- Email: [email protected]
Quantitative analysis of the spreading of the mouse trophoblastin vitro: a model for early invasion
Placenta (1996), 17, 583-590
Quantitative Analysis of the Spreading of the Mouse Trophoblast In Vitro: a Model for Early Invasion A. S u e n a g a a, C. Tachi b'c, H. Tojo b, S. T a n a k a b, O. T s u t s u m i a and Y. T a k e t a n i a a Department of Obstetrics and Gynecology, School of Medicine, The University of Tokyo, Hongo, Bunkyo-ku, Tokyo 11 3, Japan b Laboratory of Applied Genetics, Department of Animal Resource Science, School of Agriculture and Life Science, The University of Tokyo, Yayoi, Bunkyo-ku, Tokyo 113, Japan Paper accepted 9 May 1996
The outgrowth of the mouse blastocyst in culture represents an in vitro model of trophoblastic invasion. In the present study we analysed trophoblast spreading by time lapse video microscopy. Trophoblast spreading consists of (1) the migration and (2) the giant cell transformation of trophoblast cells, (3) the proliferation of ectoplacental cone (EPC) cells and (4) the subsequent transformation of EPC cells into the secondary giant cells. During migration, ruffling of the trophoblast cell membrane is followed by the formation of lamellipodia. The mean surface areas of the spreading trophoblast, measured in more than 100 cultured blastocysts, increased linearly from 48 to 96 h of culture, while the linear migratory speed at the periphery of the outgrowth declined as the time of culture advanced. The EPC cells increased in size approximately eightfold during the giant cell transformation. The apparent nuclear:cytoplasmic ratios, i.e., ratios between the size of nucleus and that of the cytoplasm, measured as the surface areas on the photomicrographs, of EPC cells increased between 40--46 h of culture, but a sharp decline in the ratio occurred between 50 and 51 h of culture, reflecting either the sudden and tremendous increase in the cellular volume and/or spreading of the cytoplasm. The rates of trophoblast spreading varied considerably among the blastocysts of different genetic constitution examined (ICR, C57BL/6, C3H/He and (B6 x C3)F1. It was fastest in blastocysts obtained from matings of males and females of (B6 x C3)F1, and slowest in the C57BL/6 embryos. The differences in the rate of outgrowth observed may not simply be ascribed to difference in the developmental speed of the early embryos, because the rate of outgrowth reached a plateau at about 96-120 h and no 'catch-up' was observed by leaving the blastocysts in culture longer. Our results strongly suggest the possible presence of genetic regulatory mechanisms underlying trophoblast outgrowth; further analysis of the phenomenon may provide clues to understand the molecular mechanisms of trophoblastic invasion during the early phase of implantation, hopefully leading to improved success rates of in vitro fertilization-embryo transfer. © 1996 W. B. Saunders Company Ltd Placenta (1996), 17, 583-590
INTRODUCTION
The percentage of pregnancies going to term from human in vitro fertilization and embryo transfer (IVF-ET), averages from 15-20 per cent (The American Fertility Society, 1992; Society for Assisted Reproductive Technology, 1993). Although the reasons for the relatively high failure rate of human 1VF-ET are not known, the peri-implantation period is thought to be crucial, as indicated by the high incidence of the post-implantation intrauterine deaths (Leridon, 1977) and 'chemical' or 13-human choronic gonadotrophin (13-hCG') pregnancies (Miller et al., 1980; Lenton, Neal and Sulaiman 1982; Edmonds et al., 1982; Jones et al., 1983). Thus to achieve greater success through the assisted reproductive technologies, more basic research on cellular and molecular bases of implantation must be done. To whom correspondence sliould be addressed. 0143-4004/96/080583 + 08 $12.00/0
The cellular processes of implantation may be grossly classified into three distinct consecutive stages, on the basis of ultrastructural features (Ports, 1966, 1968, 1969; Enders, 1964; Enders & Schlafke,. 1967; Reinius, 1967; Tachi, Tachi and Linder, 1970), i.e., the early attachment stage, the late attachment stage and the invasion stage (Tachi, Tachi and Linder, 1970). Each of these stages involves surprisingly complex biological events and the failure of any single step may lead to the failure of implantation. Since the earliest period of gestation when I3-hCG becomes detectable in the serum of pregnant women (day 10 or 11 counting the day of lutenzing hormone LH peak as day 0) approximately corresponds to the early phase of trophoblastic invasion (Lenton, Neal and Sulaiman, 1982), it may be suspected that a considerable number of the post-implantation losses of conceptuses might be accounted for by the failure of trophoblastic invasion, although no definitive data demonstrating this idea are available in the literature. © 1996w. B. SaundersCompanyLtd
Placenta (1996), Vol. 17
584
J.nvasion of the endometrium by trophoblast cells is an extremely intricate cellular process, and probably one of the most critical episodes in the early phase of haemochorial placentation. Recently, the regulatory mechanisms underlying trophoblastic invasion have attracted the attention of researchers (for recent reviews, see Strickland and Richards, 1992; Graham and Lala, 1992; Fisher and Damsky, 1993; Enders and Welsh, 1993; Hoffman and Wooding, 1993; Damsky et al., 1993; McMaster, Bass and Fisher, 1994), though many important aspects of this complex phenomenon remain to be clarified. Because investigation of the invasion phase of implantation in vivo is difficult, studies using an appropriate in vitro model are essential. We have developed an in vitro culture system suitable for the analysis of trophoblast spreading which is considered to represent the early phase of the trophoblastic invasion in vitro (Tachi, 1992). Using this system, we analysed, temporally and quantitatively, the morphological changes associated with the trophoblast spreading by the time-lapse video microscopy. Furthermore, we examined the possible presence of genetic control mechanisms in trophoblast spreading by comparing the process using blastocysts from different strains of mice.
MATERIALS A N D M E T H O D S Animals Mice of ICR, C57BL/6, C3H/He strains and (B6 x C3)F1 were purchased from a local dealer (Sankyo Lab Service & Co., Tokyo, Japan). They were housed under controlled temperature and lighting cycles (12 h light and 12 h dark) until death. Female mice approximately 5 weeks old were injected with 5 IU of pregnant mare's serum gonadotropin (PMSG; Teikoku Hormone Manufacturing & Co., Tokyo, Japan) and 48h later with 5 IU of hCG (Teikoku HormOne Manufacturing). Immediately after the hCG injection, they were caged with male mice of proven fertility. The day when a copulatory plug was found, was counted as Day 1 of pregnancy.
Culture media Powdered Eagle's minimal essential medium (EMEM) without bicarbonate, glutamine, antibiotics and phenol red, was purchased from Nissui & Co. (Tokyo, Japan), and a concentrated solution ( x 100) of non-essential amino acid, from GIBCO Oriental & Co. (Tokyo, Japan). Fetal bovine serum (FBS) was obtained from Nichirei & Co. (Tokyo, Japan), heat-inactivated at 56°C for 30 min and stored frozen at - 2 0 ° C until use. Other chemicals were purchased from Wako Pure Chemical Indusl~ries, Osaka. Powdered EMEM was dissolved in quartzdistilled water and added NaHCO 3 (2.0 g/l), L-glutamine (292mg/1), and the non-essential amino acid solution
(10ml/1). The other components added were as follows: cysteine HCI'HzO (200 mg/l), sodium pyruvate (110 mg/1), ascorbic acid (50 rag/I), co-carboxylase (1.0 mg/l), thymidine (5.0mg/I), uridine (5.0mg/l), penicillin (100000IU) and streptomycin (50mg/l). This medium was designated 'F0 medium' (Tachi, 1992) in our laboratory and will be referred to as such in the following. CMRL1066 medium without L-glutamine was purchased from GIBCO Oriental. Immediately before the experiments, F0 medium was mixed with an equal volume of CMRL1066 medium (F0-CMRL). Then, heat-inactivated FBS was added to F0-CMRL medium at a concentration of 20 per cent (v/v) (F0-CMRL/20). This medium was used as the standard medium for culture of mouse blastocysts in all experiments.
Culture of blastocysts Mice were killed by cervical dislocation in the afternoon of day 4 of pregnancy. Blastocysts were collected by flushing the excised uterus with PBS, and were cultured in quartz wells (Tachi, 1992) prepared as follows. A quartz ring (HayashiRikagaku Co. Ltd., Tokyo) was laid on the bottom of a Falcon dish (3001, Becton, Dickinson Overseas Inc., Tokyo) and fastened to the centre of the dish with paraffin of high melting point (62°C). Although the specially-treated glass surface was used as the substratum in the original method described by Tachi (1990), the plastic surface of the Falcon dish was used as the substratum for the cultured embryos in the present series of experiments. The inside of the well was washed with a single change of F0, and filled with 200 lal of F0-CMRL/20. Then, liquefied paraffin of low melting point (approximately 35"C) was pored into the dish around the well so as to prevent the seepage of the medium. Before the blastocysts were added, the dishes with the medium in the well were incubated at 37"C under an atmosphere of 5 per cent CO z in air for about 30 min.
Time-lapse video recordings For time-lapse video recording, the blastocysts were cultured at 37"C, under an atmosphere of 5 per cent CO 2 in air using a specially-designed culture chamber (IMT2-IBC, Olympus, Tokyo, Japan) fastened to the specimen stage of an inverted phase contrast microscope (IMT2, Olympus). The images of the cultured blastocysts were video-recorded at the rate of one frame per 4 sec using a model BR-9060 time-lapse video recorder (Victor & Co., Tokyo, Japan).
Photomicrography Photomicrographs were taken using an inverted microscope (Model IMT2, Olympus) equipped with an automated camera (Model C35AD4, Olympus). The surface areas of the
Suenaga et al.: Mouse BlastocystsIn Vitro
585
@
.@
19:2.~ 2,E
20 00 41
:~,
O 21 C2~ EZ
~, .t:;:~:.'~i.
.'~-~ ;,
S:CO:/-'S 9.3- B - S
.
'
fi~ e~
7:DO
", ~
I':0D:4'5 $3- 6 9
Z~_ r-E' 7-v
..... "t~
5 CD 2~ S~-- 6 - S
"~
E ~uZ E2 93 E ~
.
.$b?
44
9 3 - E- 8
~0
6:0D:33 93- 6= 9
o
9 D0:35
$3- 6
9
.:,
IC,:CD 4-~:
E%- ,S- ,~
,:,,
Representativephotomicrographsfrom the framesof time-lapsevideorecordingsof a blastocystcultured in vitro, showingthe process of trophoblast spreading. The recordingswere started after 24 h of pre-incubation(frame 1). Intervalbetweenframe 1 and 2, 30 min; intervalsbetweenthe rest of consecutive frames, 2 h. One of the early front of migration(0. Ruffling(frame 2 and 3) is followedby formationof lamellipodiaand migrationof the cells(frame 4-7). The migratoryactivitygraduallyspread all around the embryo, i, inner-cellmass;e, ectoplacentalcone. Actualtime of the day when the imagesamplingwas done is indicated at the bottom of each frame (hour: minute:second). Scale indicates 100 ~tm. F i g u r e 1.
trophoblast spreads were quantitatively evaluated using a digitizer tablet (Model DT1000, Watanabe Sokki, Tokyo, Japan) connected to a personal computer (Model 9801, NEC, Tokyo r Japan).
RESULTS
Sequence
of events
of trophoblast
spreading
Time-lapse video recordings of cultured blastocysts were started after 24 h of pre-incubation; in the following, the time of culture is counted from the time of initiation of preincubation. By this time, the blastocysts had already hatched from the zona pellucida and lightly adhered to the substratum. Although slight ruming of the edge of the trophoblast was noted, no migration of trophoblast cells had yet began (Figure 1, frame no. 1). The first sign o f trophoblastic migration
(primary giant trophoblast cells) was seen at 26 h (corresponds to day 5 afternoon in vivo) (Figure 1, frame 2). The ruffling of the trophoblast cell membrane and the initiation of migratory behaviour of the cells gradually spread along the entire edge of the blastocyst by 32 h (Figure 1, frame no. 9) (corresponds to day 5 evening in vivo). After ruffling of the cell membrane, the trophoblast cells extended large lamellipodia; then they gradually migrated away from the blastocyst. The inner-cell mass (ICM) continued to grow covered with trophoblast cell layer until about 40 h (Figure 1, frame no. 16) (see Tachi, 1992 and Naruse, Somata and Shoji, 1982); it emerged from the trophoblast cell layer at around 44-48 h [Figure 2(A), frame no. 1] (corresponds to day 6 afternoon in vivo). Differentiation of the germ layers in the ICM becomes recognizable between 48 and 52 h of culture [Figure 2(A), frame no. 1 and 2]. After 54 h [Figure 2(A), frame no. 4], the two
Placenta (1996), Vol. 17
586
(a!. +:~
"
":')i
-
"
i
....: : ,L.
+
++
'
+
+
'
Y:~',~:~,~::~+i
:
"
'
+~.~ ,,y ", +'-
+
!
(
/
-'~+-~;,'.&,,-:- . ' "
:
~ c ~::~,,::---
~/~
':':-~'
"
.
',+,.,~+',(.) -," ' ,u
' (:~
,7"
i
i¸
,
,
~,
,
(b)
,,.. ~.,., ..... c?-~~',;t~ .......
.
. ..,,~i,~j.;4<.: "
"~':~'~,-.. u:..~,,-~.& ':'K' ~"i..
,~J~L;-~'+~:,';L , ._'.>.~ . ~ .~
'::,u ....
:',- ;;
;C2.j..,I:,,:
\-+-,t~~
~ ,
7
.
...:,/ ..
~ .~,.->;,.,~,~;f~. ~<~ .'..-%~+'~i+'~.. , ,-&.'' ,~.:4".~+; ,', .
, .
~.+v.,'.y.',:'-,~.W~.~-, "~ :~,++%"+>~:,i::~ ' ~, .'
'
. ; "~:i" '-~"-:
~ "
I!iF--Z;
iZ
?..~ ..
~
~<--~---_--
•
::
:.... ..;~:, .
i , :_+~_7:__.<~ -" . .:~,. ,
"*"
•
:+' :
!
'
--
-:.?~..-I
,~
: ,.- :
": :?~+'!"h:~'i
..~_~-:..:?~5.
i
.
.~
Figure 2. Representative photomicrographsfrom the frames of time-lapse video recordings of a blastocyst cultured 0 h, showing the process of germ layer differentiation(a) and growth of the embryoniccylinder (b). Recordings of this series were started at 48 h of culture (for the definition of the time of culture, see the legend for Figure 1). Interval between frame I and 2, 1 h and 10 min: intervals between the rest of consecutive frames, 2 h. The inner-cell mass emerges from the overlyingtxophoblast(frame 2), and the primitive endoderm becomes recognizable(frame 3). The well-developedectoplacentalcone supports the growth and differentiationof the cylinder (c). e, ectoplaeentalcone; i, inner-cell mass; g, giant trophoblast cells. Scale indicates 100 Ftm.
germ layers, the endoderm and the epiblast, were clearly visible. By this time, the ectoplacental cone (EPC) is well developed. In / the EPC, vigorous cellular movement could be observed reflecting active proliferation and migratory activity of the cells in this region. T h e primitive endoderm in the I C M increased its thickness from 6"2 h [Figure 2(A), frame no. 8] on, and the formation of a cylinder was observed around 72 h (corresponds to day 7 afternoon in vivo). T h e cylinder grew in size markedly
between 80 to 92 h [Figure 2(B), frame no. 17-20] (corresponds to day 8 morning in vivo). Meanwhile, the entire surface area of the trophoblast spread continued to expand during this period. Some EPC cells were seen to migrate toward the edge of the trophoblast spread and to differentiate into giant trophoblast cells. After 96 h of culture (corresponds to day 8 afternoon in vivo), the giant trophoblast cells tended to dissociate from each other forming a mesh-like structure, probably representing the
Suenaga et al.: Mouse Blastocysts In Vitro
587
3.0 1.0 2.5 0.8 2.0 0.6
x 1.5
0.4 r/2
1.0
0.2
0.5
I
0
24
48 72 96 Time in culture (h)
120
Figure 3. Changes in the surface area of the trophoblast outgrowths according to the time of culture. Values are the means of more than 100 independent measurements. Vertical bars indicate s.d.
formation of the spongiotrophoblast. At the same time, both the size and the number of trophoblast cells increased remarkably.
Morphornetric analysis of t r o p h o b l a s t outgrowths The contours of the trophoblast spread were traced from the stills taken at 24 h intervals. The surface area of the outgrowths was measured by planimetry. As shown in Figure 3, the surface area increased linearly from 24 h to about 100 h of culture. Similar tracings were made from the time-lapse video cinematographs taken at every 2 h of culture and the changes in the surface area are graphically presented in Figure 4. A good linear relationship was observed between the surface area and the time of culture.
Giant cell transformation of trophoblast cells The cells located at the edge of the EPC were seen to be transformed into the secondary giant cells after 38 h of culture (Figure 1, frame no. 15, 16); the size of both cytoplasm and the nucleus increased tremendously. After about 72 h of culture, however, the giant trophoblast cells ceased to increase in cellular volume. In Figure 5, the process of giant cell transformation of small EPC cells is shown in sequential frames from the video recordings. As clearly seen in this figure, a substan-
0
24
36 48 Time in culture (h)
60
72
Figure 4. Changes in the surface area of the outgrowths as measured on the frames of time lapse video recordings of a single cultured blastocyst according to the time in culture.
tial increase in the size of cytoplasm occurred, preceded by an increase in nuclear size; the changes in the cytoplasmic and the nuclear sizes were measured on the photographs and the results are graphically shown in Figure 6. The ratio between the nuclear size and that of cytoplasm measured as the two-dimensional surface areas increased from 0.05 at 40 h to 0.35 at 50 h of culture, i.e., within 10 h after their emergence as small EPC cells. Thereafter, the ratio decreased slightly. After 60 h of culture, the microscopic appearance of the cells exhibited no sign of further changes either in the cytoplasm or in the nucleus.
Genetic differences in the rate of trophoblast o u t g r o w t h according to the strain of the mouse The rates of trophoblast outgrowth of blastocysts of different genetic constitution, i.e., ICR, C57B1/6, C3H, (B6 x C3)F1, (C3 x B6)F1 and (B6 x C3)F2, were quantitatively analysed. The results are shown in Figure 7. It was revealed that a clear difference exists in the rates of trophoblast outgrowth, according to the genetic constitution of the embryos. Among the blastocysts examined, those of inter-strain crosses, i.e., (B6 x C3)F1, (C3 x B6)F1 and (B6 × C3)F2 exhibited a significantly faster rate of the trophoblast spreading compared with the embryos of the inbred strains. Particularly, the rate of trophoblast spreading in (B6 × C3)F1 blastocysts was greater than any others. Conversely the rate of trophoblast spreading in the B6 blastocysts was slowest among the six different genetic constitutions of embryos examined.
588
mmmm///
Placenta (1996), Vol. 17
Figure 5. Photomicrographs showing the process of giant trophoblast cell formation. Two small EPC cells (frame 2) gradually increase their size (frame 4-6). Then, the cells increase their size abrupdy and tremendously, pushing the peripheral cells outward (frames 7-10). e, ectoplacental cone. Scale indicates 100 ~m.
0.35 0.6 0.3
--
0.5
o 0.25-
~
~
I. 0.4
0.2-
0.15
0.3
-
Z 0.1-
0.2
0.05 -
0.1
0---ll
I
40
q
I
44
~
I
~
I
~
I
q
I
48 52 56 60 Time in culture (h)
Figure" 6. A representative graph showing changes in the ratios between the size of nucleus and that of the cytoplasm (nuclear:cytoplasmic ratios), measured as the surface areas on the photomicrographs, in a trophoblast cell undergoing giant--celltransformation, according to the time in culture.
I
0
24
S I
I
I
48 72 96 Time in culture (h)
I
120
Figure 7. The rate of trophoblastic spreading in blastocysts of different genetic constitution. (O), (B6 x C3)F2; (O), ICR; (11), (B6 x C3)FI; (A), C57BL/6; (~), (C3 x B6)F1; (O) C3H. Vertical bars indicate s.d. All the measurements were actually done on the hour, although some of the plots are slightly dislodged to either the plus or the minus direction so as to avoid overlapping of the s.d. bars.
DISCUSSION T r o p h o b l a s t spreading or outgrowth, which takes place in the mouse blastocyst cultured in vitro, is considered to reflect the early process o f trophoblastic invasion in vivo (e.g. Gwatkin, 1966; Spindle & Pedersen, 1973; Kanai-Azuma et al., 1993; Koi et al., 1995). It consists of, as clearly seen in the timelapse video recordings, several different cell biological phenomena; i.e., (1) migration and (2) giant cell transformation
o f trophoblast cells, (3) proliferation o f E P C cells and (4) the subsequent transformation o f E P C cells into secondary giant cells. D u r i n g the process o f migration, ruffling o f t h e cell m e m brane, which is induced by the interaction o f the trophoblast cells with the substratum is followed by the extension o f lamellipodia. T h e general sequence o f events conforms with
Suenaga et al.: Mouse BlastocystsIn Vitro that described for other cells cultured in vitro (e.g., Abercrombie, Heaysman and Pegrum, 1970; Eckstein and Shur, 1989; Stossel, 1993; Appedu and Shur, 1994). The mean surface areas of trophoblast outgrowths, measured on more than 100 cultured blastocysts, increased linearly from 48 h to 96 h of culture; the results agreed well with those reported previously by Tanaka, Tojo and Tachi (1993) and by Koi et al. (1995). The contour of the trophoblast outgrowths as observed in the sequential frames of video recordings, showed that spreading takes place equally all around the cultured blastocysts and the linear migratory speed at the periphery of the spread declined as the time of culture advanced (data not shown). The detailed sequential study of the process revealed the possible presence of two phases in the outgrowth process. As shown in Figure 4 after 38-40 h of culture, the rate of increase in the surface area slowed down slightly and the contour of the spread became smoother; this phase of trophoblast outgrowth may be designated as the primary phase. Thereafter, the 'secondary' phase of spreading began. During this phase, cells migrating from the EPC contributed to the trophoblast spreading. Further analysis of the different phases of trophoblastic spreading at a molecular level, such as the possible involvement of surface galactosyl transferase in the maintenance of the stability of lamellipodia (Appedu and Shur, 1994) or effects of fibronectin or laminin upon the proliferation and migration of EPC cells (Koi et al., 1995), will give us crucial knowledge to understand the mechanisms of trophoblastic invasion in vivo. In the present report, the morphological sequence of events of giant cell transformation was documented (Figure 5 and 6). The EPC cells increased their size approximately eightfold during the process. The apparent nuclear:cytoplasmic ratios of EPC cells calculated from the measurements of the areas occupied by the nucleus and the cytoplasm on the pictures exhibited linear increases from 40 to 46 h of culture. A sharp decline in the ratio took place between 50 and 51 h of culture, reflecting either the sudden and tremendous increase in the cellular volume and/or spreading of the cytoplasm. Although the measurement of thickness of living cells is notoriously
589
difficult, further analysis of the changes in actual nuclear and cellular volumes might yield important information concerning the mechanisms of giant cell transformation. Although the giant cell transformation of trophoblast cells is one of the most striking cellular changes which accompanies trophoblast outgrowths, relatively little is known about the mechanisms underlying this phenomenon. Kanai-Azuma et al. (1993) proposed that while insulin-like growth factor-I (IGF-I) stimulates EPC cell proliferation and migration, IGF-II enhances giant cell transformation of the trophoblast cells. Their observations, however, were based upon study of the EPC cells collected from the gestational uteri of day 10.5 of pregnancy. It might be possible that each of the different steps of trophoblast outgrowth is governed by different growth factors and/or cytokines. A curious observation was made in the present work on the differences in the rate of trophoblast outgrowth according to the genetic background of the blastocysts. Thus, the rate of trophoblast outgrowth was fastest in blastocysts obtained by mating males and females of (B6 x C3)F1, and slowest in the embryos of B6. While genetic variation in the speed of development of cleavage stage embryos has been well-documented (Whitten and Dagg, 1962; McLaren & Bowman, 1973; Verbanac & Warner, 1981),-the differences in the rate of outgrowth observed in the present work may not simply be ascribed to difference in the developmental speed of the early embryos, because the rate of outgrowth reaches a plateau at about 96-120 h and leaving the blastocysts in culture longer does not lead to a 'catch-up' in the surface areas of outgrowths. Although the mechanisms underlying the observed differences are entirely unknown, our results strongly suggest the possible presence of genetic regulatory mechanisms underlying trophoblast outgrowth. Because trophoblast spreading is, as mentioned before, generally considered to be an in vitro model of trophoblastic invasion in vivo, further analysis of the genetic regulatory mechanisms underlying trophoblast outgrowth may provide important clues to understand the molecular mechanisms of trophoblastic invasion during the early phase of implantation, hopefully leading to improved success rates of IVF-ET.
ACKNOWLEDGEMENTS
We thank ProfessorJerome F. Strauss I11,Centre for Research on Women'sHealth and Reproduction,Departmentof Obstetricsand Gynecology,Universityof Pennsylvania,for reading the manuscriptand for makingvaluablesuggestions.This work is financiallysupported in part by Grants-in-Aidfrom Ministry of Education, Culture and Sports of Japan (#04556040, #06281107 to C.T.). A.S. is a recipientof JSPS (Japan Society for Promotionof Science) fellowship.
REFERENCES
Abercrombie, M., Heaysman, J. E. M. & Peg'rum, S. (1970) The locomotion of fibroblasts in culture. I. Movement of the leading edge. Ea'perimental Cell Research, 59, 393-398. Appeddu, P. A. & Shut, B. D. (1994) Control of stable lamellipodia formationby expressionof cell surface [[31,4-galactosyltransferasecytoplasmic domains.Journal of Cell Science, 107, 2535-2545.
Behrendtsen, O., Alexander, C. M. & Werb, Z. (1992)Metalloproteinases mediate extracellularmatrix degradation by cells from mouse blastocyst outgrowths. Development, 114, 3447-456. Dallenbaeh, F. D. (1975)Woraufberuhtdie Fahigkeityon Trophoblastzellen wie Krebszelleninvasivzu wachsen?Verhandlungender Deutschengesellschaft fur Inhere Medizin (Munchen), 59, 573. Damsky, C., Sutherland, A. & Fisher, S. (1993) Extracellularmatrix 5: adhesive interactionsin early mammalianembryogenesis,implantation,and placentation.FASEBJournal, 7, 1320-1329.
590
Eckst~in, D.J. & Shur, D. D. (1989) Laminin induces the stable expression of surface galactosyltransferase on lamellipodia of migrating cells. Journal of Cell Biology, 108, 2507-2517. " Edmonds, D. K., Lindsay, K. S., Miller, J. F., Williamson, E. & Wood, P. J. (1982) Early embryonic mortality in women. Fertility and Sterility, 38, 447-453. Enders, .A.C. (1964) Electron microscopy of an early implantation stage with a postulated mechanism of implantation. Developmental Biology,, 10, 395410. Enders, A. C. (1969) Cytological aspects of trophohlast-uterine interaction in early implantation. American Journal of Anatomy, 125, 1-29. Enders, A. C. (1974) Surface coats of the mouse blastocyst and uterus during the preimplantation period. Anatomical Record, 180, 31-45. Enders, A. C. (1975) Cellular basis interaction between trophoblast and uterus at implantation. Biolog), of Reproduction, 12, 41-65. Enders, A. C. & Schlat~e, S. J. (1967) A morphological analysis of the early implantation stage in the rat. American Journal of Anatomy, 120, 185-226. Enders A. C. & Welsh, A. O. (1993) Structural interactions of truphoblast and uterus during hemochorial placenta formation. Journal of Experimental Zoology, 266, 578-587. Fisher, S. J. & Damsky, C. H. (1993) human cytotrophoblast invasion. Semin. Cell Biology, 4, 183-188. Graham, C. H. & Lala, P. K. (1992) Mechanisms of placental invasion of the uterus and their control. Biochem. Cell Biology, 70, 867-874. Gwatkin, R. B. L. (1966) Amino acid requirements for attachment and outgrowth of the mouse blastocysts in vitro. Journal of Cell Physiologd,, 68, 335-344. Hoffman, L. H. & Wooding, F. B. (1993) Giant and binucleate trophoblast cells of mammals. Journal of Experimental Zoology, 266, 559-577. Jones, H. W., Jr., McDowell, J., Acosta, A. A., Sandow, B. A., Andrews, M. C., Veeck, L., Garcia, J. E., Whibley, T. W., Jones, G. S., Wilkes, C. A., Mantzavinos, T. & Wright, G. L., Jr. (1983) What is a pregnancy? A question for programs of in vitro fertilization. Fertility and Sterilio,, 40, 728-733. Kanai-Azuma, M., Kanai, Y., Kurohmarn, M., Sakai, S. & Hayashi, Y. (1993) Insulin-like growth factor (IGF)-I stimulates proliferation and migration of mouse ectoplacental cone cells, while IGF-II transforms them into trophoblast giant cells in vitro. Biology of Reproduction, 48, 252-261. Koi, H., Taehi, C., To)o, H., Kubota, T. & Aso, T. (1995) Effects of matrix proteins and heparin-binding components in fetal bovine serum upon the proliferation of ectoplacental cone cells in mouse blastocysts cultured in vitro. Biology of Reproduction, 52, 759-770. Lenton, E. A., Neal, L. M. & Sulaiman, R. (1982) Plasma concentrations of human chorionic gonadotropin from the time of implantation until the second week of pregnancy. Fertility and Sterilio,, 37, 773-778. Leridon, H. (1977) Human Fertility: The Basic Components Chicago, The Chicago University Press. Librach, C. L., Werb, Z., Fitzgerald, M. L., Chiu, K., Corwin, N. M , Esteves, R. A., Grobelny, D., Galardy, R., Damsky, C. H. & Fisher, S. J. (1991) 92-kD type IV collagenase mediates invasion of human cytotrophoblasts. Journal of Cell Science, 113, 437-449.
Placenta (1996), Vol. 17 MeLaren, A. & Bowman, P. (1973) Genetic effects on the timing of early development in the mouse. Journal of Embryology, atul Experimental Morpholog3,, 30, 491-498. McMaster, M. T., Bass, K. E. & Fisher, S. J. (1994) Human trophoblast invasion. Autocrine control and paracrine modulation. Ann. AT. Y. Aaulemy of Sciet~ces, 734, 122-131. Miller, J. E., Williamson, E., Glue, J., Gordon, Y. B., Grdzinskas, J. G. & Sykes, A. (1980) Fetal loss after implantation: a prospective study. Lancet, 2, 554-556. Naruse, I., Somata, S. & Shoji, R. (1982) Mammalian embryos culture from the pre-implantation stage, and analysis of developmental failures due to chromosomal abnormalities in vitro. Cong. Anom., 25, 319-326. Potts, D. M. (1968) The ultrastructure of implantation in the mouse. Journal of Anatomy, 103, 77-90. Potts, M. (1966) The attachment phase of ovoimplantation. American Journal of Obstetrics and Gynecologl,, 96, 1122-1128. Potts, M. (1969) The ultrastructure of egg implantation. Advances in Reproductive Plo,siolog~,,4, 241-267 Society for Assisted Reproductive Technology (1993) Assisted reproductive technology in the United States and Canada: 1991 results from the Society for Assisted Reproductive Technology generated from the American Fertility Society Registry. Fertility and Sterility, 59, 956-962. Spindle, A. I. & Pedersen, R. A. (1973) Hatching, attachment, and outgrowth of mouse blastocysts in vitro: fixed nitrogen requirements. Journal of Experimental Zoologj,, 186, 305-318. Stossel, T. P. (1993) On the crawling of animal cells. Science, 260, 1086-1094. Strickland, S. & Riehards, W. G. (1992) Invasion of the trophoblasts. Cell, 71,355-357 Tachi, C. (1992) Partial characterization of macromelecular components in fetal bovine serum required for development of mouse blastocysts cultured in vitro. Development Growth & Differentiation, 34, 69-77. Tachi, S., Taehi, C. & Lindner, H. R. (1970) Uhrastructural features of blastocyst attachment and trophoblastic invasion in the rat. Journal of Reproduction and Fertilio,, 21, 37-56. Tantaka, S. S., To)o, H. & Tachi, C. (1993) Possible involvement of amiroride-sensitive N a + / H + exchanger and Na+ transport systems in hatching and the trophoblast spreading in mouse blastocysts cultured in vitro, Journal of Reprodnction and Development, 39, 97-107. The American Fertility Society (1992) In vitro fertilization-embryo transfer (IVF-ET) in the United States: 1990 results from the IVF-ET registry. Fertility and Sterilio, , 57, 15-24. Verbanac, K. M. & Warner, C. M. (1981) Role of the major histocompatibility complex in the timing of early mammalian development. In Cellular and Molecular Aspects of Implantation (Eds) S. R. Glasser and D. W. Bullock pp. 467-472, New York, Plenum Press. Whitten, W. K. & Dagg, C. P. (1962) Influence of spermatozoa on the cleavage rate of mouse eggs. Journal of Experimental Zoolo~,, 148, 173-183. Wegmann, T. (1992) Immune signalling at the maternal fetal interface and trophoblast differentiation. Developmentaland Comparative Immunology, 16, 425-430.
Quantitative Analysis of the Spreading of the Mouse Trophoblast In Vitro: a Model for Early Invasion A. S u e n a g a a, C. Tachi b'c, H. Tojo b, S. T a n a k a b, O. T s u t s u m i a and Y. T a k e t a n i a a Department of Obstetrics and Gynecology, School of Medicine, The University of Tokyo, Hongo, Bunkyo-ku, Tokyo 11 3, Japan b Laboratory of Applied Genetics, Department of Animal Resource Science, School of Agriculture and Life Science, The University of Tokyo, Yayoi, Bunkyo-ku, Tokyo 113, Japan Paper accepted 9 May 1996
The outgrowth of the mouse blastocyst in culture represents an in vitro model of trophoblastic invasion. In the present study we analysed trophoblast spreading by time lapse video microscopy. Trophoblast spreading consists of (1) the migration and (2) the giant cell transformation of trophoblast cells, (3) the proliferation of ectoplacental cone (EPC) cells and (4) the subsequent transformation of EPC cells into the secondary giant cells. During migration, ruffling of the trophoblast cell membrane is followed by the formation of lamellipodia. The mean surface areas of the spreading trophoblast, measured in more than 100 cultured blastocysts, increased linearly from 48 to 96 h of culture, while the linear migratory speed at the periphery of the outgrowth declined as the time of culture advanced. The EPC cells increased in size approximately eightfold during the giant cell transformation. The apparent nuclear:cytoplasmic ratios, i.e., ratios between the size of nucleus and that of the cytoplasm, measured as the surface areas on the photomicrographs, of EPC cells increased between 40--46 h of culture, but a sharp decline in the ratio occurred between 50 and 51 h of culture, reflecting either the sudden and tremendous increase in the cellular volume and/or spreading of the cytoplasm. The rates of trophoblast spreading varied considerably among the blastocysts of different genetic constitution examined (ICR, C57BL/6, C3H/He and (B6 x C3)F1. It was fastest in blastocysts obtained from matings of males and females of (B6 x C3)F1, and slowest in the C57BL/6 embryos. The differences in the rate of outgrowth observed may not simply be ascribed to difference in the developmental speed of the early embryos, because the rate of outgrowth reached a plateau at about 96-120 h and no 'catch-up' was observed by leaving the blastocysts in culture longer. Our results strongly suggest the possible presence of genetic regulatory mechanisms underlying trophoblast outgrowth; further analysis of the phenomenon may provide clues to understand the molecular mechanisms of trophoblastic invasion during the early phase of implantation, hopefully leading to improved success rates of in vitro fertilization-embryo transfer. © 1996 W. B. Saunders Company Ltd Placenta (1996), 17, 583-590
INTRODUCTION
The percentage of pregnancies going to term from human in vitro fertilization and embryo transfer (IVF-ET), averages from 15-20 per cent (The American Fertility Society, 1992; Society for Assisted Reproductive Technology, 1993). Although the reasons for the relatively high failure rate of human 1VF-ET are not known, the peri-implantation period is thought to be crucial, as indicated by the high incidence of the post-implantation intrauterine deaths (Leridon, 1977) and 'chemical' or 13-human choronic gonadotrophin (13-hCG') pregnancies (Miller et al., 1980; Lenton, Neal and Sulaiman 1982; Edmonds et al., 1982; Jones et al., 1983). Thus to achieve greater success through the assisted reproductive technologies, more basic research on cellular and molecular bases of implantation must be done. To whom correspondence sliould be addressed. 0143-4004/96/080583 + 08 $12.00/0
The cellular processes of implantation may be grossly classified into three distinct consecutive stages, on the basis of ultrastructural features (Ports, 1966, 1968, 1969; Enders, 1964; Enders & Schlafke,. 1967; Reinius, 1967; Tachi, Tachi and Linder, 1970), i.e., the early attachment stage, the late attachment stage and the invasion stage (Tachi, Tachi and Linder, 1970). Each of these stages involves surprisingly complex biological events and the failure of any single step may lead to the failure of implantation. Since the earliest period of gestation when I3-hCG becomes detectable in the serum of pregnant women (day 10 or 11 counting the day of lutenzing hormone LH peak as day 0) approximately corresponds to the early phase of trophoblastic invasion (Lenton, Neal and Sulaiman, 1982), it may be suspected that a considerable number of the post-implantation losses of conceptuses might be accounted for by the failure of trophoblastic invasion, although no definitive data demonstrating this idea are available in the literature. © 1996w. B. SaundersCompanyLtd
Placenta (1996), Vol. 17
584
J.nvasion of the endometrium by trophoblast cells is an extremely intricate cellular process, and probably one of the most critical episodes in the early phase of haemochorial placentation. Recently, the regulatory mechanisms underlying trophoblastic invasion have attracted the attention of researchers (for recent reviews, see Strickland and Richards, 1992; Graham and Lala, 1992; Fisher and Damsky, 1993; Enders and Welsh, 1993; Hoffman and Wooding, 1993; Damsky et al., 1993; McMaster, Bass and Fisher, 1994), though many important aspects of this complex phenomenon remain to be clarified. Because investigation of the invasion phase of implantation in vivo is difficult, studies using an appropriate in vitro model are essential. We have developed an in vitro culture system suitable for the analysis of trophoblast spreading which is considered to represent the early phase of the trophoblastic invasion in vitro (Tachi, 1992). Using this system, we analysed, temporally and quantitatively, the morphological changes associated with the trophoblast spreading by the time-lapse video microscopy. Furthermore, we examined the possible presence of genetic control mechanisms in trophoblast spreading by comparing the process using blastocysts from different strains of mice.
MATERIALS A N D M E T H O D S Animals Mice of ICR, C57BL/6, C3H/He strains and (B6 x C3)F1 were purchased from a local dealer (Sankyo Lab Service & Co., Tokyo, Japan). They were housed under controlled temperature and lighting cycles (12 h light and 12 h dark) until death. Female mice approximately 5 weeks old were injected with 5 IU of pregnant mare's serum gonadotropin (PMSG; Teikoku Hormone Manufacturing & Co., Tokyo, Japan) and 48h later with 5 IU of hCG (Teikoku HormOne Manufacturing). Immediately after the hCG injection, they were caged with male mice of proven fertility. The day when a copulatory plug was found, was counted as Day 1 of pregnancy.
Culture media Powdered Eagle's minimal essential medium (EMEM) without bicarbonate, glutamine, antibiotics and phenol red, was purchased from Nissui & Co. (Tokyo, Japan), and a concentrated solution ( x 100) of non-essential amino acid, from GIBCO Oriental & Co. (Tokyo, Japan). Fetal bovine serum (FBS) was obtained from Nichirei & Co. (Tokyo, Japan), heat-inactivated at 56°C for 30 min and stored frozen at - 2 0 ° C until use. Other chemicals were purchased from Wako Pure Chemical Indusl~ries, Osaka. Powdered EMEM was dissolved in quartzdistilled water and added NaHCO 3 (2.0 g/l), L-glutamine (292mg/1), and the non-essential amino acid solution
(10ml/1). The other components added were as follows: cysteine HCI'HzO (200 mg/l), sodium pyruvate (110 mg/1), ascorbic acid (50 rag/I), co-carboxylase (1.0 mg/l), thymidine (5.0mg/I), uridine (5.0mg/l), penicillin (100000IU) and streptomycin (50mg/l). This medium was designated 'F0 medium' (Tachi, 1992) in our laboratory and will be referred to as such in the following. CMRL1066 medium without L-glutamine was purchased from GIBCO Oriental. Immediately before the experiments, F0 medium was mixed with an equal volume of CMRL1066 medium (F0-CMRL). Then, heat-inactivated FBS was added to F0-CMRL medium at a concentration of 20 per cent (v/v) (F0-CMRL/20). This medium was used as the standard medium for culture of mouse blastocysts in all experiments.
Culture of blastocysts Mice were killed by cervical dislocation in the afternoon of day 4 of pregnancy. Blastocysts were collected by flushing the excised uterus with PBS, and were cultured in quartz wells (Tachi, 1992) prepared as follows. A quartz ring (HayashiRikagaku Co. Ltd., Tokyo) was laid on the bottom of a Falcon dish (3001, Becton, Dickinson Overseas Inc., Tokyo) and fastened to the centre of the dish with paraffin of high melting point (62°C). Although the specially-treated glass surface was used as the substratum in the original method described by Tachi (1990), the plastic surface of the Falcon dish was used as the substratum for the cultured embryos in the present series of experiments. The inside of the well was washed with a single change of F0, and filled with 200 lal of F0-CMRL/20. Then, liquefied paraffin of low melting point (approximately 35"C) was pored into the dish around the well so as to prevent the seepage of the medium. Before the blastocysts were added, the dishes with the medium in the well were incubated at 37"C under an atmosphere of 5 per cent CO z in air for about 30 min.
Time-lapse video recordings For time-lapse video recording, the blastocysts were cultured at 37"C, under an atmosphere of 5 per cent CO 2 in air using a specially-designed culture chamber (IMT2-IBC, Olympus, Tokyo, Japan) fastened to the specimen stage of an inverted phase contrast microscope (IMT2, Olympus). The images of the cultured blastocysts were video-recorded at the rate of one frame per 4 sec using a model BR-9060 time-lapse video recorder (Victor & Co., Tokyo, Japan).
Photomicrography Photomicrographs were taken using an inverted microscope (Model IMT2, Olympus) equipped with an automated camera (Model C35AD4, Olympus). The surface areas of the
Suenaga et al.: Mouse BlastocystsIn Vitro
585
@
.@
19:2.~ 2,E
20 00 41
:~,
O 21 C2~ EZ
~, .t:;:~:.'~i.
.'~-~ ;,
S:CO:/-'S 9.3- B - S
.
'
fi~ e~
7:DO
", ~
I':0D:4'5 $3- 6 9
Z~_ r-E' 7-v
..... "t~
5 CD 2~ S~-- 6 - S
"~
E ~uZ E2 93 E ~
.
.$b?
44
9 3 - E- 8
~0
6:0D:33 93- 6= 9
o
9 D0:35
$3- 6
9
.:,
IC,:CD 4-~:
E%- ,S- ,~
,:,,
Representativephotomicrographsfrom the framesof time-lapsevideorecordingsof a blastocystcultured in vitro, showingthe process of trophoblast spreading. The recordingswere started after 24 h of pre-incubation(frame 1). Intervalbetweenframe 1 and 2, 30 min; intervalsbetweenthe rest of consecutive frames, 2 h. One of the early front of migration(0. Ruffling(frame 2 and 3) is followedby formationof lamellipodiaand migrationof the cells(frame 4-7). The migratoryactivitygraduallyspread all around the embryo, i, inner-cellmass;e, ectoplacentalcone. Actualtime of the day when the imagesamplingwas done is indicated at the bottom of each frame (hour: minute:second). Scale indicates 100 ~tm. F i g u r e 1.
trophoblast spreads were quantitatively evaluated using a digitizer tablet (Model DT1000, Watanabe Sokki, Tokyo, Japan) connected to a personal computer (Model 9801, NEC, Tokyo r Japan).
RESULTS
Sequence
of events
of trophoblast
spreading
Time-lapse video recordings of cultured blastocysts were started after 24 h of pre-incubation; in the following, the time of culture is counted from the time of initiation of preincubation. By this time, the blastocysts had already hatched from the zona pellucida and lightly adhered to the substratum. Although slight ruming of the edge of the trophoblast was noted, no migration of trophoblast cells had yet began (Figure 1, frame no. 1). The first sign o f trophoblastic migration
(primary giant trophoblast cells) was seen at 26 h (corresponds to day 5 afternoon in vivo) (Figure 1, frame 2). The ruffling of the trophoblast cell membrane and the initiation of migratory behaviour of the cells gradually spread along the entire edge of the blastocyst by 32 h (Figure 1, frame no. 9) (corresponds to day 5 evening in vivo). After ruffling of the cell membrane, the trophoblast cells extended large lamellipodia; then they gradually migrated away from the blastocyst. The inner-cell mass (ICM) continued to grow covered with trophoblast cell layer until about 40 h (Figure 1, frame no. 16) (see Tachi, 1992 and Naruse, Somata and Shoji, 1982); it emerged from the trophoblast cell layer at around 44-48 h [Figure 2(A), frame no. 1] (corresponds to day 6 afternoon in vivo). Differentiation of the germ layers in the ICM becomes recognizable between 48 and 52 h of culture [Figure 2(A), frame no. 1 and 2]. After 54 h [Figure 2(A), frame no. 4], the two
Placenta (1996), Vol. 17
586
(a!. +:~
"
":')i
-
"
i
....: : ,L.
+
++
'
+
+
'
Y:~',~:~,~::~+i
:
"
'
+~.~ ,,y ", +'-
+
!
(
/
-'~+-~;,'.&,,-:- . ' "
:
~ c ~::~,,::---
~/~
':':-~'
"
.
',+,.,~+',(.) -," ' ,u
' (:~
,7"
i
i¸
,
,
~,
,
(b)
,,.. ~.,., ..... c?-~~',;t~ .......
.
. ..,,~i,~j.;4<.: "
"~':~'~,-.. u:..~,,-~.& ':'K' ~"i..
,~J~L;-~'+~:,';L , ._'.>.~ . ~ .~
'::,u ....
:',- ;;
;C2.j..,I:,,:
\-+-,t~~
~ ,
7
.
...:,/ ..
~ .~,.->;,.,~,~;f~. ~<~ .'..-%~+'~i+'~.. , ,-&.'' ,~.:4".~+; ,', .
, .
~.+v.,'.y.',:'-,~.W~.~-, "~ :~,++%"+>~:,i::~ ' ~, .'
'
. ; "~:i" '-~"-:
~ "
I!iF--Z;
iZ
?..~ ..
~
~<--~---_--
•
::
:.... ..;~:, .
i , :_+~_7:__.<~ -" . .:~,. ,
"*"
•
:+' :
!
'
--
-:.?~..-I
,~
: ,.- :
": :?~+'!"h:~'i
..~_~-:..:?~5.
i
.
.~
Figure 2. Representative photomicrographsfrom the frames of time-lapse video recordings of a blastocyst cultured 0 h, showing the process of germ layer differentiation(a) and growth of the embryoniccylinder (b). Recordings of this series were started at 48 h of culture (for the definition of the time of culture, see the legend for Figure 1). Interval between frame I and 2, 1 h and 10 min: intervals between the rest of consecutive frames, 2 h. The inner-cell mass emerges from the overlyingtxophoblast(frame 2), and the primitive endoderm becomes recognizable(frame 3). The well-developedectoplacentalcone supports the growth and differentiationof the cylinder (c). e, ectoplaeentalcone; i, inner-cell mass; g, giant trophoblast cells. Scale indicates 100 Ftm.
germ layers, the endoderm and the epiblast, were clearly visible. By this time, the ectoplacental cone (EPC) is well developed. In / the EPC, vigorous cellular movement could be observed reflecting active proliferation and migratory activity of the cells in this region. T h e primitive endoderm in the I C M increased its thickness from 6"2 h [Figure 2(A), frame no. 8] on, and the formation of a cylinder was observed around 72 h (corresponds to day 7 afternoon in vivo). T h e cylinder grew in size markedly
between 80 to 92 h [Figure 2(B), frame no. 17-20] (corresponds to day 8 morning in vivo). Meanwhile, the entire surface area of the trophoblast spread continued to expand during this period. Some EPC cells were seen to migrate toward the edge of the trophoblast spread and to differentiate into giant trophoblast cells. After 96 h of culture (corresponds to day 8 afternoon in vivo), the giant trophoblast cells tended to dissociate from each other forming a mesh-like structure, probably representing the
Suenaga et al.: Mouse Blastocysts In Vitro
587
3.0 1.0 2.5 0.8 2.0 0.6
x 1.5
0.4 r/2
1.0
0.2
0.5
I
0
24
48 72 96 Time in culture (h)
120
Figure 3. Changes in the surface area of the trophoblast outgrowths according to the time of culture. Values are the means of more than 100 independent measurements. Vertical bars indicate s.d.
formation of the spongiotrophoblast. At the same time, both the size and the number of trophoblast cells increased remarkably.
Morphornetric analysis of t r o p h o b l a s t outgrowths The contours of the trophoblast spread were traced from the stills taken at 24 h intervals. The surface area of the outgrowths was measured by planimetry. As shown in Figure 3, the surface area increased linearly from 24 h to about 100 h of culture. Similar tracings were made from the time-lapse video cinematographs taken at every 2 h of culture and the changes in the surface area are graphically presented in Figure 4. A good linear relationship was observed between the surface area and the time of culture.
Giant cell transformation of trophoblast cells The cells located at the edge of the EPC were seen to be transformed into the secondary giant cells after 38 h of culture (Figure 1, frame no. 15, 16); the size of both cytoplasm and the nucleus increased tremendously. After about 72 h of culture, however, the giant trophoblast cells ceased to increase in cellular volume. In Figure 5, the process of giant cell transformation of small EPC cells is shown in sequential frames from the video recordings. As clearly seen in this figure, a substan-
0
24
36 48 Time in culture (h)
60
72
Figure 4. Changes in the surface area of the outgrowths as measured on the frames of time lapse video recordings of a single cultured blastocyst according to the time in culture.
tial increase in the size of cytoplasm occurred, preceded by an increase in nuclear size; the changes in the cytoplasmic and the nuclear sizes were measured on the photographs and the results are graphically shown in Figure 6. The ratio between the nuclear size and that of cytoplasm measured as the two-dimensional surface areas increased from 0.05 at 40 h to 0.35 at 50 h of culture, i.e., within 10 h after their emergence as small EPC cells. Thereafter, the ratio decreased slightly. After 60 h of culture, the microscopic appearance of the cells exhibited no sign of further changes either in the cytoplasm or in the nucleus.
Genetic differences in the rate of trophoblast o u t g r o w t h according to the strain of the mouse The rates of trophoblast outgrowth of blastocysts of different genetic constitution, i.e., ICR, C57B1/6, C3H, (B6 x C3)F1, (C3 x B6)F1 and (B6 x C3)F2, were quantitatively analysed. The results are shown in Figure 7. It was revealed that a clear difference exists in the rates of trophoblast outgrowth, according to the genetic constitution of the embryos. Among the blastocysts examined, those of inter-strain crosses, i.e., (B6 x C3)F1, (C3 x B6)F1 and (B6 × C3)F2 exhibited a significantly faster rate of the trophoblast spreading compared with the embryos of the inbred strains. Particularly, the rate of trophoblast spreading in (B6 × C3)F1 blastocysts was greater than any others. Conversely the rate of trophoblast spreading in the B6 blastocysts was slowest among the six different genetic constitutions of embryos examined.
588
mmmm///
Placenta (1996), Vol. 17
Figure 5. Photomicrographs showing the process of giant trophoblast cell formation. Two small EPC cells (frame 2) gradually increase their size (frame 4-6). Then, the cells increase their size abrupdy and tremendously, pushing the peripheral cells outward (frames 7-10). e, ectoplacental cone. Scale indicates 100 ~m.
0.35 0.6 0.3
--
0.5
o 0.25-
~
~
I. 0.4
0.2-
0.15
0.3
-
Z 0.1-
0.2
0.05 -
0.1
0---ll
I
40
q
I
44
~
I
~
I
~
I
q
I
48 52 56 60 Time in culture (h)
Figure" 6. A representative graph showing changes in the ratios between the size of nucleus and that of the cytoplasm (nuclear:cytoplasmic ratios), measured as the surface areas on the photomicrographs, in a trophoblast cell undergoing giant--celltransformation, according to the time in culture.
I
0
24
S I
I
I
48 72 96 Time in culture (h)
I
120
Figure 7. The rate of trophoblastic spreading in blastocysts of different genetic constitution. (O), (B6 x C3)F2; (O), ICR; (11), (B6 x C3)FI; (A), C57BL/6; (~), (C3 x B6)F1; (O) C3H. Vertical bars indicate s.d. All the measurements were actually done on the hour, although some of the plots are slightly dislodged to either the plus or the minus direction so as to avoid overlapping of the s.d. bars.
DISCUSSION T r o p h o b l a s t spreading or outgrowth, which takes place in the mouse blastocyst cultured in vitro, is considered to reflect the early process o f trophoblastic invasion in vivo (e.g. Gwatkin, 1966; Spindle & Pedersen, 1973; Kanai-Azuma et al., 1993; Koi et al., 1995). It consists of, as clearly seen in the timelapse video recordings, several different cell biological phenomena; i.e., (1) migration and (2) giant cell transformation
o f trophoblast cells, (3) proliferation o f E P C cells and (4) the subsequent transformation o f E P C cells into secondary giant cells. D u r i n g the process o f migration, ruffling o f t h e cell m e m brane, which is induced by the interaction o f the trophoblast cells with the substratum is followed by the extension o f lamellipodia. T h e general sequence o f events conforms with
Suenaga et al.: Mouse BlastocystsIn Vitro that described for other cells cultured in vitro (e.g., Abercrombie, Heaysman and Pegrum, 1970; Eckstein and Shur, 1989; Stossel, 1993; Appedu and Shur, 1994). The mean surface areas of trophoblast outgrowths, measured on more than 100 cultured blastocysts, increased linearly from 48 h to 96 h of culture; the results agreed well with those reported previously by Tanaka, Tojo and Tachi (1993) and by Koi et al. (1995). The contour of the trophoblast outgrowths as observed in the sequential frames of video recordings, showed that spreading takes place equally all around the cultured blastocysts and the linear migratory speed at the periphery of the spread declined as the time of culture advanced (data not shown). The detailed sequential study of the process revealed the possible presence of two phases in the outgrowth process. As shown in Figure 4 after 38-40 h of culture, the rate of increase in the surface area slowed down slightly and the contour of the spread became smoother; this phase of trophoblast outgrowth may be designated as the primary phase. Thereafter, the 'secondary' phase of spreading began. During this phase, cells migrating from the EPC contributed to the trophoblast spreading. Further analysis of the different phases of trophoblastic spreading at a molecular level, such as the possible involvement of surface galactosyl transferase in the maintenance of the stability of lamellipodia (Appedu and Shur, 1994) or effects of fibronectin or laminin upon the proliferation and migration of EPC cells (Koi et al., 1995), will give us crucial knowledge to understand the mechanisms of trophoblastic invasion in vivo. In the present report, the morphological sequence of events of giant cell transformation was documented (Figure 5 and 6). The EPC cells increased their size approximately eightfold during the process. The apparent nuclear:cytoplasmic ratios of EPC cells calculated from the measurements of the areas occupied by the nucleus and the cytoplasm on the pictures exhibited linear increases from 40 to 46 h of culture. A sharp decline in the ratio took place between 50 and 51 h of culture, reflecting either the sudden and tremendous increase in the cellular volume and/or spreading of the cytoplasm. Although the measurement of thickness of living cells is notoriously
589
difficult, further analysis of the changes in actual nuclear and cellular volumes might yield important information concerning the mechanisms of giant cell transformation. Although the giant cell transformation of trophoblast cells is one of the most striking cellular changes which accompanies trophoblast outgrowths, relatively little is known about the mechanisms underlying this phenomenon. Kanai-Azuma et al. (1993) proposed that while insulin-like growth factor-I (IGF-I) stimulates EPC cell proliferation and migration, IGF-II enhances giant cell transformation of the trophoblast cells. Their observations, however, were based upon study of the EPC cells collected from the gestational uteri of day 10.5 of pregnancy. It might be possible that each of the different steps of trophoblast outgrowth is governed by different growth factors and/or cytokines. A curious observation was made in the present work on the differences in the rate of trophoblast outgrowth according to the genetic background of the blastocysts. Thus, the rate of trophoblast outgrowth was fastest in blastocysts obtained by mating males and females of (B6 x C3)F1, and slowest in the embryos of B6. While genetic variation in the speed of development of cleavage stage embryos has been well-documented (Whitten and Dagg, 1962; McLaren & Bowman, 1973; Verbanac & Warner, 1981),-the differences in the rate of outgrowth observed in the present work may not simply be ascribed to difference in the developmental speed of the early embryos, because the rate of outgrowth reaches a plateau at about 96-120 h and leaving the blastocysts in culture longer does not lead to a 'catch-up' in the surface areas of outgrowths. Although the mechanisms underlying the observed differences are entirely unknown, our results strongly suggest the possible presence of genetic regulatory mechanisms underlying trophoblast outgrowth. Because trophoblast spreading is, as mentioned before, generally considered to be an in vitro model of trophoblastic invasion in vivo, further analysis of the genetic regulatory mechanisms underlying trophoblast outgrowth may provide important clues to understand the molecular mechanisms of trophoblastic invasion during the early phase of implantation, hopefully leading to improved success rates of IVF-ET.
ACKNOWLEDGEMENTS
We thank ProfessorJerome F. Strauss I11,Centre for Research on Women'sHealth and Reproduction,Departmentof Obstetricsand Gynecology,Universityof Pennsylvania,for reading the manuscriptand for makingvaluablesuggestions.This work is financiallysupported in part by Grants-in-Aidfrom Ministry of Education, Culture and Sports of Japan (#04556040, #06281107 to C.T.). A.S. is a recipientof JSPS (Japan Society for Promotionof Science) fellowship.
REFERENCES
Abercrombie, M., Heaysman, J. E. M. & Peg'rum, S. (1970) The locomotion of fibroblasts in culture. I. Movement of the leading edge. Ea'perimental Cell Research, 59, 393-398. Appeddu, P. A. & Shut, B. D. (1994) Control of stable lamellipodia formationby expressionof cell surface [[31,4-galactosyltransferasecytoplasmic domains.Journal of Cell Science, 107, 2535-2545.
Behrendtsen, O., Alexander, C. M. & Werb, Z. (1992)Metalloproteinases mediate extracellularmatrix degradation by cells from mouse blastocyst outgrowths. Development, 114, 3447-456. Dallenbaeh, F. D. (1975)Woraufberuhtdie Fahigkeityon Trophoblastzellen wie Krebszelleninvasivzu wachsen?Verhandlungender Deutschengesellschaft fur Inhere Medizin (Munchen), 59, 573. Damsky, C., Sutherland, A. & Fisher, S. (1993) Extracellularmatrix 5: adhesive interactionsin early mammalianembryogenesis,implantation,and placentation.FASEBJournal, 7, 1320-1329.
590
Eckst~in, D.J. & Shur, D. D. (1989) Laminin induces the stable expression of surface galactosyltransferase on lamellipodia of migrating cells. Journal of Cell Biology, 108, 2507-2517. " Edmonds, D. K., Lindsay, K. S., Miller, J. F., Williamson, E. & Wood, P. J. (1982) Early embryonic mortality in women. Fertility and Sterility, 38, 447-453. Enders, .A.C. (1964) Electron microscopy of an early implantation stage with a postulated mechanism of implantation. Developmental Biology,, 10, 395410. Enders, A. C. (1969) Cytological aspects of trophohlast-uterine interaction in early implantation. American Journal of Anatomy, 125, 1-29. Enders, A. C. (1974) Surface coats of the mouse blastocyst and uterus during the preimplantation period. Anatomical Record, 180, 31-45. Enders, A. C. (1975) Cellular basis interaction between trophoblast and uterus at implantation. Biolog), of Reproduction, 12, 41-65. Enders, A. C. & Schlat~e, S. J. (1967) A morphological analysis of the early implantation stage in the rat. American Journal of Anatomy, 120, 185-226. Enders A. C. & Welsh, A. O. (1993) Structural interactions of truphoblast and uterus during hemochorial placenta formation. Journal of Experimental Zoology, 266, 578-587. Fisher, S. J. & Damsky, C. H. (1993) human cytotrophoblast invasion. Semin. Cell Biology, 4, 183-188. Graham, C. H. & Lala, P. K. (1992) Mechanisms of placental invasion of the uterus and their control. Biochem. Cell Biology, 70, 867-874. Gwatkin, R. B. L. (1966) Amino acid requirements for attachment and outgrowth of the mouse blastocysts in vitro. Journal of Cell Physiologd,, 68, 335-344. Hoffman, L. H. & Wooding, F. B. (1993) Giant and binucleate trophoblast cells of mammals. Journal of Experimental Zoology, 266, 559-577. Jones, H. W., Jr., McDowell, J., Acosta, A. A., Sandow, B. A., Andrews, M. C., Veeck, L., Garcia, J. E., Whibley, T. W., Jones, G. S., Wilkes, C. A., Mantzavinos, T. & Wright, G. L., Jr. (1983) What is a pregnancy? A question for programs of in vitro fertilization. Fertility and Sterilio,, 40, 728-733. Kanai-Azuma, M., Kanai, Y., Kurohmarn, M., Sakai, S. & Hayashi, Y. (1993) Insulin-like growth factor (IGF)-I stimulates proliferation and migration of mouse ectoplacental cone cells, while IGF-II transforms them into trophoblast giant cells in vitro. Biology of Reproduction, 48, 252-261. Koi, H., Taehi, C., To)o, H., Kubota, T. & Aso, T. (1995) Effects of matrix proteins and heparin-binding components in fetal bovine serum upon the proliferation of ectoplacental cone cells in mouse blastocysts cultured in vitro. Biology of Reproduction, 52, 759-770. Lenton, E. A., Neal, L. M. & Sulaiman, R. (1982) Plasma concentrations of human chorionic gonadotropin from the time of implantation until the second week of pregnancy. Fertility and Sterilio,, 37, 773-778. Leridon, H. (1977) Human Fertility: The Basic Components Chicago, The Chicago University Press. Librach, C. L., Werb, Z., Fitzgerald, M. L., Chiu, K., Corwin, N. M , Esteves, R. A., Grobelny, D., Galardy, R., Damsky, C. H. & Fisher, S. J. (1991) 92-kD type IV collagenase mediates invasion of human cytotrophoblasts. Journal of Cell Science, 113, 437-449.
Placenta (1996), Vol. 17 MeLaren, A. & Bowman, P. (1973) Genetic effects on the timing of early development in the mouse. Journal of Embryology, atul Experimental Morpholog3,, 30, 491-498. McMaster, M. T., Bass, K. E. & Fisher, S. J. (1994) Human trophoblast invasion. Autocrine control and paracrine modulation. Ann. AT. Y. Aaulemy of Sciet~ces, 734, 122-131. Miller, J. E., Williamson, E., Glue, J., Gordon, Y. B., Grdzinskas, J. G. & Sykes, A. (1980) Fetal loss after implantation: a prospective study. Lancet, 2, 554-556. Naruse, I., Somata, S. & Shoji, R. (1982) Mammalian embryos culture from the pre-implantation stage, and analysis of developmental failures due to chromosomal abnormalities in vitro. Cong. Anom., 25, 319-326. Potts, D. M. (1968) The ultrastructure of implantation in the mouse. Journal of Anatomy, 103, 77-90. Potts, M. (1966) The attachment phase of ovoimplantation. American Journal of Obstetrics and Gynecologl,, 96, 1122-1128. Potts, M. (1969) The ultrastructure of egg implantation. Advances in Reproductive Plo,siolog~,,4, 241-267 Society for Assisted Reproductive Technology (1993) Assisted reproductive technology in the United States and Canada: 1991 results from the Society for Assisted Reproductive Technology generated from the American Fertility Society Registry. Fertility and Sterility, 59, 956-962. Spindle, A. I. & Pedersen, R. A. (1973) Hatching, attachment, and outgrowth of mouse blastocysts in vitro: fixed nitrogen requirements. Journal of Experimental Zoologj,, 186, 305-318. Stossel, T. P. (1993) On the crawling of animal cells. Science, 260, 1086-1094. Strickland, S. & Riehards, W. G. (1992) Invasion of the trophoblasts. Cell, 71,355-357 Tachi, C. (1992) Partial characterization of macromelecular components in fetal bovine serum required for development of mouse blastocysts cultured in vitro. Development Growth & Differentiation, 34, 69-77. Tachi, S., Taehi, C. & Lindner, H. R. (1970) Uhrastructural features of blastocyst attachment and trophoblastic invasion in the rat. Journal of Reproduction and Fertilio,, 21, 37-56. Tantaka, S. S., To)o, H. & Tachi, C. (1993) Possible involvement of amiroride-sensitive N a + / H + exchanger and Na+ transport systems in hatching and the trophoblast spreading in mouse blastocysts cultured in vitro, Journal of Reprodnction and Development, 39, 97-107. The American Fertility Society (1992) In vitro fertilization-embryo transfer (IVF-ET) in the United States: 1990 results from the IVF-ET registry. Fertility and Sterilio, , 57, 15-24. Verbanac, K. M. & Warner, C. M. (1981) Role of the major histocompatibility complex in the timing of early mammalian development. In Cellular and Molecular Aspects of Implantation (Eds) S. R. Glasser and D. W. Bullock pp. 467-472, New York, Plenum Press. Whitten, W. K. & Dagg, C. P. (1962) Influence of spermatozoa on the cleavage rate of mouse eggs. Journal of Experimental Zoolo~,, 148, 173-183. Wegmann, T. (1992) Immune signalling at the maternal fetal interface and trophoblast differentiation. Developmentaland Comparative Immunology, 16, 425-430.