Rat Sertoli Cells: A Rapid Method for Obtaining Viable Cells MICHAEL J. WELSH AND JOHN P. WIEBE Regulatory Mechanisms Laboratory, Department of Zoology, University of Western London, Ontario, Canada ABSTRACT. A method is described for obtaining populations of viable Sertoli cells from rat testes. Minced whole testes from rats of 15 to 29 days of age are sequentially treated with collagenase and pancreatin. The resulting suspension of cells is sedimented through a sucrose density gradient. Preparations are produced consisting of from 60% to 82% Sertoli cells, an enrichment of 2 to 5 times the proportion of Sertoli cells in whole testes of these ages. The preparations are free of interstitial cells, are essentially free of peritubular cells and
W
HILE many activities have been suggested for mammalian Sertoli cells (1-3), only recently have researchers been able to report evidence of functions or responses that may be specifically attributed to these cells (4-7). To determine the functions of the Sertoli cells it would appear ideal to have them in normal and isolated condition—free of the other cells in the testis. However, although the interstitial tissues may be separated from the tubules by teasing (8) and preparations of spermatocytes or spermatids may be made with relative ease (9), several characteristics have made the production of pure Sertoli cell preparations difficult. First, the Sertoli cells form a continuous cellular sheet as a result of extensive tight junctions (2). Also, germ cells embedded within the Sertoli cells of the mammal may form a second cellular net as a result of elaborate and abundant cytoplasmic bridges (10). The two interwoven cellular nets are difficult to separate while still maintaining the integrity of the cells. Moreover, in the laboratory rat, mitotic activity of Sertoli cells is not evident after about 16 days of age (11,12) and differentiated mammalian SerReceived September 5, 1974. 1 This research was supported by NRC grant No. A6865 and by a U.S. National Science Foundation Graduate Fellowship to M.J.W.
Ontario,
contain reduced numbers of germinal cells; the main contaminating cell types are spermatogonia and spermatocytes. The Sertoli cells are considered to be 95% viable by their ability to exclude trypan blue and by subsequent culturing in vitro. The entire procedure requires 3 h. Maintenance of the Sertoli-enriched fraction in modified Eagle's minimal essential medium temporarily at 41 C allows preparations of Sertoli cell monolayer cultures consisting of 95%-98% Sertoli cells within 3 days. (Endocrinology 96: 618, 1975)
toli cells are generally considered not to undergo mitosis (13,14) except perhaps rarely under abnormal conditions (15, personal observations). Therefore, the cells cannot be propagated in vitro for study. Most research concerning Sertoli cell function has relied on experimental systems using seminiferous tubules which also contain germ cells and peritubular cells. Elimination or reduction of the germinal cell component of the tubule has been utilized as a method of obtaining Sertolienriched preparations. However, in order to reduce or eliminate the germ cell population of the tubules, long-term abnormal conditions such as hypophysectomy (4,5), natural or surgical cryptorchidism and heat treatment (3,4,5,16), irradiation (4,5) or chemical or hormonal interference with germ cells development (3) have been employed. The physiological conditions within the seminiferous tubules—and therefore the functions of the Sertoli cells—need not be assumed to be "normal" after these treatments, especially if an interaction between the Sertoli cells and the germ cells exists as has been suggested (3,17). And, although efforts have been made to purify different testicular cell types from untreated mammals via culturing methods (14-21), none of these have produced a satisfactory rapid enrichment of Sertoli cells.
618
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 08 June 2015. at 14:00 For personal use only. No other uses without permission. . All rights reserved.
RAPID ISOLATION OF RAT SERTOLI CELLS
In the present report we detail a rapid method for the separation and enrichment of Sertoli cells from normal rat testes. The concomitant separation of interstitial, peritubular and germ cells is also indicated.
619
Xinco Tl»sut
1
Wash 3 Tic-!,
i
Filter o
(Washed T i s s u e )
Materials and Methods Animals
1 1
FRACTION 2 ( F r e e r.e rm Tellti)
Co] lng.'nasv 60 t i n
Supernatant
i
(Free Tubule Pie:es)
Rats were obtained from our own colony of Sprague-Dawley rats (original breeders purchased from Bio Breeding Laboratories of Canada, Ltd., Ottawa). They are maintained on a 12-h light, 12-h dark photoperiod, at 24 C and fed Purina Rat Chow and water ad libitum. Only males born on the same day were used for each Sertoli preparation.
FRACTION (Interst l t l o l
Wash Twice
1
i.
(Sertoli and
Tissue)
0 min ARRresat ed Tissue Maaa
1
Corn Cells)
1
FRACTION 6 (Pcrltub ular
Cells)
Wash 3 Times (51 Calf Serum)
I
1
21-6S Sucros e Gradient Sediment 20 min 4C
Enzyme
treatments
During both collagenase and pancreatin treatments the tissue or cells were suspended in a calcium and magnesium-free phosphate buffered saline (solution A) composed of NaCl 97.6 mM, KC1 25 mM, Na2HPO4 3.7 mM, glucose 8.3 mM, KH2PO4 0.3 mM, penicillin 50 U/ml, streptomycin sulfate 50 ju,g/ml, and phenol red 0.008 mM in deionized, glass distilled water. The pH of solutions during enzyme treatments was maintained at about pH 7.3 by addition of O.lM Na2HPO4. Treatments were carried out at 28 C in a constant temperature shaker bath at 135
oscillations per minute. Sertoli cell separation procedure (Fig. 1) Six to 8 rats of the same age were ether anesthetized. Both testes were removed by abdominal incision in very young rats or by an incision in the scrotal sac in more mature animals. Each testis was placed in solution A on ice while the tunica albuginea was removed. The tissue was rinsed in solution A and placed on a number 12 neoprene stopper where the tissue was cut into pieces about 2 mm on a side using a stainless-steel razor blade. The minced tissue was washed three times for 5 min each, in 20 ml of solution A in a glass stoppered 50 ml erlenmeyer flask in the shaker bath. No more than 1 g of minced tissue was contained in one flask. The supernatants from the three washings of the minced whole testis were filtered through a silk screen with a
I I I
Collect
Fast Fract ion
CentrifuRc smd Wa«h FRACTION 7
I
•Itn
FIG. 1. Flow chart outlining complete, procedure for Sertoli cell preparation.
pore size of 0.08-0.10 mm. The tissue retained on the screen was returned to theflaskcontaining the minced tissue (fraction 1, Fig. 1). The filtrate (fraction 2) consists primarily of germ cells. Fraction 1 was treated with 20 ml of solution A containing 4 mg collagenase (Sigma type I, 140 U/mg) for 1 h in the shaker bath (28 C). At the end of this time a reaggregated mass of tissue could be removed leaving tubule pieces (fraction 3). The mass of reaggregated tissue, composed primarily of the interstitial tissue, was washed twice in solution A to remove more tubule fragments. These fragments were returned to the flask (fraction 3). The tubule fragments of fraction 3 were allowed to settle and the collagenase solution was decanted. The tubules were washed twice in 20 ml' of solution A, then resuspended in 20 ml of solution A containing 4 mg pancreatin (Sigma, grade VI, porcine). After 20 min another mass of reaggregated tissue (fraction 6) was removed from the remaining tubule fragments (fraction 5).' Fraction 6; composed of peritubular cells, was washed once with solution A and the
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 08 June 2015. at 14:00 For personal use only. No other uses without permission. . All rights reserved.
620
Endo • 1975 Vol 96 • No 3
WELSH AND WIEBE
supernatant from this wash, containing more tubule fragments, was added to fraction 5. Fraction 5, composed of Sertoli cells and germ cells, was washed twice in 15 ml of solution A and once with 15 ml solution A containing 5% fetal calf serum (Grand Island Biological Co.). The cells were resuspended in 10 ml solution A with 5% fetal calf serum. Fraction 5 was passed through a syringe needle (no. 17 gauge, 7.5 cm) at a rate of 5 ml/sec to break the tubular fragments into groups of Sertoli cells containing embedded spermatogonia and spermatocytes. The cell suspension was shaken moderately on a vortex mixer (ColeParmer Super-Mixer set at 4.5) for 15 sec to loosen and detach embedded germ cells from the Sertoli cell clusters. The cell suspension was immediately layered over a 400 ml gradient of 2%-6% sucrose in solution A at 4 C in a siliconized glass column of 5 cm diameter and 26 cm depth. Twenty minutes after the cell suspension containing Sertoli-germ cell groups and free germ cells was introduced into the column, the bottom 15% of the gradient was rapidly collected and discarded. The next 60% of the gradient volume was collected and then centrifuged at 150 x g for 5 min (4 C). The pellet was washed in solution A and recentrifuged. The remainder of the gradient volume was discarded. The purity of the pelleted Sertoligerm cell fraction (fraction 7) was estimated by aceto-orcein smear preparations and the viability, by trypan blue exclusion. (22). Cell culture
13.2, L-aspartate 13.3, L-glycine 7.5, L-glutamine 692.0, L-proline 11.5, L-serine 10.5, i-inositol 14.0, glutathione 2.0, cyanocobalamin 0.1, biotin 0.2, p-aminobenzoic acid 0.2, ascorbic acid 8.5, and sodium pyruvate 50.0. All ingredients were dissolved in deionized, glass distilled water and sterilized by filtration through a 0.22 /x pore filter. For culturing, the Sertoli preparation (fraction 7) was resuspended in 5 ml of solution A. Cultures were started in 50 mm sterile petri dishes with 0.5 ml of Sertoli cell suspension and 10 ml of cell culture medium containing 5% fetal calf serum. The cultures were maintained in an incubator with humidified air atmosphere. The cultures were maintained at 32 C for 20 h, then the temperature was raised to 41 C for 40 h after which time the temperature was lowered to and maintained at 32 C. Histochemistry Small pieces of the live tissue preparations were incubated for 40 min at 27 C in histochemical incubation medium. This medium was composed of 2 ml solution A, 0.25 ml nitro BT (0.25 mg/0.25 ml water), 0.5 ml NAD (2.5 mg/0.5 ml water), 0.25 ml 3/3-hydroxy5j8-androstan-17-one (0.4 mg/ml propylene glycol-ethanol, 1:1), with the pH adjusted to pH 7.7 with a few drops of 0.1M Na2HPO4. Controls lacked substrate.
Results
Using aceto-orcein smear preparations, we identified and counted the cells of fraction 7. Sertoli cells comprised from 60% to over 80% of fraction 7 (Table 1, Fig. 2f). Viability of greater than 95% was indicated by trypan blue exclusion. The contaminants were spermatogonia and primary spermatocytes, and to a much lesser degree, peritubular cells. A final preparation of several hundred milligrams resulted from 2 Abbreviations: NAD = nicotinamide adenine di- 2-3 g of testis tissue. nucleotide; Nitro BT = nitro blue tetrazolium [2,2'The germ cell fraction (fraction 2) condi-p-nitrophenyl - 5,5'-diphenyl-3,3' - (3,3' - dimethoxy- tained 95%-97% germ cells. The germ cells 4,4'-diphenylene) ditetrazolium chloride]; HEPES were mainly spermatocytes with a lesser = N-2-hydroxyethylpiperazine-N'-2-ethane sulfonic acid; BES = N,N-bis[2-hydroxyethyl]-2-aminoethane number of spermatogonia. Histological sulfonic acid; PIPES = piperazine-N,N'-bis[2 ethane preparations showed the peritubular fracsulfonic acid]. tion (fraction 6) to be composed of almost Final Sertoli cell preparations (fraction 7) were maintained in what is basically Eagle's minimal essential medium (24), but with the following modifications. The inorganic salts were NaCl 100 mM, KC1 15 mM, MgSO4 0.06 mM, CaCl2 0.4 mM, Na2HPO4 0.2 mM, with the zwitterionic buffers HEPES 8 mM,2 BES 8 mM and PIPES 8 mM (25). The medium also contained (in mg/liter): L-alanine 8.9, L-asparagine
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 08 June 2015. at 14:00 For personal use only. No other uses without permission. . All rights reserved.
RAPID ISOLATION OF RAT SERTOLI CELLS
80% peritubular cells, about 15% Sertoli and germ cells and a small percentage of connective tissue and Leydig cells. The interstitial fraction (fraction 4) contained vessels and capillaries of the circulatory system, connective tissue, Leydig cells and small tubule fragments. Enzyme histochemistry was utilized to estimate the amount of Leydig cell contamination in fractions other than the interstitial fraction. Most of the 3/3-hydroxysteroid dehydrogenase activity usually associated with Leydig cells was found in the interstitial fraction (fraction 4) (Fig. 2a); some activity was also found in the peritubular fraction and could be localized histochemically in Leydig cells contained in fraction 6 (Fig. 2b). Significantly, the Sertoli fraction (fraction 5) was found to be completely free of Leydig cell contamination (Fig. 2c). Maintenance of the final Sertoli preparation in culture also revealed no Leydig cell contamination, and in most cultures no cells with the appearance of fibroblasts were seen. Culturing allowed preparations consisting of 95%-98% Sertoli cells to be realized within 3 days (Fig. 2 g,h). Cultures were maintained for as long as 28 days with cells maintaining morphological characteristics normally associated with differentiated Sertoli cells as seen at the light microscope level. Discussion After the testis tissue is minced, the germ cells which are not deeply embedded in the Sertoli cells escape from the tubules. The germ cells appear to be expelled from the tubule as if by an internal pressure, possibly a result of contractile tension produced by the peritubular cells (26). The tubules are freed from the interstitial tissue matrix by collagenase treatment and agitation. The interstitial fragments exhibit an adhesive quality which the tubule fragments lack. The interstitial fragments tend to reaggregate into one mass that may
621
TABLE 1. Sertoli cells in final cell preparations (fraction 7) from rats 15 to 29 days of age Cells counted8 Rat age (days)
Total
Sertoli
15 22 23 24 25 27 28 29
732 785 637 984 1,428 712 2,067 778
574 609 424
Percent
Percent Sertoli in wlinl*»
593
869 584 1,703 467
cells
tubule"
78 78 67 60 61 82 82 60
43 31 29 27
25 23 22 20
a Cells from fraction 7 were stained with 1% Orcein in 45% acetic acid. Cells were identified by nuclear morphology according to Clermont's system of classification (23). Cells identified as not Sertoli were mainly primary spermatocytes. b Percentage Sertoli cells in tubule estimated from sections prepared from samples of initial testis tissue used for each Sertoli cells preparation. Tissue samples fixed in Bouin's fluid for 36 h, embedded in paraffin, cut into sections 5 fim thick and stained in hematoxylineosin. A minimum of 10 tubule cross sections or 800 cells identified was used for each value.
be removed using forceps. After washing, the remaining tubule fragments are treated with pancreatin which frees the peritubular cell layer (Fig. 2d,e) from the Sertoli and germ cells. The peritubular cells also display adhesive characteristics and often reaggregate into one or several masses. Following removal of the outer peritubular cell layer, the peripherally located germ cells, such as the type A spermatogonia, are also detached from the Sertoli cells. The Sertoli-germ cell pieces are subjected to mechanical disruption; sufficient sheer force to break up the large cell groups can be generated by passing the fragments through a syringe needle. Moderate agitation produced by a vortex-mixer dislodges additional germ cells. Using a suspension of individual Sertoli cells amid pachytene spermatocytes, we have not found it possible to separate the two cell types. Separation was not accomplished by either velocity sedimentation (27,28) or density gradient centrifugation. If, however, the Sertoli cells are sepa-
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 08 June 2015. at 14:00 For personal use only. No other uses without permission. . All rights reserved.
622
WELSH AND WIEBE
Endo • 1975 Vol 96 • No 3
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 08 June 2015. at 14:00 For personal use only. No other uses without permission. . All rights reserved.
RAPID ISOLATION OF RAT SERTOLI CELLS
623
FIG. 2. a) to c) Histochemical demonstration of 3/3-hydroxysteroid dehydrogenase activity in cell fractions from 29-day-old rat. Histochemistry used nitro BT, NAD, and 3/3-hydroxy-5/3-androstan-17-one, for 40 min at 27C and pH 7.7. a) fraction 4, interstitial tissue. (xlOO) b) fraction 6, peritubular tissue. (xlOO) c) fraction 5, Sertoli-germ cells. (xlOO) d) Aceto-orcein stained whole-mount of tubule fragments (fraction 3) of 29-day-old rat. Note peritubular cell layer (arrows) around tubules. (xl60) e) Aceto-orcein stained whole-mount of tubule fragments (fraction 5) of 29-day-old rat. Note absence of peritubular cells after pancreatin treatment. (xl60) f) Aceto-orcein stained Sertoli cell preparation from 29-day-old rat (fraction 7) showing Sertoli cells (pale nuclei) with embedded primary spermatocytes (dark nuclei). (x500) g) Sertoli cell monolayer from 27-dayold rat after 66 h in culture under conditions described under Cell culture in text. Culture fixed in Bouin's fluid for 15 min, stained in Harris' hematoxylin and eosin, (x450) h) Sertoli cells from 30-day-old rat after 9 days in culture. Fixed in Bouin's, stained in Harris' hematoxylin and sudan IV, mounted in aqueous glycerin-jelly medium. Arrows indicate lipid droplets. (xlOOO).
rated into small groups of cells instead of individual cells, a rapid sedimentation can be used to separate the Sertoli cell groups from the free germ cells. A gradient is used to prevent convection currents in the column and reduce mixing. Maintenance of the cells in vitro, temporarily at an elevated temperature, destroys almost all of the germ cells which are then phagocytized by the Sertoli cells. The Sertoli cells appear to be unharmed by the temporary temperature increase and form a monolayer (Fig. 2g,h) which has been maintained for 28 days. Chowdhury and Steinberger (29) have indicated that higher temperature (43 C) for a shorter period of time, although injurious to germ cells, apparently is not harmful to Sertoli cells in vivo. It would be of value to mention other procedures that were tried before the present method was accepted into routine use in our laboratory. In addition to the enzymes finally selected, we also employed crude and purified trypsin, lysozyme, hyaluronidase, elastase, pepsin, peptidase, a-chymotrypsin and deoxyribonuclease I (all enzymes from Sigma Chemical Co.). These enzymes were used in various concentrations, combinations and sequences. The results of using two different chelating agents as well as the inclusion of cations (Mg++, Mn ++ , Ca ++ ) were also evaluated. Many of these treatments had little noticeable effect on cell separation. While certain concentrations of some of the enzymes allowed tissue separation, viability suffered greatly. All preparations of trypsin resulted
in great amounts of a mucoid-like substance from which the cells could only be separated with great difficulty. Neither elastase nor deoxyribonuclease I alleviated this problem satisfactorily. Testis tissue from other species including toad (Bufo marinus); Japanese quail (Coturnix coturnix japonica) and sunfish (Lepomis megalotus) was given similar treatment as described in procedure and outlined in Fig. 1; none of these tissues separated to any degre'e as a result of our procedure or slight modifications of it. Apparently only rat testis tissue, and presumably tissue of other mammalian species, responds to this method. This method for rapid preparation of a population of cells from the rat testis enriched in Sertoli cells has several advantages over other methods. The method is relatively fast and can be performed on normal rats. Hormonal alterations of the animal are not required nor are treatments with agents of unknown actions such as radiation or chemicals. The final cell preparation can be demonstrated to be free of Leydig cell contamination and, in culture, free of fibroblasts. The method can allow comparison of function of the various tissues of the testis because the interstitial fraction, peritubular cell fraction, germ cell fraction and Sertoli cell fraction can be derived from the same initial tissue. References 1. Steinberger, E., Physiol Rev 51: 1, 1971. 2. Dym, M., and D. W. Fawcett, Biol Reprod 3: 308, 1970. 3. Lacy, D., G. P. Vinson, P. Collins, J. Bell, P.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 08 June 2015. at 14:00 For personal use only. No other uses without permission. . All rights reserved.
624
4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14.
15.
WELSH AND WIEBE Fyson, J. Pudney, and A. J. Pettitt, In Gual, C. (ed.), Progress in Endocrinology, Proc. 3rd Int. Congr. Endocr. Mexico, 1968, Excerpta Medica Foundation, Amsterdam, 1969, p. 1019. Dorrington, J. H., and I. B. Fritz, Endocrinology 94: 395, 1974. Means, A. R., In Greep, R. O., and D. W. Hamilton (eds.), Handbook of Physiology, American Physiological Society, Bethesda, Md. (In press). Davies, A. G.J Reprod Fertil 25: 21, 1971. Murphy, H. D., Proc Soc Exp Biol Med 48: 1202, 1965. Christensen, A. K., and N. R. Mason, Endocrinology 76: 646, 1965. Lam, D. K. M., R. Furrer, and W. R. Bruce, Proc Natl Acad Sci USA 65: 192, 1970. Fawcett, D. W., S. Ito, and D. Slautterback, J Biophysic Biochem Cytol 5: 453, 1959. Clermont, Y., and B. Perey, Am J Anat 100: 241, 1957. Nagy, F.J Reprod Fertil 28: 389, 1972. Roosen-Runge, E. C., Biol Rev 37: 343, 1962. Steinberger, E., and A. Steinberger, In Gual, C. (ed.), Progress in Endocrinology, Proc. 3rd Int. Congr. Endocr., Mexico, 1968, Excerpta Medica Foundation, Amsterdam, 1969, p. 1038. Esaki, S., Z Mikrosk Anat Forsch 15: 365, 1928.
Endo • 1975 Vol 96 • No 3
16. Nagy, F., Fertil Steril 24: 805, 1973. 17. Johnsen, S. G., In Rosenberg, E., and C. A. Paulsen (eds.), The Human Testis, Plenum Press, New York, 1970 p. 231. 18. Kodani, M., and K. Kodani, Proc Natl Acad Sci USA 56: 1200, 1966. 19. DeKretser, D. M., K. J. Catt, M. L. Dufau, and B. Hudson, 7 Reprod Fertil 24: 311, 1971. 20. Dufau, M., D. M. deKretser, and B. Hudson, Endocrinology 88: 825, 1971. 21. Yamada, M., S. Yasue, and K. Matsumoto, Endocrinology 93: 81, 1973. 22. Merchant, D. J., R. H. Kahn, and W. H. Murphy Jr., Handbook of Cell and and Organ Culture, Burgess Publishing Co., Minneapolis, Minn., 1964, p. 157. 23. Clermont, Y., Physiol Rev 52: 198, 1972. 24. Eagle, H., Science 130: 432, 1959. 25. , Science 174: 500, 1971. 26. Leeson, C. R., and T. S. Leeson, Anat Rec 147: 243, 1963. 27. Miller, R. G., and R. A. Phillips, J Cell Physiol 73: 191, 1969. 28. Meistrich, M. L., W. R. Bruce, and Y. Clermont, Exptl Cell Res 79: 213, 1973. 29. Chowdhury, A. K., and E. Steinberger, J Reprod Fertil 22: 205, 1970.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 08 June 2015. at 14:00 For personal use only. No other uses without permission. . All rights reserved.
W
HILE many activities have been suggested for mammalian Sertoli cells (1-3), only recently have researchers been able to report evidence of functions or responses that may be specifically attributed to these cells (4-7). To determine the functions of the Sertoli cells it would appear ideal to have them in normal and isolated condition—free of the other cells in the testis. However, although the interstitial tissues may be separated from the tubules by teasing (8) and preparations of spermatocytes or spermatids may be made with relative ease (9), several characteristics have made the production of pure Sertoli cell preparations difficult. First, the Sertoli cells form a continuous cellular sheet as a result of extensive tight junctions (2). Also, germ cells embedded within the Sertoli cells of the mammal may form a second cellular net as a result of elaborate and abundant cytoplasmic bridges (10). The two interwoven cellular nets are difficult to separate while still maintaining the integrity of the cells. Moreover, in the laboratory rat, mitotic activity of Sertoli cells is not evident after about 16 days of age (11,12) and differentiated mammalian SerReceived September 5, 1974. 1 This research was supported by NRC grant No. A6865 and by a U.S. National Science Foundation Graduate Fellowship to M.J.W.
Ontario,
contain reduced numbers of germinal cells; the main contaminating cell types are spermatogonia and spermatocytes. The Sertoli cells are considered to be 95% viable by their ability to exclude trypan blue and by subsequent culturing in vitro. The entire procedure requires 3 h. Maintenance of the Sertoli-enriched fraction in modified Eagle's minimal essential medium temporarily at 41 C allows preparations of Sertoli cell monolayer cultures consisting of 95%-98% Sertoli cells within 3 days. (Endocrinology 96: 618, 1975)
toli cells are generally considered not to undergo mitosis (13,14) except perhaps rarely under abnormal conditions (15, personal observations). Therefore, the cells cannot be propagated in vitro for study. Most research concerning Sertoli cell function has relied on experimental systems using seminiferous tubules which also contain germ cells and peritubular cells. Elimination or reduction of the germinal cell component of the tubule has been utilized as a method of obtaining Sertolienriched preparations. However, in order to reduce or eliminate the germ cell population of the tubules, long-term abnormal conditions such as hypophysectomy (4,5), natural or surgical cryptorchidism and heat treatment (3,4,5,16), irradiation (4,5) or chemical or hormonal interference with germ cells development (3) have been employed. The physiological conditions within the seminiferous tubules—and therefore the functions of the Sertoli cells—need not be assumed to be "normal" after these treatments, especially if an interaction between the Sertoli cells and the germ cells exists as has been suggested (3,17). And, although efforts have been made to purify different testicular cell types from untreated mammals via culturing methods (14-21), none of these have produced a satisfactory rapid enrichment of Sertoli cells.
618
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 08 June 2015. at 14:00 For personal use only. No other uses without permission. . All rights reserved.
RAPID ISOLATION OF RAT SERTOLI CELLS
In the present report we detail a rapid method for the separation and enrichment of Sertoli cells from normal rat testes. The concomitant separation of interstitial, peritubular and germ cells is also indicated.
619
Xinco Tl»sut
1
Wash 3 Tic-!,
i
Filter o
(Washed T i s s u e )
Materials and Methods Animals
1 1
FRACTION 2 ( F r e e r.e rm Tellti)
Co] lng.'nasv 60 t i n
Supernatant
i
(Free Tubule Pie:es)
Rats were obtained from our own colony of Sprague-Dawley rats (original breeders purchased from Bio Breeding Laboratories of Canada, Ltd., Ottawa). They are maintained on a 12-h light, 12-h dark photoperiod, at 24 C and fed Purina Rat Chow and water ad libitum. Only males born on the same day were used for each Sertoli preparation.
FRACTION (Interst l t l o l
Wash Twice
1
i.
(Sertoli and
Tissue)
0 min ARRresat ed Tissue Maaa
1
Corn Cells)
1
FRACTION 6 (Pcrltub ular
Cells)
Wash 3 Times (51 Calf Serum)
I
1
21-6S Sucros e Gradient Sediment 20 min 4C
Enzyme
treatments
During both collagenase and pancreatin treatments the tissue or cells were suspended in a calcium and magnesium-free phosphate buffered saline (solution A) composed of NaCl 97.6 mM, KC1 25 mM, Na2HPO4 3.7 mM, glucose 8.3 mM, KH2PO4 0.3 mM, penicillin 50 U/ml, streptomycin sulfate 50 ju,g/ml, and phenol red 0.008 mM in deionized, glass distilled water. The pH of solutions during enzyme treatments was maintained at about pH 7.3 by addition of O.lM Na2HPO4. Treatments were carried out at 28 C in a constant temperature shaker bath at 135
oscillations per minute. Sertoli cell separation procedure (Fig. 1) Six to 8 rats of the same age were ether anesthetized. Both testes were removed by abdominal incision in very young rats or by an incision in the scrotal sac in more mature animals. Each testis was placed in solution A on ice while the tunica albuginea was removed. The tissue was rinsed in solution A and placed on a number 12 neoprene stopper where the tissue was cut into pieces about 2 mm on a side using a stainless-steel razor blade. The minced tissue was washed three times for 5 min each, in 20 ml of solution A in a glass stoppered 50 ml erlenmeyer flask in the shaker bath. No more than 1 g of minced tissue was contained in one flask. The supernatants from the three washings of the minced whole testis were filtered through a silk screen with a
I I I
Collect
Fast Fract ion
CentrifuRc smd Wa«h FRACTION 7
I
•Itn
FIG. 1. Flow chart outlining complete, procedure for Sertoli cell preparation.
pore size of 0.08-0.10 mm. The tissue retained on the screen was returned to theflaskcontaining the minced tissue (fraction 1, Fig. 1). The filtrate (fraction 2) consists primarily of germ cells. Fraction 1 was treated with 20 ml of solution A containing 4 mg collagenase (Sigma type I, 140 U/mg) for 1 h in the shaker bath (28 C). At the end of this time a reaggregated mass of tissue could be removed leaving tubule pieces (fraction 3). The mass of reaggregated tissue, composed primarily of the interstitial tissue, was washed twice in solution A to remove more tubule fragments. These fragments were returned to the flask (fraction 3). The tubule fragments of fraction 3 were allowed to settle and the collagenase solution was decanted. The tubules were washed twice in 20 ml' of solution A, then resuspended in 20 ml of solution A containing 4 mg pancreatin (Sigma, grade VI, porcine). After 20 min another mass of reaggregated tissue (fraction 6) was removed from the remaining tubule fragments (fraction 5).' Fraction 6; composed of peritubular cells, was washed once with solution A and the
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 08 June 2015. at 14:00 For personal use only. No other uses without permission. . All rights reserved.
620
Endo • 1975 Vol 96 • No 3
WELSH AND WIEBE
supernatant from this wash, containing more tubule fragments, was added to fraction 5. Fraction 5, composed of Sertoli cells and germ cells, was washed twice in 15 ml of solution A and once with 15 ml solution A containing 5% fetal calf serum (Grand Island Biological Co.). The cells were resuspended in 10 ml solution A with 5% fetal calf serum. Fraction 5 was passed through a syringe needle (no. 17 gauge, 7.5 cm) at a rate of 5 ml/sec to break the tubular fragments into groups of Sertoli cells containing embedded spermatogonia and spermatocytes. The cell suspension was shaken moderately on a vortex mixer (ColeParmer Super-Mixer set at 4.5) for 15 sec to loosen and detach embedded germ cells from the Sertoli cell clusters. The cell suspension was immediately layered over a 400 ml gradient of 2%-6% sucrose in solution A at 4 C in a siliconized glass column of 5 cm diameter and 26 cm depth. Twenty minutes after the cell suspension containing Sertoli-germ cell groups and free germ cells was introduced into the column, the bottom 15% of the gradient was rapidly collected and discarded. The next 60% of the gradient volume was collected and then centrifuged at 150 x g for 5 min (4 C). The pellet was washed in solution A and recentrifuged. The remainder of the gradient volume was discarded. The purity of the pelleted Sertoligerm cell fraction (fraction 7) was estimated by aceto-orcein smear preparations and the viability, by trypan blue exclusion. (22). Cell culture
13.2, L-aspartate 13.3, L-glycine 7.5, L-glutamine 692.0, L-proline 11.5, L-serine 10.5, i-inositol 14.0, glutathione 2.0, cyanocobalamin 0.1, biotin 0.2, p-aminobenzoic acid 0.2, ascorbic acid 8.5, and sodium pyruvate 50.0. All ingredients were dissolved in deionized, glass distilled water and sterilized by filtration through a 0.22 /x pore filter. For culturing, the Sertoli preparation (fraction 7) was resuspended in 5 ml of solution A. Cultures were started in 50 mm sterile petri dishes with 0.5 ml of Sertoli cell suspension and 10 ml of cell culture medium containing 5% fetal calf serum. The cultures were maintained in an incubator with humidified air atmosphere. The cultures were maintained at 32 C for 20 h, then the temperature was raised to 41 C for 40 h after which time the temperature was lowered to and maintained at 32 C. Histochemistry Small pieces of the live tissue preparations were incubated for 40 min at 27 C in histochemical incubation medium. This medium was composed of 2 ml solution A, 0.25 ml nitro BT (0.25 mg/0.25 ml water), 0.5 ml NAD (2.5 mg/0.5 ml water), 0.25 ml 3/3-hydroxy5j8-androstan-17-one (0.4 mg/ml propylene glycol-ethanol, 1:1), with the pH adjusted to pH 7.7 with a few drops of 0.1M Na2HPO4. Controls lacked substrate.
Results
Using aceto-orcein smear preparations, we identified and counted the cells of fraction 7. Sertoli cells comprised from 60% to over 80% of fraction 7 (Table 1, Fig. 2f). Viability of greater than 95% was indicated by trypan blue exclusion. The contaminants were spermatogonia and primary spermatocytes, and to a much lesser degree, peritubular cells. A final preparation of several hundred milligrams resulted from 2 Abbreviations: NAD = nicotinamide adenine di- 2-3 g of testis tissue. nucleotide; Nitro BT = nitro blue tetrazolium [2,2'The germ cell fraction (fraction 2) condi-p-nitrophenyl - 5,5'-diphenyl-3,3' - (3,3' - dimethoxy- tained 95%-97% germ cells. The germ cells 4,4'-diphenylene) ditetrazolium chloride]; HEPES were mainly spermatocytes with a lesser = N-2-hydroxyethylpiperazine-N'-2-ethane sulfonic acid; BES = N,N-bis[2-hydroxyethyl]-2-aminoethane number of spermatogonia. Histological sulfonic acid; PIPES = piperazine-N,N'-bis[2 ethane preparations showed the peritubular fracsulfonic acid]. tion (fraction 6) to be composed of almost Final Sertoli cell preparations (fraction 7) were maintained in what is basically Eagle's minimal essential medium (24), but with the following modifications. The inorganic salts were NaCl 100 mM, KC1 15 mM, MgSO4 0.06 mM, CaCl2 0.4 mM, Na2HPO4 0.2 mM, with the zwitterionic buffers HEPES 8 mM,2 BES 8 mM and PIPES 8 mM (25). The medium also contained (in mg/liter): L-alanine 8.9, L-asparagine
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 08 June 2015. at 14:00 For personal use only. No other uses without permission. . All rights reserved.
RAPID ISOLATION OF RAT SERTOLI CELLS
80% peritubular cells, about 15% Sertoli and germ cells and a small percentage of connective tissue and Leydig cells. The interstitial fraction (fraction 4) contained vessels and capillaries of the circulatory system, connective tissue, Leydig cells and small tubule fragments. Enzyme histochemistry was utilized to estimate the amount of Leydig cell contamination in fractions other than the interstitial fraction. Most of the 3/3-hydroxysteroid dehydrogenase activity usually associated with Leydig cells was found in the interstitial fraction (fraction 4) (Fig. 2a); some activity was also found in the peritubular fraction and could be localized histochemically in Leydig cells contained in fraction 6 (Fig. 2b). Significantly, the Sertoli fraction (fraction 5) was found to be completely free of Leydig cell contamination (Fig. 2c). Maintenance of the final Sertoli preparation in culture also revealed no Leydig cell contamination, and in most cultures no cells with the appearance of fibroblasts were seen. Culturing allowed preparations consisting of 95%-98% Sertoli cells to be realized within 3 days (Fig. 2 g,h). Cultures were maintained for as long as 28 days with cells maintaining morphological characteristics normally associated with differentiated Sertoli cells as seen at the light microscope level. Discussion After the testis tissue is minced, the germ cells which are not deeply embedded in the Sertoli cells escape from the tubules. The germ cells appear to be expelled from the tubule as if by an internal pressure, possibly a result of contractile tension produced by the peritubular cells (26). The tubules are freed from the interstitial tissue matrix by collagenase treatment and agitation. The interstitial fragments exhibit an adhesive quality which the tubule fragments lack. The interstitial fragments tend to reaggregate into one mass that may
621
TABLE 1. Sertoli cells in final cell preparations (fraction 7) from rats 15 to 29 days of age Cells counted8 Rat age (days)
Total
Sertoli
15 22 23 24 25 27 28 29
732 785 637 984 1,428 712 2,067 778
574 609 424
Percent
Percent Sertoli in wlinl*»
593
869 584 1,703 467
cells
tubule"
78 78 67 60 61 82 82 60
43 31 29 27
25 23 22 20
a Cells from fraction 7 were stained with 1% Orcein in 45% acetic acid. Cells were identified by nuclear morphology according to Clermont's system of classification (23). Cells identified as not Sertoli were mainly primary spermatocytes. b Percentage Sertoli cells in tubule estimated from sections prepared from samples of initial testis tissue used for each Sertoli cells preparation. Tissue samples fixed in Bouin's fluid for 36 h, embedded in paraffin, cut into sections 5 fim thick and stained in hematoxylineosin. A minimum of 10 tubule cross sections or 800 cells identified was used for each value.
be removed using forceps. After washing, the remaining tubule fragments are treated with pancreatin which frees the peritubular cell layer (Fig. 2d,e) from the Sertoli and germ cells. The peritubular cells also display adhesive characteristics and often reaggregate into one or several masses. Following removal of the outer peritubular cell layer, the peripherally located germ cells, such as the type A spermatogonia, are also detached from the Sertoli cells. The Sertoli-germ cell pieces are subjected to mechanical disruption; sufficient sheer force to break up the large cell groups can be generated by passing the fragments through a syringe needle. Moderate agitation produced by a vortex-mixer dislodges additional germ cells. Using a suspension of individual Sertoli cells amid pachytene spermatocytes, we have not found it possible to separate the two cell types. Separation was not accomplished by either velocity sedimentation (27,28) or density gradient centrifugation. If, however, the Sertoli cells are sepa-
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 08 June 2015. at 14:00 For personal use only. No other uses without permission. . All rights reserved.
622
WELSH AND WIEBE
Endo • 1975 Vol 96 • No 3
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 08 June 2015. at 14:00 For personal use only. No other uses without permission. . All rights reserved.
RAPID ISOLATION OF RAT SERTOLI CELLS
623
FIG. 2. a) to c) Histochemical demonstration of 3/3-hydroxysteroid dehydrogenase activity in cell fractions from 29-day-old rat. Histochemistry used nitro BT, NAD, and 3/3-hydroxy-5/3-androstan-17-one, for 40 min at 27C and pH 7.7. a) fraction 4, interstitial tissue. (xlOO) b) fraction 6, peritubular tissue. (xlOO) c) fraction 5, Sertoli-germ cells. (xlOO) d) Aceto-orcein stained whole-mount of tubule fragments (fraction 3) of 29-day-old rat. Note peritubular cell layer (arrows) around tubules. (xl60) e) Aceto-orcein stained whole-mount of tubule fragments (fraction 5) of 29-day-old rat. Note absence of peritubular cells after pancreatin treatment. (xl60) f) Aceto-orcein stained Sertoli cell preparation from 29-day-old rat (fraction 7) showing Sertoli cells (pale nuclei) with embedded primary spermatocytes (dark nuclei). (x500) g) Sertoli cell monolayer from 27-dayold rat after 66 h in culture under conditions described under Cell culture in text. Culture fixed in Bouin's fluid for 15 min, stained in Harris' hematoxylin and eosin, (x450) h) Sertoli cells from 30-day-old rat after 9 days in culture. Fixed in Bouin's, stained in Harris' hematoxylin and sudan IV, mounted in aqueous glycerin-jelly medium. Arrows indicate lipid droplets. (xlOOO).
rated into small groups of cells instead of individual cells, a rapid sedimentation can be used to separate the Sertoli cell groups from the free germ cells. A gradient is used to prevent convection currents in the column and reduce mixing. Maintenance of the cells in vitro, temporarily at an elevated temperature, destroys almost all of the germ cells which are then phagocytized by the Sertoli cells. The Sertoli cells appear to be unharmed by the temporary temperature increase and form a monolayer (Fig. 2g,h) which has been maintained for 28 days. Chowdhury and Steinberger (29) have indicated that higher temperature (43 C) for a shorter period of time, although injurious to germ cells, apparently is not harmful to Sertoli cells in vivo. It would be of value to mention other procedures that were tried before the present method was accepted into routine use in our laboratory. In addition to the enzymes finally selected, we also employed crude and purified trypsin, lysozyme, hyaluronidase, elastase, pepsin, peptidase, a-chymotrypsin and deoxyribonuclease I (all enzymes from Sigma Chemical Co.). These enzymes were used in various concentrations, combinations and sequences. The results of using two different chelating agents as well as the inclusion of cations (Mg++, Mn ++ , Ca ++ ) were also evaluated. Many of these treatments had little noticeable effect on cell separation. While certain concentrations of some of the enzymes allowed tissue separation, viability suffered greatly. All preparations of trypsin resulted
in great amounts of a mucoid-like substance from which the cells could only be separated with great difficulty. Neither elastase nor deoxyribonuclease I alleviated this problem satisfactorily. Testis tissue from other species including toad (Bufo marinus); Japanese quail (Coturnix coturnix japonica) and sunfish (Lepomis megalotus) was given similar treatment as described in procedure and outlined in Fig. 1; none of these tissues separated to any degre'e as a result of our procedure or slight modifications of it. Apparently only rat testis tissue, and presumably tissue of other mammalian species, responds to this method. This method for rapid preparation of a population of cells from the rat testis enriched in Sertoli cells has several advantages over other methods. The method is relatively fast and can be performed on normal rats. Hormonal alterations of the animal are not required nor are treatments with agents of unknown actions such as radiation or chemicals. The final cell preparation can be demonstrated to be free of Leydig cell contamination and, in culture, free of fibroblasts. The method can allow comparison of function of the various tissues of the testis because the interstitial fraction, peritubular cell fraction, germ cell fraction and Sertoli cell fraction can be derived from the same initial tissue. References 1. Steinberger, E., Physiol Rev 51: 1, 1971. 2. Dym, M., and D. W. Fawcett, Biol Reprod 3: 308, 1970. 3. Lacy, D., G. P. Vinson, P. Collins, J. Bell, P.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 08 June 2015. at 14:00 For personal use only. No other uses without permission. . All rights reserved.
624
4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14.
15.
WELSH AND WIEBE Fyson, J. Pudney, and A. J. Pettitt, In Gual, C. (ed.), Progress in Endocrinology, Proc. 3rd Int. Congr. Endocr. Mexico, 1968, Excerpta Medica Foundation, Amsterdam, 1969, p. 1019. Dorrington, J. H., and I. B. Fritz, Endocrinology 94: 395, 1974. Means, A. R., In Greep, R. O., and D. W. Hamilton (eds.), Handbook of Physiology, American Physiological Society, Bethesda, Md. (In press). Davies, A. G.J Reprod Fertil 25: 21, 1971. Murphy, H. D., Proc Soc Exp Biol Med 48: 1202, 1965. Christensen, A. K., and N. R. Mason, Endocrinology 76: 646, 1965. Lam, D. K. M., R. Furrer, and W. R. Bruce, Proc Natl Acad Sci USA 65: 192, 1970. Fawcett, D. W., S. Ito, and D. Slautterback, J Biophysic Biochem Cytol 5: 453, 1959. Clermont, Y., and B. Perey, Am J Anat 100: 241, 1957. Nagy, F.J Reprod Fertil 28: 389, 1972. Roosen-Runge, E. C., Biol Rev 37: 343, 1962. Steinberger, E., and A. Steinberger, In Gual, C. (ed.), Progress in Endocrinology, Proc. 3rd Int. Congr. Endocr., Mexico, 1968, Excerpta Medica Foundation, Amsterdam, 1969, p. 1038. Esaki, S., Z Mikrosk Anat Forsch 15: 365, 1928.
Endo • 1975 Vol 96 • No 3
16. Nagy, F., Fertil Steril 24: 805, 1973. 17. Johnsen, S. G., In Rosenberg, E., and C. A. Paulsen (eds.), The Human Testis, Plenum Press, New York, 1970 p. 231. 18. Kodani, M., and K. Kodani, Proc Natl Acad Sci USA 56: 1200, 1966. 19. DeKretser, D. M., K. J. Catt, M. L. Dufau, and B. Hudson, 7 Reprod Fertil 24: 311, 1971. 20. Dufau, M., D. M. deKretser, and B. Hudson, Endocrinology 88: 825, 1971. 21. Yamada, M., S. Yasue, and K. Matsumoto, Endocrinology 93: 81, 1973. 22. Merchant, D. J., R. H. Kahn, and W. H. Murphy Jr., Handbook of Cell and and Organ Culture, Burgess Publishing Co., Minneapolis, Minn., 1964, p. 157. 23. Clermont, Y., Physiol Rev 52: 198, 1972. 24. Eagle, H., Science 130: 432, 1959. 25. , Science 174: 500, 1971. 26. Leeson, C. R., and T. S. Leeson, Anat Rec 147: 243, 1963. 27. Miller, R. G., and R. A. Phillips, J Cell Physiol 73: 191, 1969. 28. Meistrich, M. L., W. R. Bruce, and Y. Clermont, Exptl Cell Res 79: 213, 1973. 29. Chowdhury, A. K., and E. Steinberger, J Reprod Fertil 22: 205, 1970.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 08 June 2015. at 14:00 For personal use only. No other uses without permission. . All rights reserved.
