Paper - Studies on the human oocyte and its follicle 1: Difference between revisions

From Embryology
mNo edit summary
Line 13: Line 13:
=Studies on the Human Oocyte and its Follicle=
=Studies on the Human Oocyte and its Follicle=


I. Ultrastructural and Histochemical Observations on the Primordial Follicle Stage
==I. Ultrastructural and Histochemical Observations on the Primordial Follicle Stage==
 
 
 
Arthur T. Hertig and Eleanor C. Adams
 
From the Department of Pathology, the Harvard Medical School, Boston, Massachusetts 062115
 
==Abstract==
 
Oocytes in primordial (“resting”) follicles in adult human ovaries contain a complex paranuclear structure identified by light microscopists as Balbiani’s vitelline body. By electron
microscopy this structure is composed of a mass of mitochondria with associated endoplasmic
reticulum, multiple compound aggregates which form a ring around the cytocentrum, and
a single stack or coil of annulate lamellae either attached to the nuclear membrane or free
in the cytoplasm. The compound aggregates contain vacuoles and finely divided electronopaque material. Evidence is presented for the probable transport of this material between
the oocyte and its environment. The cytocentrum contains a central aggregate of amorphous
electron—opaque deposits which appear to become periodically aligned on fine fibrils to form
the long coarse fibers at the periphery of the cytocentrum. The apparent prevalence of
annulate lamellae attached or adjacent to the nuclear membrane of oocytes in ovaries removed during the mid—follicular (estrogenic) phase of the cycle indicates the need for further
study of a possible hormonal influence on the resting oocyte. By light microscopy phosphatases were not found within the oocyte, but adenosine—monophosphatase activity is present in the cortical cells surrounding primordial follicles, and also at the periphery of each
primitive follicle cell, most prominently at the oocyte side. Glucose—6—phosphate dehydrogenase activity is present within the oocyte cytoplasm.
 
INTRODUCTION
 
In the adult human ovary, there are only two
stages in the growth of the oocyte and its follicle
that can be confidently recognized histologically
as normal phases of the maturation process. These
stages are at opposite ends of the maturation
spectrum: one is the resting oocyte in the primordial follicle (40 ,u in diameter); the other is the
oocyte with a meiotic spindle forming the first
polar body in the fully matured preovulation
follicle (1.5 cm). The normal histology of all the
intervening stages cannot be recognized since there
are as yet no agreed morphological criteria for
 
differentiating follicles that are continuing to
develop from those in the earliest stages of atresia.
(Certainly mitotic activity in the follicle wall or
cumulus is no criterion of normal preovulatory
development. In both the human and guinea pig
we have seen numerous examples of obviously
atretic follicles showing mitoses in their granulosa
cells. It is unknown, however, at what stage in the
development of any given follicle atresia had
begun.) The duration of the developmental processes and its possible correlation with cyclic endocrine events also are unknown in the human
 
female. Moreover the interrelationship of the two
ovaries with respect to follicular development is
some mammals, however, the
morphology of follicular and oocyte developmental
processes has been correlated with events in the
estrous cycle. For instance, Myers et al. (1936)
 
unknown. In
 
made a careful histological examination of follicles
in ovaries removed at known times of the 16-17
day estrous cycle of the guinea pig. They reported
that soon after estrus many multilaminar primary
(nonvesicular) follicles had begun to form an
antrum thus becoming a secondary or vesicular
follicle, and that from these only two or three
follicles in each ovary complete their development
and ovulate at the next estrus; the remainder
becoming atretic at various stages of their growth.
In addition to the normal duration of these
vesicular follicle stages (one estrous cycle), Myers
et al. reported their histological and size criteria in
relation to the various succeeding days of the
cycle. We have recently described some of the
histochemical aspects of oocytes in developing and
atretic follicles throughout the estrous cycle of the
guinea pig (Adams et al., 1966).
 
In order to determine the morphological correlation of human follicle and oocyte developmental
stages with known events in the menstrual cycle
and thus to estimate follicular growth rate, we have
begun a careful examination of the total follicular
population in human ovaries removed at various
times in the menstrual cycle. An electron microscopic study of the first stage examined, oocytes in
primordial follicles, is reported here together with
the results of a few histochemical studies. In these
oocytes from adult human ovaries we have noted
ultrastructural differences from oocytes in similar
follicles studied in the human prenatal ovary
(Lanzavecchia and Mangioni, 1964; Stegner and
Wartenberg, 1963) and in ovaries of a variety of
other mammalian species. These differences are
chiefly characterized by the complexity and organization of the paranuclear structure previously
 
identified by light microscopists as Balbiani’s
vitelline body.
 
MATERIALS AND METHODS
 
The follicles were obtained from
normal ovaries removed from five fertile women,
28-37 yr of age, who had gynecologic and /or obstetric indications for hysterectomy such as carcinoma in situ of the cervix or leiomyomata of the
uterus. One ovary was removed in four instances and
both 1n one instance. In addition, we examined
 
 
primordial
 
 
primordial follicles from an ovary removed from a
patient who had been on the antifertility medication
Orthonovum. Also follicles from a wedge biopsy of
an ovary, obtained at laparotomy for infertility
investigation, were studied. From this material
corpora lutea and intact large vesicular follicles
are dissected out for other histological, histochemical, or electron microscopic studies. The remaining
ovary is then sliced serially at 2-3 mn1 either crosswise or sagitally and fixed for histological, histochemical, and electron microscopic observation of
smaller developing or atretic follicles.
 
For izz'rz‘0clzemz'5try, some slices were fixed 3-12
hr in 4% paraformaldehyde in 0.1 M phosphate
buffer pH 7.6, in 2 -,’/'5 glutaraldehyde, or in a combined 2% paraformaldehyde—2%% glutaraldehyde
(one-half strength Karnovsky fixative) (1965), both
in 0.1 M eacodylate buffer pH 7.4. These were subsequently washed in their respective buffers and
rapidly frozen at —70°C. Cryostat sections were
reacted for various phosphatases, dehydrogenases,
or diaphorases or were stained with Sudan black.
Other slices were fresh frozen, sectioned in a cryostat,
and reacted for similar enzymes. Although acid
phosphatase activity, as visualized by the Gomori
method (Becker et al., 1960), was maintained by all
three fixatives used, ovarian nucleoside polyphos—
phatase, adenosine—5-monophosphatase (VVachstein
and Meisel, 1957), and alkaline-phosphatase (Pearse,
1960) activities were maintained only by fixation
in 4% paraformaldehyde or in fresh frozen cryostat
sections. In view of the doubts recently cast on the
reliability of dehydrogenase methods which employ
unfixed cryostat sections (Fahimi and Karnovsky,
1966), we have used both 4% paraformaldehydefixed (3 hr) and fresh-frozen cryostat sections for the
study of glucose-6—phosphate dehydrogenase and both
NADH and NADPH diaphorases by using the methods of Pearse as modified slightly by Karnovsky
(1961). In addition, we have added 5 mg/100 ml Nmethyl phenazonium methosulfate (PMS) to some
glucose-6-phosphate dehydrogenase incubations in
order to bypass diaphorase activity in the production
of the formazan.
 
For electron mz'cro5copz'c studies, the strips of cortex
were fixed in 2% glutaraldehyde or in the combined
2% paraformaldehyde——21/2% glutaraldehyde both
in 0.1 M eacodylate buffer pH 7.4. After 5 hr fixation,
the strips are washed in repeated changes of chilled
0.1 M caeodylate buffer with CaC1g added and are
stored in the refrigerator. The slices of ovary thereafter are dissected so that the medulla is separated
from a strip of cortex. These are then examined under
a dissecting microscope as they are cut into 0.5-0.1
mm transverse slices, each of which is again searched
by a dissecting microscope using oblique transillumiprimordial (0.04 mm in diameter),
nonvesicular or primary (0.08—0.28 mm in diameter),
 
nation for
 
or small vesicular or secondary (0.30 -1- mm in
diameter) follicles. After recording the measurements
of individual follicles they are further dissected into
orientable blocks of 1.5 mm greatest size, are postosmicated, rapidly dehydrated, and embedded in
Epon 812. Before thin sectioning, the preselected
cortical blocks are “thick”-sectioned (lp) serially,
and successive groups of five serial, thick sections
are examined with phase microscopy to search for
the presence of primordial follicles. When the edge
of a primordial follicle is identified, the block is
and serial, thin sections are cut and
mounted on 100—mesh grids. After six to eight serial
grids of thin-sectioned ribbons are collected, another
thick section usually is mounted on a glass slide
before collecting the next series of six to eight grids
of serial sections. The seriated thick-sections are
examined and subsequently photographed for
orientation of individual primordial follicles which,
in our material, are often present in clusters of
two to five oocytes per section. This orientation
facilitates the electron microscopic study of serial
depths through any one oocyte in a cluster. The thin
sections, stained with lead (Venable and Coggeshall,
1965) or with uranyl followed by lead were examined
with an RCA EMU-3F or 3H microscope.
 
For fzistalogical study, one cortical strip, fixed in
4% paraformaldehyde or in 2% glutaraldehyde
followed by Bouin’s fluid, is embedded in paraffin
for serial sectioning and is stained with hematoxylin
and eosin or periodic—acid Schiff.
 
trimmed,
 
OBSERVATIONS
 
Light llficroscopy
 
In the young adult, primordial follicles about
40 ,u. in diameter can be found in clusters lying in
the narrow band of dense cortex just beneath the
serosal surface and the tunica albuginea. Primordial follicles, whose oocytes are in the prolonged diplotene phase of meiotic prophase (Baker,
1963), are in a resting stage which can be found
continuously from the earliest postnatal ovary
(when oogenesis has ceased and earlier meiotic
prophase stages have been completed) through
the adult reproductive period. Their number is
gradually decreased, particularly from adolescence
to the menopause, owing to the repeated initiation
of development of some of them, first to the
primary (nonvesicular) and subsequently to the
secondary (vesicular) follicular stages which terminate either with maturation and ovulation or
with atresia.
 
A striking cytoplasmic characteristic of oocytes
in primordial follicles is a large paranuclear
complex identified previously by light microscopists as the “corps vitelline” surrounded by a
“couche mitochondrial” (van der Stricht, 1923),
as the “yolk nucleus complex,” (Beams and
Sheehan, 1941), or preferably (Raven, 1961) as
Ba1biani’s vitelline body (1864). This structure has
been studied in our histological and histochemical
preparations: with hematoxylin and eosin staining
(Fig. 1), it is an eosinophilic complex closely
applied to the nuclear membrane; with periodicacid Schiff staining both with and without diastase digestion it is a ring of positive granules
surrounding a paler pink central sphere; in frozen
sections stained with Sudan black, sudanophilic
granules of various sizes are found in this area. In
both 4% paraforma1dehyde—fixed frozen sections
with and without the addition of PMS in the
incubation medium and in unfixed sections without PMS, the entire cytoplasm of the oocyte is
strongly reactive for glucose-6-phosphate dehydrogenase activity (Fig. 2). When PMS is added
to the incubation medium for unfixed sections, the
solubility of this enzyme, G—6—P dehydrogenase, is
indicated by the production of the formazan in
the incubation medium. The oocyte cytoplasm is
moderately reactive for both NADH and NADPH
diaphorases in both fixed and unfixed sections.
Alkaline and acid phosphatases, as well as neutral
phosphatases such as nucleoside poly— or monophosphatases are negative in the oocyte.
 
Some of the primitive follicle cells form a
crescent—shaped cap of thicker cells over part of
the oocyte; the remaining cells form a much
thinner wall around the rest of the oocyte. Mitotic
activity has not been observed in the follicle wall of
this stage. Adenosine—monophosphatase activity,
present in the cortical cells which surround the
primordial follicles, also appears at the periphery
of the primitive follicle cells, most prominently at
the surface facing the oocyte (Fig. 3). Acid phosphatase has not been identified in the primordial
follicle wall, although it does appear at subsequent
stages of development. According to our criteria,
when any follicle has thick or cuboidal cells
around its entire surface, development to the
primary (nonvesicular) follicle has been initiated.
Although the oocyte within this early developing
follicle is not much larger in diameter than that in
the primordial follicle, it begins to show changes
at the ultrastructural level.
 
Electron Microscopy
 
In thin—sectioning blocks of cortex which contain
dispersed or clustered primordial follicles, the
plane of section with respect to the polarity of any
oocyte is necessarily random. Of the 30-40 oocytes
serially sectioned in our study, only a very few
fortunate micrographs show the interrelationship
of all of the cytoplasmic organelles found in the
mid—polar region (Fig. 4). (By mid—polar is meant
a section which passes through or near the center
of the nucleus, through or near the center of the
cytocentrum and in addition includes the significant part of the annulate lamellae. This structure
is single and eccentric with respect to the true
polar axis and may be either attached or detached
from the nuclear membrane.) In most oocytes it is
necessary to reconstruct mentally the three—dimen—
sional relationships by superimposing micrographs
from serial depths through an oocyte.
 
Study of primordial follicles in ovaries obtained
 
 
Abbreviations
 
AL, annulate lamellae
 
C, Cortex
 
CA, compound aggregates
CF, coarse fiber
 
CH, el1roma.tin
 
F, follicle cell
 
G, Golgi complex
 
H, heterochroinatin
 
from a variety of phases of the menstrual cycle and
from a term pregnancy reveal a constant general
relationship of organelles. The nucleus of the
oocyte is large and usually spherical and contains
a large reticulated nucleolus. Mitochondria, elements of the endoplasmic reticulum and Golgi
complexes, are found in variable numbers around
the entire circumference of the nucleus. A large
paranuclear complex is formed by a concentration
of these organelles surrounding large compound
aggregates which form a ring around a large
cytocentrum. Annulate lamellae are found near
or within this paranuclear complex which in its
entirety comprises the structure previously described as Balbiani’s vitelline body. The variations
between individual oocytes which appear to reflect
either metabolic changes or a response to fixation
difference will be noted below in the following
detailed description of organelle arrangements.
The spherical nucleus (Figs. 4 and 25), varying
 
I, inte1'chro1na.tin granules
ill T, n1icrotubules
 
N, nucleus
 
NU, nucleolus
 
0, oocyte
 
P, projection
 
VA, vesicular aggrega.tes
 
650
 
NOTE: In the legends the II number which identifies each patient is followed by another
number identifying the particular oocyte represented.
 
FIGURE 1 Mid—polar section through an oocyte in a primordial follicle. Closely applied to
the spherical vesicular nucleus is the cresce11t—shaped Balbia.ni’s vitelline body (between
asterisks at lower right against nucleus) formed by peripheral dense granules (mitochondria) and a central paler area (cytoeentrum) surrounding a dense body. H 48, H —l— 14}.
X 560.
 
FIGURE Q Note that the cytoplasm of these three oocytes in primordial follicles is strongly
reactive for glucose—6—phosphate dehydrogenase. II 54; paraformaldehyde fixative. >( 350.
 
FIGURE 3 The cortical cells and the periphery of the cells in the Wall of the primitive
follicles at right are reactive for adenosine~Inon0phosphatase. The pronounced activity at
the oocyte—follicle cell junction apparently is located in the numerous projections of the
follicle cells present in this intereellula.r space. The activity of the enzyme is not maintained
at the periphery of differentiating follicular cells of the multilaminous primary follicle at
left. H 53; paraformaldehyde fixative. X 350.
 
FIGURE 4 A low—power survey micrograph through the mid-polar axis of an oocyte in a
primordial follicle. Note that most of the organelles are concentrated in one pole of the
oocyte; compare with Fig. 9.5. II 44-1; Karnovsky fixative; uranyl and lead stain. X $2700.
 
 
FIGURE 5 Nucleus of oocyte showing nucleolonema and associated clump of heterochromatin, aggregates of chromatin with associated perichromatin granules and fine fibrils, and dispersed clumps of interchromatin granules. Pores are prominent on the nuclea.r membrane. H ‘M-1; Karnovsky fixative;
lead stain X 15,000.
 
from 22 to 24 p. in diameter, is enclosed by a
smooth envelope whose two leaflets exhibit prominent pores (Figs. 5-7). Aggregates of filamentous
chromatin with associated perichromatin granules
and fine fibrils are distributed throughout the
nucleus in wispy strands that apparently represent
sections through the diplotene configuration of
chromosomes. Clumps of interchromatin granules
are also dispersed ‘throughout the nucleus. A large
irregularly shaped nucleolus with a nucleolonema
is prominent in these oocytes (Fig. 5). Multiple,
large clumps of fine, granular material interpreted
as heterochromatin are distributed around the periphery of the nucleus or are located near the
nucleolus (Figs. 5 and 6). They are surrounded by
chromatin and perichromatin granules.
 
The cytocentrum (Figs. 8 and 10) is similar to that
in other cell types in that it contains few, if any,
mitochondria. At its periphery, a network of
sparsely granular endoplasmic reticulum appears
to be continuous with many small Golgi complexes
and with the more granular endoplasmic reticulum
of the peripheral cytoplasm. Elements of this
 
- network often are also closely applied to the large
 
compound aggregates. Coarse, long fibers of
electron—opaque material are found in the cyto
THE JOURNAL or CELL BIOLOGY - VOLUME 34, 1967
FIGURE 6 Multiple clumps of heterochromatin are also found near the periphery of the nucleus. H 48L2; Karnovsky fixative; lead stain. >( 21,000.
 
FIGURE 7 Bundles of spiral filaments are occasionally attached to the nucleus or lie free in the cytoplasm. H 35-1; 2%. glutaraldehyde. )( 952,000. Inset from same oocyte but adjacent section. X 99,000.
 
plasmic matrix of this network and extend just
beyond the periphery of the cytocentrum where
they usually appear to terminate bluntly but
occasionally may abut a compound aggregate or a
Golgi complex. When serial sections have afforded
an opportunity to estimate their three—dimensional
relationship, the coarsest fibers appear to form an
interwoven basket—like structure at the periphery
of the cytocentrum. Smaller caliber fibrils in some
oocytes radiate out from near the center of the
cytocentrum to join the coarser ones. Although the
maximum length of each fiber could not be determined, the greatest diameter is about 0.13 M and
each appears to be straight or only slightly curved
(Fig. 11). Fixation differences may affect their
structure since on cross—section some are homogeneous, some appear as rings with pale centers (Fig.
9), and others as closely packed tetrads. N0 substructures such as tubules have been identified.
The microtubules seen within the cytocentrum do
not appear to originate from or to be concentrated
:near these fibers. A feltwork of very fine fibrils
often visualized in the matrix of the cytocentrum,
appears to merge with these coarse fibers.
 
The large, prominent center of the cytocentrum
i.s composed of deposits of amorphous, electronopaque material (Figs. 8 and 10). These deposits
frequently are seen in periodic alignments which
radiate from the center (Fig. 9). When this
 
periodicity is present, the amorphous deposits
appear to be aligned on very fine fibrils; this
suggests the formation of a transition between the
central concentration of amorphous deposits and
the more peripheral fibers. Although on occasion
the amorphous deposits in the center of the
cytocentrum may be reminiscent of the ultra-structural aspect of glycogen, the presence of
glycogen in oocytes could not be confirmed by our
histochemical studies. Also within the cytocentrum
of most oocytes one, or more, large (3.5 p. in
diameter), compact aggregate of closely packed
vesicles is found (Figs. 8 and 9). Such vesicles are
smooth—wal1ed, often appear to contain slightly
electron-opaque material, and frequently can be
seen to be continuous with the endoplasmic reticulum of this area. In other oocytes these vesicles
seem to be dispersed among the amorphous deposits to contribute to a single, central, dense area
of the cytocentrum measuring up to 4.5 p. (Figs.
10 and 11). This last arrangement may be the
ultrastructural basis for the dense granule visible
in the center of the cytocentrum in some H (hematoxylin) and E (eosin) preparations of oocytes
(Fig. I).
 
The resemblance of some of the periodic arrays
of deposits and of some coarse fibers to peri»centriolar structures described by others in a
variety of somatic cells, led us to hunt systernati-—
 
 
FIGURE 8 A section through the center of Ba1biani’s vitelline body, showing a cytoplasmic stack of
annulate lamellae, ballooned compound aggregates, and a Vesicular aggregate in the cytocentrum near
a concentration of clumps of dense amorphous material. For higher magnification see Fig. 9. Straight
 
fibers appear to radiate from this area. Coarser fibers are seen in cross-section at the periphery of the
cytocentrum. H 44-1; Karnovsky fixative. X 10,000.
 
\
FIGURE 9 A higher power of Fig. 8 to show the clumps of dense amorphous material, some of which
appear to be periodically aligned on fine fibrils suggesting a transition between the amorphous deposits
and the peripheral coarse fibers at right center. Note feltwork of fine fibrils in the cytoplasmic matrix. A
portion of the vesicular aggregate is seen at top left. Note the two instances of helical formation of ribo—
somes (asterisk) on flat View of endoplasmic reticulum. H sH—1; Karnovsky fixative. )( 32,000.
 
 
FIGURE 10 A mid—section through Balbiani’s vitelline body to show the peripheral concentration of
mitochondria, the variable ballooning of compound aggregates and the eytocentrum with a dense central
area surrounded by endoplasmic reticulum and coarse fibers. A coiled configuration of annulate lamellae,
detached from the nuclear inembrane, was seen. beginning in. the adjacent and continuing in subsequent
grids. H 48—R1; Karnovsky fixative. X 3800.
 
FIGURE 1.1 A higher power of Fig. 10, showing in this oocyte, the apparent mingling of the elements
of the vesicular aggregates with the clumps of dense amorphous material. Note further the Central cluster
of vesicles. At the periphery a.re seen longitudinal and tangential sections through coarse fibers. H 48
R1; Karnovsky fixative. X 91,000,
FIGURE 12 Compound aggregates showing variability’ in their structure. Those that are membranebounded are ballooned to a variable degree and in addition to vacuoles and electron-opaque granules
also contain Vesicles enclosing electron—opaque material. Others that are not 1'ne1nbrane—bounded frequently have elements of endoplasmic reticulum closely applied to their surface (right). Note mitochondrial rosette at right. H 48-R1; Karnovsky fixative. X £21,000.
 
cally for centrioles, particularly in this area of the
ooeyte. Since no centriole had been seen in any
oocyte routinely examined throughout its entirety,
a series of 75 serial grids containing four oocytes
was specially prepared without intervening thick
sections. In spite of careful search throughout the
cytocentrum and elsewhere in the cytoplasm of
these four oocytes, no centrioles could be identified.
 
Golgi compfexes (Figs. 8, 23, and 24) are closely
associated with the periphery of the cytocentrum
and in some oocytes appear to form a fenestrated shell around it. They are also interspersed between
the mitochondria that form the peripheral region
of Balbiani’s vitelline body and in addition are
frequently associated with endoplasmic reticulum
near the oolemma (Fig. 19). As noted above, their
membranes appear to be continuous with the
endoplasmic reticulum. Although Golgi complexes
frequently are located adjacent to the large compound aggregates, no continuity between these
structures has been noted. In some oocytes, Golgi
complexes along with endoplasmic reticulum and
 
 
FIGURES 13 and 14 These microgra.phs are from two sections on the same grid through the same area
at the periphery of the oocyte. Vacuoles are present tha.t contain either material sin1ilar to myelin figures
or elements such as finely divided electron-opaque material (Fig. 13) or small vacuoles (Fig. 14) similar
to those within. the compound aggregates. Note coated vesicles at oocyte membrane in Fig. 14. II 44-1;
Karnovsky fixative. X 21,000.
 
mitochondria are found closely applied to the
entire circumference of the nucleus (Fig. 25).
The large electron—opaque compound aggregates
(Figs. 8 and 10) at the periphery of the cytocentrum are composed of large vacuoles containing
a homogeneous, slightly electron—opaque material
surrounded by closely packed, very electronopaque, membranous or granular substance. They
 
present a Variable appearance since some are
compact and may or may not be membrane
enclosed, while others in the same oocyte are
enclosed by a membrane, are ballooned to a
variable degree, and appear to have imbibed a
clear fluid in which the substructures are dispersed.
The latter form contains many small vesicles
enclosing dense material (Fig. 12). Smaller com
FIGURES 15 and 16 Membranous formations are frequently seen at the intercellular
junction between oocyte (top) and follicle cell (bottom). These frequently contain small
vacuoles or finely divided electron-opaque material similar to those within the compound
aggregates and may represent a mechanism for the bulk transfer of this material between
follicle wall and oocyte. H 48-111 and H 44-5; Karnovsky fixative. X 32,000.
 
FIGURES 17 and 18
 
Compound aggregates are also seen within the cytoplasm of the
 
follicle cell. They frequently appear to bulge into the oocyte and in such areas the membranes of both cells appear indistinct. H 44-1 and H 44-5; Karnovsky fixative. X 21,000.
 
FIGURE 19 A compound aggregate in a follicle cell. Note similarity to aggregates in
oocyte. The intercellular space between oocyte and follicle conta.ins projections from both
cells. Note Golgi complex in close association with endoplasmic reticulum at periphery of
oocyte. H 44-3; Karnovsky fixative. X 21,000.
 
FIGURE 20 Compound a.ggregates are also seen within the cytoplasm of cortical cells
outside the primordial follicle wall. Note desmosome at junction between two follicle cells
 
and also concentration of projections i11 intercellular space at the j unction of oocyte membrane and two follicle cells. II 43-1; Karnovsky fixative. X 21,000.
 
 
 
FIGURE 21 Mitochondria to show effect of fixation. With the slightly hypertonic 2% glutaraldehyde
fixative, the mitochondria contain arched cristae a.nd a distended channel interpreted to be caused by
the infolding of the inner lamina. of the peripheral membranes with or without rupture of the external
lamina. The internal vesicle frequently seen within this distended channel could be a cross—section of
endoplasmic reticulum which apparently artifactitiously invades the mitochondria (see mitochondrion at
 
lower left). Note distended channels of endopla.smic reticulum with this fixative. H 35-4; 2% gluta.ral—
dehyde. X 36,000.
 
FIGURE 22 With the more hypertonic one-half strength Karnovsky fixative, the arched cristae are
prominent a.nd, in addition, multiple secondary tubular cristae are formed by the infolding of the inter—
nal lamina. of the-periplieral n_1en'1branes. Note close association of endoplasmic reticulum with mitochondria. Note also electron-opaque material at center of mitochondrial rosette. H 44-3; Karnovsky fixative.
X 49,000.
 
pact aggregates, usually without enclosing mem- 17—20 and 37). Similar compound aggregates are
 
branes, may be seen in the more peripheral cyto- also present in the serosal epithelium. Vacuoles
plasm, within the primitive follicle cells and in containing elements similar to those in the ag
many cells of the surrounding cortex (Figs. 4, gregates are present in the peripheral cytoplasm
 
660 THE JOURNAL OF CELL BIOLOGY - VOLUME 34, 1967
FIGURE ‘23
 
Golgi complexes are seen in the peripheral cytoplasm of the oocyte where they frequently
 
are associated with vacuoles containing multiple vesicles. H 44-1; Karnovsky fixative. >< Q1,000.
 
FIGURE +24 Golgi complexes (right) a.re also prominent at the periphery of the cytocentrum (left) where
they appear t.o be in continuity with the network of endoplasmic reticulum surrounding a.nd intermingled
with the clumps of amorphous electron-opa.que material. H 44-3; Karnovsky fixative. X 21,000.
 
of the oocyte (Figs. 13 and 14-). Frequently a
dense, membranous structure is seen lying across
and apparently merged with the oocyte membrane
(Figs. 15 and 16). This rather amorphous structure, when followed in serial sections, often is
associated with vacuoles in the oocyte cytoplasm.
When seen in thinner areas of the cytoplasm of the
follicle wall, the aggregates often appear to bulge
into the cytoplasm of the oocyte (Figs. 17 and 18).
 
A mass of mitochondria (Fig. 10) form the periphery of Balbiani’s vitelline body and a variable
number are scattered around the nucleus and in
the peripheral cytoplasm. They are spherical—to—
ovoid in shape, of variable size, 0.5—l .8 p., and are
intimately associated with the endoplasmic reticulum. Both the mitochondria and the endoplasmic
 
reticulum appear to be unusually responsive to
differences in tonicity of the fixing fluid. With only
slightly hypertonic glutaraldehyde—fixation (Fig.
21), the mitochondria contain a few, prominent,
peripheral arched cristae in a pale, homogeneous
matrix within which are only a few small granules.
In addition, they also contain a small, distended
channel interpreted as caused by the swelling of a
single crista or by a local infolding of the inner
lamina of the peripheral membrane with or without the rupture of the outer lamina. This distortion
is frequently associated with the close proximity
of endoplasmic reticulum which on occasion even
appears (apparently artifactitiously) to invaginate
the mitochondria. With the more hypertonic
one—half strength Karnovsky paraformaldehyde—
 
ARTHUR T. HERTIG AND ELEANOR C. ADAMS Human Oocyte and Its Follicle 661
FIGURE 25 This l0W—p0wer micrograph shows a section that is not mid-polar through the oocyte since
the paranuclear Balbiani Vitelline body was present in deeper sections. Note that a stack of annulate
lamellae is attached to nuclear mernbrane, that a Variety of organelles are closely applied to the entire
circuinference of the oocyte nucleus and that. endoplasmic reticulum is present as distended channels.
ll 35-4; 2% glutaraldehyde. X 2800.
 
FIGURE 26 A higher power of Fig. 25 to show continuity of membranes of a1111ulate lainellae with those
of the peripheral endoplasmic reticulum. Note evagination of outer leaflet of nuclear membraile at left
of annulate lamcllae. See Figs. 27 a.11d 28 for higher power View of this stack of annulate lamellae as observed in nearby sections. H 35-4; 2% glutaraldeliyde. X 7300.
 
662
gluteraldehyde-fixation (Fig. 22) the very close
structural relationship of mitochondria to endoplasmic reticulum is maintained, but the mitochondrial matrix is more electron—opaque. Furthermore, in addition to the usual prominent,
arched cristae, the peripheral inner membrane
often appears to be infolded to form multiple
secondary vesicular or tubular cristae. The membranes forming these structures are not so electronopaque as those of the apparently more stable,
arched cristae. Mitochondria in Balbiani’s vitelline
body or elsewhere around the nucleus frequently
are clustered in rosettes around a dense amorphous, somewhat reticulated substance (Figs. 12
and 22). Although this material may appear to
form an adherent cap on one or more of the
mitochondria in the rosette formation, examination of serial sections through a rosette indicates
that this image is caused by a superimposition of
structures on a tangential cut.
 
When closely seriated grids through Balbiani’s
vitelline body are available for study, a single mass
of stacked or concentric annulate lamellae can be
found (Figs. 4, 8, and 29—33). In some oocytes
these are in continuity with the nuclear membrane
(Figs. 25-28); in others they are found at some
distance from the nucleus (Figs. 33 and 34).
Whether they are attached to the nucleus or not,
the smooth—surfaced paired membranes of the
annulate lamellae are continuous at their periphery with the more granular peripheral endoplasmic
reticulum (Figs. 27, 28, 31-34) which in turn is
closely associated with mitochondria (Fig. 26). In
one of a series of sections cut in a fortunate plane
through annulate lamellae attached to the nucleus,
a continuity of membranes could be clearly visualized (Fig. 27). The outer leaflet of the nuclear
membrane is evaginated to form, by infoldings, a
large stack of paired membranes whose pore
formations are similar to but more closely spaced
than those found on the nuclear membrane at a
distance from the annulate lamellae (Figs. 29-32).
It is noted that in the area of the evagination and
infolding of the outer nuclear membrane, the
outer loops of the paired annulate membranes are
double whereas they are single in areas of the
stack where the paired membranes become continuous with the peripheral endoplasmic reticulum. Unfortunately, micrographs through the
entire stack of these annulate lamellae were not
available for complete three—dimensional study
since some of the stack was contained within a
thick section. Study of five sections on three serial
 
ARTHUR T. IIERTIG AND ELEANOR. C. A1JAMs H'u.ma.n Uocyte and Its Follicle
 
grids suggests that the evaginating nuclear membrane forms folds that interdigitate either with
similar, looped membranes from a second unidentified nuclear evagination or with membranes
from the peripheral endoplasmic reticulum. If the
latter is the case, the nuclear evagination and
infoldings may form a structural template for the
induction of “nuclear” pores on the interdigitating
membranes of the endoplasmic reticulum. When
the pores on adjacent lamellae are aligned, the
dense, amorphous substance contained within and
between these aligned pores may form a continuous
band from the inner nuclear membrane through
several layers of the stack (Figs. 27 and 29). The
largest mass of annulate lamellae measures 5-7 it
in longest diameter. Those that are unattached to
the nuclear membranes are often smaller and in
some oocytes only one or two fragmented remnants
of the larger structure could be found (Fig. 34).
Even as they fragment, their continuity with endoplasmic reticulum appears to be maintained since
in several oocytes we have seen pores either singly
or in clusters on a profile of endoplasmic reticulum
(Fig. 35).
 
In some oocytes, particularly those with annulate lamellae attached or adjacent to the nucleus, the membranes of the endoplasmic reticulum
may communicate with the outer leaflet of the
nuclear membrane at intervals around its entire
circumference (Figs. 25 and 28) and with Golgi
complexes scattered throughout the cytoplasm.
Other oocytes have most of their endoplasmic
reticulum as well as Golgi complexes and mitochondria confined to the area of Balbiani’s vitelline
body and to the peripheral cytoplasm in this pole
of the oocyte (Fig. 4). Vesicles of endoplasmic
reticulum in the peripheral cytoplasm frequently
contain an electron—opaque amorphous material
(Fig. 35). Although similar vesicles have been
observed adjacent to a ballooned, compound
aggregate (Fig. 39) we have no clear—cut morphological evidence that the electron—opaque material
contained within the endoplasmic reticulum is
related to that within the aggregates. The granularity of the endoplasmic reticulum can only be
seen in patches on cut view, although a helical
configuration (Fig. 9) of polyribosomes can often
be seen on fiat view. Single or clustered ribosomes
are numerous in the cytoplasmic matrix. The
proximity of endoplasmic reticulum to mitochondria was noted above. There appears to be some
morphological relationship between the endoplasmic reticulum and the membrane-enclosed com
663
pound aggregates which also is revealed as a
fixation eflect. Although the membrane around an
aggregate does not appear to be continuous with
the endoplasmic reticulum, channels of endoplasmic reticulum often appear to encircle portions of
aggregates that are not membrane enclosed (Figs.
12 and 37). The dilation and constriction of the
endoplasmic reticulum in any one oocyte appears
to reflect differences in the osmolality of the
fixation used (see figs. 4 and 25). With 2%
glutaraldehyde (only slightly hypertonic) the endoplasmic reticulum is seen as dilated, continuous
channels, whereas with the more hypertonic
paraformaldehyde-glutaraldehyde solution, it appears constricted and is usually seen as chains of
large vesicles. In addition, there is frequently a
reciprocal inverse response in some compound
aggregates that are membrane bounded. Those
oocytes with more constricted endoplasmic retic~
ulum contain some large aggregates so ballooned
that they appear to distort the cytoplasm, whereas
those with dilated channels of endoplasmic reticulum contain more compact aggregates whose
contents are dispersed in more moderate amounts
of clear fluid.
 
Vacuoles containing a variety of structures are
seen in the peripheral cytoplasm. Some containing
multiple small vesicles and often surrounded by a
halo of similar free vesicles are frequently associated with Golgi complexes (Figs. 23 and 38).
Others contain either dense material similar to
myelin figures (Fig. 39) or a variety of dispersed
elements similar to those contained in the compound aggregates: electron—opaque granular material or less electron—opaque small vacuoles (Figs.
13 and 14).
 
M icrotubules present throughout the oocyte cytoplasm are most prevalent around the circum
ference of the nuclear membrane where they
apparently support the mitochondria and endoplasmic reticulum. Bundles of spiral filaments
occasionally abut the nuclear membrane (Fig. 7)
or are seen in the more peripheral cytoplasm.
Although the fibrils usually appear to possess only
one coil they occasionally appear to possess two
(see inset, Fig. 7). These fibrils are easily distinguishable from the dispersed microtubules which
are present in the nearby cytoplasmic matrix. At
the extreme periphery of a few oocytes, the cytoplasmic matrix contains a meshwork of extremely
fine fibrils which extend into the evaginations
projecting from the cell membrane of the oocyte
(Fig. 19).
 
The cellular junction between the oocyte and the
follicle cells is complex. Although a narrow stalk
frequently appears to attach the endoplasmic
reticulum of the oocyte to a desmosome at the
oocyte—follicle cell junction, the neck of this stalk
has not been seen to open into the intercellular
space (Fig. 36). Occasionally pinocytic vesicles
apparently of the “coated” variety are seen on the
oolemma (Fig. 14). At intervals, particularly near
junctions of two or three follicle cells, the oolemma
is evaginated into clusters of projections that
interdigitate with the tips of longer projections
from the follicle cells (Figs. 20 and 38). Since these
projections run at a variety of angles to the oocytecut curface, it is often impossible to determine
whether they originate from oocyte or follicle cell.
The AMPase activity that is pronounced in this
area may indicate the predominance of follicular
projections (Fig. 3).
 
The follicle wall contains both lighter and darker
cells (Fig. 4-). Within the cells of the crescentic
thicker area is a conspicuous Golgi complex and
cytocentrum containing centrioles. These para
 
FIGURE 27 A micrograpll from a section on the same grid as that in Fig. 26. N 0t.e evagi—
nation of outer leaflet of nuclear membrane (bottom) that infolds to interdigitate with
membranes from the peripheral cytoplasm thus to form a double loop at this area of the
periphery of annulate lamellae. Note single loops (top) where annulate lamellae become
continuous with endoplasmic reticulum. Note another evagination of outer leaflet of
nuclear membrane at top. H 35-4; 2% glutaraldehyde. )( 252,000.
 
FIGURE 28 A section on a grid adjacent to that shown in Figs. 96 and 97. Note continuity
of distended channels of endoplasmic reticulum with a.nnulate lamellae. Note also that
 
these channels are also continuous with outer leaflet of nuclear membrane (bottom left).
H 35-4; 2% glutaraldehyde. X 22,000.
 
664
 
THE JOURNAL or CELL BIOLOGY - VOLUME 34:, 1967ARTHUR T. HERTIG AND ELEANOR C. ADAMS Human Oocyte and Its Follicle 665
nuclear organelles are oriented toward the oocyte.
These cells frequently contain a large aggregate of
fine filaments. Granular endoplasmic reticulum
and small mitochondria with transverse cristae
are characteristic of the thicker, lighter cells (Fig.
41). As noted above, compound aggregates smaller
than but similar to those in Balbiani’s vitelline
body are present also in the follicular cells (Figs.
17-19). The denser cells of the follicle wall appear
to resemble those in the cortex: they contain many
ribosomes free in the cytoplasmic matrix, often
have many pinocytic vesicles on their peripheral
free surface, and have long extensions that overlap
their neighbors. A very thin, amorphous, basal
lamina intervenes between the follicle wall and
the collagen—containing matrix of the cortex (Figs.
17, 18, and 41). The spindle—shaped cells of the
cortex have long, thin projections that are closely
applied to, and on occasion invade, this basal
lamina. In a few instances peripheral cells within
the follicle wall may overlap their neighbors but
be partially separated from them by a lake of
homogeneous substance morphologically similar to
the material of the basal lamina (Fig. 41). At the
immediate periphery of some follicle walls, a coneshaped arrangement of cells has been observed
that contains at its base a cortical cell that appears
 
to be partially incorporated into the follicle wall
(Fig. 40).
 
DISCUSSION
 
Light microscopists have carefully studied the
large paranuclear structure in human oocytes in
the primordial follicle stage (van der Stricht, 1923;
Aykroyd, 1938; Beams and Sheehan, 1941). Raven
(1961) in a volume summarizing comparative
studies of developing oocytes describes the cytoplasm of this stage as follows: “the earliest oocyte
 
often contains a distinct cytocentrum, situated on
one side against the nuclear membrane, sometimes
in an indentation of the latter. It consists of one or
two small granules, the centrioles, surrounded by a
sphere of dense cytoplasm, the arc/zoplasm or
z'a’z'o5ome. Sometimes there is an indication of astral
radiations in the latter. The archoplasm is often
encircled by a ring of Golgi bodies, which in turn
is surrounded by a dense cloud of mitochondria.
The whole structure thus outlined is known as
Balbianfs vitelline body, often wrongly called yolk
nucleus.” The complexity of the organelles that
form Balbiani’s vitelline body is the distinguishing
ultrastructural characteristic of the oocytes in
primordial follicles obtained from adult human
ovaries. Electron microscopic studies of the unilaminar follicle in other mammalian species have
shown a rather simple, large paranuclear aggregate of vesicles or saccules interpreted as a large
Golgi complex: rat, Sotelo(1959); guinea pig,
Anderson and Beams (1960) and Adams and
Hertig (1964); rabbit, Blanchette (1961) and
Zamboni and Mastroianni (1966); hamster, Odor
(1965) and Weakley (1966). In studying fetal
human ovaries, Lanzavecchia and Mangioni
(1964) and Stegner and Wartenberg (1963) reported a perinuclear concentration of mitochondria and Golgi complexes in the oocytes in the
early stages of meiotic prophase.
 
It is important to recall in examining the ultrastructural characteristics of the oocytes reported
here, that they were obtained from women in their
middle reproductive years, and therefore are cells
that have remained in an arrested meiotic prophase for 25-35 yr. Nevertheless, they would have
continued to be capable of responding to stimuli
that would cause them to resume development:
to enlarge and differentiate into a mature oocyte,
 
 
FIGURES Q9w3‘:2 Sections througli the same concentric arrangement of annulate lamellac
which appears to be attached to nuclear nlenibrane (Fig. 392). Note association of pores 011
annulate lamellae with those on nuclear menibranes (Figs. 29 and 32). Also note continuity
of membranes of annulate lamellae with endoplasmic reticulum (Figs. 31-32). H 43-1;
 
Karnovsky fixative. X 91,000.
 
FIGURE 33
 
These concentrically ar1'a.ng‘ed annulate lamellae are detached from the
 
nuclear meiribrane (right). Closely seriated sections never showed any closer morphologic
proximity to the nucleus. II 44-9.; Karnovsky fi.\'at.ivc. X 10,000.
 
FIGURE 34 In some oocytes only fraginents of the larger annulate lamellar structures can
be found in the cytoplasm. H 43-2; Karnovsky fixative. X 14,000.
 
 
FIGURE 35 Vesicles of endoplasmic reticulum in the peripheral cytoplasm of some oocytes contain an
elect.ron—opaque material. The pore formations on the endoplasmic reticulum at right are interpreted to
be the residuum of a disintegrated annulate lamellar structure. H 44-1; Karnovsky fixative. X 92Q,000.
 
to complete the maturation divisions, and thus to
become fertilizable. In addition to being aged in
terms of somatic cell life span, oocytes in primordial follicles are somewhat unusual cells in that
they have no apparent function other than to
maintain this readiness to respond to stimulation.
 
The almost complete localization of organelles,
except for profiles of endoplasmic reticulum, to a
paranuclear position in many oocytes indicates
a predominantly localized metabolism. Lehninger
(1965) suggests that, in unfertilized, sea—urchin
oocytes whose rate of respiration is low and whose
mitochondria are aggregated into bundles, the rate
of respiration is limited by the rate of diffusion of
oxygen or substrates into the interior of the bundle.
In human oocytes, however, the intimate association of endoplasmic reticulum with mitochondria
suggests a mechanism for the transport of substrates into the aggregation of mitochondria. We
have noted an unusually close morphological relationship of endoplasmic reticulum to mitochondria
in these human oocytes regardless of the fixative
employed. Whether the variable morphology of the
inner lamina of the peripheral membrane of mitochondria, noted as an effect of variation in osmolality of the fixative, reflects any physiological
adaptation related to mechanochemical phenomena (Lehninger, 1965) is unknown. Vacuoles
containing a variety of internal structures also
appear to be evidence of transport to the para
668
 
nuclear metabolic center from the oocyte surface
or vice versa. In some oocytes, the dispersal of
limited numbers of organelles around the circumference of the nucleus together with annulate
lamellae attached or adjacent to the nuclear
membrane may be morphological evidence of some
fluctuating distribution of activity in the oocyte
cytoplasm. Localized projections from both the
oocyte and the follicle cells interdigitate in the
intercellular space and appear to reflect a constant
active interchange between the oocyte and its
follicle wall. The absence of phosphatase activities
within the oocyte is similar to our findings in
guinea pig oocytes at this stage (Adams et al.,
1966). The presence of glucose—6—phosphate dehydrogenase in the oocyte cytoplasm indicates that
the pentose—phosphate pathway of carbohydrate
metabolism may be operative in this resting phase.
 
The presence of annulate lamellae in human
oocytes of this stage was not unexpected since
they have been reported in the cytoplasm of later
stages of human oocyte development by Tardini
et al. (1960) and by Wartenberg and Stegner
(1960) as well as in a human ovum in the pronuclear stage (Zamboni et al., 1966). Annulate
lamellae were not observed in guinea pig oocytes
in primordial or primary follicles (Adams and
Hertig, 1964) nor do they appear to be structures
characteristic of most mammalian oocytes other
than human ones. The derivation of cytoplasmic
 
THE JOURNAL or CELL BIOLOGY - VOLUME 34, 1967
annulate lamellae by vesiculation or blebbing of finding a few oocytes with a large stack of annulate
the nuclear membrane has been described in lamellae attached to the nucleus, but of a greater
oocytes of Necturus and of the echinoderm, T/zyone number with stacked or concentric annulate
briareus by Kessel (1963, 1964, and 1966). Our lamellae lying free in the cytoplasm, in or near
 
FIGURE 36 A narrow stalk of endoplasmic reticulum is in apposition to a desmosolne at the intercellular
junctions between the follicle wall (top) and the oocyte (bottom). H 35-3; 2% glutaraldehyde. X 21,000.
 
FIGURE 37 Endoplasmic reticulum is frequently associated with the compound aggregates in the peripheral cytoplasm of the oocyte. H 35-3; 2% gluta.raldehyde. X 22,000.
 
FIGURE 38 Vacuoles containing multiple small vesicles (center) and often surrounded by a halo of
similar vesicles (left) are present in the peripheral cytoplasm of the oocyte. H 44-3; Karnovsky fixative.
X 15,000.
 
FIGURE 39 Structures resembling myelin figures are separately contained within vacuoles (left) and
are not found within the compound aggregates (_rigl1t). Note that a vesicle of endoplasmic reticulum
containing electr0n—opa.que material is adjacent to the edge of a ballooned aggregate (lower right). H 444; Karnovsky fixative. X 22,000.
 
 
FIGURE 40
 
Balbiani’s vitelline body, shows that they are
repeatedly formed at the nuclear membrane and
are subsequently released into the cytoplasm
where they apparently fragment. The suggested
role for their participation in nucleo~cytoplasmic
transfer (Swift, 1956) was based on the resemblance of the membranes of annulate lamellae in
the cytoplasm to the ultrastructure of the nuclear
membrane. When occasional pores of the nuclear
membranes of these human oocytes have been
observed in register with those in an attached
stack of annulate lamellae, the dense, amorphous
substance contained within and between the
aligned pores appears to form a continuous band
from the inner nuclear membrane toward the base
of the stack. Whether this represents transfer of
material from the nucleus to either the cytoplasmic
matrix or membranes is unknown. When annulate
lamellae are attached to the nuclear membrane, it
is the presence of their pores and their greatly
increased membrane area that distinguish them
 
 
A cone~shaped configuration of cortical cells is frequently seen at the immediate periphery of
the follicle wall. The tip of the cone lies between two arrows. A projection (P) of the cortical cell at the
base of the cone illustrated here appears to be becoming incorporated into the follicle wall. H 4.«4~1;
Karnovsky fixative. X 5400.
 
from the membranes of the slightly granular
endoplasmic reticulum that in the same oocyte
also may be continuous with the outer leaflet of
the nuclear membrane at intervals around its
entire circumference.
 
The varying location of annulate lamellae within
Balbiani’s vitelline body became of interest to us
since, for a considerable period of our study of
primordial follicles, the only oocytes with annulate
lamellae attached to the nuclear membrane were
three (Adams and Hertig, 1965) from two patients
both in the follicular or estrogenic phase of the
menstrual cycle, the endometria showing histologically a typical mid-proliferative pattern. In this
preliminary series other oocytes from the luteal
and menstrual stages of the cycle showed both
detached annulate lamellae and poorer development of the peripheral endoplasmic reticulum.
Histologically these endometria showed 19-20
day secretory criteria and those of early menstruation respectively. There have been many previous suggestions that the cyclic production of estrogen
and /or progesterone in the ovary may have a
local influence on a variety of ovarian structures
and our initial finding suggested to us that a
fluctuating participation by the oocyte nucleus in
the cytoplasmic arrangement of organelles might
be evidence of such a response to cyclic endocrine
events. A hormonal effect on the prevalence and
position of annulate lamellae has been noted in
human endometrium by Ancla and De Brux
(1965). In a series of 40 patients subjected to various forms of hormonal therapy for primary infertility, they reported that in the glandular cells
“annulate lamellae may appear before ovulation,
but are more frequently seen during the secretory
phase if there is hyperestrogenism.” After studying
intracellular localization of estrogen in so—called
target cells, such as rat uterus, as opposed to cells
of the body in general, Segal (1967) has recently
formulated concepts of estrogen action which
may involve an action at the nuclear level. The
difference we noted between oocytes could be
 
easily explained either by our random sampling of
the entire population of primordial follicles in any
one ovary or by our failure to locate the annulate
lamellae in some oocytes examined at only a few
levels in their cytoplasm. Serial grids, although
arduous to study, are required even to locate the
annulate lamellae and a single fortunate section
within a series may produce the only evidence of
their attachment to the nuclear membrane. The
results of a closer inspection of oocytes from the
proliferative, secretory, and menstrual phases of
the cycle and from pregnancy near term are shown
in Table I.
 
This is admittedly too small a series for any
statistical validity, but in these 27 oocytes studied
with great care, there appears to be some evidence
that annulate lamellae attached or adjacent to the
nuclear membrane are more prevalent when
estrogen production is dominant in the normal
menstrual cycle. Whether this in fact reflects any
environmental hormonal influence will require
further detailed study of oocytes in primordial
 
FIGURE 41 In a few follicles, a peripheral cell may overlap its neighbors but be separated from them by
a homogeneous, extracellular material similar to that in the basal lamina at the periphery of the follicle
wall. H 43-1; Karnovsky fixative, X 10,000.
 
TABLE I
 
The Location of Amzulate Lamellae with Respect to the Nuclear in Phase: of
the rlzfenstruai Cycle
 
Position of the annulate
 
Case ' . gicéejytifis lamellar s‘t)1(')\(1:ctture in each
N0_ Phase of cycle by endometrial dating Cycle day C105 eh, Seriated y e_ ______
sections At nuclear Free in
membrane cytoplasm
H6* Mid—proliferative l 1 attached 0
H35 Mid-proliferative 4 4 attached (J
H5411 Mid-proliferative l 1 adjacent 0
H48 Secretory (19~20th day) 23§ Rt ovary 0 3
(CL side)
3
H48 Secretory (l9—20th day) 23§ Left ovary 1 adjacent 2
3
H37 Menstrual 8 2 attached 6
H44 Menstrual 5 1 adjacent 4
H43 Pregnant, caesarian section 35% wk 2 1 attached 1
 
* Infertility patient; wedge resection of normal ovarian tissue.
 
ii This patient had been on a course of antifertility medication for 4 yr, interrupted
by a pregnancy. She had been on same drug, Orthonovum, for 5 months previous to
surgery including the 6th and 7th days of the cycle in which surgery was performed.
 
§ The endometrial dating may or may not coincide with day of cycle owing to normal
variation in the length of the menstrual cycle in the human female.
 
follicles from many more ovaries obtained not only
from various cyclic hormonal phases in the adult
but also from prepubertal age groups.
 
The compound aggregates that surround the
cytocentrum of these human oocytes in primordial
follicles are subject to at least two interpretations:
(a) that they are reservoirs of nutritive material
analagous to yolk material in oocytes of other
species, or (b) that they represent waste deposits of
degradation Of organelles similar to lipofuscin
pigment commonly found in aging cells. Some of
the ultrastructural characteristics of the aggregates, finely divided membranous material and
vacuoles, resemble those found
pigment in neurons, another nondividing cell
(Samarajski et al., 1965). The absence of compound aggregates in oocytes in primordial follicles
of the human prenatal ovaries reported by
Lanzavecchia and Mangioni (1964) and Stegner
and Wartenberg (1963) suggests that the examination of similar follicles in various postnatal and
preadolescent years might be of value in evaluating the effects of aging on the oocyte. The lack of
demonstrable acid-phosphatase activity, however,
indicates that the aggregates may not be lysosomal
and, therefore, may not represent autophagic
processes by which lipofuscin pigment commonly
 
in lipofuscin
 
 
is localized. Furthermore, myelin figures, usually
indicative of membrane degradation, have not
been observed among their heterogeneous components but are frequently seen to be separately
contained within a vacuole in the peripheral
cytoplasm. Thus, it is possible that compound
aggregates represent nutritive material. Dense
aggregates were not observed in our previous
study of guinea pig oocytes (Adams and Hertig,
1964) but have been noted in the cytoplasm of
some other immature mammalian oocytes (rabbit:
Blanchette, 1961; hamster: Weakley, 1966, and
Odor, 1965; rhesus monkey: Hope, 1965). Those
observed in the rhesus oocytes have some ultrastructural similarities to the compound aggregates
we see in the human oocytes, and it was suggested
that they represent developing yolk. The presence
of membranes around some aggregates in these
human oocytes and the absence of a membrane
around others within the same oocyte suggests that
not all compound aggregates are in the same
physiological state. Those that are membrane
enclosed often contain dispersed vesicles of varying size which contain electron—opaque material.
Whether this material is related to that seen
within the endoplasmic reticulum has not been
determined. The close morphological association of endoplasmic reticulum to these aggregates,
 
however, may be related to the inverse reciprocal
effects on these two organelles of variations in
 
osmolality of the fixatives used. The presence of
 
some ballooned aggregates, when the more
peripheral endoplasmic reticulum is constricted,
and the absence of gross ballooning, when the
endoplasmic reticulum is dilated, suggests a
possible pump mechanism between these two
organelles. Merriam (1966) has presented evidence that frog oocytes consist of two osmotically
active “compartments.” One, the protoplasmic
mass, owes its osmotic properties to the plasma
membrane, and the other consisting of interior
vesicles owes its osmotic properties to their investing membranes. Because of the magnitudes of
experimental swelling in the interior vesicles, he
suggests that ion-pumping takes place in the
vesicles of the cytoplasm of the frog oocyte.
 
Whatever the nature of the aggregates seen in
a paranuclear location within the oocyte, the
presence in the periphery of the oocyte of a variety
of vacuoles containing elements of the aggregates
together with the presence of compound aggregates in most follicle cells and in some cortical
cells suggests their transport to the oocyte from
its environment or vice versa. The mechanism
for their transfer to the follicle cell from the cytoplasm of the oocyte, or vice versa, is not clear
from these studies since in micrographs where a
follicular aggregate bulges into the oocyte, complete intervening cell membranes have not been
clearly visualized. The repeated observation of
dense membranous material which appears to
bridge the oocyte-follicle cell membranes may
indicate that there is some mechanism for bulk
transfer of the elements of the compound aggregates. Such membranous forms frequently are
closely associated either in the follicle cell with a
compound aggregate or in the oocyte cytoplasm
with vacuoles containing elements of the aggregates. The presence in the peripheral oocyte
cytoplasm of elements of aggregates within a
vacuolar structure suggests that portions of follicular aggregates may be taken up by a phagocytic mechanism at the oocyte surface to be subsequently transported toward the paranuclear
location.
 
The presence of similar compound aggregates
in some cortical cells may be evidence that these
cells are related to those within the follicle wall.
A histochemical similarity between the cortical cells and those of the primitive follicle wall was
observed with respect to adenosine—5-phosphatase
activity. There is an analogous histochemical
similarity in guinea pig ovaries, although the
differentiated granulosa cells of developing follicles
 
do not maintain this activity in the human as they
do in the guinea pig (Adams et al., 1966). The
absence of observed mitotic activity in primitive
follicle cells could indicate either that they are as
old as the contained oocyte or that they are renewed by some process other than mitosis. In a
study of guinea pig follicles in a similar stage
(Adams and Hertig, 1964) there was suggestive
histological evidence that cortical cells contribute
to the follicle wall. In the human primordial
follicle, it is possible that cortical cells may move
into the thickened, crescentic cap of the follicle
wall and gradually slide around the oocyte to
move out to the cortex again. Although never
clearly visualized in the actual act of becoming
incorporated into the follicle wall, occasional
follicle Cells overlap the periphery of their neighbors and are partially separated from them by a
homogeneous substance morphologically similar
to the material of the basal lamina. In addition,
many dark peripheral cells in the follicle wall
closely resemble cells of the cortex. The coneshaped arrangement of cortical cells noted at the
periphery of some primordial follicles has suggested that these may be the areas where cortical
cells become incorporated into the follicle wall.
Further histochemical study of the nature of the
material within the compound aggregates is
indicated. Such studies might well lead to a clarification of the relationship of such aggregates
within the oocyte to those of the various cells of
the surrounding ovarian tissue.
 
Although light microscopists have over the
years identified structures in oocytes as centrioles
(Raven, 1961), others have believed that the
centriolar apparatus for postfertilization cleavage
is introduced into the egg by the spermatozoan
(see Mazia, 1961 for discussion). The absence of
identifiable centrioles in most electron microscope studies of a variety of oocytes is not surprising in view of the necessarily random sampling of such a large cell. We know of only two
reports of the clear visualization of centrioles.
Sotelo (1959) illustrated a pair of centrioles in an
oocyte from a 2—day—old rat, and Greenfield
(1966) has demonstrated centrioles in oocytes of the domestic chicken shortly after hatching. It
may be of importance that these oocytes in which
centrioles have been identified were from ovaries
of the immediate postnatal stage when mitotic
activity of oogenesis had recently ceased. Even
after specific search in all of our human oocytes,
centrioles could not be identified. Recently in a
report of spindle—associated chromosomes of
meiosis in the rat, Zamboni and Mastroianni
(1966) reported that centrioles are not observed.
Whether centrioles in oocytes disappear with
aging or simply are not found is still unknown.
In the human oocyte, at least, it is possible that
the one or two areas of dense amorphous deposits
found in the center of the cytocentrum, as well
as the long coarse fibers supporting the periphery
of the cytocentrum, may be adaptations of the
centriolar apparatus. Electron microscopic investigations of human oocytes from prenatal and
various postnatal stages might clarify this concept.
Vesicular aggregates either adjacent to or inter
mingled with the central aggregate of dense
 
BIBLIOGRAPHY
 
ADAMS, E. C. and A. T. HERTIG. 1964. Studies on
guinea pig oocytes. I. Electron microscopic observations on the development of cytoplasmic organelles
in oocytes in primordial and primary follicles.
J. Cell Biol. 21:397-427.
 
ADAMS, E. C., and A. T. HERTIG. 1965. Annulate
laniellae in human oocytes in primordial and
primary follicles. J. Cell Biol. 2T:119A.
 
ADAMS, E. C., A. T. HERTIG, and S. FosTER. 1966.
Studies on guinea pig oocytes. II. Histochemical
observations on some phosphatases and lipid in
 
developing and in atretic oocytes and follicles.
 
Am. J. Anal. ]_l9:303—3-40.
 
ANCLA, M. and  DE BRUX. 1965. Occurrence of
intranuclear tubular structures in the
endometrium during the secretory phase, and of
annulate lamellae in hyperestrogenic
Obslet. Gynecol. 26:23+33.
 
ANDERSON, E. and H. V-V. BEAMS. 1960. Cytological
observations on the fine structure of the guinea pig
 
human
 
states.
 
ovary with special reference to the oogonium,
primary oocyte, and associated follicle cells. J.
Ultra-struet. Res. 3;432—446.
 
AYKROYD, O. E. 1938. The cytoplasmic inclusions
in the oogenesis of man. Z. Zellforsefz. 11/Iz'/rroskop.
Andi. 27:691-710.
 
BAKER, T. G. 1963. A quantitative and cytological
study of germ cells in human ovaries. Prue. Roy.
 
Soc. (London) Ser. B. 158:-‘l17—-=1-33.
 
 
amorphous deposits were a consistent finding.
 
This may indicate that the material within these
vesicles is related either to the formation of the
deposits or to their periodic alignment on "line
fibrils which apparently becomes a transition to
the peripheral coarse fibers.
 
This work was supported entirely by grant HD-00137,
Division of Human Development and Child Health,
formerly C-2451 of the National Institutes of Health.
 
The authors are greatly indebted to Doctors
Charles Easterday, Shirley Driscoll, Christopher
Duncan, Sumner Gochburg, Charles Horne, Lorna
Johnson, and Alan Rosenfield of the Boston Hospital
for Women for their cooperation in providing us
with these human ovaries. In addition, grateful
thanks are due to Miss Harriet Jopson and Mrs.
Barbara Barton for their skilled technical assistance,
to Mrs. Audrey Hadfield for the preparation of
photomicrographs and electron micrographs, and
to Mrs. Nancy Cote for typing the manuscript.
 
Received for pubIz'eatz'on 6 February 1967.
 
BALBIANI, E. G. 1864. Sur la constitution du germe
dans 1’oeuf animal avant la fécondation. Compt.
Renal. 58 584-588.
 
BEAMS, H. W. and J. F. SHEEHAN. 1941. The yolk
nucleus complex of the human ovum. Amzt. Record.
81:5-415F554.
 
BECKER, N. H., S. GOLDFISHER, VV. SHIN, and A. B.
NOVIKOFF. 1960. The localization of enzyme activities in the rat brain. J. Bz'o}JIzys‘. Bioefzem. Cytol.
8:649—663.
 
BLANCHETTE, E. J. 1961. A study of the fine structure of the rabbit primary oocyte. J. Ultrastruct.
Res. 5:349-363.
 
FAHIMI, H. D., and M. J. KARNOVSKY. 1966. Cytochemical localization of two glycolytic dehydrogenases in white skeletal muscle.J. Cell Biol. 29:1 13128.
 
GREENFIELD, M. L. 1966. The oocyte of the domestic
chicken shortly after hatching, studied by electron
microscopy. J. Embryoi. Exptf. Moryb/20!. l5:297—
316.
 
Hope, J. 1965. The fine structure of the developing
follicle of the Rhesus ovary. J. Ultrastruet. Res.
12:592-610.
 
KARNOVSKY, M.  1961. The use of ethylene diamine
tetraacetic acid disodium in the histochemical
demonstration of triphosphopyridine nucleotide
linked dehydrogenases. J. Histoehem. Cytoehem.
9:203—20-<1.
 
KARNOVSKY, M. J. 1965. A formaldehyde-glutaraL
dehyde fixative of high osmolality for use in electron microscopy. J. Cell Biol. 2T:137A—138A.
 
KESSEL, R. G. 1963. Electron microscopic studies on
the origin of annulate lamellae in oocytes of Necturus. J. Cell Biol. 19:391-414.
 
KESSEL, R. G. 1964. Electron microscopic studies on
oocytes of an echinoderm, Tlzyone briareus, with
special reference to the origin and structure of
the annulate lamellae. J. Ultrastruct. Res. 10:498514.
 
KESSEL, R. G. 1966. Some observations on the ultrastructure of the oocyte of Thyone briareus with
special reference to the relationship of the Golgi
complex and endoplasmic reticulum in the formation of yolk. J. Ultrastruct. Res. 16:305-319.
 
LANZAVECCHIA, G., and C. MANGIONI. 1964. Etude
de la structure et des constituants du follicle
humain dans l’ovaire foetal I Le follicle primordial.
J. M icroscop. 3 :447—464.
 
LEHNINGER, A. L. 1965. The mitochondrion. Molecular Basis of Structure and Function. W. A.
Beniamin, Inc., New York.
 
MAZIA, D. 1961. Mitosis and the physiology of cell
division. In The Cell. J. Brachet and A. E. Mirsky,
editors. Academic Press Inc., New York. 3.
 
MERRIAM, R. W. 1966. The role of cytoplasmic
membranes in the regulation of water concentration within frog oocytes. Exptl. Cell Res. 41:34-48.
 
MYERS, H. I., W. C. YOUNG, and E. W. DEMPSEY.
1936. Graafian follicle development throughout
the reproductive cycle in the guinea pig, with
especial reference to changes during estrus (sexual
receptivity). Anal. Record. 65:381-401.
 
ODOR, D. L. 1965. The ultrastructure of unilaminar follicles of the hamster ovary. Am. J. Anal. 116:493—
522.
 
PEARSE, A. G. E. 1960. Histochemistry, Theoretical
and Applied. Little, Brown and Company, Boston.
 
RAVEN, C. P. 1961. Oogenesis: The Storage of Developmental Information Pergamon Press, New
York,
 
SAMARAJSKI, T., J. M. ORDY, and J. R. KFEFE. 1965.
The fine structure of lipofuscin age pigment in
 
ARTHUR T. HERTIG AND ELEANOR C. ADAMS Human Oocyte and Its Follicle
 
the nervous system of aged mice. J. Cell Biol.
26:779-795.
 
SEGAL, S.  1967. Fetal homeostasis. 11. Arm. N. Y.
Acad. Sci. In press.
 
SOTELO, J. R. 1959. An electron microscopic study
on the cytoplasmic and nuclear components of
rat primary oocytes. Z. Zellforselz. ;'1-filcroskop. Anal.
50: 749-765.
 
STEGNER, H. E., and H. H. WARTENBERG. 1963.
Elektronenmikroskopische Untersuchungen an
Eizellen des Menchen in verschiedenen Stadien
der Oogenese. Arclz. Gynaekol. 1992151-172.
 
STRICHT, O. VAN DER. 1923. Etude comparée des
ovules des mammiferes aux différentes périodes de
Povogenese, d’aprés les travaux du laboratoire
d’histologie et d’embryologie de l’Université de
Gand. Arch. Biol. 33: 229-300.
 
SWIFT, H. 1956. The fine structure of annulate
larnellae. J. Bioplzys. Bioclzem. Cytol. 2 (Suppl.):
415-418.
 
TARDINI, A., L. VITAL:-MAZZA, and F. E. MANSANI.
1960. Ultrastruttura de1l’ovocita umano maturo.
1. Rapporti fra cellule della corona radiata, pellucida ed ovoplasma. Arch “ale l/ecclzz” Anal‘.
Pallzol. .-Med. Clin. 33:281-305.
 
VENABLE, J. H., and R. COGGESHALL. 1965. A simplified lead citrate stain for use in electron microscopy.
J. Cell Biol. 25:407-408.
 
VVACHSTEIN, M., and E. MEISEL. 1957. Histochemistry
of hepatic phosphatases at a physiologic pH. Am.
J. Clin. Patlzol. 27:13-23.
 
W'ARTENBERo, H., and H. E. STEGNER. 1960. Uber
die elektronenmikroskopische feinstruktur des
menschlichen ovarialeies. Z. Zellforrclz. Mz'kroskop.
Anal. 52:450-474.
 
WEAKLEY, B. S. 1966. Electron microscopy of the
oocyte and granulosa cells in the developing
ovarian follicles of the golden hamster (Mes0crz'cetus aumtus). J. Anal. l00:503-534.
 
ZAMBONI, L., and L. MASTROIANNI, JR. 1966. Electron microscopic studies on rabbit ova. I. The
follicular oocyte. J. Ultmsmcet. Res. 14:95-117.
 
ZAMBONI, L., D. R. MISHELL, JR., J. H. BELL, and
M. BACA. 1966. Fine structure of the human ovum
in the pronuclear stage. J. Cell Biol. 30:579-600.

Revision as of 18:08, 1 May 2018

Embryology - 29 Mar 2024    Facebook link Pinterest link Twitter link  Expand to Translate  
Google Translate - select your language from the list shown below (this will open a new external page)

العربية | català | 中文 | 中國傳統的 | français | Deutsche | עִברִית | हिंदी | bahasa Indonesia | italiano | 日本語 | 한국어 | မြန်မာ | Pilipino | Polskie | português | ਪੰਜਾਬੀ ਦੇ | Română | русский | Español | Swahili | Svensk | ไทย | Türkçe | اردو | ייִדיש | Tiếng Việt    These external translations are automated and may not be accurate. (More? About Translations)

Hertig AT. and Adams EC. Studies on the human oocyte and its follicle. I. Ultrastructural and histochemical observations on the primordial follicle stage. (1967) J Cell Biol. 34(2):647-75. PMID 4292010

Online Editor  
Mark Hill.jpg
This historic 1967 paper by Hertig and Adams uses electron microscopy to study the human oocyte.



Modern Notes: oocyte

Historic Disclaimer - information about historic embryology pages 
Mark Hill.jpg
Pages where the terms "Historic" (textbooks, papers, people, recommendations) appear on this site, and sections within pages where this disclaimer appears, indicate that the content and scientific understanding are specific to the time of publication. This means that while some scientific descriptions are still accurate, the terminology and interpretation of the developmental mechanisms reflect the understanding at the time of original publication and those of the preceding periods, these terms, interpretations and recommendations may not reflect our current scientific understanding.     (More? Embryology History | Historic Embryology Papers)

Studies on the Human Oocyte and its Follicle

I. Ultrastructural and Histochemical Observations on the Primordial Follicle Stage

Arthur T. Hertig and Eleanor C. Adams

From the Department of Pathology, the Harvard Medical School, Boston, Massachusetts 062115

Abstract

Oocytes in primordial (“resting”) follicles in adult human ovaries contain a complex paranuclear structure identified by light microscopists as Balbiani’s vitelline body. By electron microscopy this structure is composed of a mass of mitochondria with associated endoplasmic reticulum, multiple compound aggregates which form a ring around the cytocentrum, and a single stack or coil of annulate lamellae either attached to the nuclear membrane or free in the cytoplasm. The compound aggregates contain vacuoles and finely divided electronopaque material. Evidence is presented for the probable transport of this material between the oocyte and its environment. The cytocentrum contains a central aggregate of amorphous electron—opaque deposits which appear to become periodically aligned on fine fibrils to form the long coarse fibers at the periphery of the cytocentrum. The apparent prevalence of annulate lamellae attached or adjacent to the nuclear membrane of oocytes in ovaries removed during the mid—follicular (estrogenic) phase of the cycle indicates the need for further study of a possible hormonal influence on the resting oocyte. By light microscopy phosphatases were not found within the oocyte, but adenosine—monophosphatase activity is present in the cortical cells surrounding primordial follicles, and also at the periphery of each primitive follicle cell, most prominently at the oocyte side. Glucose—6—phosphate dehydrogenase activity is present within the oocyte cytoplasm.

INTRODUCTION

In the adult human ovary, there are only two stages in the growth of the oocyte and its follicle that can be confidently recognized histologically as normal phases of the maturation process. These stages are at opposite ends of the maturation spectrum: one is the resting oocyte in the primordial follicle (40 ,u in diameter); the other is the oocyte with a meiotic spindle forming the first polar body in the fully matured preovulation follicle (1.5 cm). The normal histology of all the intervening stages cannot be recognized since there are as yet no agreed morphological criteria for

differentiating follicles that are continuing to develop from those in the earliest stages of atresia. (Certainly mitotic activity in the follicle wall or cumulus is no criterion of normal preovulatory development. In both the human and guinea pig we have seen numerous examples of obviously atretic follicles showing mitoses in their granulosa cells. It is unknown, however, at what stage in the development of any given follicle atresia had begun.) The duration of the developmental processes and its possible correlation with cyclic endocrine events also are unknown in the human

female. Moreover the interrelationship of the two ovaries with respect to follicular development is some mammals, however, the morphology of follicular and oocyte developmental processes has been correlated with events in the estrous cycle. For instance, Myers et al. (1936)

unknown. In

made a careful histological examination of follicles in ovaries removed at known times of the 16-17 day estrous cycle of the guinea pig. They reported that soon after estrus many multilaminar primary (nonvesicular) follicles had begun to form an antrum thus becoming a secondary or vesicular follicle, and that from these only two or three follicles in each ovary complete their development and ovulate at the next estrus; the remainder becoming atretic at various stages of their growth. In addition to the normal duration of these vesicular follicle stages (one estrous cycle), Myers et al. reported their histological and size criteria in relation to the various succeeding days of the cycle. We have recently described some of the histochemical aspects of oocytes in developing and atretic follicles throughout the estrous cycle of the guinea pig (Adams et al., 1966).

In order to determine the morphological correlation of human follicle and oocyte developmental stages with known events in the menstrual cycle and thus to estimate follicular growth rate, we have begun a careful examination of the total follicular population in human ovaries removed at various times in the menstrual cycle. An electron microscopic study of the first stage examined, oocytes in primordial follicles, is reported here together with the results of a few histochemical studies. In these oocytes from adult human ovaries we have noted ultrastructural differences from oocytes in similar follicles studied in the human prenatal ovary (Lanzavecchia and Mangioni, 1964; Stegner and Wartenberg, 1963) and in ovaries of a variety of other mammalian species. These differences are chiefly characterized by the complexity and organization of the paranuclear structure previously

identified by light microscopists as Balbiani’s vitelline body.

MATERIALS AND METHODS

The follicles were obtained from normal ovaries removed from five fertile women, 28-37 yr of age, who had gynecologic and /or obstetric indications for hysterectomy such as carcinoma in situ of the cervix or leiomyomata of the uterus. One ovary was removed in four instances and both 1n one instance. In addition, we examined


primordial


primordial follicles from an ovary removed from a patient who had been on the antifertility medication Orthonovum. Also follicles from a wedge biopsy of an ovary, obtained at laparotomy for infertility investigation, were studied. From this material corpora lutea and intact large vesicular follicles are dissected out for other histological, histochemical, or electron microscopic studies. The remaining ovary is then sliced serially at 2-3 mn1 either crosswise or sagitally and fixed for histological, histochemical, and electron microscopic observation of smaller developing or atretic follicles.

For izz'rz‘0clzemz'5try, some slices were fixed 3-12 hr in 4% paraformaldehyde in 0.1 M phosphate buffer pH 7.6, in 2 -,’/'5 glutaraldehyde, or in a combined 2% paraformaldehyde—2%% glutaraldehyde (one-half strength Karnovsky fixative) (1965), both in 0.1 M eacodylate buffer pH 7.4. These were subsequently washed in their respective buffers and rapidly frozen at —70°C. Cryostat sections were reacted for various phosphatases, dehydrogenases, or diaphorases or were stained with Sudan black. Other slices were fresh frozen, sectioned in a cryostat, and reacted for similar enzymes. Although acid phosphatase activity, as visualized by the Gomori method (Becker et al., 1960), was maintained by all three fixatives used, ovarian nucleoside polyphos— phatase, adenosine—5-monophosphatase (VVachstein and Meisel, 1957), and alkaline-phosphatase (Pearse, 1960) activities were maintained only by fixation in 4% paraformaldehyde or in fresh frozen cryostat sections. In view of the doubts recently cast on the reliability of dehydrogenase methods which employ unfixed cryostat sections (Fahimi and Karnovsky, 1966), we have used both 4% paraformaldehydefixed (3 hr) and fresh-frozen cryostat sections for the study of glucose-6—phosphate dehydrogenase and both NADH and NADPH diaphorases by using the methods of Pearse as modified slightly by Karnovsky (1961). In addition, we have added 5 mg/100 ml Nmethyl phenazonium methosulfate (PMS) to some glucose-6-phosphate dehydrogenase incubations in order to bypass diaphorase activity in the production of the formazan.

For electron mz'cro5copz'c studies, the strips of cortex were fixed in 2% glutaraldehyde or in the combined 2% paraformaldehyde——21/2% glutaraldehyde both in 0.1 M eacodylate buffer pH 7.4. After 5 hr fixation, the strips are washed in repeated changes of chilled 0.1 M caeodylate buffer with CaC1g added and are stored in the refrigerator. The slices of ovary thereafter are dissected so that the medulla is separated from a strip of cortex. These are then examined under a dissecting microscope as they are cut into 0.5-0.1 mm transverse slices, each of which is again searched by a dissecting microscope using oblique transillumiprimordial (0.04 mm in diameter), nonvesicular or primary (0.08—0.28 mm in diameter),

nation for

or small vesicular or secondary (0.30 -1- mm in diameter) follicles. After recording the measurements of individual follicles they are further dissected into orientable blocks of 1.5 mm greatest size, are postosmicated, rapidly dehydrated, and embedded in Epon 812. Before thin sectioning, the preselected cortical blocks are “thick”-sectioned (lp) serially, and successive groups of five serial, thick sections are examined with phase microscopy to search for the presence of primordial follicles. When the edge of a primordial follicle is identified, the block is and serial, thin sections are cut and mounted on 100—mesh grids. After six to eight serial grids of thin-sectioned ribbons are collected, another thick section usually is mounted on a glass slide before collecting the next series of six to eight grids of serial sections. The seriated thick-sections are examined and subsequently photographed for orientation of individual primordial follicles which, in our material, are often present in clusters of two to five oocytes per section. This orientation facilitates the electron microscopic study of serial depths through any one oocyte in a cluster. The thin sections, stained with lead (Venable and Coggeshall, 1965) or with uranyl followed by lead were examined with an RCA EMU-3F or 3H microscope.

For fzistalogical study, one cortical strip, fixed in 4% paraformaldehyde or in 2% glutaraldehyde followed by Bouin’s fluid, is embedded in paraffin for serial sectioning and is stained with hematoxylin and eosin or periodic—acid Schiff.

trimmed,

OBSERVATIONS

Light llficroscopy

In the young adult, primordial follicles about 40 ,u. in diameter can be found in clusters lying in the narrow band of dense cortex just beneath the serosal surface and the tunica albuginea. Primordial follicles, whose oocytes are in the prolonged diplotene phase of meiotic prophase (Baker, 1963), are in a resting stage which can be found continuously from the earliest postnatal ovary (when oogenesis has ceased and earlier meiotic prophase stages have been completed) through the adult reproductive period. Their number is gradually decreased, particularly from adolescence to the menopause, owing to the repeated initiation of development of some of them, first to the primary (nonvesicular) and subsequently to the secondary (vesicular) follicular stages which terminate either with maturation and ovulation or with atresia.

A striking cytoplasmic characteristic of oocytes in primordial follicles is a large paranuclear complex identified previously by light microscopists as the “corps vitelline” surrounded by a “couche mitochondrial” (van der Stricht, 1923), as the “yolk nucleus complex,” (Beams and Sheehan, 1941), or preferably (Raven, 1961) as Ba1biani’s vitelline body (1864). This structure has been studied in our histological and histochemical preparations: with hematoxylin and eosin staining (Fig. 1), it is an eosinophilic complex closely applied to the nuclear membrane; with periodicacid Schiff staining both with and without diastase digestion it is a ring of positive granules surrounding a paler pink central sphere; in frozen sections stained with Sudan black, sudanophilic granules of various sizes are found in this area. In both 4% paraforma1dehyde—fixed frozen sections with and without the addition of PMS in the incubation medium and in unfixed sections without PMS, the entire cytoplasm of the oocyte is strongly reactive for glucose-6-phosphate dehydrogenase activity (Fig. 2). When PMS is added to the incubation medium for unfixed sections, the solubility of this enzyme, G—6—P dehydrogenase, is indicated by the production of the formazan in the incubation medium. The oocyte cytoplasm is moderately reactive for both NADH and NADPH diaphorases in both fixed and unfixed sections. Alkaline and acid phosphatases, as well as neutral phosphatases such as nucleoside poly— or monophosphatases are negative in the oocyte.

Some of the primitive follicle cells form a crescent—shaped cap of thicker cells over part of the oocyte; the remaining cells form a much thinner wall around the rest of the oocyte. Mitotic activity has not been observed in the follicle wall of this stage. Adenosine—monophosphatase activity, present in the cortical cells which surround the primordial follicles, also appears at the periphery of the primitive follicle cells, most prominently at the surface facing the oocyte (Fig. 3). Acid phosphatase has not been identified in the primordial follicle wall, although it does appear at subsequent stages of development. According to our criteria, when any follicle has thick or cuboidal cells around its entire surface, development to the primary (nonvesicular) follicle has been initiated. Although the oocyte within this early developing follicle is not much larger in diameter than that in the primordial follicle, it begins to show changes at the ultrastructural level.

Electron Microscopy

In thin—sectioning blocks of cortex which contain dispersed or clustered primordial follicles, the plane of section with respect to the polarity of any oocyte is necessarily random. Of the 30-40 oocytes serially sectioned in our study, only a very few fortunate micrographs show the interrelationship of all of the cytoplasmic organelles found in the mid—polar region (Fig. 4). (By mid—polar is meant a section which passes through or near the center of the nucleus, through or near the center of the cytocentrum and in addition includes the significant part of the annulate lamellae. This structure is single and eccentric with respect to the true polar axis and may be either attached or detached from the nuclear membrane.) In most oocytes it is necessary to reconstruct mentally the three—dimen— sional relationships by superimposing micrographs from serial depths through an oocyte.

Study of primordial follicles in ovaries obtained


Abbreviations

AL, annulate lamellae

C, Cortex

CA, compound aggregates CF, coarse fiber

CH, el1roma.tin

F, follicle cell

G, Golgi complex

H, heterochroinatin

from a variety of phases of the menstrual cycle and from a term pregnancy reveal a constant general relationship of organelles. The nucleus of the oocyte is large and usually spherical and contains a large reticulated nucleolus. Mitochondria, elements of the endoplasmic reticulum and Golgi complexes, are found in variable numbers around the entire circumference of the nucleus. A large paranuclear complex is formed by a concentration of these organelles surrounding large compound aggregates which form a ring around a large cytocentrum. Annulate lamellae are found near or within this paranuclear complex which in its entirety comprises the structure previously described as Balbiani’s vitelline body. The variations between individual oocytes which appear to reflect either metabolic changes or a response to fixation difference will be noted below in the following detailed description of organelle arrangements. The spherical nucleus (Figs. 4 and 25), varying

I, inte1'chro1na.tin granules ill T, n1icrotubules

N, nucleus

NU, nucleolus

0, oocyte

P, projection

VA, vesicular aggrega.tes

650

NOTE: In the legends the II number which identifies each patient is followed by another number identifying the particular oocyte represented.

FIGURE 1 Mid—polar section through an oocyte in a primordial follicle. Closely applied to the spherical vesicular nucleus is the cresce11t—shaped Balbia.ni’s vitelline body (between asterisks at lower right against nucleus) formed by peripheral dense granules (mitochondria) and a central paler area (cytoeentrum) surrounding a dense body. H 48, H —l— 14}. X 560.

FIGURE Q Note that the cytoplasm of these three oocytes in primordial follicles is strongly reactive for glucose—6—phosphate dehydrogenase. II 54; paraformaldehyde fixative. >( 350.

FIGURE 3 The cortical cells and the periphery of the cells in the Wall of the primitive follicles at right are reactive for adenosine~Inon0phosphatase. The pronounced activity at the oocyte—follicle cell junction apparently is located in the numerous projections of the follicle cells present in this intereellula.r space. The activity of the enzyme is not maintained at the periphery of differentiating follicular cells of the multilaminous primary follicle at left. H 53; paraformaldehyde fixative. X 350.

FIGURE 4 A low—power survey micrograph through the mid-polar axis of an oocyte in a primordial follicle. Note that most of the organelles are concentrated in one pole of the oocyte; compare with Fig. 9.5. II 44-1; Karnovsky fixative; uranyl and lead stain. X $2700.


FIGURE 5 Nucleus of oocyte showing nucleolonema and associated clump of heterochromatin, aggregates of chromatin with associated perichromatin granules and fine fibrils, and dispersed clumps of interchromatin granules. Pores are prominent on the nuclea.r membrane. H ‘M-1; Karnovsky fixative; lead stain X 15,000.

from 22 to 24 p. in diameter, is enclosed by a smooth envelope whose two leaflets exhibit prominent pores (Figs. 5-7). Aggregates of filamentous chromatin with associated perichromatin granules and fine fibrils are distributed throughout the nucleus in wispy strands that apparently represent sections through the diplotene configuration of chromosomes. Clumps of interchromatin granules are also dispersed ‘throughout the nucleus. A large irregularly shaped nucleolus with a nucleolonema is prominent in these oocytes (Fig. 5). Multiple, large clumps of fine, granular material interpreted as heterochromatin are distributed around the periphery of the nucleus or are located near the nucleolus (Figs. 5 and 6). They are surrounded by chromatin and perichromatin granules.

The cytocentrum (Figs. 8 and 10) is similar to that in other cell types in that it contains few, if any, mitochondria. At its periphery, a network of sparsely granular endoplasmic reticulum appears to be continuous with many small Golgi complexes and with the more granular endoplasmic reticulum of the peripheral cytoplasm. Elements of this

- network often are also closely applied to the large

compound aggregates. Coarse, long fibers of electron—opaque material are found in the cyto THE JOURNAL or CELL BIOLOGY - VOLUME 34, 1967 FIGURE 6 Multiple clumps of heterochromatin are also found near the periphery of the nucleus. H 48L2; Karnovsky fixative; lead stain. >( 21,000.

FIGURE 7 Bundles of spiral filaments are occasionally attached to the nucleus or lie free in the cytoplasm. H 35-1; 2%. glutaraldehyde. )( 952,000. Inset from same oocyte but adjacent section. X 99,000.

plasmic matrix of this network and extend just beyond the periphery of the cytocentrum where they usually appear to terminate bluntly but occasionally may abut a compound aggregate or a Golgi complex. When serial sections have afforded an opportunity to estimate their three—dimensional relationship, the coarsest fibers appear to form an interwoven basket—like structure at the periphery of the cytocentrum. Smaller caliber fibrils in some oocytes radiate out from near the center of the cytocentrum to join the coarser ones. Although the maximum length of each fiber could not be determined, the greatest diameter is about 0.13 M and each appears to be straight or only slightly curved (Fig. 11). Fixation differences may affect their structure since on cross—section some are homogeneous, some appear as rings with pale centers (Fig. 9), and others as closely packed tetrads. N0 substructures such as tubules have been identified. The microtubules seen within the cytocentrum do not appear to originate from or to be concentrated

near these fibers. A feltwork of very fine fibrils

often visualized in the matrix of the cytocentrum, appears to merge with these coarse fibers.

The large, prominent center of the cytocentrum i.s composed of deposits of amorphous, electronopaque material (Figs. 8 and 10). These deposits frequently are seen in periodic alignments which radiate from the center (Fig. 9). When this

periodicity is present, the amorphous deposits appear to be aligned on very fine fibrils; this suggests the formation of a transition between the central concentration of amorphous deposits and the more peripheral fibers. Although on occasion the amorphous deposits in the center of the cytocentrum may be reminiscent of the ultra-structural aspect of glycogen, the presence of glycogen in oocytes could not be confirmed by our histochemical studies. Also within the cytocentrum of most oocytes one, or more, large (3.5 p. in diameter), compact aggregate of closely packed vesicles is found (Figs. 8 and 9). Such vesicles are smooth—wal1ed, often appear to contain slightly electron-opaque material, and frequently can be seen to be continuous with the endoplasmic reticulum of this area. In other oocytes these vesicles seem to be dispersed among the amorphous deposits to contribute to a single, central, dense area of the cytocentrum measuring up to 4.5 p. (Figs. 10 and 11). This last arrangement may be the ultrastructural basis for the dense granule visible in the center of the cytocentrum in some H (hematoxylin) and E (eosin) preparations of oocytes (Fig. I).

The resemblance of some of the periodic arrays of deposits and of some coarse fibers to peri»centriolar structures described by others in a variety of somatic cells, led us to hunt systernati-—


FIGURE 8 A section through the center of Ba1biani’s vitelline body, showing a cytoplasmic stack of annulate lamellae, ballooned compound aggregates, and a Vesicular aggregate in the cytocentrum near a concentration of clumps of dense amorphous material. For higher magnification see Fig. 9. Straight

fibers appear to radiate from this area. Coarser fibers are seen in cross-section at the periphery of the cytocentrum. H 44-1; Karnovsky fixative. X 10,000.

\ FIGURE 9 A higher power of Fig. 8 to show the clumps of dense amorphous material, some of which appear to be periodically aligned on fine fibrils suggesting a transition between the amorphous deposits and the peripheral coarse fibers at right center. Note feltwork of fine fibrils in the cytoplasmic matrix. A portion of the vesicular aggregate is seen at top left. Note the two instances of helical formation of ribo— somes (asterisk) on flat View of endoplasmic reticulum. H sH—1; Karnovsky fixative. )( 32,000.


FIGURE 10 A mid—section through Balbiani’s vitelline body to show the peripheral concentration of mitochondria, the variable ballooning of compound aggregates and the eytocentrum with a dense central area surrounded by endoplasmic reticulum and coarse fibers. A coiled configuration of annulate lamellae, detached from the nuclear inembrane, was seen. beginning in. the adjacent and continuing in subsequent grids. H 48—R1; Karnovsky fixative. X 3800.

FIGURE 1.1 A higher power of Fig. 10, showing in this oocyte, the apparent mingling of the elements of the vesicular aggregates with the clumps of dense amorphous material. Note further the Central cluster of vesicles. At the periphery a.re seen longitudinal and tangential sections through coarse fibers. H 48 R1; Karnovsky fixative. X 91,000, FIGURE 12 Compound aggregates showing variability’ in their structure. Those that are membranebounded are ballooned to a variable degree and in addition to vacuoles and electron-opaque granules also contain Vesicles enclosing electron—opaque material. Others that are not 1'ne1nbrane—bounded frequently have elements of endoplasmic reticulum closely applied to their surface (right). Note mitochondrial rosette at right. H 48-R1; Karnovsky fixative. X £21,000.

cally for centrioles, particularly in this area of the ooeyte. Since no centriole had been seen in any oocyte routinely examined throughout its entirety, a series of 75 serial grids containing four oocytes was specially prepared without intervening thick sections. In spite of careful search throughout the cytocentrum and elsewhere in the cytoplasm of these four oocytes, no centrioles could be identified.

Golgi compfexes (Figs. 8, 23, and 24) are closely associated with the periphery of the cytocentrum and in some oocytes appear to form a fenestrated shell around it. They are also interspersed between the mitochondria that form the peripheral region of Balbiani’s vitelline body and in addition are frequently associated with endoplasmic reticulum near the oolemma (Fig. 19). As noted above, their membranes appear to be continuous with the endoplasmic reticulum. Although Golgi complexes frequently are located adjacent to the large compound aggregates, no continuity between these structures has been noted. In some oocytes, Golgi complexes along with endoplasmic reticulum and


FIGURES 13 and 14 These microgra.phs are from two sections on the same grid through the same area at the periphery of the oocyte. Vacuoles are present tha.t contain either material sin1ilar to myelin figures or elements such as finely divided electron-opaque material (Fig. 13) or small vacuoles (Fig. 14) similar to those within. the compound aggregates. Note coated vesicles at oocyte membrane in Fig. 14. II 44-1; Karnovsky fixative. X 21,000.

mitochondria are found closely applied to the entire circumference of the nucleus (Fig. 25). The large electron—opaque compound aggregates (Figs. 8 and 10) at the periphery of the cytocentrum are composed of large vacuoles containing a homogeneous, slightly electron—opaque material surrounded by closely packed, very electronopaque, membranous or granular substance. They

present a Variable appearance since some are compact and may or may not be membrane enclosed, while others in the same oocyte are enclosed by a membrane, are ballooned to a variable degree, and appear to have imbibed a clear fluid in which the substructures are dispersed. The latter form contains many small vesicles enclosing dense material (Fig. 12). Smaller com FIGURES 15 and 16 Membranous formations are frequently seen at the intercellular junction between oocyte (top) and follicle cell (bottom). These frequently contain small vacuoles or finely divided electron-opaque material similar to those within the compound aggregates and may represent a mechanism for the bulk transfer of this material between follicle wall and oocyte. H 48-111 and H 44-5; Karnovsky fixative. X 32,000.

FIGURES 17 and 18

Compound aggregates are also seen within the cytoplasm of the

follicle cell. They frequently appear to bulge into the oocyte and in such areas the membranes of both cells appear indistinct. H 44-1 and H 44-5; Karnovsky fixative. X 21,000.

FIGURE 19 A compound aggregate in a follicle cell. Note similarity to aggregates in oocyte. The intercellular space between oocyte and follicle conta.ins projections from both cells. Note Golgi complex in close association with endoplasmic reticulum at periphery of oocyte. H 44-3; Karnovsky fixative. X 21,000.

FIGURE 20 Compound a.ggregates are also seen within the cytoplasm of cortical cells outside the primordial follicle wall. Note desmosome at junction between two follicle cells

and also concentration of projections i11 intercellular space at the j unction of oocyte membrane and two follicle cells. II 43-1; Karnovsky fixative. X 21,000.


FIGURE 21 Mitochondria to show effect of fixation. With the slightly hypertonic 2% glutaraldehyde fixative, the mitochondria contain arched cristae a.nd a distended channel interpreted to be caused by the infolding of the inner lamina. of the peripheral membranes with or without rupture of the external lamina. The internal vesicle frequently seen within this distended channel could be a cross—section of endoplasmic reticulum which apparently artifactitiously invades the mitochondria (see mitochondrion at

lower left). Note distended channels of endopla.smic reticulum with this fixative. H 35-4; 2% gluta.ral— dehyde. X 36,000.

FIGURE 22 With the more hypertonic one-half strength Karnovsky fixative, the arched cristae are prominent a.nd, in addition, multiple secondary tubular cristae are formed by the infolding of the inter— nal lamina. of the-periplieral n_1en'1branes. Note close association of endoplasmic reticulum with mitochondria. Note also electron-opaque material at center of mitochondrial rosette. H 44-3; Karnovsky fixative. X 49,000.

pact aggregates, usually without enclosing mem- 17—20 and 37). Similar compound aggregates are

branes, may be seen in the more peripheral cyto- also present in the serosal epithelium. Vacuoles plasm, within the primitive follicle cells and in containing elements similar to those in the ag many cells of the surrounding cortex (Figs. 4, gregates are present in the peripheral cytoplasm

660 THE JOURNAL OF CELL BIOLOGY - VOLUME 34, 1967 FIGURE ‘23

Golgi complexes are seen in the peripheral cytoplasm of the oocyte where they frequently

are associated with vacuoles containing multiple vesicles. H 44-1; Karnovsky fixative. >< Q1,000.

FIGURE +24 Golgi complexes (right) a.re also prominent at the periphery of the cytocentrum (left) where they appear t.o be in continuity with the network of endoplasmic reticulum surrounding a.nd intermingled with the clumps of amorphous electron-opa.que material. H 44-3; Karnovsky fixative. X 21,000.

of the oocyte (Figs. 13 and 14-). Frequently a dense, membranous structure is seen lying across and apparently merged with the oocyte membrane (Figs. 15 and 16). This rather amorphous structure, when followed in serial sections, often is associated with vacuoles in the oocyte cytoplasm. When seen in thinner areas of the cytoplasm of the follicle wall, the aggregates often appear to bulge into the cytoplasm of the oocyte (Figs. 17 and 18).

A mass of mitochondria (Fig. 10) form the periphery of Balbiani’s vitelline body and a variable number are scattered around the nucleus and in the peripheral cytoplasm. They are spherical—to— ovoid in shape, of variable size, 0.5—l .8 p., and are intimately associated with the endoplasmic reticulum. Both the mitochondria and the endoplasmic

reticulum appear to be unusually responsive to differences in tonicity of the fixing fluid. With only slightly hypertonic glutaraldehyde—fixation (Fig. 21), the mitochondria contain a few, prominent, peripheral arched cristae in a pale, homogeneous matrix within which are only a few small granules. In addition, they also contain a small, distended channel interpreted as caused by the swelling of a single crista or by a local infolding of the inner lamina of the peripheral membrane with or without the rupture of the outer lamina. This distortion is frequently associated with the close proximity of endoplasmic reticulum which on occasion even appears (apparently artifactitiously) to invaginate the mitochondria. With the more hypertonic one—half strength Karnovsky paraformaldehyde—

ARTHUR T. HERTIG AND ELEANOR C. ADAMS Human Oocyte and Its Follicle 661 FIGURE 25 This l0W—p0wer micrograph shows a section that is not mid-polar through the oocyte since the paranuclear Balbiani Vitelline body was present in deeper sections. Note that a stack of annulate lamellae is attached to nuclear mernbrane, that a Variety of organelles are closely applied to the entire circuinference of the oocyte nucleus and that. endoplasmic reticulum is present as distended channels. ll 35-4; 2% glutaraldehyde. X 2800.

FIGURE 26 A higher power of Fig. 25 to show continuity of membranes of a1111ulate lainellae with those of the peripheral endoplasmic reticulum. Note evagination of outer leaflet of nuclear membraile at left of annulate lamcllae. See Figs. 27 a.11d 28 for higher power View of this stack of annulate lamellae as observed in nearby sections. H 35-4; 2% glutaraldeliyde. X 7300.

662 gluteraldehyde-fixation (Fig. 22) the very close structural relationship of mitochondria to endoplasmic reticulum is maintained, but the mitochondrial matrix is more electron—opaque. Furthermore, in addition to the usual prominent, arched cristae, the peripheral inner membrane often appears to be infolded to form multiple secondary vesicular or tubular cristae. The membranes forming these structures are not so electronopaque as those of the apparently more stable, arched cristae. Mitochondria in Balbiani’s vitelline body or elsewhere around the nucleus frequently are clustered in rosettes around a dense amorphous, somewhat reticulated substance (Figs. 12 and 22). Although this material may appear to form an adherent cap on one or more of the mitochondria in the rosette formation, examination of serial sections through a rosette indicates that this image is caused by a superimposition of structures on a tangential cut.

When closely seriated grids through Balbiani’s vitelline body are available for study, a single mass of stacked or concentric annulate lamellae can be found (Figs. 4, 8, and 29—33). In some oocytes these are in continuity with the nuclear membrane (Figs. 25-28); in others they are found at some distance from the nucleus (Figs. 33 and 34). Whether they are attached to the nucleus or not, the smooth—surfaced paired membranes of the annulate lamellae are continuous at their periphery with the more granular peripheral endoplasmic reticulum (Figs. 27, 28, 31-34) which in turn is closely associated with mitochondria (Fig. 26). In one of a series of sections cut in a fortunate plane through annulate lamellae attached to the nucleus, a continuity of membranes could be clearly visualized (Fig. 27). The outer leaflet of the nuclear membrane is evaginated to form, by infoldings, a large stack of paired membranes whose pore formations are similar to but more closely spaced than those found on the nuclear membrane at a distance from the annulate lamellae (Figs. 29-32). It is noted that in the area of the evagination and infolding of the outer nuclear membrane, the outer loops of the paired annulate membranes are double whereas they are single in areas of the stack where the paired membranes become continuous with the peripheral endoplasmic reticulum. Unfortunately, micrographs through the entire stack of these annulate lamellae were not available for complete three—dimensional study since some of the stack was contained within a thick section. Study of five sections on three serial

ARTHUR T. IIERTIG AND ELEANOR. C. A1JAMs H'u.ma.n Uocyte and Its Follicle

grids suggests that the evaginating nuclear membrane forms folds that interdigitate either with similar, looped membranes from a second unidentified nuclear evagination or with membranes from the peripheral endoplasmic reticulum. If the latter is the case, the nuclear evagination and infoldings may form a structural template for the induction of “nuclear” pores on the interdigitating membranes of the endoplasmic reticulum. When the pores on adjacent lamellae are aligned, the dense, amorphous substance contained within and between these aligned pores may form a continuous band from the inner nuclear membrane through several layers of the stack (Figs. 27 and 29). The largest mass of annulate lamellae measures 5-7 it in longest diameter. Those that are unattached to the nuclear membranes are often smaller and in some oocytes only one or two fragmented remnants of the larger structure could be found (Fig. 34). Even as they fragment, their continuity with endoplasmic reticulum appears to be maintained since in several oocytes we have seen pores either singly or in clusters on a profile of endoplasmic reticulum (Fig. 35).

In some oocytes, particularly those with annulate lamellae attached or adjacent to the nucleus, the membranes of the endoplasmic reticulum may communicate with the outer leaflet of the nuclear membrane at intervals around its entire circumference (Figs. 25 and 28) and with Golgi complexes scattered throughout the cytoplasm. Other oocytes have most of their endoplasmic reticulum as well as Golgi complexes and mitochondria confined to the area of Balbiani’s vitelline body and to the peripheral cytoplasm in this pole of the oocyte (Fig. 4). Vesicles of endoplasmic reticulum in the peripheral cytoplasm frequently contain an electron—opaque amorphous material (Fig. 35). Although similar vesicles have been observed adjacent to a ballooned, compound aggregate (Fig. 39) we have no clear—cut morphological evidence that the electron—opaque material contained within the endoplasmic reticulum is related to that within the aggregates. The granularity of the endoplasmic reticulum can only be seen in patches on cut view, although a helical configuration (Fig. 9) of polyribosomes can often be seen on fiat view. Single or clustered ribosomes are numerous in the cytoplasmic matrix. The proximity of endoplasmic reticulum to mitochondria was noted above. There appears to be some morphological relationship between the endoplasmic reticulum and the membrane-enclosed com 663 pound aggregates which also is revealed as a fixation eflect. Although the membrane around an aggregate does not appear to be continuous with the endoplasmic reticulum, channels of endoplasmic reticulum often appear to encircle portions of aggregates that are not membrane enclosed (Figs. 12 and 37). The dilation and constriction of the endoplasmic reticulum in any one oocyte appears to reflect differences in the osmolality of the fixation used (see figs. 4 and 25). With 2% glutaraldehyde (only slightly hypertonic) the endoplasmic reticulum is seen as dilated, continuous channels, whereas with the more hypertonic paraformaldehyde-glutaraldehyde solution, it appears constricted and is usually seen as chains of large vesicles. In addition, there is frequently a reciprocal inverse response in some compound aggregates that are membrane bounded. Those oocytes with more constricted endoplasmic retic~ ulum contain some large aggregates so ballooned that they appear to distort the cytoplasm, whereas those with dilated channels of endoplasmic reticulum contain more compact aggregates whose contents are dispersed in more moderate amounts of clear fluid.

Vacuoles containing a variety of structures are seen in the peripheral cytoplasm. Some containing multiple small vesicles and often surrounded by a halo of similar free vesicles are frequently associated with Golgi complexes (Figs. 23 and 38). Others contain either dense material similar to myelin figures (Fig. 39) or a variety of dispersed elements similar to those contained in the compound aggregates: electron—opaque granular material or less electron—opaque small vacuoles (Figs. 13 and 14).

M icrotubules present throughout the oocyte cytoplasm are most prevalent around the circum ference of the nuclear membrane where they apparently support the mitochondria and endoplasmic reticulum. Bundles of spiral filaments occasionally abut the nuclear membrane (Fig. 7) or are seen in the more peripheral cytoplasm. Although the fibrils usually appear to possess only one coil they occasionally appear to possess two (see inset, Fig. 7). These fibrils are easily distinguishable from the dispersed microtubules which are present in the nearby cytoplasmic matrix. At the extreme periphery of a few oocytes, the cytoplasmic matrix contains a meshwork of extremely fine fibrils which extend into the evaginations projecting from the cell membrane of the oocyte (Fig. 19).

The cellular junction between the oocyte and the follicle cells is complex. Although a narrow stalk frequently appears to attach the endoplasmic reticulum of the oocyte to a desmosome at the oocyte—follicle cell junction, the neck of this stalk has not been seen to open into the intercellular space (Fig. 36). Occasionally pinocytic vesicles apparently of the “coated” variety are seen on the oolemma (Fig. 14). At intervals, particularly near junctions of two or three follicle cells, the oolemma is evaginated into clusters of projections that interdigitate with the tips of longer projections from the follicle cells (Figs. 20 and 38). Since these projections run at a variety of angles to the oocytecut curface, it is often impossible to determine whether they originate from oocyte or follicle cell. The AMPase activity that is pronounced in this area may indicate the predominance of follicular projections (Fig. 3).

The follicle wall contains both lighter and darker cells (Fig. 4-). Within the cells of the crescentic thicker area is a conspicuous Golgi complex and cytocentrum containing centrioles. These para


FIGURE 27 A micrograpll from a section on the same grid as that in Fig. 26. N 0t.e evagi— nation of outer leaflet of nuclear membrane (bottom) that infolds to interdigitate with membranes from the peripheral cytoplasm thus to form a double loop at this area of the periphery of annulate lamellae. Note single loops (top) where annulate lamellae become continuous with endoplasmic reticulum. Note another evagination of outer leaflet of nuclear membrane at top. H 35-4; 2% glutaraldehyde. )( 252,000.

FIGURE 28 A section on a grid adjacent to that shown in Figs. 96 and 97. Note continuity of distended channels of endoplasmic reticulum with a.nnulate lamellae. Note also that

these channels are also continuous with outer leaflet of nuclear membrane (bottom left). H 35-4; 2% glutaraldehyde. X 22,000.

664

THE JOURNAL or CELL BIOLOGY - VOLUME 34:, 1967ARTHUR T. HERTIG AND ELEANOR C. ADAMS Human Oocyte and Its Follicle 665 nuclear organelles are oriented toward the oocyte. These cells frequently contain a large aggregate of fine filaments. Granular endoplasmic reticulum and small mitochondria with transverse cristae are characteristic of the thicker, lighter cells (Fig. 41). As noted above, compound aggregates smaller than but similar to those in Balbiani’s vitelline body are present also in the follicular cells (Figs. 17-19). The denser cells of the follicle wall appear to resemble those in the cortex: they contain many ribosomes free in the cytoplasmic matrix, often have many pinocytic vesicles on their peripheral free surface, and have long extensions that overlap their neighbors. A very thin, amorphous, basal lamina intervenes between the follicle wall and the collagen—containing matrix of the cortex (Figs. 17, 18, and 41). The spindle—shaped cells of the cortex have long, thin projections that are closely applied to, and on occasion invade, this basal lamina. In a few instances peripheral cells within the follicle wall may overlap their neighbors but be partially separated from them by a lake of homogeneous substance morphologically similar to the material of the basal lamina (Fig. 41). At the immediate periphery of some follicle walls, a coneshaped arrangement of cells has been observed that contains at its base a cortical cell that appears

to be partially incorporated into the follicle wall (Fig. 40).

DISCUSSION

Light microscopists have carefully studied the large paranuclear structure in human oocytes in the primordial follicle stage (van der Stricht, 1923; Aykroyd, 1938; Beams and Sheehan, 1941). Raven (1961) in a volume summarizing comparative studies of developing oocytes describes the cytoplasm of this stage as follows: “the earliest oocyte

often contains a distinct cytocentrum, situated on one side against the nuclear membrane, sometimes in an indentation of the latter. It consists of one or two small granules, the centrioles, surrounded by a sphere of dense cytoplasm, the arc/zoplasm or z'a’z'o5ome. Sometimes there is an indication of astral radiations in the latter. The archoplasm is often encircled by a ring of Golgi bodies, which in turn is surrounded by a dense cloud of mitochondria. The whole structure thus outlined is known as Balbianfs vitelline body, often wrongly called yolk nucleus.” The complexity of the organelles that form Balbiani’s vitelline body is the distinguishing ultrastructural characteristic of the oocytes in primordial follicles obtained from adult human ovaries. Electron microscopic studies of the unilaminar follicle in other mammalian species have shown a rather simple, large paranuclear aggregate of vesicles or saccules interpreted as a large Golgi complex: rat, Sotelo(1959); guinea pig, Anderson and Beams (1960) and Adams and Hertig (1964); rabbit, Blanchette (1961) and Zamboni and Mastroianni (1966); hamster, Odor (1965) and Weakley (1966). In studying fetal human ovaries, Lanzavecchia and Mangioni (1964) and Stegner and Wartenberg (1963) reported a perinuclear concentration of mitochondria and Golgi complexes in the oocytes in the early stages of meiotic prophase.

It is important to recall in examining the ultrastructural characteristics of the oocytes reported here, that they were obtained from women in their middle reproductive years, and therefore are cells that have remained in an arrested meiotic prophase for 25-35 yr. Nevertheless, they would have continued to be capable of responding to stimuli that would cause them to resume development: to enlarge and differentiate into a mature oocyte,


FIGURES Q9w3‘:2 Sections througli the same concentric arrangement of annulate lamellac which appears to be attached to nuclear nlenibrane (Fig. 392). Note association of pores 011 annulate lamellae with those on nuclear menibranes (Figs. 29 and 32). Also note continuity of membranes of annulate lamellae with endoplasmic reticulum (Figs. 31-32). H 43-1;

Karnovsky fixative. X 91,000.

FIGURE 33

These concentrically ar1'a.ng‘ed annulate lamellae are detached from the

nuclear meiribrane (right). Closely seriated sections never showed any closer morphologic proximity to the nucleus. II 44-9.; Karnovsky fi.\'at.ivc. X 10,000.

FIGURE 34 In some oocytes only fraginents of the larger annulate lamellar structures can be found in the cytoplasm. H 43-2; Karnovsky fixative. X 14,000.


FIGURE 35 Vesicles of endoplasmic reticulum in the peripheral cytoplasm of some oocytes contain an elect.ron—opaque material. The pore formations on the endoplasmic reticulum at right are interpreted to be the residuum of a disintegrated annulate lamellar structure. H 44-1; Karnovsky fixative. X 92Q,000.

to complete the maturation divisions, and thus to become fertilizable. In addition to being aged in terms of somatic cell life span, oocytes in primordial follicles are somewhat unusual cells in that they have no apparent function other than to maintain this readiness to respond to stimulation.

The almost complete localization of organelles, except for profiles of endoplasmic reticulum, to a paranuclear position in many oocytes indicates a predominantly localized metabolism. Lehninger (1965) suggests that, in unfertilized, sea—urchin oocytes whose rate of respiration is low and whose mitochondria are aggregated into bundles, the rate of respiration is limited by the rate of diffusion of oxygen or substrates into the interior of the bundle. In human oocytes, however, the intimate association of endoplasmic reticulum with mitochondria suggests a mechanism for the transport of substrates into the aggregation of mitochondria. We have noted an unusually close morphological relationship of endoplasmic reticulum to mitochondria in these human oocytes regardless of the fixative employed. Whether the variable morphology of the inner lamina of the peripheral membrane of mitochondria, noted as an effect of variation in osmolality of the fixative, reflects any physiological adaptation related to mechanochemical phenomena (Lehninger, 1965) is unknown. Vacuoles containing a variety of internal structures also appear to be evidence of transport to the para 668

nuclear metabolic center from the oocyte surface or vice versa. In some oocytes, the dispersal of limited numbers of organelles around the circumference of the nucleus together with annulate lamellae attached or adjacent to the nuclear membrane may be morphological evidence of some fluctuating distribution of activity in the oocyte cytoplasm. Localized projections from both the oocyte and the follicle cells interdigitate in the intercellular space and appear to reflect a constant active interchange between the oocyte and its follicle wall. The absence of phosphatase activities within the oocyte is similar to our findings in guinea pig oocytes at this stage (Adams et al., 1966). The presence of glucose—6—phosphate dehydrogenase in the oocyte cytoplasm indicates that the pentose—phosphate pathway of carbohydrate metabolism may be operative in this resting phase.

The presence of annulate lamellae in human oocytes of this stage was not unexpected since they have been reported in the cytoplasm of later stages of human oocyte development by Tardini et al. (1960) and by Wartenberg and Stegner (1960) as well as in a human ovum in the pronuclear stage (Zamboni et al., 1966). Annulate lamellae were not observed in guinea pig oocytes in primordial or primary follicles (Adams and Hertig, 1964) nor do they appear to be structures characteristic of most mammalian oocytes other than human ones. The derivation of cytoplasmic

THE JOURNAL or CELL BIOLOGY - VOLUME 34, 1967 annulate lamellae by vesiculation or blebbing of finding a few oocytes with a large stack of annulate the nuclear membrane has been described in lamellae attached to the nucleus, but of a greater oocytes of Necturus and of the echinoderm, T/zyone number with stacked or concentric annulate briareus by Kessel (1963, 1964, and 1966). Our lamellae lying free in the cytoplasm, in or near

FIGURE 36 A narrow stalk of endoplasmic reticulum is in apposition to a desmosolne at the intercellular junctions between the follicle wall (top) and the oocyte (bottom). H 35-3; 2% glutaraldehyde. X 21,000.

FIGURE 37 Endoplasmic reticulum is frequently associated with the compound aggregates in the peripheral cytoplasm of the oocyte. H 35-3; 2% gluta.raldehyde. X 22,000.

FIGURE 38 Vacuoles containing multiple small vesicles (center) and often surrounded by a halo of similar vesicles (left) are present in the peripheral cytoplasm of the oocyte. H 44-3; Karnovsky fixative. X 15,000.

FIGURE 39 Structures resembling myelin figures are separately contained within vacuoles (left) and are not found within the compound aggregates (_rigl1t). Note that a vesicle of endoplasmic reticulum containing electr0n—opa.que material is adjacent to the edge of a ballooned aggregate (lower right). H 444; Karnovsky fixative. X 22,000.


FIGURE 40

Balbiani’s vitelline body, shows that they are repeatedly formed at the nuclear membrane and are subsequently released into the cytoplasm where they apparently fragment. The suggested role for their participation in nucleo~cytoplasmic transfer (Swift, 1956) was based on the resemblance of the membranes of annulate lamellae in the cytoplasm to the ultrastructure of the nuclear membrane. When occasional pores of the nuclear membranes of these human oocytes have been observed in register with those in an attached stack of annulate lamellae, the dense, amorphous substance contained within and between the aligned pores appears to form a continuous band from the inner nuclear membrane toward the base of the stack. Whether this represents transfer of material from the nucleus to either the cytoplasmic matrix or membranes is unknown. When annulate lamellae are attached to the nuclear membrane, it is the presence of their pores and their greatly increased membrane area that distinguish them


A cone~shaped configuration of cortical cells is frequently seen at the immediate periphery of the follicle wall. The tip of the cone lies between two arrows. A projection (P) of the cortical cell at the base of the cone illustrated here appears to be becoming incorporated into the follicle wall. H 4.«4~1; Karnovsky fixative. X 5400.

from the membranes of the slightly granular endoplasmic reticulum that in the same oocyte also may be continuous with the outer leaflet of the nuclear membrane at intervals around its entire circumference.

The varying location of annulate lamellae within Balbiani’s vitelline body became of interest to us since, for a considerable period of our study of primordial follicles, the only oocytes with annulate lamellae attached to the nuclear membrane were three (Adams and Hertig, 1965) from two patients both in the follicular or estrogenic phase of the menstrual cycle, the endometria showing histologically a typical mid-proliferative pattern. In this preliminary series other oocytes from the luteal and menstrual stages of the cycle showed both detached annulate lamellae and poorer development of the peripheral endoplasmic reticulum. Histologically these endometria showed 19-20 day secretory criteria and those of early menstruation respectively. There have been many previous suggestions that the cyclic production of estrogen and /or progesterone in the ovary may have a local influence on a variety of ovarian structures and our initial finding suggested to us that a fluctuating participation by the oocyte nucleus in the cytoplasmic arrangement of organelles might be evidence of such a response to cyclic endocrine events. A hormonal effect on the prevalence and position of annulate lamellae has been noted in human endometrium by Ancla and De Brux (1965). In a series of 40 patients subjected to various forms of hormonal therapy for primary infertility, they reported that in the glandular cells “annulate lamellae may appear before ovulation, but are more frequently seen during the secretory phase if there is hyperestrogenism.” After studying intracellular localization of estrogen in so—called target cells, such as rat uterus, as opposed to cells of the body in general, Segal (1967) has recently formulated concepts of estrogen action which may involve an action at the nuclear level. The difference we noted between oocytes could be

easily explained either by our random sampling of the entire population of primordial follicles in any one ovary or by our failure to locate the annulate lamellae in some oocytes examined at only a few levels in their cytoplasm. Serial grids, although arduous to study, are required even to locate the annulate lamellae and a single fortunate section within a series may produce the only evidence of their attachment to the nuclear membrane. The results of a closer inspection of oocytes from the proliferative, secretory, and menstrual phases of the cycle and from pregnancy near term are shown in Table I.

This is admittedly too small a series for any statistical validity, but in these 27 oocytes studied with great care, there appears to be some evidence that annulate lamellae attached or adjacent to the nuclear membrane are more prevalent when estrogen production is dominant in the normal menstrual cycle. Whether this in fact reflects any environmental hormonal influence will require further detailed study of oocytes in primordial

FIGURE 41 In a few follicles, a peripheral cell may overlap its neighbors but be separated from them by a homogeneous, extracellular material similar to that in the basal lamina at the periphery of the follicle wall. H 43-1; Karnovsky fixative, X 10,000.

TABLE I

The Location of Amzulate Lamellae with Respect to the Nuclear in Phase: of the rlzfenstruai Cycle

Position of the annulate

Case ' . gicéejytifis lamellar s‘t)1(')\(1:ctture in each N0_ Phase of cycle by endometrial dating Cycle day C105 eh, Seriated y e_ ______ sections At nuclear Free in membrane cytoplasm H6* Mid—proliferative l 1 attached 0 H35 Mid-proliferative 4 4 attached (J H5411 Mid-proliferative l 1 adjacent 0 H48 Secretory (19~20th day) 23§ Rt ovary 0 3 (CL side) 3 H48 Secretory (l9—20th day) 23§ Left ovary 1 adjacent 2 3 H37 Menstrual 8 2 attached 6 H44 Menstrual 5 1 adjacent 4 H43 Pregnant, caesarian section 35% wk 2 1 attached 1

  • Infertility patient; wedge resection of normal ovarian tissue.

ii This patient had been on a course of antifertility medication for 4 yr, interrupted by a pregnancy. She had been on same drug, Orthonovum, for 5 months previous to surgery including the 6th and 7th days of the cycle in which surgery was performed.

§ The endometrial dating may or may not coincide with day of cycle owing to normal variation in the length of the menstrual cycle in the human female.

follicles from many more ovaries obtained not only from various cyclic hormonal phases in the adult but also from prepubertal age groups.

The compound aggregates that surround the cytocentrum of these human oocytes in primordial follicles are subject to at least two interpretations: (a) that they are reservoirs of nutritive material analagous to yolk material in oocytes of other species, or (b) that they represent waste deposits of degradation Of organelles similar to lipofuscin pigment commonly found in aging cells. Some of the ultrastructural characteristics of the aggregates, finely divided membranous material and vacuoles, resemble those found pigment in neurons, another nondividing cell (Samarajski et al., 1965). The absence of compound aggregates in oocytes in primordial follicles of the human prenatal ovaries reported by Lanzavecchia and Mangioni (1964) and Stegner and Wartenberg (1963) suggests that the examination of similar follicles in various postnatal and preadolescent years might be of value in evaluating the effects of aging on the oocyte. The lack of demonstrable acid-phosphatase activity, however, indicates that the aggregates may not be lysosomal and, therefore, may not represent autophagic processes by which lipofuscin pigment commonly

in lipofuscin


is localized. Furthermore, myelin figures, usually indicative of membrane degradation, have not been observed among their heterogeneous components but are frequently seen to be separately contained within a vacuole in the peripheral cytoplasm. Thus, it is possible that compound aggregates represent nutritive material. Dense aggregates were not observed in our previous study of guinea pig oocytes (Adams and Hertig, 1964) but have been noted in the cytoplasm of some other immature mammalian oocytes (rabbit: Blanchette, 1961; hamster: Weakley, 1966, and Odor, 1965; rhesus monkey: Hope, 1965). Those observed in the rhesus oocytes have some ultrastructural similarities to the compound aggregates we see in the human oocytes, and it was suggested that they represent developing yolk. The presence of membranes around some aggregates in these human oocytes and the absence of a membrane around others within the same oocyte suggests that not all compound aggregates are in the same physiological state. Those that are membrane enclosed often contain dispersed vesicles of varying size which contain electron—opaque material. Whether this material is related to that seen within the endoplasmic reticulum has not been determined. The close morphological association of endoplasmic reticulum to these aggregates,

however, may be related to the inverse reciprocal effects on these two organelles of variations in

osmolality of the fixatives used. The presence of

some ballooned aggregates, when the more peripheral endoplasmic reticulum is constricted, and the absence of gross ballooning, when the endoplasmic reticulum is dilated, suggests a possible pump mechanism between these two organelles. Merriam (1966) has presented evidence that frog oocytes consist of two osmotically active “compartments.” One, the protoplasmic mass, owes its osmotic properties to the plasma membrane, and the other consisting of interior vesicles owes its osmotic properties to their investing membranes. Because of the magnitudes of experimental swelling in the interior vesicles, he suggests that ion-pumping takes place in the vesicles of the cytoplasm of the frog oocyte.

Whatever the nature of the aggregates seen in a paranuclear location within the oocyte, the presence in the periphery of the oocyte of a variety of vacuoles containing elements of the aggregates together with the presence of compound aggregates in most follicle cells and in some cortical cells suggests their transport to the oocyte from its environment or vice versa. The mechanism for their transfer to the follicle cell from the cytoplasm of the oocyte, or vice versa, is not clear from these studies since in micrographs where a follicular aggregate bulges into the oocyte, complete intervening cell membranes have not been clearly visualized. The repeated observation of dense membranous material which appears to bridge the oocyte-follicle cell membranes may indicate that there is some mechanism for bulk transfer of the elements of the compound aggregates. Such membranous forms frequently are closely associated either in the follicle cell with a compound aggregate or in the oocyte cytoplasm with vacuoles containing elements of the aggregates. The presence in the peripheral oocyte cytoplasm of elements of aggregates within a vacuolar structure suggests that portions of follicular aggregates may be taken up by a phagocytic mechanism at the oocyte surface to be subsequently transported toward the paranuclear location.

The presence of similar compound aggregates in some cortical cells may be evidence that these cells are related to those within the follicle wall. A histochemical similarity between the cortical cells and those of the primitive follicle wall was observed with respect to adenosine—5-phosphatase activity. There is an analogous histochemical similarity in guinea pig ovaries, although the differentiated granulosa cells of developing follicles

do not maintain this activity in the human as they do in the guinea pig (Adams et al., 1966). The absence of observed mitotic activity in primitive follicle cells could indicate either that they are as old as the contained oocyte or that they are renewed by some process other than mitosis. In a study of guinea pig follicles in a similar stage (Adams and Hertig, 1964) there was suggestive histological evidence that cortical cells contribute to the follicle wall. In the human primordial follicle, it is possible that cortical cells may move into the thickened, crescentic cap of the follicle wall and gradually slide around the oocyte to move out to the cortex again. Although never clearly visualized in the actual act of becoming incorporated into the follicle wall, occasional follicle Cells overlap the periphery of their neighbors and are partially separated from them by a homogeneous substance morphologically similar to the material of the basal lamina. In addition, many dark peripheral cells in the follicle wall closely resemble cells of the cortex. The coneshaped arrangement of cortical cells noted at the periphery of some primordial follicles has suggested that these may be the areas where cortical cells become incorporated into the follicle wall. Further histochemical study of the nature of the material within the compound aggregates is indicated. Such studies might well lead to a clarification of the relationship of such aggregates within the oocyte to those of the various cells of the surrounding ovarian tissue.

Although light microscopists have over the years identified structures in oocytes as centrioles (Raven, 1961), others have believed that the centriolar apparatus for postfertilization cleavage is introduced into the egg by the spermatozoan (see Mazia, 1961 for discussion). The absence of identifiable centrioles in most electron microscope studies of a variety of oocytes is not surprising in view of the necessarily random sampling of such a large cell. We know of only two reports of the clear visualization of centrioles. Sotelo (1959) illustrated a pair of centrioles in an oocyte from a 2—day—old rat, and Greenfield (1966) has demonstrated centrioles in oocytes of the domestic chicken shortly after hatching. It may be of importance that these oocytes in which centrioles have been identified were from ovaries of the immediate postnatal stage when mitotic activity of oogenesis had recently ceased. Even after specific search in all of our human oocytes, centrioles could not be identified. Recently in a report of spindle—associated chromosomes of meiosis in the rat, Zamboni and Mastroianni (1966) reported that centrioles are not observed. Whether centrioles in oocytes disappear with aging or simply are not found is still unknown. In the human oocyte, at least, it is possible that the one or two areas of dense amorphous deposits found in the center of the cytocentrum, as well as the long coarse fibers supporting the periphery of the cytocentrum, may be adaptations of the centriolar apparatus. Electron microscopic investigations of human oocytes from prenatal and various postnatal stages might clarify this concept. Vesicular aggregates either adjacent to or inter mingled with the central aggregate of dense

BIBLIOGRAPHY

ADAMS, E. C. and A. T. HERTIG. 1964. Studies on guinea pig oocytes. I. Electron microscopic observations on the development of cytoplasmic organelles in oocytes in primordial and primary follicles. J. Cell Biol. 21:397-427.

ADAMS, E. C., and A. T. HERTIG. 1965. Annulate laniellae in human oocytes in primordial and primary follicles. J. Cell Biol. 2T:119A.

ADAMS, E. C., A. T. HERTIG, and S. FosTER. 1966. Studies on guinea pig oocytes. II. Histochemical observations on some phosphatases and lipid in

developing and in atretic oocytes and follicles.

Am. J. Anal. ]_l9:303—3-40.

ANCLA, M. and DE BRUX. 1965. Occurrence of intranuclear tubular structures in the endometrium during the secretory phase, and of annulate lamellae in hyperestrogenic Obslet. Gynecol. 26:23+33.

ANDERSON, E. and H. V-V. BEAMS. 1960. Cytological observations on the fine structure of the guinea pig

human

states.

ovary with special reference to the oogonium, primary oocyte, and associated follicle cells. J. Ultra-struet. Res. 3;432—446.

AYKROYD, O. E. 1938. The cytoplasmic inclusions in the oogenesis of man. Z. Zellforsefz. 11/Iz'/rroskop. Andi. 27:691-710.

BAKER, T. G. 1963. A quantitative and cytological study of germ cells in human ovaries. Prue. Roy.

Soc. (London) Ser. B. 158:-‘l17—-=1-33.


amorphous deposits were a consistent finding.

This may indicate that the material within these vesicles is related either to the formation of the deposits or to their periodic alignment on "line fibrils which apparently becomes a transition to the peripheral coarse fibers.

This work was supported entirely by grant HD-00137, Division of Human Development and Child Health, formerly C-2451 of the National Institutes of Health.

The authors are greatly indebted to Doctors Charles Easterday, Shirley Driscoll, Christopher Duncan, Sumner Gochburg, Charles Horne, Lorna Johnson, and Alan Rosenfield of the Boston Hospital for Women for their cooperation in providing us with these human ovaries. In addition, grateful thanks are due to Miss Harriet Jopson and Mrs. Barbara Barton for their skilled technical assistance, to Mrs. Audrey Hadfield for the preparation of photomicrographs and electron micrographs, and to Mrs. Nancy Cote for typing the manuscript.

Received for pubIz'eatz'on 6 February 1967.

BALBIANI, E. G. 1864. Sur la constitution du germe dans 1’oeuf animal avant la fécondation. Compt. Renal. 58 584-588.

BEAMS, H. W. and J. F. SHEEHAN. 1941. The yolk nucleus complex of the human ovum. Amzt. Record. 81:5-415F554.

BECKER, N. H., S. GOLDFISHER, VV. SHIN, and A. B. NOVIKOFF. 1960. The localization of enzyme activities in the rat brain. J. Bz'o}JIzys‘. Bioefzem. Cytol. 8:649—663.

BLANCHETTE, E. J. 1961. A study of the fine structure of the rabbit primary oocyte. J. Ultrastruct. Res. 5:349-363.

FAHIMI, H. D., and M. J. KARNOVSKY. 1966. Cytochemical localization of two glycolytic dehydrogenases in white skeletal muscle.J. Cell Biol. 29:1 13128.

GREENFIELD, M. L. 1966. The oocyte of the domestic chicken shortly after hatching, studied by electron microscopy. J. Embryoi. Exptf. Moryb/20!. l5:297— 316.

Hope, J. 1965. The fine structure of the developing follicle of the Rhesus ovary. J. Ultrastruet. Res. 12:592-610.

KARNOVSKY, M. 1961. The use of ethylene diamine tetraacetic acid disodium in the histochemical demonstration of triphosphopyridine nucleotide linked dehydrogenases. J. Histoehem. Cytoehem. 9:203—20-<1.

KARNOVSKY, M. J. 1965. A formaldehyde-glutaraL dehyde fixative of high osmolality for use in electron microscopy. J. Cell Biol. 2T:137A—138A.

KESSEL, R. G. 1963. Electron microscopic studies on the origin of annulate lamellae in oocytes of Necturus. J. Cell Biol. 19:391-414.

KESSEL, R. G. 1964. Electron microscopic studies on oocytes of an echinoderm, Tlzyone briareus, with special reference to the origin and structure of the annulate lamellae. J. Ultrastruct. Res. 10:498514.

KESSEL, R. G. 1966. Some observations on the ultrastructure of the oocyte of Thyone briareus with special reference to the relationship of the Golgi complex and endoplasmic reticulum in the formation of yolk. J. Ultrastruct. Res. 16:305-319.

LANZAVECCHIA, G., and C. MANGIONI. 1964. Etude de la structure et des constituants du follicle humain dans l’ovaire foetal I Le follicle primordial. J. M icroscop. 3 :447—464.

LEHNINGER, A. L. 1965. The mitochondrion. Molecular Basis of Structure and Function. W. A. Beniamin, Inc., New York.

MAZIA, D. 1961. Mitosis and the physiology of cell division. In The Cell. J. Brachet and A. E. Mirsky, editors. Academic Press Inc., New York. 3.

MERRIAM, R. W. 1966. The role of cytoplasmic membranes in the regulation of water concentration within frog oocytes. Exptl. Cell Res. 41:34-48.

MYERS, H. I., W. C. YOUNG, and E. W. DEMPSEY. 1936. Graafian follicle development throughout the reproductive cycle in the guinea pig, with especial reference to changes during estrus (sexual receptivity). Anal. Record. 65:381-401.

ODOR, D. L. 1965. The ultrastructure of unilaminar follicles of the hamster ovary. Am. J. Anal. 116:493— 522.

PEARSE, A. G. E. 1960. Histochemistry, Theoretical and Applied. Little, Brown and Company, Boston.

RAVEN, C. P. 1961. Oogenesis: The Storage of Developmental Information Pergamon Press, New York,

SAMARAJSKI, T., J. M. ORDY, and J. R. KFEFE. 1965. The fine structure of lipofuscin age pigment in

ARTHUR T. HERTIG AND ELEANOR C. ADAMS Human Oocyte and Its Follicle

the nervous system of aged mice. J. Cell Biol. 26:779-795.

SEGAL, S. 1967. Fetal homeostasis. 11. Arm. N. Y. Acad. Sci. In press.

SOTELO, J. R. 1959. An electron microscopic study on the cytoplasmic and nuclear components of rat primary oocytes. Z. Zellforselz. ;'1-filcroskop. Anal. 50: 749-765.

STEGNER, H. E., and H. H. WARTENBERG. 1963. Elektronenmikroskopische Untersuchungen an Eizellen des Menchen in verschiedenen Stadien der Oogenese. Arclz. Gynaekol. 1992151-172.

STRICHT, O. VAN DER. 1923. Etude comparée des ovules des mammiferes aux différentes périodes de Povogenese, d’aprés les travaux du laboratoire d’histologie et d’embryologie de l’Université de Gand. Arch. Biol. 33: 229-300.

SWIFT, H. 1956. The fine structure of annulate larnellae. J. Bioplzys. Bioclzem. Cytol. 2 (Suppl.): 415-418.

TARDINI, A., L. VITAL:-MAZZA, and F. E. MANSANI. 1960. Ultrastruttura de1l’ovocita umano maturo. 1. Rapporti fra cellule della corona radiata, pellucida ed ovoplasma. Arch “ale l/ecclzz” Anal‘. Pallzol. .-Med. Clin. 33:281-305.

VENABLE, J. H., and R. COGGESHALL. 1965. A simplified lead citrate stain for use in electron microscopy. J. Cell Biol. 25:407-408.

VVACHSTEIN, M., and E. MEISEL. 1957. Histochemistry of hepatic phosphatases at a physiologic pH. Am. J. Clin. Patlzol. 27:13-23.

W'ARTENBERo, H., and H. E. STEGNER. 1960. Uber die elektronenmikroskopische feinstruktur des menschlichen ovarialeies. Z. Zellforrclz. Mz'kroskop. Anal. 52:450-474.

WEAKLEY, B. S. 1966. Electron microscopy of the oocyte and granulosa cells in the developing ovarian follicles of the golden hamster (Mes0crz'cetus aumtus). J. Anal. l00:503-534.

ZAMBONI, L., and L. MASTROIANNI, JR. 1966. Electron microscopic studies on rabbit ova. I. The follicular oocyte. J. Ultmsmcet. Res. 14:95-117.

ZAMBONI, L., D. R. MISHELL, JR., J. H. BELL, and M. BACA. 1966. Fine structure of the human ovum in the pronuclear stage. J. Cell Biol. 30:579-600.