close

Вход

Забыли?

вход по аккаунту

?

Stage dependent transfer of systemically injected foreign protein antigen and radiolabel into mouse ovarian follicles.

код для вставкиСкачать
Stage Dependent Transfer of Systemically Injected
Foreign Protein Antigen and Radiolabel into
Mouse Ovarian Foll&les’
LAUREL E. GLASS AND JEAN M. CONS
Department of Anatomy, University of California Medical School,
San Francisco, California
ABSTRACT
The appearance of albumin-like antigen and of radioisotope in mouse
ovarian follicles following the systemic injection of 10 mg human serum albumin 11s’
(HSAIl31) was studied by immunohistological and autoradiographic methods.
Both isotope and antigen appeared in granulosa cell cytoplasm, ooplasm and oocyte
nuclei in similar (or identical) positions. Qualitative impressions from fluorescent
antibody preparations were that follicle stages (I, smallest; IX, largest) differed in
relative amounts of detectable albumin-like antigen. In addition, autoradiographic
grain counts over 1421 follicles, compared by Sign Test, demonstrated statistically
significant stage differences in amount of isotope detected. Radiolabel (and HSA-like
antigen) in ooplasm was highest at Stages I and VII-IX and lowest at Stage 111.
Granulosa cell amounts were highest at Stage I11 and higher than or equal to ooplasm
amounts at every other stage (except, perhaps IX). The relative amount of ‘label,”
antigenic or isotopic, in oocyte nuclei was always less than in ooplasm and increase or
decrease occurred one stage later than i n ooplasm. The relative amount of radiolabel
in ovarian blood vessels always exceeded that in granulosa cells and oocytes.
The data are consistent with the interpretation that: (a) systemically injected
foreign protein was transferred without major degradation from the blood to constituent cells of the mouse ovarian follicle; ( b ) transfer was stage dependent; (c) transfer
occurred both before granulosa cells completely surrounded the oocyte and after
oocyte microvilli and granulosa cell processes interdigitated; and, ( d ) transfer followed
a concentration gradient.
Normal embryogenesis depends upon
normal synthetic pathways and normal
macromolecular interactions within the
developing organism. In addition, normal
development may depend upon the availability and biological activity of other,
externally synthesized macromolecules.
These would originate elsewhere in the
maternal body and be transferred from her
blood into the developing oocyte or embryo.
For semantic convenience, such maternally-synthesized, transferred macromolecules have been called “heterosynthetic”
to distinguish them from molecules
synthesized in situ by the egg or embryo
itself (Schechtman, ’ 5 5 ) . It has been
demonstrated that heterosynthetic molecules are transferred from female blood
into the ovarian oocytes of insects (Telfer,
’54; Telfer and Melius, ’63), amphibians
(Flickinger and Rounds, ’56; Glass, ’59),
birds (Knight and Schechtman, ’54;
Schj eide, Wilkens, McCandless, Munn ,
Peterson and Carlson, ’63), and mammals
ANAT. REC.,162: 139-156.
(Glass, ’61a,b, ’66a) and into tubal embryos in mammals (Glass, ’63). Since the
transfer of heterosynthetic molecules has
such a long phylogenetic history and since
it affects the female gamete so intimately,
there is inferential support for the hypothesis that such transfer may be obligatory
for normal oogenesis and embryonic development.
The fluorescent antibody and pilot autoradiographic studies reported here concern the stage dependence of the transfer
process. Despite methodological problems,
analysis of the autoradiographic data
yielded information demonstrating si@cant correlations between transfer and
particular stages of oocyte and follicle
growth. These data support and expand
the qualitative fluorescent antibody data
and are pertinent to an understanding of
the transfer process.
1Supported by United States Atomic Energy Commission contract No. AT( 11-1)-34. Project Agreement
No.53.
139
140
LAUREL E. GLASS AND JEAN M. CONS
MATERIALS AND METHODS
Animals. Twenty-seven female mice
(12-16 weeks of age) of the Cal A strain
were injected intravenously (19 animals)
or intraperitoneally (8 animals) with
human serum albumin (HSA) isotopically
labelled by 1131 (Risa-M, Abbott)3; no difference attributable to the route of injection was observed subsequently. Earlier
work showed that serum-to-oocyte transfer of intravenously injected heterologous
serum antigens was detected most conveniently when 10-20 mg of protein were
used (Glass, ’61b); therefore, in the present experiment, 10 mg albumin were injected. Doses of radioactivity were dependent on the protein concentration desired
and ranged roughly between 20 and 50 pc.
The labelled albumin was administered
after intraperitoneal injection of 2-3 International Units (I.U.) of pregnant mare’s
serum (PMS: ‘Xquinex,” Ayerst) or of
PMS followed 60 hours later by 4-6 I.U.
of human chorionic gonadotropin (HCG:
“A.P.L.,” Ayerst), and at various known
intervals after ovulation, mating and during cleavage (Glass, ’61b, ’63). At the expected time of ovulation (12-14 hours
after HCG administration), the animals
were placed with males for two hours
after which they were checked for vaginal
plugs. Differences in fluorescent antibody
localization and in autoradiographic counts
could not be correlated with these treatments but, instead, were related as described below to follicle stage. In addition,
seven non-pregnant mice were injected
intravenously with unlabelled HSA. Six
additional mice were injected intraperitoneally with NaIx3Iin saline (Cambridge
Nuclear Corp.); three received 50 pc and
three 500 pc of isotope.
Animals were sacrificed two hours after
the isotope-labelled or unlabelled protein
was administered; ovaries and oviducts
were removed, fixed in Carnoy’s, embedded in low melting point Tissuemat
(50-53°C M.P.) and sectioned serially.
Autoradiographs were made by dipping
the slides into Kodak 50% NTB-3 liquid
emulsion in distilled water; slides were
exposed five to ten days. The emulsion
was developed for ten minutes in the cold
using half strength Kodak D-19developer.
Dried slides were stained with buffered
Azure A-basic fuchsin. Control slides contained “cold ovarian sections from uninjected animals.
Fluorescent antibody procedures. Fluorescent antibody studies of HSA localization in the mouse ovary utilized sections
from four mice injected with HSAIiS1 and
from seven mice injected intravenously
with 10 mg HSA in saline.
The indirect fluorescent antibody procedure was used (Mellors, ’59; Glass, op.
cit., and in press, b). Several drops of unlabelled, specific antiserum were placed on
a slide and the preparation was incubated
in a moist chamber for several minutes.
After rinsing, a fluorescein-labelled sheep
anti-rabbit globulin serum (Progressive
Laboratories) was placed on the sections.
Following incubation and rinsing, a cover
slip was mounted in 50% glycerol. Sections were viewed with a Zeiss fluorescence
OG 4 and OG 5
m i c r o s ~ ~ pUsing
e
primary and BG 12 ( 2 m) as secondary
filters.
High titer rabbit antisera to HSA (18
mg a.ntibody/ml precipitated 2.1 mg antigen) were obtained commercially (Antibodies Incorporated). The antiserum was
diluted 50:50 with 0.15 M saline and characterized in microimmunoelectrophoresis
runs against several concentrations (ranging from 25-10 mg protein/ml) of HSA
or of adult female mouse serum (MS). A
single line was present in reactions with
HSA and no cross reactions were observed
between MS and the anti HSA. However,
when the anti HSA antisera were used on
ovarian sections from uninjected mice,
fluorescence of moderate intensity indicated a slight cross-reaction. Therefore, the
specificity of the anti HSA was refined
further by absorption of several samples
with 6 mg/ml MS; a small amount of precipitate formed and was centrifuged out.
The MS absorbed-anti HSA reacted only
with ovaries from HSA-injected mice and
did not react with ovaries from uninjected
mice : therefore, these antisera were used
in all experimental fluorescent antibody
runs.
2 Derived from A/Crgl 2.
aA sterile, pyrogen
solution which contained
not less than 10 mg of HSA per milliliter and less
than two percent of loosely bound radio-iodine: each
gram molecule of albumin contamed no more than
one gram-atom of iodine.
4-
141
ANTIGEN AND ISOTOPE TRANSFER TO OVARY
Controls for the immunological specificity of the observed fluorescence included
the following: (a) Autofluorescence was
determined by mounting a cover slip over
the hydrated sections; (b) control anti
HSA sera were derived by absorption with
excess HSA; six mg antigen protein were
added per ml of half strength antiserum
and the precipitate was centrifuged out.
These HSA absorbed-anti HSA sera were
used on ovarian sections from HSA-injected and from uninjected mice; (c) MS
absorbed-anti HSA samples were placed on
ovarian sections from uninjected mice in
procedures identical with those used in
experimental tests with ovaries from HSAinjected animals.
Ovarian follicle stages. Ovarian follicles were classified into nine stages according to oocyte size, number of layers of
granulosa cells and antnun formation; the
classification roughly follows that of
Pincus (’36). These are described in table
1 and diagrammed in figure 1.
Autoradiographic counts. Autoradiographic counts were made at 640 magnification using a Whipple disc. Counts were
taken over the ooplasm, nucleus and grandosa cells of six follicles for each of the
nine follicle stages in each of the 27 animals. For large follicles, a known unit area
was counted. For small follicles, all grains
overlying the follicle cells, ooplasm and
nucleus were counted and the area of the
whole follicle, the oocyte and the nucleus
was calculated from two measurements
made at right angles to each other.4 Grain
counts in the total area were converted to
the same unit used for the large follicles.
Background was counted on tissue-free
regions of the slide adjacent to each
counted follicle. All the background
counts on each slide were averaged and
this average background was subtracted
from each tissue count made on the same
slide.
The amount of isotope present in 6Ued
ovarian blood vessels was defined arbitrarily as the isotope “available” to the
ovary in that animal. To obtain this figure,
counts were made over a filled blood vessel
(BV) near each follicle counted; a total
4A
= r/16 (D1 + D2)2 where D is the measured
diameter.
TABLE 1
Characteristics of mouse ovarian follicle stages 1
Follicle
Stage
I
I1
I11
IV
V
VI
VII
VIII
Ix
Description
Primordial ova with incomplete
layer of granulosa cells.
One layer of flattened or round
granulosa cells surrounding
oocyte.
One layer of cuboidal granulosa cells.
Two layers of cuboidal granulosa cells.
Three layers of cuboidal granulosa cells; small ragged spaces
may be present between them.
Small tertiary. Antrum present
as large slit between 3 or 4
layers of granulosa cells.
Small tertiary. Antrum larger;
3 to 10 layers granulosa cells
around oocyte.
Middle tertiary Antrium larger;
5 to 20 layers of granulosa
cells.
Large tertiary. May be 10 to
30 layers of granulosa cells at
cumulus oophorus.
1 Carnoy
2
(PI2
Oocyte
nucleus
(PI2
(10.5- 16.5)
14.4
( 13.5-2.5)
15.5
-
(7.5- 10.5)
8.9
(7.5- 12.0)
9.9
-
(27.0 46.5)
36.2
(43.5 - 82.5)
67.4
__
(70.5 100.5)
82.7
-
(16.5- 28.5)
21.8
(27.0T46.5)
37.4
)
0
.
1
5
:
5
.
7
3
(
43.6
-
(6.0- 15.0)
10.9
(13.5- 22.5)
17.3
(13.5- 24.0)
18.7
-
(99.0- 147.0)
124.5
-
(54.0- 64.5)
59.7
(18.0- 27.0)
22.7
(117.0- 234.0)
166.6
-
(51.0 - 66.0)
61.4
-
(19.5- 27.0)
23.7
-
(216.0 360.0)
283.5
-
(63.0- 70.5)
67.6
(21.0- 30.0)
25.7
-
(334.5- 438.0)
382.0
-
(66.0- 76.5)
69.8
(22.5 31.5)
27.5
Incomplete
(18.0- 34.5)
23.7
-
-
-
-
fixed ovaries from “Cal A” strain mice (derived from
( ) Range. -Average.
Oocyte
-
-
A/Crgl/Z).
-
-
142
LAUREL E. GLASS AND JEAN M. CONS
Fig. 1 Diagrammatic representation of relative sizes of mouse oocytes and follicles at various
ovarian follicle stages. Description in table 1.
of approximately 54 BV counts was obtained for each animal. All the BV counts
for an animal were averaged and a series
of ratios were calculated between each
corrected tissue count and the BV average
for that a n i m a l . 5 Use of such ratios minimized the difference between animals due
to difference in the original dose of isotope
and in the route of administration.
Statistical procedures.6 Because different amounts of isotope were injected and
because different injection routes were
used, all comparisons of the position and
amount of isotope were made within an
experimental animal. Moreover, since our
autoradiography used only carefully applied standard procedures rather than
rigidly controlled quantitative techniques,
the Sign Test was used in all statistical
analyses. In this procedure, the direction of
deviation between two observations was
scored (A > B =
, or A < B =
)
rather than the amount of the difference
between them; plusses and minuses were
+
-
summed and the probability was estimated
by calculation of 2.7 The two-tailed test
was used since our advance prediction did
not state for most follicle stages which of
A > B or A < B would occur most frequently, but only that their frequencies
would differ.
Using the Sign Test, two sorts of comparisons were made: (1) between the
components (granulosa cells, ooplasm, and
oocyte nucleus), within each follicle
counted, and (2) between follicle stages
(Stage I, I1
IX) within each animal.
In the first instance, counts (corrected
for background) in the granulosa cells,
ooplasm and oocyte nucleus were compared within each of the 1421 follicles
counted. These data were summed by fol-
...
corrected tissue count
Ratio =
blood vessel average for that animal
6 Dr. Robert Elash&. Assistant Research Biostatistician, served as statistical consultant.
(X
. 5 ) - VZ N
7z=
46
where T is the number of
fewer signs (Siegel, ' 5 6 ) .
5
+
J
r
143
ANTIGEN AND ISOTOPE TRANSFER TO OVARY
licle stage and the probability that the
count in one component was significantly
greater or less than in another was calculated.
In the second instance, the count/blood
vessel ratios were used. In general, six follicles at each of the nine follicle stages
were counted for each of the 27 experimental mice. Ratios were calculated for
each count and the mean of the six ratios
at each stage was calculated for each
mouse. Then, for each mouse, the mean
ratio for granulosa cells (or ooplasm or
oocyte nucleus) at each follicle stage was
compared with the mean ratio for that
component at each other follicle stage.
These data were collected by follicle stage
and probability estimates calculated.
IBM computers 1620 and 7070 were
used in the data analysis.
shape of the curves is similar to that reported by Pincus (’36) for rabbit ovarian
follicles. When oocyte growth was plotted
against follicle growth, the curve was consistent with that for mouse follicles reported by Brambell (’29). The nine follicle stages are diagrammed in figure 1.
Fluorescent antibody observations. As
evidenced by both negative and positive
control and test reactions, fluorescence
which was immunologically specific was
present in the oocytes and granulosa cells
of ovaries from HSA-injected mice.
Characteristic autofluorescence was observed on hydrated control sections (figured in Glass, ’61b); many corpus luteum
cells had a very low intensity yellowish
autofluorescence, erythrocytes in blood vessels fluoresced a very pale orangish color
and small regions of bright yellow-orange
fluorescence were scattered irregularly in
RESULTS
the ovarian stroma. HSA absorbed-anti
Follicle stages. The range and average HSA did not react with ovaries from HSAdiameter of the follicle, the oocyte and the injected or uninjected mice; these sections
oocyte nucleus at each follicle stage in showed a generalized fluorescence of an
Carnoy-fixed, paraffin-embedded ovaries intensity only slightly higher than on the
from Cal A mice are shown in table 1. autofluorescent sections. Similarly, in conThese data are plotted in figure 2. The trol runs, MS absorbed-anti HSA did not
s
70
--O-- Follicle
--o- Oocyte
***-U**-*Oocyte Nucleus
5
z
P
z
a
c
Ly
>
U
0
0
sol-
t5
01
I
I
I
I
I
I
I
’
II I I m E n n r s n m J x
Jo
FOLLICLE STAGE
Fig. 2 Growth curves for mouse ovarian follicles. Ovaries were Carnoy-fixed from “Cal A”
strain mice (derived from A/Crgl/2).
144
LAUREL E. GLASS AND JEAN M. CONS
react with ovaries from uninjected mice.
This very low level of non-specific fluorescence on the ovary from an uninjected
mouse is illustrated in figure 5; the ovary
of an HSA-injected mouse treated identic d y with MS absorbed-anti HSA is pictured in figure 6. Photographic treatment
for figures 5 and 6 was identical. These
con~ols
in
plusthe fact that
consistent and characteristic fluorescence
was observed when ovarian sections from
HSA-injected animals Were treated with
MS absorbed-anti HSA, demonstrated that
the binding of fluorescent antibody was
due to immunologically specific reactions.
Fluorescence was due to the reactions of
anti HSA bodies with antigen in the
ovaries which Was S h n h to Or identical
with the SYstedcaUY injected f o r e i s
albumin.
Consistent differences in the intensity
and location of fluorescence were observed
when ovaries from HSA-injected mice
were reacted with MS absorbed-anti HSA.
The cytoplasm of oocytes in Stage I (figs.
4,7) and Stage VII-IX (figs. 6,s) follicles
fluoresced brightly. Fluorescence in the
ooplasm in Stage 111 and IV (fig. 7) follicles seemed less intense and that in Stage
IX follicles (fig. 8) seemed slightly more
intense than at other follicle stages. This
subjective impression was confirmed
roughly by photographic record. Compare,
for example, the intensity of ooplasmic
fluorescence in follicles at the various
stages figured in 4 or in 7 and 8. Figures
7 and 8 were from the same ovarian section and photographic treatment was
identical.
The cytoplasm of the granulosa cells
fluoresced also and, in larger follicles, the
rows of cells nearer the antrum fluoresced
most intensely (figs. 68). In Some of the
intermediate Stages Of Doc@ growth the
granU1OSa Ceu cytoplasm Was more intensely fluorescent than was the ooplasm
(fig.7)*
Autoradiographic data. The fluorescent
antibody method does not allow ready
quantitation. Analysis of the autoradiographic data provided independent confirmation and refinement of the fluorescent
antibody results.
Autora&ogaphs of control ovaries from
did not showlocalNaI13I-injected
background level.
ization of isotope
c~~~~~~~~~
bemeen the corrected
counts of each folliclecomponent - Danulosa cell, oocyte, and omyte nucleus
within each follicle - in HSA1131-injected
mice demonstrated that there were sign&
cant differences in the relative amount of
radiolabel jn each component and that the
relationship between components varied
with follicle stage. These data are recorded in table 2 and probability cdculat k m from them
summarized under
the line graph in figure 3 as “greater than,”
‘less than” or “equal to” statements for
each component at each follicle stage.
The Sign Test comparisons between the
mean ratios at different follicle stages in
the same mouse are recorded in tables 3,
41 and 411 for granulosa cells, ooplasm,
and oocyte nucleus respectively. These
analyses showed that there were simcant differences between follicle stages in
the relative amount of radiolabel in the
TABLE 2
Sign test comparisons of grain counts of granulosa cells, ooplasm, and
oocyte nucleus in HSA1’31-injectednice 1
Nusber
A. Oocyte
Follicle
stage follicles
T
q
111
162
IV
V
VI
159
VIII
vll
154
Total
IX
160
161
1421
149
-B. Nucleus
A>B
A=EA<B
I
I
I
33
32
26
26
25
35
25
25
37
1 Sign Test comparisons
2 Two-tailed probability
75/52
65/62
79/57
88/45
90/45
72/54
88/44
82/47
73/39
A>BZl
P
.05
.86
.07
,0003
,0002
.12
,0002
,002
.002
-
A. Oacyte B. Granulosa cells A. Nucleus - B. Granulosa cells
L x L pA > B Z
A>B
*>Bz1
p---A < B
A=B A < B
A C R
~
14
13
19
32
36
49
35
45
28
~
=
.za
56/90
24/119
41/86
47/77
42/70
55/67
54/55
66/55
.OD6
.00006
.0001
.QO9
41
.32
--
.07
between follicle components in each of 1421 follicles.
of summed sign differences as great as those observed.
17
17
16
16
21
33
33
34
22
~
54/89
60182
40/106
29/114
33/106
36/92
33/91
42/78
50/77
m
,003
.08
.O0006
,00006
.Q0006
.00006
.00006
.001
.02
145
ANTIGEN AND ISOTOPE TRANSFER TO OVARY
--&-Granuloso
-0oplarm
.c
Cells
.....0.....Oocyte Nucleus
O>N O = N O>,N O>N O>N O = N O>N O>N O>N
O<G O<G O<G O<G O<G O = G O = G O>,G
N(G
N<G N<G N<G N<G N<G N<G N<G
I
l
I
m
E
~
Y
I
~
m
l
x
FOLLICLE STAGE
Fig. 3 Collation of Sign Test data on presence of radiolabel in mouse ovarian follicles.
The mean of mean ratios (table 5 ) was plotted to give a rough estimate of magnitude;
however, the relationships drawn depend upon the statistically significant Sign Text comparisons described in the text and not upon the highly variable mean ratios. The relationships
between the curves are consistent with all data from tables 2, 3 and 4 excepting only that
the graph does not shown I11 = VIII nor VIII < IX for ooplasm.
TABLE 3
Sign test comparisons of granulosa cell mean ratios compared by follicle stage
in ffSA1I3l -injected mice 1. 2. 3. 4
Follicle stage
I1
I
I
I1
I11
b
Z I V
Q V
3“c
VI
VII
VIII
M
1 Ratio
21/6 **
19/8 *
20/7*
16/11 =
17/10=
17/10=
1819-
IV
V
6/21
8/19’
23/3 **
a p *
7/20 *
19/8 *
11/15 =
i5/ii=
14/13 =
16/10=
15/12=
16/11 =
10/17 =
16/11 =
12/15 =
14/13 =
I11
**
3/23 **
1O/17 =
7/20*
8/19*
10/17=
VI
VII
VIII
IX
11/16=
17/10=
13/14 =
17/10 =
10/17=
20/7*
10/16 =
11/16 =
12/15 =
10/17=
igp*
12/15 =
15/12 =
15/12 =
17/10 =
10/17 =
11/16 =
9/18=
i7/1011/16
13/14
14/13
16/11 =
11/16
16/11 =
15/12 =
12/15 =
13/14 =
--
-
= Tissue Count/Blood Vessel Count.
W counts corrected for background.
Sign Test comparisons by mouse; granulosa cell mean ratio at each follicle stage in each mouse compared
to nranulosa cell mean ratio at each other follicle stage in the same mouse.
8-Recorded as A B/A < B.
4Symbols in upper right hand comer of each box represent the two-tailed probability of summed sign differences as great as those observed: =, no significant dewation; *, p < 0.05; **, p < 0.01.
2
>
granulosa cells. Similarly, ooplasm ratios
differed significantly at different follicle
stages as did oocyte nucleus ratios.
A line graph summary of the relationships at each follicle stage between radiolabel in the follicle cells, the ooplasm and
the oocyte nucleus is presented in figure 3.
The mean of mean ratios at each stage
(table 5) is indicated so as to give a rough
estimate of magnitude. Except as stated
in the legend to the figure, relationships
between the lines in figure 3 are consistent
with all the probability statements derived
from the Sign Test data in tables 2, 3 and
4. (The means of the mean ratios and their
standard deviations are given in table 5;
the variability within the data is indicated
by the size of the standard deviations.
146
LAUREL E. GLASS AND JEAN M. CONS
TABLE 4
I. Sign test comparisons of ooplasm mean ratios compared by follicle stage
in HSAI131-injected mice 2, 3. 4
1 8
Follicle stage
I
I1
I11
VI
V
IV
VII
VIII
IX
11. Sign test comparisons of oocyte nucleus mean ratios compared by follicle stage
in HSAIlzLinjected mice
1 Ratio
11 4s 3 , 4
= Tissue Count/Blood Vessel Count.
A l l counts corrected for background.
3 Sign Test comparisons by mouse: ooplasm mean ratio at each follicle stage in each mouse compared to
ooplasm mean ratio at each other follicle stage in the same mouse.
8 Recorded as A
B/A < B.
4 Symbols in the up er ri ht hand comer of each box represent the two-tailed probability of summed sign
ferences as great as &ose o%served: =, no significant deviation;
p < 0.05; **, p < 0.01.
>
*,
TABLE 5
Mean ratios by follicle stage of granulosa cells, ooplasm
and oocyte nucleus i n HSAI131-injected mice
Follicle cells
Mean f Standard deviation
Ix
.309 .331
.363 -c .33a
.540 2.359
.424 f ,303
.45a -c 353
.457f .334
-486f -344
,401-c .2a7
.446f -358
Therefore, all conclusions were based on
Sign Test comparisons using the methods
detailed above.)
In summary, these data show that relatively more isotope was present in the
granulosa cells than in the oocytes from
Stages I1 through Stage VI. By Stages VII
and VIII, isotope in the ooplasm equaled
that in the follicle cells. At Stage IX, relatively more isotope appeared to be present
in the ooplasm than in the granulosa cells.
In the last instance, the two-tailed probability, p < 0.07, was suggestive of difference; this observation is strengthened by
the fluorescent antibody work which suggested that the oocyte in Stage IX follicles
contained more serum-like antigens than
did the grandosa cells.
There was a relatively large amount of
isotope in the cytoplasm of the Stage I
oocytes; these eggs were only partially sur-
Ooplasm
Oocyte nucleus
Mean C Standard deviation Mean -C Standard deviation
.423-c .4ao
,323rt .4a7
.2552 .277
.317f .306
.390f .369
.3aa 2 ,317
,488-c .390
.4342.529
.496 .394
*
.334f ,463
.325r+ ,419
.263 .368
.211f ,306
.2a3r?: .3i5
.310f .329
.3562.343
.313 2.284
.393f .385
*
rounded by granulosa cells. Less isotope
was present in Stage I1 ooplasm and still
less was detected in Stage I11 ooplasm; a
single layer of flattened granulosa cells at
Stage I1 and of cuboidal granulosa cells at
Stage I11 surrounded the oocyte completely. Radiolabel in the cytoplasm of
Stage IV oocytes was relatively higher than
at Stages I1 and I11 though still not so high
as at Stage I; although the granulosa cell
layer was complete in these follicles, a thin
zona pellucida had appeared. The amount
of isotope in the ooplasm at Stages VII,
VIII, and IX was relatively higher than at
Stages 11 and 111; statistically this was
equal to the amount in the ooplasm at
Stage I.
Adequate photographs of these autoradiographs were difficult to obtain because of section thickness and of overstaining which obscured the grains over-
ANTIGEN AND ISOTOPE TRANSFER TO OVARY
lying the granulosa cells in black and
white pictures. The characteristic relative
density of grains overlying the ooplasm at
several stages is shown in figures 9-14.
These figures are illustrative only; as already stated, our conclusions are based
upon the statistical analysis of the Sign
Test comparisons detailed above.
Isotope in the nucleus in Stage I was
relatively less than in either the ooplasm
or the granulosa cells and in Stages I1 and
111 was approximately equal to ooplasmic
counts. Relative to other stages, nuclear
isotope reached its lowest point at Stage
IV, increased slightly at Stage V and continued to increase slightly from Stage V
through Stage IX.
The mean ratios for each follicle component at each follicle stage were all less
than one; this indicates that there was less
radiolabel in the follicles at all stages than
was “available” in the blood vessels.
DISCUSSION
Human serum albumin labelled with
1”’ (or unlabelled human serum albumin)
was injected systemically into female
mice. Two hours later the mice were sacrificed and immunohistological or autoradiographic preparations were made of
their sectioned ovaries.
Antigens similar to or identical with the
systemically injected human serum
albumin1131 (or of the unlabelled HSA)
were detected by the fluorescent antibody
method in the oocytes and granulosa cells
of mouse ovarian follicles. Control studies
demonstrated that the fluorescence was
immunologically specific. Since the intensity of fluorescence Mered in oocytes
and granulosa cells at m e r e n t follicle
stages it appeared that the presence of the
albumin-like antigen in the follicle was
stage dependent.
Despite variability in the autoradiographic data, Sign Test comparisons detected significant differences in the relative amount of radiolabel in ovarian
follicles: (a) between the granulosa cells,
the ooplasm and the oocyte nucleus, when
compared to each other within each of the
counted follicles and, (b) between follicle
stages when each follicle component granulosa cells or ooplasm or oocyte
nucleus - was compared at a particular
follicle stage to the same component at
147
each different follicle stage. Thus, as with
antigenic reactivity, the presence in the
ovarian follicle of the radioisotopic constituent of the injected foreign albumin
appeared to be stage dependent.
Size of the transferred molecules. As
discussed previously (Glass, ’61, ’63, ’66a;
Glass and McClure, ’65) de nouo synthesis
cannot account easily for the appearance
in the ovarian follicle of foreign protein
antigen which is introduced systemically;
molecular transfer from the plasma to the
ovum is a more likely hypothesis. Moreover, chemical, immunological, ultracentrifugal and electrophoretic data on the
transfer of blood proteins to the egg during insect, amphibian and avian vitellogenesis (op. cit.) support the hypothesis
that the transferred blood molecules are
essentially intact blood macromolecules. In
an extensive review of this literature,
Glass (in press, a ) concluded that: ( a )
blood macromolecules are transferred to
the oocyte without major degradation; (b)
dthough they are very similar to, the
serum-like oocyte molecules are not identical with their blood precursors; (c) in
general, the transferred molecules retain
their biological activity; and (d) molecular
species differ from one another in the extent to which they are modified during
transfer to and storage within the ovarian
egg.
While it cannot be concluded unequivocally that the radiolabel and the antigenic
‘label” we detected in the mouse ovary was
identical with the HSAIl31 injected systemically, our fluorescent antibody and
autoradiographic data are consistent with
this hypothesis. Within methodologicd
limits, both ‘labels” were present at similar, perhaps identical, locations within the
oocytes and granulosa cells. This can be
interpreted as indicating the simultaneous
transfer of both molecular labels and increases the probability that the protein
molecules were not degraded sigmficantly
before, during or after passage into the
oocytes and granulosa cells.
The transfer hypothesis is consistent
also with methodological considerations.
Macromolecules (proteins or polysaccharides) rather than haptens or small
molecular weight antigens probably are
localized by the fluorescent antibody
method (Nairns, ’62, p. 100; Glass and
148
LAUREL E. GLASS AND JEAN M. CONS
Cons, '65). Moreover, in control ovaries
from animals injected systemically with
NaI131, the isotope did not mimic the
ovarian localization of the albumin-I131.
These data agree with observations from
many laboratories that small radiolabelled
molecules are washed from tissue sections
by standard histological and autoradiographic procedures (e.g., Rogers,. '67,
p. 121, 223). Abundant undegraded
albumin1131 probably was present in the
plasma throughout the two hours duration
of our experiment since the in v i m halflife for injected HSA1131 in the serum of
the mouse is reported to be 1.5 days
(Dixon and Weigle, '57).
In the absence of evidence to the contrary, in the context of evidence from
several laboratories that blood-to-oocyte
transfer of macromolecules occurs and on
the basis of our combined fluorescent
antibody and autoradiographic data, it is
concluded provisionally that the injected
human serum albumin1131 molecules were
transferred intact, or nearly so, from the
plasma into mouse ovarian oocytes and
granulosa cells.
Morphological correlations with transfer.
Comparison of the data graphed in figure
3 with the curves of oocyte and follicle
growth graphed in figure 2 suggests some
morphological correlations with the transfer process.
The simplest is an association of transfer both with the absence of granulosa
cells and with morphological adaptations
at the granulosa cell-oocyte interface.
Thus, the Stage I oocyte which lacked a
complete granulosa cell layer appeared to
be freely accessible to the blood HSA. At
Stages I1 and 111, however, after the granulosa cells completely surrounded the oocyte
less HSA reached the ooplasm, being relatively lower in Stage I1 than in Stage I
and at its lowest relative point in Stage
I11 follicles; although HSA transfer to the
granulosa cells apparently was still occurring freely, the protein was not passing on
into the egg.
These data could indicate that the HSA
passing into Stage I1 and I11 oocytes was
degraded (or utilized) nearly as fast as it
was transferred. In this case, the degradation products presumably were washed
from the ooplasm during subsequent tissue
preparation procedures.
Alternatively, the data probably indicate
that HSA transfer into the oocyte was
blocked during Stages I1 and I11 with a
consequent accumulation of protein in the
granulosa cells.
Observations of Stage IV follicles were
consistent with the second intewretation
because at Stage IV, an increase of ooplasmic HSA was coincident with a decrease
of HSA in the granulosa cells. Morphological correlations also support this interpretation. In the mouse follicles studied,
small, irregular patches of PAS-positive
material appeared between the grandosa
cells and the oocyte in many late Stage I11
follicles and by Stage IV a thin zona pellucida completely encircled the egg. Electron micrographs of mouse oocytes show
that short microvilli form from the oocyte
membrane at this stage and that short
blunt processes from the surface of the
granulosa cells also make their appearance
at this time (Odor, '60). As follicle development proceeds, the slender microvilli
become more numerous and extend a third
or half the width of the zona pellucida. The
granulosa cell processes also become more
numerous and interdigitate with the oocyte
microvilli, some ending on the oocyte
membrane and some on the microvilli
(Odor, '60; see also Yamada, Muta, Motomura and Koga, '57, mouse; Sotelo and
Porter, '59, rat; Trujillo-Cenoz and Sotelo,
'59, rabbit; Anderson and Beams, '60,
guinea pig). The intimate association provided by the interdigitation and contact between the granulosa cell processes and the
oocyte microvilli may have facilitated HSA
transfer from granulosa cells to oocyte.
Oocyte growth was most rapid between
Stages I11 and IV.Even though the volume
of the egg was increasing rapidly during
this period, the concentration of transferred protein in the ooplasm increased
still more rapidly. This is evidenced by the
increase in counts per unit of ooplasm area
from Stage 111 to Stage VI. As oocyte
growth slowed in Stages VII-IX, the rate
of HSA transfer apparently decreased also
and the amount of radiolabel per unit of
ooplasm and intensity of ooplasmic fluorescence remained virtually constant (with
the possible exception of Stage IX). We
ANTIGEN AND ISOTOPE TRANSFER TO OVARY
have no clear evidence concerning the
functional meaning of this association between rapid oocyte growth and even more
rapid protein transfer. However, in fluorescent antibody studies using antisera
against mouse serum fractions, serum
components migrating in the "albumin" region were localized on oocyte nucleoli at
Stages VI, VII, and VIII; only weak and
inconstant nucleolar localizations were observed during Stages I through V (Glass,
'66b). Since the transferred serum molecules become associated with the oocyte
nucleoli near the end of rapid oocyte
growth, questions are raised concerning a
possible genetic and/or synthetic control
function for some of the transferred serum
molecules.
In addition to the presence of follicle cell
processes in close association with oocyte
microvilli during rapid oocyte growth and
to the concomitant occurrence of rapid
transfer and rapid oocyte growth, it should
be noted that the period of rapid oocyte
growth occurs mainly before antrum formation. According to Moricard ('33), Golgi
material in rabbit granulosa cells is polarized toward the oocyte before antrum formation but when follicular fluid secretion
begins, the polarization reverses in the
granulosa cells surrounding the oocyte and
the Golgi apparatus lies between the granulosa cell nucleus and the antrum. Gresson ('33) describes a similar but less regular orientation of the Golgi material in
mouse ovarian follicles. It is possible that
the Golgi may be functionally related to
the transfer of blood protein into the
oocyte.
Effect o f concentration on transfer.
Counts in the nucleus were always relatively lower than in the ooplasm and
'lagged one stage behind ooplasm counts,
whether amounts were decreasing (Stage
I > Stage I1 > Stage I11 > Stage IV) or
increasing (Stage IV < Stage V < Stage
VI < Stage VII). This suggests a (passive?) dependence of nuclear HSA on the
amount of transferred protein in the ooplasm.
HSA concentration in the blood vessels
was higher than that in the granulosa cells
or ooplasm at all follicle stages. In general,
too, more HSA was present in the granulosa cells than in the ooplasm and in the
ooplasm than in the nucleus. Thus, the
149
transfer of HSAIlsl into the mouse ovary
occurred in the direction of the concentration gradient. Nonetheless, in the absence
of other evidence, the data do not exclude
the possibility that the transfer process is
active and requires utilization of energy.
Selectivity of transfer. That blood-tooocyte transfer in the mouse is selective as
to molecular species (Glass, '61b, a ) and
as to animal species of origin (Glass, '61b)
was reported earlier. As indicated by fluorescence intensity, comparisons between
the HSA-anti HSA system and analogous
runs with a bovine plasma albumin
(BPA)-anti BPA system (Glass, '66a) suggest that BPA is transferred to mouse
ovarian granulosa cells and oocytes more
readily than is HSA.
CONCLUSION
Whatever the mechanism of transfer,
these data confirm previous research which
showed that foreign as well as native
serum protein molecules were transferred
from the blood into the mouse ovarian 00cyte (Glass, op. cit.). The present data
demonstrate, in addition, a rather specific
stage dependence for the transfer of foreign proteins, and this dependence makes
it likely that particular morphological
adaptations of the granulosa cell-oocyte
interface are functionally associated with
transfer. Probably such morphological correlations are also a valid model for the
transfer of native serum protein into the
mouse ovary. Continuing work is based on
this hypothesis.
ACKNOWLEDGMENTS
Thanks are expressed to Drs. Robert Elashoff and Mary Epling for statistical consultation, to Miss Jean Hanson for technical help and to Miss Evelyn Gabriels and
David Akers for photographic assistance.
LITERATURE CITED
Anderson, E. W., and H. W. Beams 1960 Cytological observations on the fine structure of the
guinea pig ovary with special reference to the
oogonium, primary oocyte and associated
follicle cells. J. Ultrastructure Research, 3:
432446.
Brambell, F. W. R. 1929 The development and
morphology of the gonads of the mouse. Part
111. The growth of the follicles. Proc. Roy.
sot. B, 103: 258-272.
Dixon, Frank J., and William 0. Weigle 1957
The relationship of the rates of serum protein
metabolism, heterologous serum protein catabo-
150
LAUREL E. GLASS A N D JEAN M. CONS
lism, and the rime and magnitude of the antibody response. Ann. N. Y. Acad. Sci., 70: 6972.
Flickinger, R. A,, and D. E. Rounds 1956 The
maternal synthesis of egg yolk proteins as
demonstrated by isotopic and serological
means. Biochim. Biophys. Acta, 22: 38-42.
Glass, Laurel E. 1959 Immunohistological
localization of serum-like antigens in frog
oocytes. J. Exp. Zool., 141; 257-290.
- 1961a Immuno-histological localization
of systemically injected foreign protein in the
ovary of the mouse. In Symposium on the Germ
Cells and Earliest Stages of Development.
Intern. Instit. Embryol. and Fondazione A.
Baselli, Pallanza.
1961b Localization of autologous and
heterologous serum antigens in the mouse
ovary. Developmental Biol., 3: 787-804.
- 1963 Transfer of native and foreign
serum antigens to oviducal mouse eggs. Amer.
Zool., 3: 135-156.
1966a Dissimilar localization of serum
albumin and globulins in mouse ovaries.
Fertil. Steril., 17: 226-233.
1966b Nucleolar localization of serumlike antigens in mouse ovarian oocytes. J. Cell
Biol., 31 :38A.
(in press, a ) Translocation of macromolecules. In: 0. A. Schjeide and J. de Vellis
(eds.), Cell Differentiation. V a n Nostrand,
Princeton.
(in press, b ) Fluorescence antibody
methods applicable to studies of mammalian
embryos. In: J. C. Daniel, Jr. (ed.) Methods in
mammalian embryology. W. H. Freeman, San
Francisco.
Glass, Laurel E., and Jean M. Cons 1965 Failure of immunohistological techniques to demonstrate hapten in the mouse ovary. Anat. Rec.,
151: 452.
Glass, Laurel E., and T. R. McClure 1965 Postnatal development of the mouse oviduct: Transfer of serum antigens to the tubal epithelium.
In: G. E. W. Wolstenholme and M. O’Conner
(eds.), Ciba Found. Sympos. on Preimplantation Stages of Pregnancy. Churchill, London.
Gresson, R. A. R. 1933 A study of the cytoplasmic inclusions and nucleolar phenomena
during the oogenesis of the mouse. Quart. J.
Microscop. Sci., 75:697-721.
Knieht. P. F.. and A. M. Schechtman 1954 The
passage of heterologous serum proteins from
-
the circulation into the ovum of the fowl. J.
Exp. Zool., 127: 271-304.
Mellors, R. C.
1959 Fluorescent antibody
method, p. 1-68. In: R. C. Mellore (ed.),
Analytical Cytology, 2nd ed. McGraw-HiU, N. Y.
Moncard, R. 1933 Zone de Golgi du follicule
ovarien. Ann. Anat. Pathol. Med. Chir., 10:
12221226.
Nairns, R. C. (ed.) 1962 Fluorescent protein
tracing. Livingstone, Edinburgh.
Odor, Louise D. 1960 Electron microscopic
studies on ovarian oocytes and unfertilized
tubal ova in the rat. J. Biophys. Biochem.
Cytol., 7:567-574.
Pincus, Gregory 1936 The eggs of mammals.
Chapter 3: 3241. Macmillan Co., New York.
Rogers, A. W. 1967 Techniques of autoradiography. Elsevier, Amsterdam.
Schechtman, A. M. 1955 Ontogeny of the
blood and related antigens and their significance for the theory of differentiation, p. 3-31.
In: D. M. Prescott (ed.), Biological specificity
and growth. Princeton Univ. Press, N. J.
Schjeide, 0. A,, Myrna Wilkens, Ruth G. McCandless, R. 0. Munn, Marion Peterson and
E. Carlsen 1963 Liver synthesis, plasma
transport, and structural alterations accompanying passage of yolk proteins. Amer. Zool.,
3: 167-184.
Siegel, Sidney 1956 Nonparametric statistics
for the behavioral sciences. pp. 68-75, McGrawHill, New York.
Sotelo, J. R., and K. R. Porter 1959 An electron
microscope study of the rat ovum. J. Biophys.
Biochem. Cytol.. 5: 327-342.
Telfer, William H. 1954 Immunological studies
of insect metamorphosis 11. The role of a sexlimited protein in egg formations by the
cecropia silkworm. J. Gen. Physiol., 37: 539558.
Telfer, William H., and Melvin E. Melius, Jr.
1963 The mechanism of blood protein uptake
by insect oocytes. h e r . Zool., 3: 185-192.
Trujillo-Cenoz, O., and J. R. Sotelo 1959 Relationships of the ovular surface with follicle
cells and origin of the zona pellucida in rabbit
oocytes. J. Biophys. Biochem. Cytol., 5: 347350.
Yamada, E., T. Muta, A. Motomura and H. Koga
1957 The fine structure of the oocvte in the
mouse ovary studied with electron microscope.
The Kurume Medical Journal, 7:148-160.
PLATES
PLATE 1
EXPLANATION OF FIGURES
With the exception of figure 5, the mouse ovarian sections pictured
were from mice injected intravenously with human serum albumin
(HSA). Indirect fluorescent antibody techniques were used; the sections
were treated with rabbit anti HSA absorbed with mouse serum (MS
absorbed-anti HSA) followed by fluorescein-labelled sheep anti rabbit
globulin serum (anti R*). Fluorescence indicates the presence of antigens
similar to or identical with the injected foreign protein.
4
Follicle Stages I and VI. Compare and contrast the intensity of
ooplasmic fluorescence at these stages in this animal. The granulosa
cell cytoplasm was fluorescent also. X 250.
5 Control section from ovary of uninjected mouse treated with M S
absorbed-anti HSA followed by anti R*. The section included many
antrum-containing follicles of which a Stage VI follicle was present
in the field pictured. Antibody and photographic treatment for tissue
pictured in figures 5 and 6 was identical. The differences between
these figures illustrate the specacity of the immunohistological
reaction. X 125.
6
Stage IX follicle from HSA-injected mouse. Note intense fluorescence
of ooplasm and the distribution of follicular fluorescence in the rows
of granulosa cells near the antrum. X 125.
7,8 Ovarian follicles from the same HSA-injected mouse (different
mouse than fig. 4). Compare and contrast the intensity of cytoplasmic fluorescence in the Stage I, I11 and IV (fig. 7) and IX
(fig. 8 ) oocytes pictured. Compare, too, the intensity and location
of fluorescence in the granulosa cells at these stages. Photographic
treatment was identical for figures 7 and 8. X 250.
152
ANTIGEN AND ISOTOPE TRANSFER TO OVARY
Laurel E. Glass and Jean M. Cons
PLATE 1
153
PLATE 2
EXPLANATION OF FIGURES
Autoradiographs of mouse ovarian follicles from mice injected systemically with human serum albumin labelled with 1131. Ovaries were
Carnoy-fixed, paraffin-embedded and sectioned serially at 5 p. Liquid
emulsion autoradiography was used. These figures illustrate relative grain
densities overlying the cytoplasm of oocytes a t various follicle stages;
however, conclusions were dependent on statistical analysis of Sign Test
comparisons of grain count data as described i n text. Section thickness
and overstaining prevented demonstration i n black and white pictures of
grains overlying granulosa cells.
9
Cluster of Stage I and I1 follicles. Contrast grain density overlying
ooplasm with near absence of grains overlying Stage I11 oocytes i n
figures 11 and 12.
10
Late Stage V (or early Stage VI) follicle. Grain density very similar
to that overlying oocyte in Stage VI follicle i n figure 11 and the
Stage VIII follicle in figure 13.
11
Stage 111, IV and VI follicles. Contrast the characteristic paucity of
grains overlying the Stage 111, the few grains overlying the Stage I V
and the several grains overlying the Stage V I ooplasm.
12
(Inset). Another Stage I11 follicle.
13 Oocyte and follicle cells from Stage VIII follicle. Grain density overlying ooplasm is roughly the same as that of the Stage V and VI
oocytes in figures 10 and 11 respectively.
14 Oocyte from Stage IX follicle. Grain density overlying cytoplasm
slightly higher than that over oocytes in antrum-containing follicles
pictured in figures 10, 11 and 13.
154
ANTIGEN AND ISOTOPE TRANSFER TO OVARY
Laurel E. Glass and Jean M. Cons
PLATE 2
155
Документ
Категория
Без категории
Просмотров
2
Размер файла
1 253 Кб
Теги
foreign, radiolabeled, systemically, injected, stage, ovarian, transfer, protein, antigen, mouse, dependence, follicle
1/--страниц
Пожаловаться на содержимое документа