The fine structure of the monkey (Macaca) sertoli cell and its role in maintaining the blood-testis barrier.код для вставкиСкачать
The Fine Structure of the Monkey (Macaca) Sertoli Cell and Its Role in Maintaining the Blood-Testis Barrier ’ MARTIN DYM Department of Anatomy, and Laboratory of Human Reproduction and Reproductive Biology, Harvard Medical School, Boston, Massachusetts 02115 ABSTRACT The monkey Sertoli cell, a tall columnar cell, extends from the basement membrane of the seminiferous epithelium to the tubule lumen. Its nucleus occupies a basal position and reveals extensive nuclear envelope infoldings. A zone of fine filaments, approximately 0.5 irc in thickness, invests the nucleus and appears to prevent other cell organelles from approaching it. The basal cytoplasm is characterized by numerous mitochondria and abundant smooth endoplasmic reticulum. Lipid droplets, 3 to 4 P in diameter, membrane-limited dense bodies of various shapes and densities, Golgi cisternae, scattered free ribosomes and parallel profiles of rough endoplasmic reticulum are common. The more apical portions of the cell contain longitudinally oriented microtubules and rod-shaped mitochondria, but other organelles are rare. The seminiferous tubules of monkeys are surrounded by three to five circumferentially arranged cells that overlap each other but are separated by intercellular spaces of a t least 300 to 400 A. Tracers such as horseradish peroxidase and lanthanum nitrate injected intravascularly readily pass between the peritubular cells and enter the germinal epithelium. Within the epithelium the tracers outline the spermatogonia and early spermatocytes by permeating the surrounding intercellular spaces. Further penetration toward the tubule lumen is effectively prevented by the occluding tight junctions joining adjacent Sertoli cells. Thus, in monkeys the peritubular epithelioid cells do not impede vascularly introduced tracers from penetrating into the germinal epithelium. The only morphological component of the blood-testis barrier in the macaque appears to be the SertoliSertoli occluding junction. The original work of Sertoli (1865) and a number of articles since that time (Brown, 1885; Regaud, ’00; Winiwarter, ’12; Rolshoven, ’44; Elftman, ’50, ’63; Nishida, ’54) established that the transformations taking place within the germ cell line occur in close association with the Sertoli cell. No doubt the spermatogenic process is very largely dependent upon this ramifying cell type, yet evidence of the way in which its control over the germ cells is exerted remains elusive. In mammals the Sertoli cell is found evenly distributed in the seminiferous epithelium extending from the basement membrane to the tubule lumen. The general cytological characteristics observed with the light miANAT. REC.. 175: E39-656. croscope suggest that this non-germinal element is functionally very active. Electron microscopic studies provided additional morphological data but added little conclusive experimental evidence which could assist investigators in understanding its overall role in spermatogenesis. The idea that remains prevalent today was summed up in the following way by Nishida (’54) : “diffusion currents of nutritive substances occurred in the Sertoli cell from base toward lumen, which nourished the germ cells.” More recently experimental evidence in rats and guinea Received Sept. 6, ‘72. Accepted Nov. 15, ’72. 1 Supported by contract NIH 69-2107 from the National Institute of Child Health and Human Development. 639 640 NIARTIN DYM pigs has suggested a new role for the mammalian Sertoli cell; i.e., the maintenance of the blood-testis barrier and the partitioning of the seminiferous epithelium into a basal compartment containing the spermatogonia and early spermatocytes and a n adluminul compartmepnt containing the more advanced germ cells (Dym and Fawcett, '70; Fawcett et al., '70; Dym, '72). Vascularly injected electron-opaque markers such as lanthanum nitrate readily fill the lymphatic channels and interstitial spaces in the testis and enter the basal compartment of the seminiferous epithelium but are prevented from reaching the tubule lumen mainly by the occluding junctions between adjacent Sertoli cells (Dym and Fawcett, '70). Elegant physiological studies have established that plasma proteins are found in very low concentrations in the fluid within the lumina of rat seminferous tubules (Tuck et al.. ' 7 0 ) , and it is likely that the proteins are excluded by the junctional complexes between Sertoli cells. On the other hand, Reddy and Svoboda ('67), in abstract form, and Aragon, Lustig and Mancini ('72) reported that in mice intravenously injected peroxidase was localized within the cytoplasm of the Sertoli cells and germ cells and did penetrate the blood-testis barrier to reach the tubule lumen. The purpose of this study was to determine whether there is a blood-testis barrier to vascularly infused horseradish peroxidase and lanthanum nitrate in the monkey and, if so, to localize its morphological site. In view of the contradictory report by Aragon, Lustig and Mancini ('72) on mice, the blood-testis barrier in this species was also investigated using horseradish peroxidase. In addition, the fine structure of the monkey Sertoli cell was examined in the different stages of the cycle of the seminiferous epithelium in a n attempt to correlate, if possible, Sertoli cell organelle modifications with specific developmental events of the germ cells. MATERIALS AND METHODS A total of six adult male monkeys (Macaca mulatta and Macaca n e m e s t r i n a ) were used in these experiments. The testes were fixed by vascular perfusion using a technique first described by Christensen ('65). Under deep anesthesia the processus vaginalis was opened, the remnant of the gubernaculum cut, and the testis and spermatic cord mobilized. Two ligatures (no. 0 surgical silk) were tied around the cord approximately two inches proximal to the testis. The spermatic cord was cut between the ligatures and the testis was removed. A 22 gauge needle at the end of Intramedic polyethylene tubing PE190, attached to a reservoir 130 cm high, was inserted under the tunica albuginea and into the testicular artery at the upper pole of the testis. The pampiniform plexus of veins was incised to permit egress of blood and perfusate. Ten cm" of .9% saline followed by 200 cm" of 5 % glutaraldehyde buffered with 0.2 M s-collidine were permitted to flow through the testis over a 30 minute period. One millimeter cubes of tissue were then cut with a razor blade and immersed for a n additional 60 minutes in the same fixative. Secondary fixation was carried out in a solution containing two parts of 2% osmium tetroxide and one part of 0.2 M s-collidine. The tissue was dehydrated in alcohol and embedded in Epon. One micron sections stained with toluidine blue were used to identify the stages of the cycle of the seminiferous epithelium as described by Clermont ('69). Thin sections exhibiting silver to pale gold interference colors were cut with a diamond knife and examined on a Siemens Elmiskop I or RCA EMU 3G electron microscope. One major source of variability in descriptions of the fine structure of the seminiferous epithelium has been the lack of uniformity in methods of fixation. When pieces of testicular tissue are immersed in the fixative the ultrastructural preservation of the Sertoli cell is often inadequate. However, fixation by perfusion has markedly improved the quality of preservation of the monkey seminiferous epithelium. Tracer studies The lanthanum technique employed in these experiments was similar to that described previously (Dym and Fawcett, '70) : A solution of 4% lanthanum nitrate was adjusted to pH 7.7, using 0.1 N NaOH, and added to a n equal volume of glutaraldehyde buffered collidine (.2 M ) . The BLOOD-TESTIS BARRIER final concentration of the lanthanum was 2% and that of the glutaraldehyde 5 % . In monkeys this mixture was perfused into the testicular artery as described above. In order to use peroxidase as an intercellular tracer, one 10 kg monkey, while lightly tranquilized was injected in the femoral vein with 2 gm of Sigma type I1 horseradish peroxidase dissolved in 35 cm3 of distilled HzO. In mice 20 mg of horseradish dissolved in .5 cm3 of H,O was slowly injected into the tail vein. Fifteen minutes after infusion of peroxidase the animals were anesthetized, the testes were removed, and small pieces were fixed for three hours in Karnovsky's fixative ('65). Following an overnight wash in cacodylate buffer, 50 to 100 p sections were prepared on a Smith and Farquhar Tissue Sectioner and reacted in diaminobenzidine tetrachloride and hydrogen peroxide using the method of Graham and Karnovsky ('66). Subsequently the tissue was osmicated in 1.3% osmium tetroxide, dehydrated in graded alcohols, and embedded in Epon. RESULTS Architecture of the monkey Sertoli cell Sections of seminiferous tubules embedded in paraffin, stained with routine histological dyes, and examined with the light microscope reveal much information concerning the developmental patterns of the germ cells. However, using these same techniques the Sertoli cell cytoplasm often remains inconspicuous, crowded in between the more numerous germ cells; therefore, it has proved difficult both to examine the finer morphology of this cell type under normal conditions and to study its response to experimental treatments. Examination of favorably oriented one micron sections of Epon embedded tissue permits greater resolution and reveals that the Sertoli cell is a tall columnar cell extending from the base of the seminiferous epithelium to the tubule lumen. It consists of a stem resting on the basal lamina, an intermediate portion which provides lateral processes around which the spermatocytes and spermatids are arranged, and apical projections which enclose the late spermatids just prior to their release into the tubule lumen. Stages of the cycle 641 as defined by Clermont ('69) are readily identifiable in these 1 I.1 toluidine blue stained sections (figs. 1, 2). In Sertoli cells of monkeys the basal portion of the cytoplasm is voluminous and a large variety of the common cell organelles is evident (figs. 3, 4). Lipid droplets, 3 to 5 ;FL in diameter, are common and their size and number do not appear to vary in the different stages of the cycle of the seminiferous epithelium. Numerous spherical and elongated mitochondria exhibiting transverse cristae of orthodox configuration are present in the basal portion of the cell, while elongated mitochondria predominate in the more apical portions. Perhaps the most characteristic feature of the basal Sertoli cell cytoplasm is the abundant profiles of smooth endoplasmic reticulum. Often these are aligned in parallel arrays near the nucleus. Isolated patches of flattened profiles of rough endoplasmic reticulum are present, and occasionally several such cisternae can be found arranged circularly around a single lipid droplet. Mitochondria and abundant smooth ER are frequently observed in the immediate vicinity of the lipid. Free ribosomes are scattered throughout the Sertoli cell cytoplasm. Membrane limited bodies of various sizes, shapes, and densities are characteristic features. Their origin, turnover rate, enzymatic content and ultimate fate all remain obscure. But it is likely that many of these figures are products of the degenerating germ cells and residual bodies in the seminiferous epithelium. Each Sertoli cell possesses a well developed Golgi apparatus, and often stacks of Golgi cisternae are visible in a single thin section at varying distances from each other. This suggests that a single Sertoli cell may contain several non-connected Golgi profiles. In other cell types the Golgi apparatus is the main building site for a variety of large carbohydrates. The role of the extensive Golgi in the Sertoli cell still remains to be elucidated. Electron microscopic radioautographic studies using labeled sugars such as galactose, mannose, or fucose may provide clues concerning the substances which are elaborated or packaged by this cell type. The intermediate and apical portions of 642 MARTIN DYM Figs. 1 and 2 Light micrographs of 1 + sections of monkey seminiferous tubules showing germ cells in various stages of development and Sertoli cells. The schematic drawings on the left (Clermont, '69) depict the stage of the spermatogenic cycle. x 637. the columnar Sertoli cells contain rodshaped mitochondria and numerous longitudinally oriented microtubules. Other organelles are rarely observed in large numbers in these regions. The Sertoli cell nucleus is large and is characterized by a homogenous nucleoplasm and a distinctive nucleolus when viewed with the light microscope. It occupies a basal position in the cell. In mon- BLOOD-TESTIS BARRIER 643 Fig. 3 A n electron micrograph of the basal portion of a monkey Sertoli cell. Note the filamentous zone surrounding the nucleus and the abundant agranular reticulum and mitochondria. Stage 111. x 8190. 644 MARTIN DYM Fig. 4 A n electron micrograph of a monkey Sertoli cell demonstrating deep infoldings of the nuclear envelope. A prominent Golgi apparatus and membrane-bounded dense bodies are apparent. Note the filaments in ihe inner cell layer of the tunica propria. Stage VII. x 8190. BLOOD-TESTIS BARRIER keys the nuclear envelope is elaborately infolded and fine structural analysis often reveals bizarre shapes (fig. 5). Nuclear pores are very numerous, fairly evenly distributed, and usually separated by a distance of approximately 300 A. Neither the position of the nucleus in the epithelium nor the architecture of the nuclear lobulations in the monkey appears to change in the various stages of the cycle of the seminiferous epithelium. Similarly the number of nuclear pores and their distribution over the Sertoli cell nuclear surface seem to be relatively constant along the length of the seminiferous tubules. A sheath of 70 A filaments, 0.5 cL in thickness, completely surrounds the Sertoli cell nucleus and prevents other cellular organelles from approaching it (fig. 3). The function of this filamentous zone still remains obscure but it may provide rigidity to the pleomorphic nucleus. Similar sized filaments are also scattered throughout the Sertoli cell cytoplasm, and frequently a basal web is preserved adjacent to the basal lamina (fig. 3 ) . Residual bodies (fig. 6 ) and degenerating germ cells (fig. 7 ) are identifiable with the electron microscope, and they appear to be membrane-bounded inside the cytoplasm of the Sertoli cell. An early sign of degeneration of the germ cells is the disappearance of all its membranes including the plasmalemma. Careful examination of the neighboring Sertoli cell cytoplasm fails to reveal any local changes in population of organelles, such as an increase in concentration of lysosomes, which could be correlated with the germ cell degeneration or residual body phagocytosis. Near the base of the seminiferous epithelium three types of junctional complexes join adjacent Sertoli cells to each other. These are essentially identical to the Sertoli cell junctions described previously in the rat by Dym and Fawcett ('70) and include: 1. A series o f tight occluding junctions which extend for a considerable length in either direction perpendicular to the plane of a section. 2. Nexuses or gap junctions with a 20 A interspace. 3. Narrowing of t h e intercellular space t o 70 A without local specialization of cell surfaces. The junctional complexes between Sertoli 645 cells are characterized by subsurface filaments hexagonally packed in bundles which run parallel both to the cell surface and to cisternae of endoplasmic reticulum subjacent to the filaments. T h e tunica propria of monkey seminiferous tubules The tunica propria of the monkeys is markedly different from that found in rodents and more closely resembles the pattern in humans. Whereas in the rodents two layers of flattened cells surround the seminiferous tubules, monkeys exhibit three to five circumferentially arrayed cell layers each one overlapping the other but separated by intercellular spaces of at least 300 to 400 A (fig. 8). On no occasion are occluding junctions observed between neighboring cells. The innermost cell layer immediately subjacent to the seminiferous tubules of subhuman primates contains abundant filaments and is probably contractile, whereas the more peripherally placed cells exhibit fewer filaments and are fibrocyte-like. This is in contrast to rodents where the outer cell layer is the endothelium of large interstitially placed lymphatics. Tracer experiments The horseradish peroxidase injected intravenously in vivo and the lanthanum nitrate perfused agonally with the fixative, exhibited a distribution which was essentially identical in all respects. Both of these electron opaque markers are intercellular tracers and normally do not enter into the cytoplasm of cells except in small quantities via pinocytotic vesicles. It is generally agreed that if vascularly infused peroxidase is found within the cytoplasm of cells in large amounts, this is an artifact. Although such artif acts occasionally occur with the recommended dose, artifacts are produced consistently when ten times the recommended dose of peroxidase is used. Rodent seminiferous tubules are easily teased apart in the fresh native condition (Christensen and Mason, '65; Dym and Clermont, '70), and indeed the intertubular areas consist of a very loose connective tissue and large pervasive lymphatic sinusoids. This may account for the rela- 646 MARTIN DYM Fig. 5 A n electron micrograph of a monkey Sertoli cell nucleus demonstrating the bizarre nuclear pattern. Note the filaments surrounding the nucleus. Stage X. x 22,295. BLOOD-TESTIS BARRIER 647 Fig. 6 An electron micrograph of a portion of a monkey seminiferous tubule showing four residual bodies of Regaud within the Sertoli cell cytoplasm. Note the parallel profiles of granular endoplasmic reticulum in the upper left. Stage VII. x 10,920. 648 MARTIN DYM tive ease with which exogenously added proteins such as horseradish peroxidase reach the seminiferous tubules. On the other hand, monkey and human tubules (Barn, '67) are tightly packed together and do not permit such an easy separation. In primates morphological examination of the interstitial tissue reveals abundant dense collagen fibrils. The intertubular lymph channels, more typical of such vessels in general (Fawcett et al., '73), are lined by a single layer of flattened endothelial cells and contain a fine protein precipitate. Nevertheless, in the monkey experiments the vascularly introduced tracers leave the capillaries, fill the interstitial spaces and easily enter the seminiferous epithelium by passing between the circumferential peritubular cells. In the seminiferous epithelium the tracers demarcate the spermatogonia and early spermatocytes by occupying the interspaces between these cells and the Sertoli cells, as in rodents (figs. 10, 1 1 ) . Further penetration toward the tubule lumen is effectively prevented by the occluding junctions between the SertoIi cells. The tunica propria in monkeys does not appear to prevent or retard the penetration of the tracer molecules into the seminiferous epithelium. Examination of numerous sections from one testis reveals the same pattern of penetration indicating that there is no variation in permeability along the length of a seminiferous tubule. Consequently there is no variation in the different stages of the seminiferous epithelium cycle. This confirms the suggestion (Fawcett and Ito, '72) that in monkeys the tunica propria offers no barrier to penetration of vascularly introduced substances. The only component of the blood-testis barrier in the monkey appears to be the Sertoli-Sertoli occluding junction. On rare occasions, peroxidase and lanthanum are observed in large quantities inside the cytoplasm of scattered germ cells and Sertoli cells but this is interpreted as artifact since these areas of the seminiferous tubules seem to be poorly fixed, as manifested by membrane discontinuities and organelle disruption. The results in mice confirmed our previous work on the rat concerning the location of the blood-testis barrier in rodents and the distribution of the markers (Dym and Fawcett, '70). Over large areas of the seminiferous tubules the tracers are excluded from the seminiferous epithelium by the peritubular myoid layer and their edge-to-edge tight junctions (fig. 9). In other areas of the rodent tubules where open junctions separate adjacent myoid cells the tracers penetrate between the cells of the tunica propria (fig. 10) and fill the interspaces surrounding the spermatogonia and early spermatocytes. As in the monkey, further penetrating toward the tubule lumen is blocked by the more effective component of the blood-testis barrier, the Sertoli-Sertoli occluding junction. On no accasion do tracers penetrate these junctions to reach the more advanced germ cells or the tubule lumen. DISCUSSION Spermatogenesis is a long complex process originating with the renewal and differentiation of the type A spermatogonia and concluding with the release of spermatozoa into the lumen of the seminiferous tubules. The microenvironment in which these events occur is provided by the Sertoli cell; therefore, it is reasonable to suggest that the production of sperm is largely dependent upon the normal functioning of this cell type. Yet surprisingly little is known concerning the role played by the Sertoli cell in spermatogenesis. On the basis of its shape and strategic position in the seminiferous tubules the functions of support for the germ cells and provision of bloodborne nutrients have been suggested. Recent evidence has implicated the Sertoli cell in the release of late spermatids into the tubule lumen and the retention of the residual bodies within the seminiferous epithelium (Fawcett and Phillips, '69). Lacy and coworkers (Lacy et al., '68; Lacy and Pettitt, '70) have suggested that this ~~ ~ Fig. 7 A n electron micrograph of a degenerating spermatogonium within the cytoplasm of a Sertoli cell. Note the lack of germ cell membranes including the nuclear envelope and pIasmalemma. The limiting membrane investing the degenerating figure (small arrowheads) is believed to belong to the Sertoli cell. Compare the thickness of this membrane with the nearby germ cell-Sertoli cell membranes to the right of the small arrowheads. For an enlargement of this region see "inset." The large arrowheads depict the junctional complexes between adjacent Sertoli cells. Stage VII. x 7735. BLOOD-TESTIS BARRIER Figure 7 649 650 MARTIN DYM Fig. 8 A n electrom micrograph of the tunica propria of a monkey seminiferous tubule. Wide intercellular spaces (at least 300 to 400 A ) separate the peritubular cells of the seminiferous tubules (arrowheads). x 7280. cell type produces a steroid hormone involved in local control of the cycle of the seminiferous epithelium. This latter function remains controversial. Hall et al. ('69) rejected a steroidogenic function for the Sertoli cells because the seminiferous tubules are unable to convert cholesterol7a-TI to androgens in vitro. Numerous profiles of smooth endoplasmic reticulum and mitochondria are common in Sertoli cells and in steroidogenic cells, but this similarity does not in itself substantiate a claim that the Sertoli cell produces steroids. The role of the Sertoli cell in phagocytizing residual bodies has been suggested (Rolshoven, '44; Roosen-Runge, '55; Sapsford et al, '69) and correlated with lysosomes (Dietert, '68) and acid phosphatase activity in the seminiferous tubules (Niemi and Kormano, '65), but little experimental evidence has been provided to demonstrate conclusively its phagocytic ability (Carr et al., '68). The fate of the numerous degenerating germ cells (Oaltberg, '56; Cler- mont, '62) during normal spermatogenesis has also received sparse attention. It may be recalled that one-third to one-half of the total germ cells in the testis degenerate (Clermont, '62; Clermont and BustosObregon, '68) and are presumably continuously digested by the Sertoli cell. However, the ultrastructure of the Sertoli cell cytoplasm containing these degenerating figures is not altered in any way to suggest an active phagocytosis. No doubt more experimentation will be necessary in order to determine the ultimate fate of the vast numbers of degenerating figures and the role played by the Sertoli cell. The fine structural morphology of the major cellular organelles in the monkey Sertoli cell appears identical in all stages of the cycle of the seminiferous epithelium, aDparently both in quantity and quality. This was somewhat disappointing since we had hoped to find Sertoli cell organelle differences in various stages which could be correlated with the dramatic morphological BLOOD-TESTIS BARRIER 65 1 Fig. 9 A light micrograph of several mouse seminiferous tubules showing the distribution of vascularly injected horseradish peroxidase. The black reaction product is present in the interstitial tissue and surrounds the seminiferous tubules. No peroxidase is found inside the lumen of the seminiferous tubules. x 182. Fig. 10 An electron micrograph of a mouse seminiferous tubule showing the distribution of peroxidase. The tracer penetrated the peritubular myoid cell layer via the intercellular spaces (see anow a t bottom left) and entered the germinal epithelium. Within the epithelium the tracer delineated a spermatogonium by filling the intercellular space between this germ cell and the neighboring Sertoli cells. Further penetration toward the tubule lumen is blocked by the Sertoli-Sertoli occluding junctions (arrowheads). x 9,555. (Micrograph courtesy of Dr. Roberto Vitale-Calpe). 652 MARTIN DYM Fig. 11 An electron micrograph of a monkey spermatogonium surrounded by lanthanum nitrate. The lanthanum was perfused into the monkey’s testis with the fixative. The junctional complexes between Sertoli cells (arrowheads) above the spermatogonium prevent the electron-opaque tracer from deeper penetration into the seminiferous epithelium. x 10,920. BLOOD-TESTIS BARRIER changes in germ cells during differentiation. We do not wish to imply that biochemical events along the length of the seminiferous tubules are similar. Major enzymatic alterations are well documented for various germ cell types (Blackshaw, '70; Mills and Means, '72) and both nuclear and cytoplasmic structural proteins alter quantitatively in different areas of the testis (Davis and Langford, '70). These modifications have not been assigned to specific stages of the cycle in the majority of the studies. Recently, a technique was described which permits visual identification of spermatogenic stages in long segments of freshly teased seminiferous tubules (Parvinen and Vanha-Perttula, '72) and three enzymes were correlated with certain stages of the cycle of the seminiferous epithelium. A significant step forward in understanding the physiology of the seminiferous tubules was the discovery of a bloodtestis barrier by Kormano ('67) and Setchell ('67). By cannulating the rete testis and individual seminiferous tubules of rams and rats (Waites and Setchell, '69; Tuck et al., '70), it was revealed that the composition of a variety of substances in the fluid within the lumina of the tubules is markedly different from that found in blood plasma. Proteins and certain amino acids that are abundant in in peripheral blood, were either absent or present in very low concentrations in the rete testis fluid. Certain sugars and ions also exhibited significant differences in concentration. This led Setchell ('67) to conclude that there is a barrier in or around the seminiferous epithelium which prevents many bloodborne substances including proteins from reaching the tubule lumen. Using silver nitrate staining methods, Regaud ('01) described two flattened, plate-like layers of cells surrounding the seminiferous tubules in rodents. This original observation has been confirmed repeatedly, and more recently with the advent of the electron microscope a better definition of the tunica propria was possible (Clermont, '58; Ross, '67; Dym and Fawcett, '70). The spermatogonia, early spermatocytes, and the Sertoli cells rest on a thin basal lamina separated from the innermost epithelioid layer of cells by a 653 clear zone containing sparsely distributed collagen fibrils. Outside of this is a second layer of cells described by Fawcett, I-Ieidger and Leak ('70) as lymphatic endothelial cells. The innermost cell layer contains numerous filaments, and evidence suggests that these are responsible for the contractions of the seminiferous tubules observed in vitro. In rodents adjacent myoid cells are frequently joined by tight junctions, but occasionally they are found separated by a continuous interspace of 200 A. Electron microscopic studies using lanthanum nitrate have indicated that the peritubular myoid cell layer in rodents prevents the vascularly introduced tracers from reaching the seminiferous epithlium over large areas of the tubules (Dym and Fawcett, '70). Our results have demonstrated that there is also a blood-testis barrier in monkeys. Since the peritubular cells in monkeys overlap and do not meet edge-to-edge, it was expected that the only component of the primate barrier would be an intraepithelial portion, namely the SertoliSertoli occluding junctions. This hypothesis was borne out by the tracer studies. It is likely that the location of the bloodtestis barrier in humans is similar to that observed in monkeys, since both primates have much the same organization of the tunica propria. Experiments in this laboratory have confirmed the presence of a blood-testis barrier in monkeys, rats, mice, hamsters, and guinea pigs. Other studies indicate that the barrier is present in rams, sheep, and bulls (Waites and Setchell, '69). Since the phenomenon of the blood-testis barrier is of such general occurrence in the animal kingdom it probably has a fundamental importance in the control of spermatogenesis, but the full biological significance of this relationship still remains unclear. The integrity of the barrier depends upon the normal functioning of the occluding junctions between adjacent Sertoli cells. Therefore, we can now add to the accepted functions of support and nutrition for the Sertoli cell the role of the maintenance of the blood-testis barrier and the compartmentalization of the seminiferous epithelium. The significance in rodents of a partial 654 MARTIN DYM barrier in the peritubular layer of cells to vascularly injected tracers remains obscure. Since the tracers used in these experiments are available for only very short periods of time prior to fixation (horseradish peroxidase) or are added with the fixative (lanthanum), this does not simulate the in vivo physiological condition. If it were possible to permit exogenous peroxidase or other markers to circulate in the blood for many hours or days, it is likely that the tracers would penetrate through the permeable sites in the myoid layer and then equilibrize in all directions along the length and width of the seminiferous tubules (Bennett, '71). Thus it is probable that, in rodents as in monkeys, circulating plasma constituents of a similar or smaller size to horseradish peroxidase and lanthanum nitrate have continuous access to the germ cells in the basal compartment of the seminiferous epithelium and the Sertoli cells. It is not expected that longer availability of marker substances would breach the intra-epithelial portion of the blood-testis barrier since on no occasion was the tracer observed deeper in the seminiferous epithelium. The subdivision of the seminiferous epithelium into a basal compartment, between the occluding Sertoli cell junctions and the basal lamina and an adluminal compartment between the Sertoli cell junctions and the tubule lumen may be important for a number of reasons. The germ cells in the basal compartment are the spermatogonia and preleptotene spermatocytes and bloodborne substances have direct access to these cells via the interstitial tissue and the intercellular spaces in the tunica propria. Biological processes associated with stem cell renewal and differentiation, such as DNA synthesis, which occur in the basal compartment of the epithelium may be directly under the control of circulating plasma constituents. Other renewal systems in the body such as the intestinal epithelium, stratified squamous epithelium of skin, and blood forming elements are not sequestrated behind a barrier but exposed directly to circulating substances in the plasma. It would be surprising if DNA synthesis and mitotic division in the testes occur in a different environment. Since circulating blood-borne substances can reach the spermatogonia directly without passing through the Sertoli cells, stem cell renewal in the testis may occur independently of Sertoli cell function. Stem cell renewal is not a specialized function of the testis but similar to other renewal systems in the body. On the other hand when the germ cells of the testis enter the long meiotic prophase of the first maturation division they leave the basal compartment of the seminiferous epithelium and enter the specialized environment of the adluminal compartment. Germ cells in this area of the seminiferous tubules are not exposed directly to circulating plasma substances. To reach these advancing germ cells all such materials must pass through the cytoplasm of the Sertoli cells and are subject to modification. Meiotic prophase, the two maturation divisions and spermiogenesis are processes unique to the gonads which may require the special environment provided by the Sertoli cell junctions and the blood-testis barrier. It is reasonable to suggest that these specialized biological functions of the seminiferous tubules are dependent upon the Sertoli cell. Therefore, it is not surprising that attempts to culture isolated germ cells beyond meiosis have failed (Eddy and Kahri, '71). Perhaps when we learn to simulate the environment produced by the normal Sertoli cell, further germ cell differentiation in vitro will be achieved. The results obtained by Reddy and Svoboda ('67) and Aragon, Lustig and Mancini ('72) indicated that there is penetration of peroxidase into the cytoplasm of cells of the seminiferous epithelium and into the tubule lumen. This contradicts our results. We have not observed horseradish peroxidase inside the cytoplasm of the seminiferous epithelial cells nor in the tubule lumen. Their studies were carried out with the light microscopy, and frequently the photographs were too low in magnification to determine cellular outlines. Furthermore, the fixation was not optimal and identification of germ cell types was not possible. LITERATURE CITED Aragon, J. A., L. Lustig and R. E. Mancini 1972 Uptake of horseradish peroxidase by the testis and epididymis of mice. J. Reprod. Fert., 28: 299-302. BLOOD-TESTIS BARRIER Barr, A. 1967 Human spermatogenesis. M. Sc. thesis, McGill University. Bennett, W. I. 1971 Unpublished observations. Blackshaw, A. W. 1970 Histochemical localization of testicular enzymes. In: The Testis. Vol. 11. A. D. Johnson, W. R. Gomes and N. L. Vandemark, eds. Academic Press, New York, pp. 73- 123. Brown, H. H. 1885 On spermatogenesis in the rat. Quart. J. of Micro. Sci., 25: 343-370. Carr, I., E. J. Clegg and G. A. Meek 1968 Sertoli cells as phagocytes: an electron microscopic study. J. Anat., 102: 501-509. Christensen, A. K. 1965 The fine structure of testicular interstitial cells in guinea pigs. J. Cell Biol., 26: 911-935. Christensen, A. K., and N. R. Mason 1965 Comparative ability of seminiferous tubules and interstitial tissue of rat testes to synthesize androgens from progesterone -4 - 14C in vitro. Endocrinology, 76: 646-656. Clermont, Y. 1958 Contractile elements in the limiting membrane of the seminiferous tubules of the rat. Expt. Cell Res., 15: 438-440. 1962 Quantitative analysis of spermatogenesis of the rat: A revised model for the renewal of spermatogonia. Am. J. Anat., 111: 111-129. 1969 Two classes of spermatogonial stem cells in the monkey. Am. J. Anat., 126: 57-72. Clermont, Y.,and E. Bustos-Obregon 1968 Reexamination of spermatogonial renewal in the rat by means of seminiferous tubules mounted “in toto.” Am. J. Anat., 122: 237-248. Davis, J. R., and G. A. Langford 1970 Testicular Proteins. In: The Testis. Vol. 11. A. D. Johnson, W. R. Gomes and N. L. Vandemark, eds. Academic Press, New York, pp. 259-306. Dietert, S. E. 1966 Fine structure of the formation and fate of the residual bodies of mouse spermatozoa with evidence for the participation of lysosomes. J. Morph., 120: 317-346. Dym, M. 1972 The fine structure of the monkey Sertoli cell and its role in establishing the blood-testis barrier. Biol. of Reprod., 7: 129, Abstract. Dym, M., and Y. Clermont 1970 Role of spermatogonia in the repair of the seminiferous epithelium following X-irradiation of the rat testis. Am. J. Anat., 128: 265-282. Dym, M., and D. W. Fawcett 1970 The bloodtestis barrier in the rat and the physiological compartmentation of the seminiferous epithelium, Biol. of Reprod., 3: 308-326. Eddy, E. M., and A. I. Kahri 1971 Unpublished observations. Elftman, H. 1950 The Sertoli cell cycle in the mouse. Anat. Rec., 106: 381-394. 1963 Sertoli cells and testis structure. Am. J. Anat., 113: 25-34. Fawcett, D. W.,and S . Ito 1972 Unpublished observations. Fawcett, D. W., L. V. Leak and P. M. Heidger 1970 Electron microscopic observations on thc structural components of the blood-testis barrier. J. Reprod. Fertil. Suppl., 10: 105-122. 655 Fawcett, D. W., W. B. Neaves and M. N. Flores 1973 Comparative observations of intertubular lymphatics and the organization of the interstitial tissue of the mammalian testis. Biol. of Reprod., in press. Fawcett, D. W., and D. M. Phillips 1969 Observations on the release of spermatozoa and on changes in the head during passage through the epididymis. J. Reprod. Fertil. Suppl., 6: 405-418. Graham, Jr., R. C., and M. J. Karnovsky 1966 T h e early stages of absorption of injected horseradish peroxidase in the proximal tubules of mouse kidney: ultrastructural cytochemistry by a new technique. J. Histochem. Cytochem., 14: 291-302. Hall, P. F., D. C. Irby and D. W. de Kretser 1969 The conversion of cholesterol to androgens by rat testes: Comparison of interstitial cells and seminiferous tubules. Endocrinology, 84: 488496. Karnovsky, M. J. 1965 A formaldehydsglutaraldehyde fkative of high osmolarity for use in electron microscopy. J. Cell Biol., 27: 137a, Abstract. Kormano, M. 1967 Dye permeability and alkaline phosphatase activity of testicular capillaries in the postnatal rat. Histochemie., 9: 327438. Lacy, D., and A. J. Pettitt 1970 Sites of hormone production in the mammalian testis, and their significance in the control of male fertility. Br. Med. Bull., 26: 87-91. Lacy, D., G. P. Vinson, P. Collins, J. Bell, P. Fyson, J. Pudney and A. J. Pettitt 1968 The Sertoli cell and spermatogenesis in mammals. Proc. of the 3rd International Congress of Endocrinology, Mexico, pp. 1019-1029. Mills, N. C.,A. R. Means 1972 Sorbitol dehydrogenase of rat testis: Changes of activity during development after hypophysectomy and following gonadotrophic hormone administration. Endocrinology, 91 : 147-156. Niemi, M., and M. Kormano 1965 Cyclical changes in and significance of lipids and acid phosphatase activity in the seminiferous tubules of the rat testis, Anat. Rec., 152: 159-170. Nishida, T. 1954 Cytoplasmic cytology of the Sertoli cells of mammals. Cytologia, 19: 203216. Oakberg, E. F. 1956 A description of spermiogenesis in the mouse and its use in analysis of the cycle of the seminiferous epithelium and germ cell renewal. Am. J. Anat., 99: 391-414. Parvinen, M., and T. Vanha-Perttula 1972 Visual identification in native condition and enzyme quantification of the stages of the rat seminiferous epithelial wave. Anat. Rec., 174: 435-450. Reddy, J. N., and D. J. Svoboda 1967 Peroxidase transport by Sertoli cells of the rat testis. J. Cell Biol., 35: 184a Abstract. Regaud, C. 1900 Les phases et les stades de l’onde spermatog6n6tique chez les mammifhres (rat ). Classification rationnelle des fimres -.- . - de -. Ia spermatogenSses. Comptes Rendus SOC.de Biol., 52: 1039-1042. 656 MARTIN DYM 1901 Etudes sur la structure des tubes shminifhres et sur la spermatogenhse chez les mammifhres. Arch. Anat. Micro., 4: 101-155, 231480. Rolshoven, E. 1944 Spermatogenese and Sertoli-Syncytium. Z. Zellforschung, 33: 439-460. Roosen-Runge, E.C. 1955 Untersuchungen uber die Degeneration Samenbildender Zellen in der Normalen Spermatogenese der Ratte. Z. Zellforschung, 41: 221-235. Sapsford, C. S., C. A. Rae and K. W. Cleland 1969 The fate of residual bodies and degenerating germ cells and the lipid cycle in Sertoli cells in the bandicott Perumeles Nusuta Geoffroy (Marsupialia). Aust. J. Zool., 17: 729-753. Sertoli, E. 1865 Dell 'esistenza di particolari cellule ramificate nei canalicoli seminiferi del testiculo umano. Morgagni, 7: 31-39. Setchell, B. P. 1967 The blood-testicular fluid barrier in sheep. J. Physiol., 189: 63p65p. Tuck, R. R., B. P. Setchell, G. M. H. Waites and J. A. Young 1970 The composition of fluid collected by micropuncture and catheterization from the seminiferous tubules and rete testis of rats. Pfliigers Arch., 318: 225-243. Waites, G. M. H., and B. P. Setchell 1969 Some physiological aspects of the function of the testis. In: The gonads. K. W. McKerns, ed. Appleton-Century-Crofts, New York, pp. 649714. Winiwarter, von, H. 1912 Etudes sur la spermatogenhse humaine. I. Cellule de Sertoli. I1 HCt6rochromosome et mitoses de l'Bpith6lium s6minal. Arch de Biol., 27: 91-187.