THE AMERICAN JOURNAL OF ANATOMY 192:281-292 (1991)
Structure-Function Relationships During Preovulatory Development of Porcine Follicles Following Equine Chorionic Gonadotropin Stimulation G.J. KING, C. CHAPEAU, AND C.A. ACKERLEY Department of Animal and Poultry Science, University of Guelph, Guelph, Ontario, Canada N l G 2Wl
ABSTRACT Porcine follicular maturation begins by recruitment from a continually proliferating pool of small antral follicles; those receiving the appropriate stimulus differentiate rapidly through a series of structural and functional changes. Such ovarian activity can be induced in prepubertal gilts with a single injection of equine chorionic gonadotropin (eCG). Average follicular diameter in eCG treated females increased from approximately 2 mm before stimulation to 3.5 mm by 24 hr after injection, with subsequent growth to ovulatory size (8or 9 mm) by 96 hr. Both thecia and granulosa layers increased in thickness and complexity, and a prominent capillary bed evolved immediately outside the basement membrane ,separating the two layers. Cytoplasmic organelles associated with increased metabolic activity and steroidogenesis proliferated within the fir,st 24 hr. Progressive changes included increasing amounts of lipid and rough and smooth endoplasmic reticulum, with the latter occurring in vesicular or lamellar forms and as lipid-associated whorls. Bizarre mitochondrial forms also appeared, often associated with lipids. The amount and proportion of rough and smooth endoplasmic reticulum shifted dramatically as follicles matured. By 24 hr, rough endoplasmic reticulum in thecal cells increased from 4.2 to 7% of cell volume, while the amount in granulosa cells increased from less than 3.5% to inore than 10%; the quantity remained relatively constant in the theca but declined to prestimul at'ion values in the granulosa layer. Rough endoplasmic reticulum predominated over smooth in the first 24 hr following stimulation but the proportions were then reversed, so that more than 10%of both layers was composed of smooth endoplasmic reticulum by the time ovulation was imminent. Some follicles had or were in the process of ovulating by 96 hr. Their walls were collapsed into prominent folds with the two cell types beginning to mix. Slight undulations and some regions of discontinuity were observed in basement membranes of large unovulated follicles at this time. In specimens collected at 96 hr poststimulation and processed for retention of lipid, lipid-like material was noticeable in the extracellular matrix sur0 1991 WILEY-LISS, INC
rounding cells that contained organelle configurations suggestive of steroidogenesis. INTRODUCTION
Porcine follicular maturation begins with recruitment from a proliferating pool of very small antral structures that possess thin but distinct theca and granulosa layers surrounding the central cavity. Most designated follicles develop and differentiate rapidly, proceeding from selection to ovulation in a few days time, while demonstrating substantial changes in synthesis of steroids (Ainsworth e t al., 1980), prostaglandins (Evans et al., 1983), growth factors (Hammond et al., 1988), proteins, and carbohydrates (Chang et al., 1976). The classical two-cell theory of steroidogenesis (Falck, 1959), with theca cells synthesizing androgens and transporting these to granulosa cells for aromatization, is not truly applicable in pigs. In this species, the theca cells also produce substantial estrogen; and luteinization is induced almost immediately after exposure of cultured cells to human chorionic gonadotropin (hCG) (Evans et al,. 1981). Corner (1919) presented a comprehensive light-microscopic description of structure-function relationships for porcine theca and granulosa cells during ovulation and formation of the corpora lutea. Ultrastructural information on the wall, however, is limited to a very few reports: one describes granulosa cells and their organelles in relation to steroidogenesis (Bjersing, 19671, and another describes features of only the theca interna cells (Krzysztofowicz and Stoklosowa, 1977). In a n attempt to elucidate changes in granulosa-theca structure or relationships that might be related to steroidogenesis during development, a series of pig follicles were collected, processed, and examined at various times after folliculogenesis was stimulated by a n injection of equine chorionic gonadotropin (eCG). MATERIALS AND METHODS
Prepubertal Large White-Landrace crossbred, nulliparous porcine females (gilts) approximately 6 months of age and 90 kg weight were injected intra-
Received February 22, 1991. Accepted May 31, 1991. C.A. Ackerley's current address is Department of Pathology, Hospital for Sick Children, 555 University Ave., Toronto, Ontario, Canada M5G 1x8.
282
G.J. KING ET AL.
muscularly with 750 IU eCG to stimulate follicular development. Animals were slaughtered at 12 h r (n = 3), 24 h r (n = 4), 48 hr (n = 5 ) , 72 h r (n = 4), and 96 h r (n = 6) postinjection. Three additional gilts of similar breeding age and weight were processed without injection to provide unstimulated control or 0 h r specimens. Reproductive tracts were removed immediately after exsanguination, and ovaries were perfused through the uterine artery with a solution containing 4%paraformaldehyde plus 2.5%glutaraldehyde in phosphate buffer (pH 7.2, 0.1 M). Individual follicles were dissected free and immersed in fresh fixative for 1 hr. Follicles were then cut in half, fixed for a n additional 5 hr, washed 3 times for 15 min in phosphate buffer, and stored in the same buffer at 4°C until further processing. Follicle halves were dehydrated through a n ascending series of ethanols and embedded in JB-4 methacrylate resin (Polyscience Inc., PA) oriented in such a manner to facilitate obtaining cross sections. For examination by light microscopy, 2 pm sections were stained with 1% toluidine blue 0 in 1%sodium borate. Strips of follicle walls (1 x 2 mm long) were cut prior to postfixation with 2% OsO, in water for 1hr. Specimens were either dehydrated to 70% ethanol and infiltrated with Epon resin, as described by Idelman (1964) for the retention of lipids, or dehydrated in a n ascending series of ethanols and infiltrated with Epon via propylene oxide. Samples were then flat embedded in aluminum weigh boats and polymerized a t 60°C for 48 hr. Pieces of tissue were then cut out with a jeweller’s saw and affixed to dummy Epon blocks with cyanoacrylate to obtain a cross-section of the follicle wall. U1trathin sections exhibiting a pale-gold interference color were cut on a diamond knife, mounted on grids, and stained with ethanolic uranyl acetate and lead citrate. The grids were viewed and photographed in a JEOL TEM lOOCX transmission electron microscope. For morphometric analysis of the proportion of rough and smooth endoplasmic reticulum, negative images were projected onto a digitized tablet. At least 100 theca and 100 granulosa cells from each stage were analyzed using methods described by Wiebel (1965). Data were expressed as a mean and standard deviation and plotted for individual sampling times. RESULTS
Many small antral follicles with diameters around 2 mm were present a t the surfaces of all ovaries recovered from unstimulated (0 hr) gilts and those slaughtered 12 h r after injection with eCG. By 24 hr poststimulation, these small follicles were still numerous but some had grown into more prominent structures with the largest reaching 3.5 mm in diameter. Antral follicles continued to increase in size, but numbers were reduced so that ovaries collected 48 h r after eCG injection had 3-12 obviously enlarging follicles with mean diameters about 5 mm. Rapid growth continued through the subsequent 2 days. At 96 h r postinjection, ovaries from 4 gilts had varying numbers of follicles ranging in diameter from 6 to 9 mm but no ovulations. Ovaries recovered from 2 additional gilts a t this stage had both ovulatory sized and recently ovulated follicles (corpora hemorrhagica). Those females slaughtered be-
tween 24 and 96 h r postinjection showed slight to marked vulvar edema and reddening. Follicle walls from unstimulated gilts possessed a n outer theca externa made up of compact connective tissue that blended into the theca interna. The interna region was represented by a thin stratified layer of elongated polyhedral cells interspaced with spindleshaped fibroblasts in a collagen matrix, adjacent to a basement membrane (Figs. 1, 2). The granulosa layer on the opposite side of the basement membrane completely lined the antrum and formed the cumulus mass of cells surrounding the oocyte. This general relationship remained constant throughout folliculogenesis. A few scattered cisternae of smooth and rough endoplasmic reticulum and the occasional whorl (Fig. 2) were present in both theca and granulosa cells. The stratified columnar epithelial cells forming the granulosa were diverse in shape with tortuous lateral interdigitations and somewhat loosely associated boundaries (Figs. 1, 3). Free ribosomes, inconspicuous Golgi complexes, and lipid droplets were constant features a t all stages. Some lipid was located in the basal portions of granulosa cells a t 0 hr, and it increased by 24 hr. One interesting feature of some early granulosa cells was the presence of a single cilium (Fig. 3, inset). Althrough gross examination of ovaries collected a t 12 h r poststimulation showed no obvious differences from those recovered prior to eCG injection, some microscopic changes were observed. Growth occurred in both the granulosa and theca interna regions (Fig. 4 vs. Fig. 11, and this development was even more pronounced at later stages. Extracellular spaces between granulosa cells decreased substantially by 12 hr, and numerous free ribosomes were present in cells in both layers so the cytoplasm appeared “less dense” (Figs. 5, 6,). The theca cells, first identified by their spindle shape and their location in the loose connective tissue immediately beneath the basement membrane plus the presence of some lipid material and a network of rough endoplasmic reticulum, assumed a less uniform but more rounded shape (Fig. 6). An obvious capillary bed was located immediately outside the basement membrane in all specimens obtained after stimulation. As follicular growth and differentiation continued, mitotic figures were common in the granulosa layer (Figs. 7, 8) and could also be detected in theca cells. Within the first 24 h r after stimulation, organelles associated with increased metabolic activity proliferated, resulting in the accumulation of basal lipid in granulosa cells (Fig. 9) and of abundant rough and smooth endoplasmic reticulum in vesicular and lamellar forms plus lipid-associated whorls (Fig. 10). Thecal cells became plumper and filled with abundant organelles including extensive lipid-associated smooth endoplasmic reticulum (Figs. 10, 12). Bizarre mitochondria1 forms appeared in cells of both regions (Figs. 11, 12), and these were often found in intimate association with lipids (Fig. 12). Both the theca and granulosa layers were relatively compact a t 48 h r (Fig. 13), but cellular association became more discrete a s folliculogenesis progressed (Fig. 14). By 96 hr, thecal cells had accumulated substantial lipid and assumed irregular shapes, ranging from a few that were still elongated through oval to round (Figs. 15, 16). As maturation of the follicular wall progressed and ovulation approached, an-
PORCINE FOLLICULAR DEVELOPMENT
Fig. 1. Light micrograph showing a section of porcine follicular wall prior to any hormonal stimulation. The small antral follicles have a well defined stratified columnar granulosa cell layer (G) separated from a compact theca interna (T) by a distinct basement membrane. x 200. Fig. 2. Low-power electron micrograph of thecal cells from a porcine follicle prior to hormonal stimulation. Typical cytoplasmic features are lipid droplets (L) and sparse endoplasmic reticulum arranged in whorls (W) or scattered throughout the cytoplasm. x 6,500.
tral granulosa cells commenced dissociation, isolated edematous regions formed in the theca, and slight to moderate undulations appeared in the adjacent haseinent membrane (Fig. 16). Much of the follicular fluid escaped with the oocyte and cumulus at ovulation, allowing the wall to collapse into prominent folds abliterating the antrum (Fig. 17). The postovulatory plica-
283
Fig. 3. Low-power electron micrograph of granulosa cells from a porcine follicle prior to hormonal stimulation. These cells demonstrate loose interdigitation along lateral membranes and sparse lipid droplets located predominantly in basal cytoplasm. Inset: Oblique section through a granulosa cell and basal plate of cilium. x 3,500; inset, x 15,770. Fig. 4. Light micrograph of a section of porcine follicular wall 12-hr posthormonal stimulation. Both the theca (TI and granulosa ( G ) layers have increased in thickness compared with the prestimulation follicle in Figure 1, and a prominent capillary bed is apparent immediately beneath the basement membrane. x 200.
tions resulted in thecal cells, plus their associated connective tissue matrix and capillaries, penetrating into the remaining granulosa. Increasing amounts of smooth endoplasmic reticulum, arranged in whorls or randomly (Figs. 18, 191, appeared throughout the cytoplasm a s follicular enlargement proceeded. Some regions of discontinuous
284
G.J. KING ET AL.
Fig. 5. Low-power electron micrograph illustrating basal regions of granulosa cells 12 hr posthormonal stimulation of follicular maturation. The extracellular spaces have decreased in comparison with those present in unstimulated follicles (Fig. 3). x 3,600.
Fig. 6. Low-power electron micrograph of thecal cells in various forms of differentiation, 12 hr poststimulation. The cytoplasm contains considerable endoplasmic reticulum and mitochondria with occasional lipid droplets. x 4,680.
PORCINE FOLLICULAR DEVELOPMENT
Fig. 7. Light micrograph of porcine follicular wall 24-hr pxthormonal stimulation with several prominent mitotic figures (arrowheads) in the granulosa layer (G). X 350.
285
Fig. 9. Electron micrograph of the basal portion of granulosa cells 24-hr poststimulation. There is a prominent basal lamina and the cells contain abundant basal lipid plus smooth and rough endoplasmic reticulum. x 18,500.
Fig. 8. Light micrograph of porcine follicular wall 48-hr posthormonal stimulation with numerous mitotic figures (arrowheads) in granulosa cells. x 350.
basal lamina could be observed in all postovulatory specimens (Fig. 20). Extensive vesicular endoplasmic reticulum and lipid droplets displaying varying degrees of osmiophilia were prominent features (Fig. 21). A substantial proportion of the intracellular lipid was enclosed within laminated membranes which appeared to diminish as droplets were located closer to the periphery of the cells. Some extracellular lipid appeared
a s droplets in the intercellular matrix surrounding the immediately postovulatory luteal cells in all specimens embedded after only partial dehydration (Fig. 22). One major feature observed throughout the entire process of follicular maturation was the dramatic change in the amounts and proportions of rough and smooth endoplasmic reticulum that occurred a s follicles matured. Within the first 24 hr, rough endoplas-
286
G.J. KING ET AL.
Fig. 10. Electron micrograph ofthecal cells at 24-hr poststimulation illustrating a whorl of rough endoplasmic reticulum surrounding a lipid droplet, a common feature at this stage. x 14,110.
Fig. 11. Electron micrograph illustrating several bizarre lamellar mitochondria1 forms and extensive smooth endoplasmic reticulum,
but only scattered segments of rough endoplasmic reticulum, in a granulosa cell at 48-hr poststimulation. x 17,500. Fig. 12. Electron micrograph of thecal cell cytoplasm a t 72-hr poststimulation with oval and globular vesicular mitochondria, extensive smooth endoplasmic reticulum, sparse rough endoplasmic reticulum, and large lipid droplets. x 14,525.
PORCINE F'OLLICULAR DEVELOPMENT
287
Fig. 13. Light micrograph of a section of porcine follicular wall in vicinity of the still vesicular oocyte enclosed within a compacted cumulus cell mass, a t 48-hr poststirnulation. Both granulosa ( G ) and theca (T)layers are thicker than a t previous stages. x 180.
Fig. 14. Light micrograph of a follicular wall segment and oocyte surrounded by cumulus cells at 72-hr poststimulation. The cumulus and innermost granulosa cells show signs of impending dispersion while thecal cells in this particular location have apparently assumed a more rounded appearance and less compacted association. x 180.
mic reticulum in thecal cells increased from 4.2-7% of total volume, while the amount in granulosa cells went from less than 3.5%to more than 10%. For the reinainder of the period studied, the quantity remained relatively constant in thecal cells but declined rapidly to prestimulation values in granulosa cells. After 4.8 hr, smooth predominated over rough with over 10% of' total volume for cells of both regions composed of smooth endoplasmic reticulum by the time ovulation was imminent (Figs. 23, 24).
granulosa layers preceding and at the time of ovulation and the incorporation of both components into the corpus luteum agreed with the previous description (Corner, 1919). Also, comparable increases in endoplasmic reticulum, mitochondria with tubular cristae, and lipids and their associations indicative of steroidogenesis were observed by Bjersing (1967). Hunter et al. (1989) found the granulosa layer to be of uniform thickness with no infolding prior to the LH surge, but they noted marked infolding of both granulosa and theta in some follicles just prior to ovulation. Slight to moderate undulations and some discontinuity of basement membrane separating the layers of intact follicles were present a t 96 h r in some regions of the unovulated follicles in the current study, but little obvious infolding was observed even in follicles from ovaries with some postovulatory structures already present. This discrepancy may be the result of fixation differences since Bouin's was used by Hunter et al. (1989) while vascular perfusion with paraformaldehyde-glutaraldehyde was performed in the current series. Alternately, some morphological and biochemical differences can be detected between follicles of naturally cycling and hormonally stimulated gilts, suggesting that ovaries from prepubertal gilts might function differently from those in mature animals (Wiesak et al., 1990).There is, however, abundant evidence indicating substantial heterogeneity between follicles of similar size taken from sexually mature pigs (Foxcroft and Hunter, 1985; Hunter et al., 1989; Grant et al., 1989; Xie et al., 1990). Also, follicles stimulated to ovulation in prepubertal gilts must be reasonably competent since the released oocytes can be fertilized and pregnancy established (Shaw et al., 1971). The prepubertal model is certainly
DISCUSSION
In sexually mature pigs, increased episodic Iuteinizing hormone (LH) pulses combined with adequate concentrations of follicle-stimulating hormone (FSH) promote follicular recruitment from a continually emerging pool of small antral follicles. As growthL continues, theca and granulosa layers acquire enhanced steroidogenic activities that are essential for development of feedback mechanisms necessary for maturation and ovulation. Similar events occur when ovaries of prepubertal gilts are stimulated with eCG, and in this study, all treated females responded with a progressive increase in follicular size comparable to that reported by Ainsworth et al. (1980). Size variability between follicles within animals after eCG treatment may result from a constantly changing hormonal environment, producing different rates of tissue differentiation and growth. The slight to marked vulvar edema and reddening noted in those females slaughtered between 24 and 96 h r postinjection indicated target organ response to increased plasma estrogen concentrations which would be expected in stimulated gilts. Microscopic alterations t h a t occurred in theca and
288
G.J. KING ET AL.
Fig. 15. Light micrograph of a section of wall from a n unovulated follicle a t 96-hr poststimulation. The theca (TI and granulosa t G ) layers are still separated by an intact basement membrane and prominent capillary bed. x 180.
region. The theca ( T )has considerable edema, dispersing the cells and creating undulations in the theca-granulosa interface. x 180.
Fig. 16. Light micrograph illustrating another wall segment of t h e 96-hr follicle above. Antral granulosa cells a r e dissociating in this
Fig. 17. Light micrograph of a section of wall from a recently ovulated follicle a t 96-hr poststimulation. The theca (Ti and granulosa tG1 layers have collapsed into prominent folds t h a t almost completely obliterate t h e recently evacuated antrum. x 180.
convenient for obtaining material at known times and is appropriate for many studies, but extrapolations must always be made with caution. The porcine follicular phase extends over 5 full days from recruitment until ovulation (Foxcroft and Hunter, 1985; Grant et al., 1989), with gonadotropin-receptor-
steroid modulated changes occurring throughout. Stimulated follicles developed through to ovulation over a similar time frame, and organelle modifications in both theca and granulosa cells were chronologically related to the increasing steroidogenic capabilities of these cells in vitro (Evans et al., 1981). A substantial
PORCINE FOLLICULAR DEVELOPMENT
Fig. 18. Sagittal plane electron micrograph through granulosa cells at 72-hr poststimulation showing abundance of smooth endoplasmic reticulum in whorls and random arrangements. Some necrosis was evident in the granulosa layer at this stage (arrowhead). x 5,280.
289
Fig. 19. Electron micrograph of granulosa cells desquamated from the basal lamina of a n ovulated follicle at 96-hr poststimulation. Note the shorter, stubbier microvilli and prominence of lysosome-like structures throughout the cytoplasm. X 5,280.
290
G.J. KING ET AL.
Fig. 20. Electron micrograph of postovulatory follicle wall 96 hr after stimulation. Prominent features are a thecal cell in mitosis (T*), separation of irregularly shaped granulosa cells (G), rounding of thecal cells (TI, and a n endothelial cell (El. Inset: Higher power of the upper left region illustrating discontinuity of the basal lamina. x 3,400; inset, x 5,780. Fig. 21. Electron micrograph of a section from the thecal region of a postovulatory follicular wall, 96-hr poststimulation, showing early
luteal cells containing abundant lipid droplets, vesicular smooth endoplasmic reticulum, and mitochondria. x 4,000. Fig. 22. High-power electron micrograph of the thecal-granulosa interface, 96 hr poststimulation, and immediately after ovulation, where dissolution of the basal lamina is occurring. The extracellular lipid (arrowhead) probably represents material extruded from the nearby granulosa cell. x 20,750.
291
PORCINE F'OLLICULAR DEVELOPMENT
t
T
L
R
;1 5 t -
/I
W U
0 10u
t
0
u L
0
a,
m
0
Y
5 -
U
L.
a
I:" 1
I
0
.
I
.
24
I
48
,
I
I
72
I
,-
96
Hours poststimulation
"
0
24
48
72
96
kl o ur s p 0 s ts t i m u I a t ion
Fig. 23. Proportion of total cytoplasmic area of thecal cells occupied by smooth (triangles) and rough (circles) endoplasmic reticulum.
Fig. 24. Proportion of total area of granulosa cell cytoplasm occupied by smooth (triangles) and rough (circles) endoplasmic reticulum.
increase in rough endoplasmic reticulum occurred in both layers within the first 12 h r after injection with eCG. This likely indicates a n almost immediate and dramatic increase in resources to synthesize growth and receptor proteins plus additional enzymes. Smooth endoplasmic reticulum increased steadily i n theca cells suggesting a n increasing capacity to produce steroids from soon after stimulation up to ovulation, in keeping with changes in plasma concentrations (van de Wiel e t al., 1981) and theca cell cultures (Evans et al., 1981). The elevation in proportion of smooth endoplasmic reticulum within granulosa cells occurred considerably later, but in ample time to account for the increasing concentrations of estrogens that occur in follicular fluid as ovulation approaches (Ainsworth et al., :1980; Grant et al., 1989). The onset of sexual receptivity and beginning alf the LH surge commence around the same time (van de Wiel et al., 1981). Ovulation occurs approxim,stely 40 hr after competent follicles receive the appropriate gonadotropin stimulus (Hunter, 1972). Behavioral and ovarian responses in treated gilts indicated thal; the surge should have taken place between 48 and 72 h r postinjection. Plasma gonadal steroid concentrations shift quickly from estrogens in picomolar values during the early follicular phase to progesterone in nanoniolar quantities soon after luteinization commences (va.n de Wiel et al., 1981; Evans et al., 1981), indicating a substantial increase in steroid production and release. The use of a specialized dehydration and embedding p:rocedure for the retention of lipids (Idelman, 1964) allowed demonstration of extracellular lipids (Fig. 21) that were not observed in material processed in the routine manner. This noticeable increase in extracellular lipid was located immediately adjacent to cells containing organelle configurations indicative of steroid synthesis in specimens collected after luteinization should have occurred. As stated above, steroidogenesis has sh.ifted
dramatically a t this stage from release of small quantities of estrogens to much larger quantities of progesterone. Thus, a t least some of the demonstrated lipid probably represents the latter hormone. ACKNOWLEDGMENTS
The capable technical assistance provided by Douglas Way is gratefully acknowledged. Research funds were provided by the Natural Science and Engineering Research Council of Canada and the Ontario Ministry of Agriculture and Food. LITERATURE CITED Ainsworth, L., B.K. Tsang, B.R. Downey, G.J. Marcus, and D.T. Armstrong 1980 Interrelationships between follicular fluid steroid levels, gonadotrophic stimuli, and oocyte maturation during preovulatory development of porcine follicles. Biol. Reprod., 23t621627. Bjersing, L. 1967 On the ultrastructure of follicles and isolated follicular granulosa cells of porcine ovary. Z. Zellforsch., 82t173-186. Chang, S.C.S., J.D. Jones, R.D. Ellefson, and R.J. Ryan 1976 The porcine ovarian follicle: I. Selected chemical analysis of follicular fluid at different stages of development. Biol. Reprod., 15t321328. Corner, G.W. 1919 On the origin of the corpus luteum of the sow from both granulosa and theca interna. Am. J. Anat., 26,117-183. Evans, G., M. Dobias, G.J. King, and D.T. Armstrong 1981 Estrogen, androgen and progesterone biosynthesis by theca and granulosa of preovulatory follicles in the pig. Biol. Reprod., 25:673-682. Evans, G., M. Dobias, G.J. King, and D.T. Armstrong 1983 Production of prostaglandins by porcine preovulatory follicular tissue and their role in intrafollicular function. Biol. Reprod., 28,322-328. Falck, B. 1959 Site of production of oestrogen in rat ovary as studied in micro-transplants. Acta Physiol. Scand., 47, Suppl. 163:94101. Foxcroft, G.R., and M.G. Hunter 1985 Basic physiology of follicular maturation in the pig. J . Reprod. Fertil., 33t1-19 (Suppl). Grant, S.A., M.G. Hunter, and G.R. Foxcroft 1989 Morphological and biochemical characteristics during ovarian follicular development in the pig. J. Reprod. Fertil., 86t171-183. Hammond, J.M., C.J. Hsu, J . Klindt, B.K. Tsang, and B.R. Downie 1988 Gonadotrophins increase concentrations of immunoreactive insulin-like growth factor-I in porcine follicular fluid in vivo. Biol. Reprod., 38,304-308.
292
G.J. KING ET AL.
Hunter, M.G., S.A. Grant, and G.R. Foxcroft 1989 Histological evidence for heterogeneity in the development of preovulatory pig follicles. J. Reprod. Fertil., 86:165-170. Hunter, R.H.F. 1972 Ovulation in the pig: timing of the response to injection of human chorionic gonadotrophin. Res. Vet. Sci., 13: 356-360. Idelman, S. 1964 Modification de la technique de Luft en vue de la conservation des lipides en microscopie electronique. J. Microsc., 3:715. Krzysztofowicz, A,, and S. Stoklosowa 1977 Ultrastructure of theca interna cells of porcine ovarian follicles. Anat. Hist. Embryol., 6.359-364. Shaw, G.A., B.E. McDonald, and R.D. Baker 1971 Fetal mortality in the prepubertal gilt. Can. J. Anim. Sci., 51233-236.
van de Wid, D.F.M., J. Erkens, W. Kops, E. Vos, and A.J. van Landeghem 1981 Periestrous and midluteal time courses of circulating LH, FSH, estradiol-17p and progesterone in the domestic pig. Biol. Reprod., 24:223-233. Wiebel, E.R. 1965 Stereological principals for morphometry in electron microscopic cytology. Int. Rev. Cytol., 26:235-255. Wiesak, T., M.G. Hunter, and G.R. Foxcroft 1990 Differences in follicular morphology, steroidogenesis and oocyte maturation in naturally cycling and PMSGhCG-treated prepubertal gilts. J. Reprod. Fertil., 89~633-641. Xie, S., D.M. Broermann, K.P. Nephew, J.S. Ottobre, M.L. Day, and W.F. Pope 1990 Changes in follicular endocrinology during final maturation of porcine oocytes. Domest. Anim. Endocrinol., 7:7582.
Structure-Function Relationships During Preovulatory Development of Porcine Follicles Following Equine Chorionic Gonadotropin Stimulation G.J. KING, C. CHAPEAU, AND C.A. ACKERLEY Department of Animal and Poultry Science, University of Guelph, Guelph, Ontario, Canada N l G 2Wl
ABSTRACT Porcine follicular maturation begins by recruitment from a continually proliferating pool of small antral follicles; those receiving the appropriate stimulus differentiate rapidly through a series of structural and functional changes. Such ovarian activity can be induced in prepubertal gilts with a single injection of equine chorionic gonadotropin (eCG). Average follicular diameter in eCG treated females increased from approximately 2 mm before stimulation to 3.5 mm by 24 hr after injection, with subsequent growth to ovulatory size (8or 9 mm) by 96 hr. Both thecia and granulosa layers increased in thickness and complexity, and a prominent capillary bed evolved immediately outside the basement membrane ,separating the two layers. Cytoplasmic organelles associated with increased metabolic activity and steroidogenesis proliferated within the fir,st 24 hr. Progressive changes included increasing amounts of lipid and rough and smooth endoplasmic reticulum, with the latter occurring in vesicular or lamellar forms and as lipid-associated whorls. Bizarre mitochondrial forms also appeared, often associated with lipids. The amount and proportion of rough and smooth endoplasmic reticulum shifted dramatically as follicles matured. By 24 hr, rough endoplasmic reticulum in thecal cells increased from 4.2 to 7% of cell volume, while the amount in granulosa cells increased from less than 3.5% to inore than 10%; the quantity remained relatively constant in the theca but declined to prestimul at'ion values in the granulosa layer. Rough endoplasmic reticulum predominated over smooth in the first 24 hr following stimulation but the proportions were then reversed, so that more than 10%of both layers was composed of smooth endoplasmic reticulum by the time ovulation was imminent. Some follicles had or were in the process of ovulating by 96 hr. Their walls were collapsed into prominent folds with the two cell types beginning to mix. Slight undulations and some regions of discontinuity were observed in basement membranes of large unovulated follicles at this time. In specimens collected at 96 hr poststimulation and processed for retention of lipid, lipid-like material was noticeable in the extracellular matrix sur0 1991 WILEY-LISS, INC
rounding cells that contained organelle configurations suggestive of steroidogenesis. INTRODUCTION
Porcine follicular maturation begins with recruitment from a proliferating pool of very small antral structures that possess thin but distinct theca and granulosa layers surrounding the central cavity. Most designated follicles develop and differentiate rapidly, proceeding from selection to ovulation in a few days time, while demonstrating substantial changes in synthesis of steroids (Ainsworth e t al., 1980), prostaglandins (Evans et al., 1983), growth factors (Hammond et al., 1988), proteins, and carbohydrates (Chang et al., 1976). The classical two-cell theory of steroidogenesis (Falck, 1959), with theca cells synthesizing androgens and transporting these to granulosa cells for aromatization, is not truly applicable in pigs. In this species, the theca cells also produce substantial estrogen; and luteinization is induced almost immediately after exposure of cultured cells to human chorionic gonadotropin (hCG) (Evans et al,. 1981). Corner (1919) presented a comprehensive light-microscopic description of structure-function relationships for porcine theca and granulosa cells during ovulation and formation of the corpora lutea. Ultrastructural information on the wall, however, is limited to a very few reports: one describes granulosa cells and their organelles in relation to steroidogenesis (Bjersing, 19671, and another describes features of only the theca interna cells (Krzysztofowicz and Stoklosowa, 1977). In a n attempt to elucidate changes in granulosa-theca structure or relationships that might be related to steroidogenesis during development, a series of pig follicles were collected, processed, and examined at various times after folliculogenesis was stimulated by a n injection of equine chorionic gonadotropin (eCG). MATERIALS AND METHODS
Prepubertal Large White-Landrace crossbred, nulliparous porcine females (gilts) approximately 6 months of age and 90 kg weight were injected intra-
Received February 22, 1991. Accepted May 31, 1991. C.A. Ackerley's current address is Department of Pathology, Hospital for Sick Children, 555 University Ave., Toronto, Ontario, Canada M5G 1x8.
282
G.J. KING ET AL.
muscularly with 750 IU eCG to stimulate follicular development. Animals were slaughtered at 12 h r (n = 3), 24 h r (n = 4), 48 hr (n = 5 ) , 72 h r (n = 4), and 96 h r (n = 6) postinjection. Three additional gilts of similar breeding age and weight were processed without injection to provide unstimulated control or 0 h r specimens. Reproductive tracts were removed immediately after exsanguination, and ovaries were perfused through the uterine artery with a solution containing 4%paraformaldehyde plus 2.5%glutaraldehyde in phosphate buffer (pH 7.2, 0.1 M). Individual follicles were dissected free and immersed in fresh fixative for 1 hr. Follicles were then cut in half, fixed for a n additional 5 hr, washed 3 times for 15 min in phosphate buffer, and stored in the same buffer at 4°C until further processing. Follicle halves were dehydrated through a n ascending series of ethanols and embedded in JB-4 methacrylate resin (Polyscience Inc., PA) oriented in such a manner to facilitate obtaining cross sections. For examination by light microscopy, 2 pm sections were stained with 1% toluidine blue 0 in 1%sodium borate. Strips of follicle walls (1 x 2 mm long) were cut prior to postfixation with 2% OsO, in water for 1hr. Specimens were either dehydrated to 70% ethanol and infiltrated with Epon resin, as described by Idelman (1964) for the retention of lipids, or dehydrated in a n ascending series of ethanols and infiltrated with Epon via propylene oxide. Samples were then flat embedded in aluminum weigh boats and polymerized a t 60°C for 48 hr. Pieces of tissue were then cut out with a jeweller’s saw and affixed to dummy Epon blocks with cyanoacrylate to obtain a cross-section of the follicle wall. U1trathin sections exhibiting a pale-gold interference color were cut on a diamond knife, mounted on grids, and stained with ethanolic uranyl acetate and lead citrate. The grids were viewed and photographed in a JEOL TEM lOOCX transmission electron microscope. For morphometric analysis of the proportion of rough and smooth endoplasmic reticulum, negative images were projected onto a digitized tablet. At least 100 theca and 100 granulosa cells from each stage were analyzed using methods described by Wiebel (1965). Data were expressed as a mean and standard deviation and plotted for individual sampling times. RESULTS
Many small antral follicles with diameters around 2 mm were present a t the surfaces of all ovaries recovered from unstimulated (0 hr) gilts and those slaughtered 12 h r after injection with eCG. By 24 hr poststimulation, these small follicles were still numerous but some had grown into more prominent structures with the largest reaching 3.5 mm in diameter. Antral follicles continued to increase in size, but numbers were reduced so that ovaries collected 48 h r after eCG injection had 3-12 obviously enlarging follicles with mean diameters about 5 mm. Rapid growth continued through the subsequent 2 days. At 96 h r postinjection, ovaries from 4 gilts had varying numbers of follicles ranging in diameter from 6 to 9 mm but no ovulations. Ovaries recovered from 2 additional gilts a t this stage had both ovulatory sized and recently ovulated follicles (corpora hemorrhagica). Those females slaughtered be-
tween 24 and 96 h r postinjection showed slight to marked vulvar edema and reddening. Follicle walls from unstimulated gilts possessed a n outer theca externa made up of compact connective tissue that blended into the theca interna. The interna region was represented by a thin stratified layer of elongated polyhedral cells interspaced with spindleshaped fibroblasts in a collagen matrix, adjacent to a basement membrane (Figs. 1, 2). The granulosa layer on the opposite side of the basement membrane completely lined the antrum and formed the cumulus mass of cells surrounding the oocyte. This general relationship remained constant throughout folliculogenesis. A few scattered cisternae of smooth and rough endoplasmic reticulum and the occasional whorl (Fig. 2) were present in both theca and granulosa cells. The stratified columnar epithelial cells forming the granulosa were diverse in shape with tortuous lateral interdigitations and somewhat loosely associated boundaries (Figs. 1, 3). Free ribosomes, inconspicuous Golgi complexes, and lipid droplets were constant features a t all stages. Some lipid was located in the basal portions of granulosa cells a t 0 hr, and it increased by 24 hr. One interesting feature of some early granulosa cells was the presence of a single cilium (Fig. 3, inset). Althrough gross examination of ovaries collected a t 12 h r poststimulation showed no obvious differences from those recovered prior to eCG injection, some microscopic changes were observed. Growth occurred in both the granulosa and theca interna regions (Fig. 4 vs. Fig. 11, and this development was even more pronounced at later stages. Extracellular spaces between granulosa cells decreased substantially by 12 hr, and numerous free ribosomes were present in cells in both layers so the cytoplasm appeared “less dense” (Figs. 5, 6,). The theca cells, first identified by their spindle shape and their location in the loose connective tissue immediately beneath the basement membrane plus the presence of some lipid material and a network of rough endoplasmic reticulum, assumed a less uniform but more rounded shape (Fig. 6). An obvious capillary bed was located immediately outside the basement membrane in all specimens obtained after stimulation. As follicular growth and differentiation continued, mitotic figures were common in the granulosa layer (Figs. 7, 8) and could also be detected in theca cells. Within the first 24 h r after stimulation, organelles associated with increased metabolic activity proliferated, resulting in the accumulation of basal lipid in granulosa cells (Fig. 9) and of abundant rough and smooth endoplasmic reticulum in vesicular and lamellar forms plus lipid-associated whorls (Fig. 10). Thecal cells became plumper and filled with abundant organelles including extensive lipid-associated smooth endoplasmic reticulum (Figs. 10, 12). Bizarre mitochondria1 forms appeared in cells of both regions (Figs. 11, 12), and these were often found in intimate association with lipids (Fig. 12). Both the theca and granulosa layers were relatively compact a t 48 h r (Fig. 13), but cellular association became more discrete a s folliculogenesis progressed (Fig. 14). By 96 hr, thecal cells had accumulated substantial lipid and assumed irregular shapes, ranging from a few that were still elongated through oval to round (Figs. 15, 16). As maturation of the follicular wall progressed and ovulation approached, an-
PORCINE FOLLICULAR DEVELOPMENT
Fig. 1. Light micrograph showing a section of porcine follicular wall prior to any hormonal stimulation. The small antral follicles have a well defined stratified columnar granulosa cell layer (G) separated from a compact theca interna (T) by a distinct basement membrane. x 200. Fig. 2. Low-power electron micrograph of thecal cells from a porcine follicle prior to hormonal stimulation. Typical cytoplasmic features are lipid droplets (L) and sparse endoplasmic reticulum arranged in whorls (W) or scattered throughout the cytoplasm. x 6,500.
tral granulosa cells commenced dissociation, isolated edematous regions formed in the theca, and slight to moderate undulations appeared in the adjacent haseinent membrane (Fig. 16). Much of the follicular fluid escaped with the oocyte and cumulus at ovulation, allowing the wall to collapse into prominent folds abliterating the antrum (Fig. 17). The postovulatory plica-
283
Fig. 3. Low-power electron micrograph of granulosa cells from a porcine follicle prior to hormonal stimulation. These cells demonstrate loose interdigitation along lateral membranes and sparse lipid droplets located predominantly in basal cytoplasm. Inset: Oblique section through a granulosa cell and basal plate of cilium. x 3,500; inset, x 15,770. Fig. 4. Light micrograph of a section of porcine follicular wall 12-hr posthormonal stimulation. Both the theca (TI and granulosa ( G ) layers have increased in thickness compared with the prestimulation follicle in Figure 1, and a prominent capillary bed is apparent immediately beneath the basement membrane. x 200.
tions resulted in thecal cells, plus their associated connective tissue matrix and capillaries, penetrating into the remaining granulosa. Increasing amounts of smooth endoplasmic reticulum, arranged in whorls or randomly (Figs. 18, 191, appeared throughout the cytoplasm a s follicular enlargement proceeded. Some regions of discontinuous
284
G.J. KING ET AL.
Fig. 5. Low-power electron micrograph illustrating basal regions of granulosa cells 12 hr posthormonal stimulation of follicular maturation. The extracellular spaces have decreased in comparison with those present in unstimulated follicles (Fig. 3). x 3,600.
Fig. 6. Low-power electron micrograph of thecal cells in various forms of differentiation, 12 hr poststimulation. The cytoplasm contains considerable endoplasmic reticulum and mitochondria with occasional lipid droplets. x 4,680.
PORCINE FOLLICULAR DEVELOPMENT
Fig. 7. Light micrograph of porcine follicular wall 24-hr pxthormonal stimulation with several prominent mitotic figures (arrowheads) in the granulosa layer (G). X 350.
285
Fig. 9. Electron micrograph of the basal portion of granulosa cells 24-hr poststimulation. There is a prominent basal lamina and the cells contain abundant basal lipid plus smooth and rough endoplasmic reticulum. x 18,500.
Fig. 8. Light micrograph of porcine follicular wall 48-hr posthormonal stimulation with numerous mitotic figures (arrowheads) in granulosa cells. x 350.
basal lamina could be observed in all postovulatory specimens (Fig. 20). Extensive vesicular endoplasmic reticulum and lipid droplets displaying varying degrees of osmiophilia were prominent features (Fig. 21). A substantial proportion of the intracellular lipid was enclosed within laminated membranes which appeared to diminish as droplets were located closer to the periphery of the cells. Some extracellular lipid appeared
a s droplets in the intercellular matrix surrounding the immediately postovulatory luteal cells in all specimens embedded after only partial dehydration (Fig. 22). One major feature observed throughout the entire process of follicular maturation was the dramatic change in the amounts and proportions of rough and smooth endoplasmic reticulum that occurred a s follicles matured. Within the first 24 hr, rough endoplas-
286
G.J. KING ET AL.
Fig. 10. Electron micrograph ofthecal cells at 24-hr poststimulation illustrating a whorl of rough endoplasmic reticulum surrounding a lipid droplet, a common feature at this stage. x 14,110.
Fig. 11. Electron micrograph illustrating several bizarre lamellar mitochondria1 forms and extensive smooth endoplasmic reticulum,
but only scattered segments of rough endoplasmic reticulum, in a granulosa cell at 48-hr poststimulation. x 17,500. Fig. 12. Electron micrograph of thecal cell cytoplasm a t 72-hr poststimulation with oval and globular vesicular mitochondria, extensive smooth endoplasmic reticulum, sparse rough endoplasmic reticulum, and large lipid droplets. x 14,525.
PORCINE F'OLLICULAR DEVELOPMENT
287
Fig. 13. Light micrograph of a section of porcine follicular wall in vicinity of the still vesicular oocyte enclosed within a compacted cumulus cell mass, a t 48-hr poststirnulation. Both granulosa ( G ) and theca (T)layers are thicker than a t previous stages. x 180.
Fig. 14. Light micrograph of a follicular wall segment and oocyte surrounded by cumulus cells at 72-hr poststimulation. The cumulus and innermost granulosa cells show signs of impending dispersion while thecal cells in this particular location have apparently assumed a more rounded appearance and less compacted association. x 180.
mic reticulum in thecal cells increased from 4.2-7% of total volume, while the amount in granulosa cells went from less than 3.5%to more than 10%. For the reinainder of the period studied, the quantity remained relatively constant in thecal cells but declined rapidly to prestimulation values in granulosa cells. After 4.8 hr, smooth predominated over rough with over 10% of' total volume for cells of both regions composed of smooth endoplasmic reticulum by the time ovulation was imminent (Figs. 23, 24).
granulosa layers preceding and at the time of ovulation and the incorporation of both components into the corpus luteum agreed with the previous description (Corner, 1919). Also, comparable increases in endoplasmic reticulum, mitochondria with tubular cristae, and lipids and their associations indicative of steroidogenesis were observed by Bjersing (1967). Hunter et al. (1989) found the granulosa layer to be of uniform thickness with no infolding prior to the LH surge, but they noted marked infolding of both granulosa and theta in some follicles just prior to ovulation. Slight to moderate undulations and some discontinuity of basement membrane separating the layers of intact follicles were present a t 96 h r in some regions of the unovulated follicles in the current study, but little obvious infolding was observed even in follicles from ovaries with some postovulatory structures already present. This discrepancy may be the result of fixation differences since Bouin's was used by Hunter et al. (1989) while vascular perfusion with paraformaldehyde-glutaraldehyde was performed in the current series. Alternately, some morphological and biochemical differences can be detected between follicles of naturally cycling and hormonally stimulated gilts, suggesting that ovaries from prepubertal gilts might function differently from those in mature animals (Wiesak et al., 1990).There is, however, abundant evidence indicating substantial heterogeneity between follicles of similar size taken from sexually mature pigs (Foxcroft and Hunter, 1985; Hunter et al., 1989; Grant et al., 1989; Xie et al., 1990). Also, follicles stimulated to ovulation in prepubertal gilts must be reasonably competent since the released oocytes can be fertilized and pregnancy established (Shaw et al., 1971). The prepubertal model is certainly
DISCUSSION
In sexually mature pigs, increased episodic Iuteinizing hormone (LH) pulses combined with adequate concentrations of follicle-stimulating hormone (FSH) promote follicular recruitment from a continually emerging pool of small antral follicles. As growthL continues, theca and granulosa layers acquire enhanced steroidogenic activities that are essential for development of feedback mechanisms necessary for maturation and ovulation. Similar events occur when ovaries of prepubertal gilts are stimulated with eCG, and in this study, all treated females responded with a progressive increase in follicular size comparable to that reported by Ainsworth et al. (1980). Size variability between follicles within animals after eCG treatment may result from a constantly changing hormonal environment, producing different rates of tissue differentiation and growth. The slight to marked vulvar edema and reddening noted in those females slaughtered between 24 and 96 h r postinjection indicated target organ response to increased plasma estrogen concentrations which would be expected in stimulated gilts. Microscopic alterations t h a t occurred in theca and
288
G.J. KING ET AL.
Fig. 15. Light micrograph of a section of wall from a n unovulated follicle a t 96-hr poststimulation. The theca (TI and granulosa t G ) layers are still separated by an intact basement membrane and prominent capillary bed. x 180.
region. The theca ( T )has considerable edema, dispersing the cells and creating undulations in the theca-granulosa interface. x 180.
Fig. 16. Light micrograph illustrating another wall segment of t h e 96-hr follicle above. Antral granulosa cells a r e dissociating in this
Fig. 17. Light micrograph of a section of wall from a recently ovulated follicle a t 96-hr poststimulation. The theca (Ti and granulosa tG1 layers have collapsed into prominent folds t h a t almost completely obliterate t h e recently evacuated antrum. x 180.
convenient for obtaining material at known times and is appropriate for many studies, but extrapolations must always be made with caution. The porcine follicular phase extends over 5 full days from recruitment until ovulation (Foxcroft and Hunter, 1985; Grant et al., 1989), with gonadotropin-receptor-
steroid modulated changes occurring throughout. Stimulated follicles developed through to ovulation over a similar time frame, and organelle modifications in both theca and granulosa cells were chronologically related to the increasing steroidogenic capabilities of these cells in vitro (Evans et al., 1981). A substantial
PORCINE FOLLICULAR DEVELOPMENT
Fig. 18. Sagittal plane electron micrograph through granulosa cells at 72-hr poststimulation showing abundance of smooth endoplasmic reticulum in whorls and random arrangements. Some necrosis was evident in the granulosa layer at this stage (arrowhead). x 5,280.
289
Fig. 19. Electron micrograph of granulosa cells desquamated from the basal lamina of a n ovulated follicle at 96-hr poststimulation. Note the shorter, stubbier microvilli and prominence of lysosome-like structures throughout the cytoplasm. X 5,280.
290
G.J. KING ET AL.
Fig. 20. Electron micrograph of postovulatory follicle wall 96 hr after stimulation. Prominent features are a thecal cell in mitosis (T*), separation of irregularly shaped granulosa cells (G), rounding of thecal cells (TI, and a n endothelial cell (El. Inset: Higher power of the upper left region illustrating discontinuity of the basal lamina. x 3,400; inset, x 5,780. Fig. 21. Electron micrograph of a section from the thecal region of a postovulatory follicular wall, 96-hr poststimulation, showing early
luteal cells containing abundant lipid droplets, vesicular smooth endoplasmic reticulum, and mitochondria. x 4,000. Fig. 22. High-power electron micrograph of the thecal-granulosa interface, 96 hr poststimulation, and immediately after ovulation, where dissolution of the basal lamina is occurring. The extracellular lipid (arrowhead) probably represents material extruded from the nearby granulosa cell. x 20,750.
291
PORCINE F'OLLICULAR DEVELOPMENT
t
T
L
R
;1 5 t -
/I
W U
0 10u
t
0
u L
0
a,
m
0
Y
5 -
U
L.
a
I:" 1
I
0
.
I
.
24
I
48
,
I
I
72
I
,-
96
Hours poststimulation
"
0
24
48
72
96
kl o ur s p 0 s ts t i m u I a t ion
Fig. 23. Proportion of total cytoplasmic area of thecal cells occupied by smooth (triangles) and rough (circles) endoplasmic reticulum.
Fig. 24. Proportion of total area of granulosa cell cytoplasm occupied by smooth (triangles) and rough (circles) endoplasmic reticulum.
increase in rough endoplasmic reticulum occurred in both layers within the first 12 h r after injection with eCG. This likely indicates a n almost immediate and dramatic increase in resources to synthesize growth and receptor proteins plus additional enzymes. Smooth endoplasmic reticulum increased steadily i n theca cells suggesting a n increasing capacity to produce steroids from soon after stimulation up to ovulation, in keeping with changes in plasma concentrations (van de Wiel e t al., 1981) and theca cell cultures (Evans et al., 1981). The elevation in proportion of smooth endoplasmic reticulum within granulosa cells occurred considerably later, but in ample time to account for the increasing concentrations of estrogens that occur in follicular fluid as ovulation approaches (Ainsworth et al., :1980; Grant et al., 1989). The onset of sexual receptivity and beginning alf the LH surge commence around the same time (van de Wiel et al., 1981). Ovulation occurs approxim,stely 40 hr after competent follicles receive the appropriate gonadotropin stimulus (Hunter, 1972). Behavioral and ovarian responses in treated gilts indicated thal; the surge should have taken place between 48 and 72 h r postinjection. Plasma gonadal steroid concentrations shift quickly from estrogens in picomolar values during the early follicular phase to progesterone in nanoniolar quantities soon after luteinization commences (va.n de Wiel et al., 1981; Evans et al., 1981), indicating a substantial increase in steroid production and release. The use of a specialized dehydration and embedding p:rocedure for the retention of lipids (Idelman, 1964) allowed demonstration of extracellular lipids (Fig. 21) that were not observed in material processed in the routine manner. This noticeable increase in extracellular lipid was located immediately adjacent to cells containing organelle configurations indicative of steroid synthesis in specimens collected after luteinization should have occurred. As stated above, steroidogenesis has sh.ifted
dramatically a t this stage from release of small quantities of estrogens to much larger quantities of progesterone. Thus, a t least some of the demonstrated lipid probably represents the latter hormone. ACKNOWLEDGMENTS
The capable technical assistance provided by Douglas Way is gratefully acknowledged. Research funds were provided by the Natural Science and Engineering Research Council of Canada and the Ontario Ministry of Agriculture and Food. LITERATURE CITED Ainsworth, L., B.K. Tsang, B.R. Downey, G.J. Marcus, and D.T. Armstrong 1980 Interrelationships between follicular fluid steroid levels, gonadotrophic stimuli, and oocyte maturation during preovulatory development of porcine follicles. Biol. Reprod., 23t621627. Bjersing, L. 1967 On the ultrastructure of follicles and isolated follicular granulosa cells of porcine ovary. Z. Zellforsch., 82t173-186. Chang, S.C.S., J.D. Jones, R.D. Ellefson, and R.J. Ryan 1976 The porcine ovarian follicle: I. Selected chemical analysis of follicular fluid at different stages of development. Biol. Reprod., 15t321328. Corner, G.W. 1919 On the origin of the corpus luteum of the sow from both granulosa and theca interna. Am. J. Anat., 26,117-183. Evans, G., M. Dobias, G.J. King, and D.T. Armstrong 1981 Estrogen, androgen and progesterone biosynthesis by theca and granulosa of preovulatory follicles in the pig. Biol. Reprod., 25:673-682. Evans, G., M. Dobias, G.J. King, and D.T. Armstrong 1983 Production of prostaglandins by porcine preovulatory follicular tissue and their role in intrafollicular function. Biol. Reprod., 28,322-328. Falck, B. 1959 Site of production of oestrogen in rat ovary as studied in micro-transplants. Acta Physiol. Scand., 47, Suppl. 163:94101. Foxcroft, G.R., and M.G. Hunter 1985 Basic physiology of follicular maturation in the pig. J . Reprod. Fertil., 33t1-19 (Suppl). Grant, S.A., M.G. Hunter, and G.R. Foxcroft 1989 Morphological and biochemical characteristics during ovarian follicular development in the pig. J. Reprod. Fertil., 86t171-183. Hammond, J.M., C.J. Hsu, J . Klindt, B.K. Tsang, and B.R. Downie 1988 Gonadotrophins increase concentrations of immunoreactive insulin-like growth factor-I in porcine follicular fluid in vivo. Biol. Reprod., 38,304-308.
292
G.J. KING ET AL.
Hunter, M.G., S.A. Grant, and G.R. Foxcroft 1989 Histological evidence for heterogeneity in the development of preovulatory pig follicles. J. Reprod. Fertil., 86:165-170. Hunter, R.H.F. 1972 Ovulation in the pig: timing of the response to injection of human chorionic gonadotrophin. Res. Vet. Sci., 13: 356-360. Idelman, S. 1964 Modification de la technique de Luft en vue de la conservation des lipides en microscopie electronique. J. Microsc., 3:715. Krzysztofowicz, A,, and S. Stoklosowa 1977 Ultrastructure of theca interna cells of porcine ovarian follicles. Anat. Hist. Embryol., 6.359-364. Shaw, G.A., B.E. McDonald, and R.D. Baker 1971 Fetal mortality in the prepubertal gilt. Can. J. Anim. Sci., 51233-236.
van de Wid, D.F.M., J. Erkens, W. Kops, E. Vos, and A.J. van Landeghem 1981 Periestrous and midluteal time courses of circulating LH, FSH, estradiol-17p and progesterone in the domestic pig. Biol. Reprod., 24:223-233. Wiebel, E.R. 1965 Stereological principals for morphometry in electron microscopic cytology. Int. Rev. Cytol., 26:235-255. Wiesak, T., M.G. Hunter, and G.R. Foxcroft 1990 Differences in follicular morphology, steroidogenesis and oocyte maturation in naturally cycling and PMSGhCG-treated prepubertal gilts. J. Reprod. Fertil., 89~633-641. Xie, S., D.M. Broermann, K.P. Nephew, J.S. Ottobre, M.L. Day, and W.F. Pope 1990 Changes in follicular endocrinology during final maturation of porcine oocytes. Domest. Anim. Endocrinol., 7:7582.
