR LEYDIG CELLS TABLE 3.Stereological data on m o u e Leydig cells Component Seminiferous tubules Interstitial tissue Leydig cells Parameter' V V N V S Nucleus Cytoplasm V V Endoplasmic reticulum Smooth Rough Nuclear envelope Golgi complex S Mitochondria N V Outer membr Inner membr Peroxisome Lysosome s S N V S N V Multivesicular bodies S N V Lipid droplets N V Cytoplasmic matrix V S S Mean value/ cm3 tissue SEM (n = 5) 0.8931 0.1069 24.94 x lo6 0.03822 286.82 0.00401 0.03421 0.00379 0.00379 1.28 x los 0.001528 18.700 0.000276 0.001388 0.00258 2428.13 0.00021 222.96 0.00016 38.65 0.00010 64.36 15.09 x 109 0.00388 314.63 712.19 35.96 x 109 0.00025 56.61 4.95 x 109 0.00048 41.89 1.75 x 109 0.00005 9.29 3.66 x 109 0.00194 93.76 0.02456 0.000136 119.80 0.000021 21.000 0.000015 3.304 0.000027 21.074 0.970 x lo9 0.000155 12.287 30.339 2.598 X log 0.000015 4.550 0.187 X log 0.000024 4.031 0.229 X log 0.000004 1.384 0.263 x 109 0.000292 13.411 0.001148 Per average Leydig cellZ 1533 1150 161 1372 103 9736 9 893 6 155 4 258 605 156 1262 2855 1442 10 227 198 19 168 70 2 37 147 78 376 985 'Dimension of parameters (Note: It is possible to substitute gm approximately for em3 throughout these parameters, since the specific gravity of mouse testis tissue [ = 1.0551 is near unity): Number (N):No./cm3and No./cell, respectively.Volume (V): cms/cm3and pms/cell, respectively. Surface area (S):cmz/cm3and pm2/cell,respectively. a'rhe values per cell are obtained by dividing the value per cm3 tissue by the Leydig cell number per cm' tissue. The surface areas of the outer and inner mitochondrial membranes constitute 7.4% and 16.7%,respectively, of the total membranes of the Leydig cells, and thus are 1.1and 2.5 times, respectively, greater than that of the plasma membrane. The inner membrane (including cristae) has a surface area 2.26 times greater than the outer mitochondria1 membrane. Fig. 3. Low-power electron micrograph of mouse Leydig cells, illustrating the quality of preservation of the material used for morphometry in this study. x 6,300. Fig. 4. Mouse Leydig cells are characterized by a welldeveloped smooth endoplasmic reticulum (SER), numerous lipid droplets (Lip), and mitochondria with vesicular and tubular cristae (Mit). Gol, Golgi complex; Per, peroxisome; Lys, primary lysosome; Mul, multivesicular body; Gly, glycogen. x 19,500. The Golgi complex constitutes only 0.3% of cell volume. Again, this is a measurement of the organelle itself (membrane plus contents), whereas a separate count shows that regions of cytoplasm in which the Golgi complex occurs constitute approximately 1.9% of cell volume. Since the Golgi complex is generally included with the SER in conventional fractionation centrifugation, we will give some combined values. The SER plus Golgi complex has a volume of 107 ~m~ per average Leydig cell, constituting 7.0% of the cell volume or 7.8% of cytoplasmic volume. The combined surface area of SER plus Golgi is approximately 10,000 Fm2 per cell, which is 8.7 times that of the plasma membrane or 58.4% of the total membranes of the average Leydig cell. The rough endoplasmic reticulum (RER)and the perinuclear cisternae occupy 0.6% and 0.4% 338 H. MORI ET AL. of the total cell volume, respectively, and comprise 5.2%and 0.9%of the total membranes of the average Leydig cell. Leydig cells in mouse testis have numerous lipid droplets. In an average Leydig cell, there are 147 lipid droplets, with a volume of 78 pm3, constituting 5.1%of the cell volume or 5.7%of cytoplasmic volume. The average lipid droplet has a diameter of 1.07 pm and a volume of 0.64 km3. Peroxisomes or microbodies, identified by the criteria described in a previous paper (Mori and Christensen, 1980), have an average diameter of 0.28 pm. An average Leydig cell has 1,442 peroxisomes, which occupy 0.7% of the cell volume. Each cm3 of testis tissue contains about 36 billion Leydig peroxisomes. Lysosomes occupy 1.3%of the cell volume. About one-third of this volume is comprised of primary lysosomes, with a n average diameter of 0.63 pm, while the other two-thirds is made up by secondary lysosomes (residual bodies and autophagic vacuoles), with an average diameter of 0.71 pm. Multivesicular bodies occupy only 0.1% of the cell volume, and have an average diameter of 0.41 pm. There are about 70 multivesicular bodies in the average Leydig cell. The cytoplasmic ground substance lying between major organelles constitutes 64.2% of cell volume or 71.8% of the cytoplasmic volume. It contains free ribosomes and polysomes, microfilaments, microtubules, glycogen, and other small cytoplasmic components, as well as soluble materials constituting the “cytosol” of cell fractionation. DISCUSSION The findings of this study provide quantitative information on the number, volume, and surface area of Leydig cells and their organelles in the testes of normal adult mice. In this work we have sought absolute values, determined as accurately as we are able by current stereological methods and necessary corrections. We hope these findings may be useful to biochemists and other researchers who want to know the number of Leydig cells per gm of fresh testis tissue, the surface area of the smooth endoplasmicreticulum per gm of tissue, or other values of potential use in biochemical or physiological studies on the testis. A prominent smooth endoplasmic reticulum is one of the characteristic features of steroidsecreting cells, and is known to be the site of a majority of steroidogenic enzymes (see Christensen and Gillim, 1969; and Christensen, 1975, for reviews).Early observations on Leydig cells with the electron microscope showed a conspicuous species variation in the abundance of the SER. Species with a particularly abundant SER included the guinea pig (Christensen, 1965) and opossum (Christensen and Fawcett, 19611, while the hamster (Wing and Lin, 1977) exhibited the least development of this organelle. Among species with an intermediate complement of SER in the Leydig cells were the mouse (Christensen and Fawcett, 1966), rat (Christensen and Gillim, 1969), and human (Christensen, 1975). The SER in canine Leydig cells proved to be highly variable in amount from cell to cell (Connell and Christensen, 1975). Zirkin et al. (1980) have confirmed stereologically the relative abundance of SER in the Leydig cells of the guinea pig, rabbit, dog, rat, and hamster, and have shown that the amount of SER is proportional to the level of testosterone secretion per gm of Leydig cells. Their study did not include the mouse, but it would be desirable to show how our data on the SER in mouse Leydig cells relate to their comparative findings in the other species. However, this comparison is difficult if not impossible because of differences in the goals of the two studies. In fact, it is equally difficult to relate our previous stereological studies on Leydig cells in rat (Mori and Christensen, 1980),guinea pig (Mori et al., 1980) and human (Mori et al., 19821, with the findings reported by Zirkin et al. (1980), for the same reason. As they point out in the Discussion of their paper, they do not claim that their data represent absolute values, since the findings are uncorrected for section thickness o r for tissue shrinkage during processing. As a n example of disparity between their findings and ours, they note that their value for the volume density of the SER in rat Leydig cells was almost three times as large as ours (Mori and Christensen, 1980). That degree of difference might well be expected as a possible result of two effects: First, the necessary correction for section thickness (Weibel and Paumgartner, 19781, carried out in our study but not in theirs, would reduce their uncorrected value by about half. In addition, the SER in their study may in some cases have been artifactually swollen, judging by Figure 3 of their paper. This moderate increase in SER tubule diameter would greatly increase the measured volume density of that organelle, since volume varies as the cube of the diameter. As a result of these differences in approach between our studies and the paper of Zirkin et al. (19801, it is difficult for us to relate their values to our present findings on mouse Leydig cells or our past studies on these cells in rat, guinea pig or human. 339 MORPHOMETRY OF TESTICULAR LEYDIG CELLS TABLE 4 . Surface densities of membranes involved in testosterone synthesis (em2 membrane per em3Leydig cell cytoplasm) Species Smooth ER membranes Inner mitochondria1 membranes (including cristae) Reference Mouse Aged human Rat Guinea pig 70,977 82,984 94,767 102,457 20,818 15,140 27,563 23,515 Present study Mori et al., 1982 Mori and Christensen, 1980 Mori e t al., 1980 Morphometric data from our present and previous studies are summarized in Table 4, comparing the surface areas of membranes of the SER and inner mitochondrial membranes (including cristae), which bear most of the enzymes of testosterone biosynthesis in Leydig cells of mouse, human, rat, and guinea pig. These are corrected absolute values, and are expressed in terms of surface area per unit volume of Leydig cell cytoplasm. Since the steroidogenic enzymes are tightly bound to the membranes, the membrane surface area is a much better gauge of enzyme concentration than volume density of the organelles. The value of the inner mitochondrial membranes, site of enzymes for cholesterol sidechain cleavage, is similar from species to species, although somewhat lower in the Leydig cell of aged humans. As expected, the smooth endoplasmic reticulum is most abundant in the guinea pig, although the relative amount is not as much greater than that of the rat as might have been expected, especially compared to the value reported by Zirkin et al. (1980). Whether this relates to methodological differences or is due to a difference in strains of guinea pigs available in Japan is unclear. LITERATURE CITED Christensen, A.K. (1975)Leydig cells. In: Handbook of Physiology. D.W. Hamilton and R.O. Greep, eds. American Physiological society, Washington, D.C., Section 7,5:57-94. Christensen, A.K., and D.W. Fawcett (1966) The fine structure of testicular interstitial cells in mice. Am. J. Anat., 118:551-572. Christensen, A.K., and S.W. Gillim (1969) The correlation of fine structure and function in steroid-secreting cells, with emphasis on those of the gonads. In The Gonads. K.W. McKerns, ed. Appleton-Century-Crofts, New York, pp. 415-488. Christensen, A.K., and K.C. Peacock (1980)Increase in Leydig cell number in testes of adult rats treated chronically with a n excess of human chorionic gonadotropin. Biol. Reprod., 22r383391. Connell, C.J., and A.K. Christensen (1975) The ultrastructure of the canine testicular interstitial tissue. Biol. Reprod., 12:368-382. Floderus, S. (1944) Untersuchungen uber den Bau der menechlichen Hypophyse n i t besonderer Beriicksichtigung der quantitativen mikromorphologischen Verhaltnisse. Acta Pathol. Microbiol. Scand. (Suppl.), 53:l-276. Mori, H., and A.K. Christensen (1980) Morphometric analysis of Leydig cells in the normal rat testis. J. Cell Biol., 84:340354. Mori, H., N. Hiromoto, M. Nakahara, andT. Shiraishi (1982) Stereological analysis of Leydig cell ultrastructure in aged humans. J. Clin. Endocrinol. Metab., in press. Mori, H., D. Shimizu, A. Takeda, Y. Takioka, and R. Fukunishi (1980) Stereological analysis of Leydig cells in normal guinea pig testis. J. Electron Microsc. (Tokyo), 29:8-21. Weibel, E.R., and R.P. Bolender (1973) Stereological techniques for electron microscopic morphometry. In: Principles and Techniques of Electron Microscopy. M.A. Hayat, ed. Van Nostrand Reinhold Co., New York, 3~237-296. Weibel, E.R., and D. Paumgartner (1978) Integrated stereological and biochemical studies on hepatocytic membranes. 11. Correction ofsection thickness effect on volume and surface density estimates. J. Cell Biol., 77:584-597. Wing, T.-Y., and H.-S. Lin (1977) The fine structure of testicular interstitial cells in the adult golden hamster with special reference to seasonal changes. Cell Tissue Res., 183:385-393. Zirkin, B.R., L.L. Ewing, N. Kromann, and R.C. Cochran (19801 Testosterone secretion by rat, rabbit, guinea pig, dog, and hamster testes perfused in vitro: Correlation with Leydig cell ultrastructure. Endocrinology, 107:1867-1874.