-
Proc. Natl. Acad. Sci. USAVol. 75, No. 11, pp. 5534-5538,
November 1978Cell Biology
Long-term growth and differentiation of Xenopus oocytes in
adefined medium
(oocyte growth/oocyte maturation/insulin/vitellogenin)
ROBIN A. WALLACE AND ZIVA MISULOVINBiology Division, Oak Ridge
National Laboratory, Oak Ridge, Tennessee 37830
Communicated by W. L. Russell, August 28,1978
ABSTRACT Xenopus laevis oocytes over a size range of0.15-0.78
mm3 were dissected from their follicles and culturedin a defined
medium for up to 28 days. Oocytes grew at averagerates of 0.021
mm3-day-l in the absence of insulin and 0.030mm3 day-l in the
presence of insulin. The latter average growthrate corresponds to
the fastest growth rate reported to date foroocytes in vivo.
Oocytes grown in vitro can reach a size of atleast 1.43 mm3, which
is larger than the maximum size generallyfound in vivo. During
growth in vitro, oocytes also acquire botha normal pigment pattern
and, once they reach about 0.7 mm3,the ability to undergo complete
maturation as a response toexternally applied progesterone. These
results show that Xen-opus oocytes freed of their follicular
investments are able togrow and differentiate in vitro.
The growing oocyte from any animal has traditionally been
themost difficult cell type to sustain in a physiologically
normalcondition both in vivo and in vitro. In nonmammalian
verte-brates, experimental or environmental perturbation of
femalescarrying vitellogenic oocytes has frequently led to atresia
of theoocytes (1). Likewise, early attempts to develop organ
cultureprocedures for amphibian ovaries, although successful for
mostcell types present in the ovary, failed to maintain
yolk-con-taining oocytes, which were the first and sometimes only
celltype to degenerate in vitro (2, 3).More recently, several
researchers have at least been able to
maintain Xenopus laevis oocytes in vitro for extended
periods.Stage IV/V [all oocyte stages are those described by
Dumont(4)] and Stage VI oocytes were cultured either within
theirfollicles (5) or after removal of investing follicular layers
(6, 7);several physiological criteria indicated that the oocytes
remainrelatively healthy for 2-4 weeks. However, no growth was
re-ported for such oocytes.Numerous studies have established that
oocytes of non-
mammalian vertebrates grow primarily by sequestering
theyolk-protein precursor, vitellogenin, from the
maternalbloodstream (8). Normally, vitellogenin enters the
capillarynetwork within the theca of the follicle and from there
hasaccess to the oocyte (9, 10); it cannot, however, penetrate
theouter surface epithelium of the follicle (11). These
consider-ations have led us to conclude that X. laevis oocytes can
begrown in vitro only when they are removed from their
follicularinvestments and are placed in a nutritionally adequate
andosmotically appropriate medium containing vitellogenin. Wehave
therefore developed a culture procedure for growing suchoocytes
based on this premise (12). We report here on thegrowth and
differentiation of X. laetvs oocytes individuallycultured in the
defined medium we have developed.
MATERIALS AND METHODSOocyte Culture. Females stimulated with
human chorionic
gonadotropin (hCG) were given 1000 international units (IU)of
hCG 2-3 days prior to laparotomy; unstimulated femalesreceived no
hCG. A part of the ovary was removed from eachfemale and placed in
solution O-R2(13). Qocytes within anarrow size range (10.05-mm
diameter) were manually dis-sected from their follicles and
individually placed in 50-100,gl of medium containing 50% Liebovitz
L-15 medium, 5 mgof vitellogenin per ml, 1 mM L-glutamine, 15 mM
Hepes-NaOH buffer, and the antibiotics gentamycin (100 ,ug-ml-1)and
nystatin (50 units ml-1). The final pH of the medium was7.8.
Crystalline porcine insulin (1 Ag-ml-m; 25 units -mg-1;
lot615-063-10 kindly provided by Ronald Chance, Eli Lilly Re-search
Laboratories) was also included unless specified other-wise.
Oocytes were then cultured in a dark, humidifiedchamber at 20'C.
Complete details including the isolation ofvitellogenin are
specified elsewhere (12).Other Procedures. Oocyte diameters were
measured with
an ocular micrometer in at least two directions (at 900) in
casethey were not strictly spherical. Very large oocytes tended
toflatten somewhat; in such cases three diameters were measuredby
appropriate propping and averaged. The oocyte volumesindicated
throughout are based on these diameter measure-ments.
[3H]Vitellogenin incorporation was measured as de-scribed elsewhere
(12). We determined protein by individuallydissolving oocytes in a
sodium dodecyl sulfate/diothioerythritolsolution* and placing the
protein solution (plus washings) onindividual 2.3-cm-diameter discs
punched out from Whatman42 paper. Discs were processed with
ice-cold 10% trichloroaceticacid, alcohol/ether (3:1), and ether
(14), and the processed discswere used for protein determination
(15). For a protein stan-dard, yolk platelets were isolated (16)
from X. laevis oocytes,washed in distilled H20, and extracted with
alcohol/ether (3:1)and ether. The dried powder was dissolved in
sodium dodecylsulfate/dithioerythritol, placed on discs, and
processed as above.Responsivity to progesterone was assessed by
incubating oocytesfor approximately 20 hr in media containing 1 jug
of proges-terone per ml. Ooctyes were then punctured with a fine
probeat the animal pole and gently squeezed at the equator with
aforceps; extrusion of a germinal vesicle from the wound indi-cated
that germinal vesicle breakdown (GVBD) had not oc-curred. This
procedure proved more reliable than scoring thetransient appearance
of a large "white spot" (17). Progester-one-treated oocytes were
also placed in Smith fixative (18),embedded in paraffin, sectioned
at 5 my, and stained withMayer haemalum for light microscopic
examination.
Abbreviations: hCG, human chorionic gonadotropin; GVBD,
germinalvesicle breakdown.* Wallace, R. A. & Hollinger, T. G.
(1978) Exp. Cell Res., in press.
5534
The publication costs of this article were defrayed in part by
pagecharge payment. This article must therefore be hereby marked
"ad-vertisement" in accordance with 18 U. S. C. §1734 solely to
indicatethis fact.
Dow
nloa
ded
by g
uest
on
Apr
il 2,
202
1
-
Proc. Nati. Acad. Sci. USA 75 (1978) 5535
RESULTSOocyte Growth. Oocytes ranging in volume from 0.15 to
0.78 mm3 (diameter = 0.66-1.14 mm) were isolated from a
totalnine different unstimulated or hCG-stimulated females and
placed in culture both with and without insulin. Their
increasein size was then measured over 14-28 days (Table 1).
Completedata for eight different groups are provided in Fig. 1. The
useof oocyte size as an indication of oocyte growth was validatedby
measuring the relationship between oocyte size and proteincontent
both in Vivo (freshly dissected oocytes) and after culturein vitro
for at least 5 days. These two sets of data were
essentiallysuperimposable (Fig. 2; A protein content - A mm-3 = 326
and329,ug-mm-3, respectively).
In the absence of insulin, oocytes grew at an average rate
of0.021 + 0.005 mm3-day-'; in the presence of insulin the ob-served
growth rate for all oocytes was 0.030 + 0.007 mm3.day-1(Table 1).
When oocytes of similar size from six different fe-males were
divided into two groups and cultured with orwithout insulin, those
cultured in the presence of insulin werefound to grow at a rate
that was 136 + 22% that of controls. Nosignificant differences in
growth rates were found betweenoocytes isolated from unstimulated
females and those fromhCG-stimulated females.The data provided in
Fig. 1 indicate that the growth of oo-
cytes in vitro is progressive with time and that oocytes withina
size range of 0. 15-0.78 mm3 grow at about the same rate invitro.
This was explored at a greater level of resolution bymeasuring the
rate of [3H]vitellogenin incorporation for threedifferent groups of
oocytes during culture with insulin: 0.24-mm3 oocytes from frog 17,
0.48-mm3 oocytes from frog 21, and0.78-mm3 oocytes from frog 23.
The first two groups of oocytesgrew at an average rate of 0.024
mm3.day-I and the third, ata rate of 0.038 mm3-day-I (Table 1).
This range in growth ratewas expected, because oocytes were
obtained from "wild-type"rather than genetically similar animals.
The two growth ratescorrespond to average protein increments of 329
and 521 ng-hr-1, respectively, based on the data provided in Fig.
2. Theseoverall rates of protein increment correlate well with the
av-erage rates of vitellogenin incorporation observed for the
threegroups of oocytes during their growth in vitro (Fig. 3). The
dataprovided in Fig. 3 indicate further that a maximal rate
of[3H]vitellogenin incorporation appears to occur among oocytes
7u
0
0
E
E
E
03
C-
0
0
E
E
a)
.OE0
0 10 20 30Time, days in culture
FIG. 1. Growth of X. laevis oocytes with time in vitro.
Examplesare provided of eight groups cultured either in the
presence (closedsymbol) or absence (open symbol) of 1 ,ggml-l
insulin. Each pointrepresents the average volume ISD for 8-10
oocytes. At various times,oocytes were also placed for --20 hr in 1
,g of progesterone per ml; thenumbers in parentheses indicate the
fraction which underwentGVBDas a response to progesterone
treatment. Frog: 17, circle; 18, diamond;21, square; 22, reverse
triangle; 23, triangle.
with a volume of 0.6-0.8 mm3 (diameter = approx. 1.05-1.15mm),
which is similar to what has been observed in vivo (un-published
observations).Oocyte Differentation. The smallest oocytes placed in
cul-
ture (0.15 mm3 from fkog 17) were barely pigmented at
thebeginning of the culture period. With time, they became
fullypigmented over their entire surface, and subsequently
thepigment migrated to the animal half of the oocyte, as occurs
Table 1. Average growth rates for various sizes of ooctyes in
vitro*
Without insulin With insulinFrog Days Vo Vf AV-day-' Vo Vf
AV-day-1 AV with insulinno. cultured (mm3) (mm3) (mm3) (mm3) (mm3)
(mm3) AV without insulin
Oocytes obtained from unstimulated females16 14 0.47 0.81 0.024
0.48 1.01 0.038 1.5819 14 0.48 0.84 0.026 0.43 0.86 0.031 1.1924 15
- 0.28 0.70 0.028Average 0.025 I 0.001 0.032 + 0.005
Oocytes obtained from hCG-stimulated females17 27 - 0.15 0.73
0.021t17 28 0.24- 0.91 0.024t17 14 0.44 0.81 0.026 0.43 0.89 0.033
1.2718 24 0.56 1.43 0.036t20 18 0.56 0.86 0.017 0.56 1.01 0.020
1.1821 22 0.48 0.78 0.014t 0.48 1.00 0.024t 1.7122 25 0.33 0.82
0.020t 0.33 0.95 0.025t 1.2523 17 0.78 1.43 0.038tAverage 0.019 ±
0.005 0.028 ± 0.007± indicates SD.
* Vo = volume after 24 hr of culture; Vf = volume on day of
termination.t Complete data plotted in Fig. 1.
Cell Biology: Wallace and Misulovin
Dow
nloa
ded
by g
uest
on
Apr
il 2,
202
1
-
5536 Cell Biology: Wallace and Misulovin
0.5
400-
Ii- 300-u00
.6-c
0 200-10
100-
0
0
Diameter, mm oocyte-1.0 1.1 1.2 1.3
I1.4
0.5 1.0Volume, mm3. oocyte1
FIG. 2. The relationship between protein content and oocyte
sizein vivo (@) and in vitro (0). Each point represents the average
of 8-10oocytes; five females were used as oocyte donors in each
case. Forprotein content in vivo, groups of oocytes were dissected
from theovary and immediately used for determinations; a size range
wasmeasured extending from a diameter of 0.45 mm [the smallest
oocytesundergoing pinocytosis and hence yolk deposition (4);
smaller oocytespresumably grow by some other mechanism] up to a
diameter of 1.28mm (the largest oocytes found). For protein content
in vitro, mea-surements were made on groups of oocytes cultured for
5-28 days. Thesolid and dashed lines represent the least-squares
fit to the data foroocytes in vivo and in vitro, respectively
(slopes = 326 and 329 Mlg-mm-3; Y-intercepts = -23 and -31
ng-oocyte-1).
in vivo (4). After 28 days they had reached an average size
of0.73 mm3 (Table 1), and about half underwent GVBD in re-sponse to
added progesterone (Fig. 1). Larger oocytes placedin culture were
already pigmented. The subsequent migrationand appearance of
pigment seemed normal in every respect,and most oocytes also
eventually underwent GVBD in responseto added progesterone.The
response to progesterone was explored in greater detail
because it could be more readily quantitated than the processof
pigmentation. In general, we found that oocytes with a vol-ume less
than 0.6 mm3 (diameter = 1.05 mm) did not undergoGVBD in the
presence of progesterone when initially placedin culture, whereas
the largest oocytes examined (diameter =1.14 mm) did (Fig. 1). All
groups of oocytes eventually becameresponsive to progesterone once
they reached a volume of about0.7 mm3 (diameter = 1.1 mm) (Fig. 1).
The fact that this ap-peared to be the case for oocytes cultured
either with or withoutinsulin (Fig. 1) indicates that size rather
than the presence ofinsulin was the critical determinant.
Progesterone-treated oo-cytes that appeared to have undergone GVBD
were also fixedand prepared for light microscopy. An examination of
serialsections revealed in every case the presence of a
metaphase
v 800 -u00
- 600-
caC0
2 400-0
Cc
v) 20C -CP0
0
0.5
0
Diameter, mm * oocyte11.0 1.1 1.2 1.3
I1.4
I
0.5 1.0 1.5Volume, mm3 - oocyte-1
FIG. 3. [3HJVitellogenin incorporation as a function of oocyte
sizefor three groups of oocytes grown in vitro. Cultured oocytes
from frogs17 (0), 21 (o), and 23 (&) were periodically measured
(see Fig. 1) and[3H]vitellogenin incorporation was simultaneously
determined. Eachpoint represents the average [3Hjvitellogenin
incorporation i SD for8-10 oocytes. The average protein increment
for oocytes from frogs17 and 21 was 329 ng-hr-1 and for those from
frog 23 was 521 ng-hr-1as derived from the rate of growth
calculated in Table 1 (0.024 and0.038 mm3.day-1, respectively) and
the relationship between volumeand protein content provided in Fig.
2 for oocytes cultured in vitro(329 ,g-mm-3); these values are
indicated by the horizontal dottedlines.
spindle together with a polar body, several examples of whichare
provided in Fig. 4. As also observed in Fig. 4, only an oc-casional
follicle cell was attached to the vitelline membrane.
DISCUSSIONThe volume increments we have measured in vitro (Table
1,Fig. 1) appear to be true reflections of oocyte growth rather
thanhydration, because (a) oocyte volume and protein content
haveessentially the same relationship in vivo and in vitro (Fig.
2)and (b) observed growth rates correlate well with overall
vi-tellogenin incorporation (Fig. 3). The latter process is
primarilyresponsible for oocyte growth because growing oocytes
incor-porate per hour at least eight times more vitellogenin than
thetotal protein that they synthesize (19). This discrepancy
isfurther enhanced by the turnover of endogenously
synthesizedprotein, whereas sequestered vitellogenin does not
undergoturnover (unpublished observations; ref. 19). The addition
ofvitellogenin to our culture medium is thus the most likely
reasonwe have been able to obtain oocyte growth in vitro, in
contrastto previous experiences with long-term culture of
defolliculatedoocytes (6, 7).
In the intact, unstimulated X. laevs female, oocytes havebeen
estimated to increase from 0.4 to 0.8 mm in diameter in4-8 months
(20), which corresponds to an average volume in-crement of
0.001-0.002 mm3-day'1. More recently, growthrates of oocytes in
vivo ranging from 0.6 to 1.2 mm diameterwere measured, and volume
increments of 0.004-0.014mm3.day-1 for oocytes from unstimulated
females and0.011-0.027 mm3-dayf' for those from hCG-stimulated
femaleswere found (unpublished data). In a more extreme case,
Scheer(21) removed most of an ovary from each of several
females,
0,w'O,#
/,0,/o°
*ib°
/~~~~~~~~~~~~~~~~~~~0~~~~~~~~~~~~~~~~~~~
/~~~~~~~~~~~~~~~~~~~~0~~~~~~~~~~~~~~~~~~~~~
of
*/'
,,0~~
/E
ng hr-
Proc. Natl. Acad. Sci. USA 75 (1978)
Dow
nloa
ded
by g
uest
on
Apr
il 2,
202
1
-
Proc. Natl. Acad. Sci. USA 75 (1978) 5537
S f p
FIG. 4. Light micrographs of eggs derived from oocytes grown
invitro. Oocytes with an average volume of 0.56 mm3 (unresponsive
toprogesterone) from frog 20 were grown for 14 days in vitro
(averagevolume = 0.90 mm3) and then incubated overnight in medium
con-taining 1 jig of progesterone per ml. Progesterone treatment
convertedthe oocytes into eggs as revealed by the simultaneous
presence of asecond meiotic metaphase spindle and a first polar
body. (A-D) Fourexamples of sections in which portions of the
second meiotic meta-phase spindle and first polar body were found
in the same section; (Eand F) two examples in which the spindle and
polar body were foundin separate sections. s, Second meiotic
metaphase spindle; p, polarbody; f, follicle cell. (X400.)
leaving behind only "small oocytes adhering to the rest of
themesovarium"; the females were then stimulated with hCG, andthe
oocyte diameters were periodically measured. We havetaken Scheer's
data for the steepest portion of his growth curve(diameter increase
from 0.48 to 1.10 mm), converted the di-ameter measurements to
volume measurements, and drawn,by least-squares analysis, a line to
fit the points; the slope of thisline was 0.032 mm3-day-'. The
average growth rates we haveobserved in vitro in the presence of
insulin (0.028-0.032mm3.day-l, Table 1) correspond to this rate
observed for oo-cytes in vivo. Note that our culture medium
contains vitello-genin at a concentration (5.0 mg-ml-l) that
essentially saturatessequestration by the oocyte [Km = 0.7 mg-ml-l
(22)].The largest oocytes in laboratory-maintained X. Levis
have
been reported to have a diameter between 1.3 and 1.4mm (23,24);
in most females, the largest oocytes are somewhat smaller
(4). On two occasions (Table 1, frogs 18 and 23) we found
thatoocytes had grown to a diameter of 1.4 mm (volume = 1.43mm3) by
the time the culture was terminated. Growth curvesfor these two
groups of oocytes did not indicate that a sizeplateau had been
reached by this time (Fig. 1). The proteincontent of one group of
these 1.4-mm oocytes was measuredand found to correlate with size
in a manner similar to otheroocyte groups (Fig. 2, upper right-hand
point). Thus, oocytesin vitro can grow at least as large as, if not
larger than, thosenormally found in vivo.
Both the relatively rapid growth rate and the large size
thatoocytes achieve in vitro suggest that oocyte growth and
ultimatesize in vivo may be regulated within the ovary by
nutrient(specifically vitellogenin) availability. This notion is
furtherreinforced by the observation that although oocytes grow
moreslowly in unstimulated females than in hCG-stimulated
females,they grow at essentially identical rates when removed from
theanimal and placed for 14-28 days in a nutritionally
adequatemedium containing a saturating concentration of
vitellogenin(Table 1). However, several observations have also been
madewhich tend not to support this postulate: (a) oocytes
isolatedfrom females injected with hCG 2-3 days previously have
atransiently higher rate of vitellogenin incorporation (12),
(b)oocytes from unstimulated females increase their rate of
vi-tellogenin incorporation when incubated with hCG for 48-60hr
(12, 25), and (c) oocytes in vitro do not incorporate vitello-genin
at a uniform rate throughout their growth (Fig. 3). Themodulation
of both vitellogenin incorporation and growth rateby insulin (ref.
12; Table 1; Fig. 1) may reflect a quantitativechange in oocyte
metabolism caused by the presence of thishormone rather than a
process related to vitellogenin avail-ability.
Recently, Eppig (26) was able to obtain a significant increasein
the size of apparently normal mouse oocytes grown for 7 dayseither
within their follicles or in organ culture. Isolated oocytesfailed
to grow, however, despite various attempts at cellular orhormonal
supplementation and Eppig thus concluded that "anassociation of
granulosa cells and oocytes was necessary foroocyte growth." Unlike
mouse oocytes, those of lower verte-brates normally grow to a
relatively enormous size primarilyby sequestering an external yolk
precursor (8) so that strictcomparisons cannot be made. We have
nevertheless found thatX. laevis oocytes can both substantially
increase in size anddifferentiate over a 2- to 4-week period (Table
1; Fig. 1) whileretaining an apparently normal morphology and
physiology(ref. 12; Fig. 4). When X. laevis oocytes are freshly
dissectedfrom their follicles, they are usually still invested by a
singlelayer of flattened follicle cells (27) which tend to hinder
thepassage of vitellogenin from the medium to the oocyte in
thoseregions where they remain closely applied to the surface
(seefigure 2b in ref. 28). However, we have observed that the
folliclecells normally bunch up together or slough off the oocyte
sur-face after only a few days in culture, thus completely
exposingthe oocyte surface, as has been previously reported (28).
Onlyan occasional follicle cell remains attached to the oocyte for
anylength of time (Fig. 4). Under such circumstances,
vitellogeninis maximally accessible to the oocyte and optimal
growth canoccur, apparently regardless of the hormonal history of
theoocyte donor (ref. 12; Table 1). In the intact animal,
vitellogeninnormally reaches the growing oocyte after passing
throughchannels between the follicle cells (9). These cells are
generallythought to be the target of gonadotropins in amphibians
(25,29-34); as a corollary notion, gonadotropins may thus
regulateoocyte growth in vivo by causing the follicle cells to open
orclose these channels.The smallest oocytes isolated in our
experiments had an av-
erage volume of 0.15 mm3 and were barely pigmented. After
Cell Biology: Wallace and Misulovin
Dow
nloa
ded
by g
uest
on
Apr
il 2,
202
1
-
5538 Cell Biology: Wallace and Misulovin
28 days in culture these oocytes reached an average volume
of0.73 mm3 (Table 1) and acquired a normal pigment pattern,and
about half responded to added preogesterone by under-going GVBD
(Fig. 1). Data for other groups indicated thatoocytes consistently
acqqired an ability to respond to proges-terone once they reached a
size of about 0.7 mm3 (diameter =1.1 mm), both in the presence and
in the absence of insulin (Fig.1), and that the response included
not only GVBD but thecomplete transition to an egg, because both
metaphase spindlesand polar bodies were found to be produced by
progesteronetreatment (Fig. 4). Thus, at least two aspects of
oocyte differ-entiation were observed for oocytes grown in vitro.
However,our quantitative data collected for the response to 1 ,ug
of pro-gesterone per ml (Fig. 1) does not exactly correlate with
pre-viously published observations (24). In only one case (Fig. 1,
frog22) was a positive response (four out of nine) noted for
oocyteswith a diameter less than 1.1 mm, and also in one case (Fig.
1,frog 21) not all oocytes (seven out of ten) responded to a
20-hrexposure to 1 ,tg of progesterone per ml even after reaching
adiameter of 1.2 mm. In contrast, Reynhout et al. (24) noted
thatfreshly dissected oocytes with diameters ranging from
0.90-0.99, 1.00-1.09, and 1.10-1.19 mm underwent an average of22,
77, and 98% GVBQ), respectively, when scored after a 12-hrexposure
to 0.1 ,ug of progesterone per ml. Thus, our culturedoocytes seem
to be somewhat more refractory to progesteronetreatment than
freshly dissected oocytes. This may be due tosome inadequacy of our
culture method, or it is possible thatcontinuous exposure of
oocytes to environmental steroids re-duces the threshold of
response to progesterone. Ovarian fol-licles in X. laevis produce
progesterone (32) as well as othersteroids (33, 34); therefore,
freshly dissected oocytes may bemore sensitized to externally
applied progesterone than arethose removed from the ovary and
incubated in the absence ofsteroids for 1-4 weeks.Our overall
conclusions from the results reported here are:
(a) X. laevis oocytes.can be grown in vitro at a rate
comparableto the fastest rate yet reported for oocytes in vivo, (b)
vitello-genin accessibility may be the major determinant of
oocytegrowth, (c) differentiation of oocytes also can be achieved
invitro, and (d) although vitellogenin incorporation is
undoubt-edly related to oocyte growth, it remains to be
demonstratedwhether it is also related to oocyte differentiation.
The resultspresented here and in a report describing the
development ofour culture procedure (12) establish that oocytes are
able togrow and differentiate in vitro when they are substantially
freeof associated follicular materials. Our procedure relies upon
adefined culture medium, so that it now also appears possiblein
future experimentation to obtain more definite answers toquestions
concerning the regulation of X. laevis oocyte growthand
differentiation.
This research was supported by the Division of Biomedical
andEnvironmental Research, U.S. Department of Energy, under
contractW-7405-eng-26 with the Union Carbide Corporation.
1. Barr, W. A. (1968) in Perspectives in Endocrinology, eds.
Bar-rington, E. J. W. & Jorgensen, C. B. (Academic, New York),
pp.163-237.
2. Foote, C. L. & Foote, F. M. (1957) Anat. Rec. 127,
145(abstr.).
3. Foote, C. L. & Foote, F. M. (1958) Anat. Rec.
130,553-565.4. Dumont, J. N. (1972) J. Morphol. 136, 153-179.5.
Gurdon, J. B., De Robertis, E. M. & Partington, G. (1976)
Nature
(London)-260, 116-120.6. Eppig, J. J. & Dumont, J. N. (1976)
In Vitro 12, 418-427.7. Eppig, J. J. & Steckman, M. L. (1976)
In Vitro 12, 173-179.8. Wallace, R. A. (1978) in The Vertebrate
Ovary: Comparative
Biology and Evolution, ed. Jones, R. E. (Plenum, New York),
pp.469-502.
9. Dumont, J. N. (1978) J. Exp. Zool. 204, 193-217.10. Dumont,
J. N. & Brummett, A. R. (1978) J. Morphol. 155,73-
97.11. Wallace, R. A., Jared, D. W. & Nelson, B. L. (1970)
J. Exp. Zool.
175,259-269.12. Wallace, R. A., Misulovin, Z., Jared, D. W.
& Wiley, H. S. (1978)
Gamete Res., in press.13. Wallace, R. A., Jared, D. W., Dumont,
J. N. & Sega, M. W. (1973)
J. Exp. Zool. 184,321-333.14. Mans, R. J. & Novelli, G. D.
(1961) Arch. Biochem. Biophys. 94,
48-53.15. Bramhall, S., Noack, N., Wu, M. & Loewenberg, J.
R. (1969)
Anal.. Biochem. 31, 146-148.16. Wallace, R. A. & Karasaki,
S. (1963) J. Cell Biol. 18, 153-166.17. Merriam, R. W. (1972) J.
Exp. Zool. 180,421-426.18. Davidson, M.H. (1932) Turtox News 10,
203-204.19. Wallace, R. A., Nickol, J. M., Ho, T. & Jared, D.
W. (1972) Dev.
Biol. 29, 255-272.20. Davidson, E. H. (1968) Gene Activity in
Early Development
(Academic, New York).21. Scheer, U. (1973) Dev. Biol. 30,
13-28.22. Wallace, R. A. & Jared, D. W. (1976) J. Cell Biol.
69, 345-
35L23. Wallace, R. A. & Steinhardt, R. A. (1977) Dev. Biol.
57, 305-
316.24. Reynhout, J. K., Taddei, C., Smith, L-D. & LaMarca,
M. J. (1975)
Dev. Biol. 44,375-379.25. Wiley, H. S. & Dumont, J. N.
(1978) Biol. Reprod. 18, 762-
771.26. Eppig, J. J. (1977) Dev. Biol. 60, 371-388.27. Jared, D.
W. & Wallace, R. A. (1969) Exp. Cell Res. 57, 454-
457.28. Wallace, R. A., Ho, T., Salter, D. W. & Jared, D. W.
(1973) Exp.
Cell Res. 82, 287-295.29. Masui, Y. (1967) J. Exp. Zool.
166,365-376.30. Schuetz, A. W. (1967) Proc. Soc. Exp. Biol. Med.
174, 1307-
1310.31. Smith, L. D., Ecker, R. E. & Subtelny, S. (1968)
Dev. Biol. 17,
627-643.32. Fortune, J. E., Concannon, P. W. & Hansel, W.
(1975) Biol.
Reprod. 13, 561-567.33. Redshaw, M. R. & Nicholls, T. J.
(1971) Gen. Comp. Endocrinol.
16,85-96.34. Mulner, O., Thibier, C. & Ozon, R. (1978) Gen.
Comp. Endo-
crinol. 34, 287-295.
Proc. Natl. Acad. Sci. USA 75 (1978)
Dow
nloa
ded
by g
uest
on
Apr
il 2,
202
1