1. Trophoblast development in utero in the first half of gestation is accompanied by a rise in the ploidy levels in giant cell nuclei. Giant cells constitute up to one-third of the trophoblast population over this period.

  2. The expression of polyploidy is independent of the maternal environment; the extent of polyploidy and the degree to which trophoblast develops morphologically under various conditions appear to be related.

  3. The following events are not necessary for the initiation of polyploidization in trophoblast: (a) hatching from the zona pellucida, (b) oestrogen sensitization, (c) implantation, (d) postimplantation changes in topology.

  4. That earlier developmental events such as the formation of the blastocoel cavity may in some way trigger polyploidization of trophoblast is not inconsistent with the studies described.

The mouse blastocyst consists of a single cell layer, the presumptive trophoblast, surrounding the blastocoel cavity and a group of internal cells, the inner cell mass. It is generally accepted that following implantation, the presumptive trophoblast gives rise to the ectopiacental cone and the trophoblast layer of the placenta, while the cells of the inner cell mass are the precursors of the embryo proper. A number of experiments have been carried out to investigate the lability of early embryonic cells and the effect of position upon their developmental fate. These experiments provide information about the phenomenon of determination, and have been recently reviewed by Graham (1971a). Little, however, is known about the timing of, or the events governing, the onset of differentiation of trophoblast or other embryonic cell types.

To study trophoblast differentiation at the molecular level it is necessary to show that mature trophoblast has biochemical properties not shared by the embryo ; the appearance of each new property serves as a marker of the course of differentiation. A number of trophoblast enzymes have now been described and their bearing on the question of trophoblast differentiation will be discussed elsewhere (Sherman, 1972 a, b). That the nuclei of rodent trophoblast are different from those of the embryo has been known since the early observations of Duval (1891). Using quantitative cytospectrophotometric techniques Zybina and co-workers (1961, 1963; Zybina & Mos’yan, 1967) have shown that the trophoblast of the rat has giant nuclei which contain hundreds of times the diploid DNA content. Indeed, she suggests that all trophoblast cells are polyploid by the second half of gestation (Zybina, 1970). (For the sake of convenience, we use the term ‘polyploid’ here only to denote the content in nuclei of DNA greater than 4 times the haploid amount. We do not wish to imply that within a giant trophoblast nucleus all chromosomes have necessarily been replicated the same number of times.) Zybina has proposed that this phenomenon occurs by a process of endoreduplication, i.e. replication of the nuclear material without subsequent mitosis and cell division (Zybina, 1960, 1963; Zybina & Mos’yan, 1967), resulting in the formation of polytene chromosomes (Zybina, 1970) similar to those found in the salivary glands and other tissues of dipteran flies (Beermann, 1962) and in some plant tissues (Nagl, 1963; Tschermak-Woess, 1967).

Giant nuclei are also seen in the trophoblast (but not in other embryonic cell types) of the mouse. We felt that the detection of both the onset, and the subsequent time course, of polyploidization would provide us with information useful in the study of trophoblast differentiation and complementary to the biochemical studies described elsewhere (Sherman, 1972 a, b). We have carried out measurements of the DNA content of trophoblast nuclei between the 5th and 11th days of pregnancy in the mouse (for DNA studies on earlier embryos, see Barlow, Owen & Graham, 1972). The results establish, and give a quantitative estimate of, the polyploid nature of mouse trophoblast. We have asked when polyploidy first arises and have considered possible mechanisms for its induction. In so doing, we have used blastocyst cultures to explore the extent to which the development of polyploidy is dependent upon the maternal environment; we conclude that polyploidy in trophoblast is at least to some degree an intrinsic phenomenon of mouse embryogenesis.

Preparation of animals

Embryos obtained from matings within a closed Swiss albino stock (strain PO) or from a PO×A2G cross were a source of trophoblast nuclei. No strainspecific differences were noted. Trophoblast from 6th to 11th day embryos was obtained from spontaneously mated females, while preimplantation embryos were collected from females superovulated (Runner & Palm, 1953) before mating. The day of observation of the sperm plug is considered the first day of pregnancy. For kidney capsule grafts, blastocysts were removed from the uterus on the 4th day of gestation and implanted under the kidney capsule of mature hosts (Fawcett, 1950; Kirby, 1960). The resulting haemorrhagic growths were dissected out 10 days later.

Collection of material

Blastocysts were obtained on the 4th day of pregnancy by flushing uteri with phosphate-buffered saline (PBS). Two-cell embryos were obtained by teasing apart oviducts in Whitten’s medium (1971) on the morning of the second day of pregnancy. Postimplantation trophoblast was obtained up to the 11th day of pregnancy as described elsewhere (Sherman, 1972b).

Culture conditions

Two-cell embryos were cultured to the blastocyst stage in the medium of Whitten (1971). Blastocysts removed from uteri on the 4th day of gestation were allowed to swell and enlarge by overnight culture in White’s medium (Know-land & Graham, 1972). Further development of blastocysts in vitro was effected by culture in Eagle’s medium as modified by Vogt & Dulbecco (1963), made up with four times glass-distilled water and supplemented with 10 % foetal calf serum. The percentage of blastocysts ‘implanting’ (sticking to the culture dish and spreading out along it) was well over 90 % when multiply-distilled water and exhaustively dialysed serum were used. ‘Unhatched’blastocysts, i.e. those failing to rupture their zona pellucida after 4–7 days in culture without undergoing any apparent degree of degeneracy, were obtained in small numbers from these cultures.

Preparation of materials

Trophoblast on the 7th to 9th days of development was dissected free of embryo, yolk sac and decidua in a drop of PBS directly on a microscope slide. The trophoblast cells were then teased apart with forceps, and the material was fixed by dipping the slide into a dish of absolute ethanol/glacial acetic acid (3/1, v/v). This dipping procedure appeared to remove most of the blood cells. The slide was left in fixative for 10 min. For 6th day embryos, trophoblast could not be satisfactorily separated from the embryo proper, and thus the cells on the slide were from the two tissues. This general procedure was unsatisfactory for preparations of 10th and 11th day materials as a great many cells were present and the nuclei overlapped. To obtain giant nuclei on these days, trophoblast was dissected away from the decidua. In the process, a number of giant cells were disrupted and their nuclei spilled out of the cells and sank to the bottom of the dissecting dish. Individual nuclei were then sucked up in a drawn-out capillary pipette, placed on a slide in a small drop of PBS and then fixed. Alternatively, where indicated, nuclei were teased out of intact cells following dissection of trophoblast, and fixed on slides as above. Nuclei from kidney capsule grafts were obtained in the same way. Blastocyst cultures grown on coverslips were removed from the medium, washed in PBS, fixed for 10 min and mounted on slides. Preimplantation and unimplanted embryos were prepared as described by Barlow et al. (1972).

Staining and measurements

The slides with nuclei for DNA measurements were hydrolysed in 5N-HC1 for 30 min at room temperature and then stained with Schiff’s reagent for 1–2 h. The preparations were washed with three changes of freshly prepared SO2 water, 10 min each change, then dehydrated and mounted in Euparal. Measurements of nuclear DNA contents were made with an integrating microdensitometer (Barr and Stroud, Glasgow, model GN 2) at 565 nm. All readings were corrected for background absorption. A slide of liver imprints was used to calibrate each set of measurements. Liver nuclei are either diploid, tetraploid, or octa-ploid and so contain 2C, 4C and 8C units of DNA respectively (Fig. 1) : C is the haploid amount of DNA. Nuclear contents were converted to values of C and are here presented as histograms using a log2 scale. Nuclear areas were calculated by the formula , where d1 and d2 are the major and minor diameters of the. nucleus in micrometres (μm).

Fig. 1.

The DNA content of mouse liver nuclei. Imprints were made on a slide using a cut surface of a liver. The nuclei were then stained with Schiff’s reagent and measured as indicated.

Fig. 1.

The DNA content of mouse liver nuclei. Imprints were made on a slide using a cut surface of a liver. The nuclei were then stained with Schiff’s reagent and measured as indicated.

Ploidy measurements on isolated trophoblast nuclei

As noted, by Zybina in the rat (1960), giant trophoblast nuclei in the mouse stain with Feulgen as intensely as small nuclei (Fig. 2). DNA measurements on the largest of these nuclei indicate that they contain enormous amounts of DNA, up to 850 times the haploid value (Fig. 3). In Fig. 3(a) are DNA values for nuclei which were released from cells of 11th day trophoblast during dissection, while Fig. 3(b) illustrates measurements of trophoblast nuclei teased from cells which have survived the dissection procedure intact. It may be seen that although nuclei reach the same high ploidy values in both cases, Fig. 3 (a) shows clustering of the DNA contents around euploid values whereas Fig. 3 (b) does not. Measurements on the trophoblast nuclei in Fig. 3(a) indicate that DNA content is roughly proportional to nuclear size (Fig. 4). The largest nuclei have diameters as great as 140 μm (compared with a 2C value of about 10), although this value is overestimated due to flattening of the nuclei on to the slide during fixation.

Fig. 2.

Nuclei of 11th day mouse trophoblast. Trophoblast tissue was smeared on a slide and stained with Schiff’s reagent. Nuclei from both normal and giant cells are present. Scale line= 50 μm.

Fig. 2.

Nuclei of 11th day mouse trophoblast. Trophoblast tissue was smeared on a slide and stained with Schiff’s reagent. Nuclei from both normal and giant cells are present. Scale line= 50 μm.

Fig. 3.

The DNA content of giant-cell nuclei in 11th day trophoblast. In (a), free nuclei were collected from the bottom of a dissecting dish with a capillary tube after trophoblast tissue was pulled apart from the decidua. The nuclei were then fixed on slides, stained with Schiff’s reagent and measured as indicated in Methods. In (b), trophoblast tissue was separated from those nuclei freed by the dissection procedure, and then disrupted with watchmaker’s forceps to free further nuclei. These were measured for DNA content as in (a).

Fig. 3.

The DNA content of giant-cell nuclei in 11th day trophoblast. In (a), free nuclei were collected from the bottom of a dissecting dish with a capillary tube after trophoblast tissue was pulled apart from the decidua. The nuclei were then fixed on slides, stained with Schiff’s reagent and measured as indicated in Methods. In (b), trophoblast tissue was separated from those nuclei freed by the dissection procedure, and then disrupted with watchmaker’s forceps to free further nuclei. These were measured for DNA content as in (a).

Fig. 4.

The relationship between DNA content and area of giant trophoblast nuclei. Areas of the nuclei measured in Fig. 3 (a) were determined as indicated in Methods. The area of each nucleus is plotted on a log10 scale against its DNA content.

Fig. 4.

The relationship between DNA content and area of giant trophoblast nuclei. Areas of the nuclei measured in Fig. 3 (a) were determined as indicated in Methods. The area of each nucleus is plotted on a log10 scale against its DNA content.

The nuclei of trophoblast dissected from several placentas on the 6th to 11th days of pregnancy were scanned, and the largest ones measured. The results are shown in Table 1. By the sixth day of pregnancy, nuclei were found containing between 32 and 64 times the haploid amount of DNA. The DNA content of this largest class of trophoblast nuclei roughly doubled on each day of development. The highest DNA content of a nucleus from 11th day trophoblast contained about 850 times the haploid amount of DNA.

Table 1.

Highest ploidy reached by trophoblast of different gestation age

Highest ploidy reached by trophoblast of different gestation age
Highest ploidy reached by trophoblast of different gestation age

To obtain some idea of the relative proportions of cells with different ploidies in the total trophoblast population, trophoblast from single embryos on the 7th to 11th days of gestation were dissected out and smeared on slides. DNA measurements were made on a random selection of the nuclei ; at the same time efforts were made to avoid measuring contaminating immature red blood nuclei which are distinguishable from diploid trophoblast nuclei on the basis of morphology. The results are shown in Fig. 5. On the 7th day (Fig. 5a) 8 % of the total number of nuclei have DNA contents greater than 4C; of the remaining population, one-third have a DNA value greater than 2C but no more than 4C, while the other two-thirds are 2C nuclei. By the 8th and 9th days of development (Fig. 5b, c) 30–40 % of the total nuclear population measured have DNA values greater than 4C; the distribution of DNA values within the diploid population is about the same as that of the 7th day sample. In the course of these studies, however, we suspected that the proportion of diploid cells in the total population was being constantly underestimated, as many clumps of overlapping small cells were present on the slides, and their DNA contents could not be measured. The problem of overlapping nuclei in smears of 10th and 11th day trophoblast was so marked that the measurements of a random sample was not feasible. Because of the apparent relationship between nuclear area and DNA content, at least for larger nuclei (Fig. 4), we attempted to rescore the slides visually to assess the population of polyploid nuclei. In this way, clumps of overlapping diploid cells which could not be measured spectrophotometrically could be taken into account. Mitotic index was also scored. All the mitoses were diploid as judged by microdensitometry or chromosome counts (Barlow & Sherman, unpublished) ; consequently, the mitotic index is expressed as the percentage of diploid cells in division. The results of the visual observations are presented in Table 2. By these estimates, between 10 and 15 % of the cells appear to be polyploid on the 8th to 11th days of gestation. We thus propose on the basis of the data in Fig. 5 and Table 2 that no more than one-third of trophoblast cells are polyploid between the 7th and 11th days of gestation. Furthermore, because of the paucity of nuclei with DNA contents between 2C and 4C (Fig. 5), and the very low mitotic index (Table 2), it would appear as though most of the population of diploid cells on the 7th to 9th days of gestation arein the G1 and G2 phases of the cell cycle.

Table 2.

Percentages of diploid and polyploid cells and cells in mitosis at different stages of trophoblast development

Percentages of diploid and polyploid cells and cells in mitosis at different stages of trophoblast development
Percentages of diploid and polyploid cells and cells in mitosis at different stages of trophoblast development
Fig. 5.

DNA contents in trophoblast nuclei on the 7th to 9th days of gestation. Trophoblast tissue from single conceptuses was dissected out and smeared on slides. After staining with Schiff’s reagent the DNA content of a random sample of nuclei was determined. The animals were sacrificed on the evening of the (a) 7th, (b) 8th and (c) 9th days of gestation. Each histogram represents results pooled from at least two slides.

Fig. 5.

DNA contents in trophoblast nuclei on the 7th to 9th days of gestation. Trophoblast tissue from single conceptuses was dissected out and smeared on slides. After staining with Schiff’s reagent the DNA content of a random sample of nuclei was determined. The animals were sacrificed on the evening of the (a) 7th, (b) 8th and (c) 9th days of gestation. Each histogram represents results pooled from at least two slides.

Nuclear morphology

In view of the claim by Zybina (1970) that the giant nuclei of rat trophoblast contain polytene chromosomes with a distinct chromomere banding pattern, we paid particular attention to the morphology of giant nuclei in the mouse. In preparations of giant nuclei dissected from the trophoblast and stained with Feulgen, Giemsa or fluorochromes we have been unable to find any evidence of polytene chromosomes. Some nuclei possess chromatin that has a fibrillar appearance, but for the most part these fibrils show no particular orientation within the nucleus, nor is there any apparent lateral attraction to form a single polytene chromosomal unit. Unlike Zybina (1970) we have not seen any nucleus that could be interpreted as an endomitotic nucleus (in sensuGeitler, 1953). These observations will be elaborated more fully elsewhere (in preparation).

Dependence upon the maternal environment for the expression of polyploidy

Experiments were designed to test whether polyploidy could develop and, if so, to what extent, under conditions independent of the maternal environment. In the first experiment blastocysts were removed from the uterus on the 4th day of pregnancy and transplanted under the kidney capsule of a non-pregnant adult female. After 10 days incubation, the haemorrhagic growths were removed and the cells teased apart. As in the case with mature uterine trophoblast, many large nuclei fall to the bottom of the dissecting dish during this procedure. These nuclei are highly poplyploid (Fig. 6), the largest being equivalent in DNA content to the maximum values found in uterine trophoblast on the 11th day of development (cf. Fig. 3). This result confirms the finding of Hunt & Avery (1971) that trophoblast nuclei from kidney capsule implants contain large amounts of DNA, and illustrates that implantation in the uterus is not required for the expression of polyploidy.

Fig. 6.

DNA content in giant nuclei from an ectopic pregnancy. Ten days after implanting a blastocyst under a kidney capsule of an adult mouse the haemorrhagic growth was removed and disrupted with watchmaker’s forceps. Following this, giant nuclei were collected, fixed on slides, stained, and subjected to DNA determination.

Fig. 6.

DNA content in giant nuclei from an ectopic pregnancy. Ten days after implanting a blastocyst under a kidney capsule of an adult mouse the haemorrhagic growth was removed and disrupted with watchmaker’s forceps. Following this, giant nuclei were collected, fixed on slides, stained, and subjected to DNA determination.

To determine the extent to which contact with tissue is responsible for the onset of polyploidy, blastocysts were removed on the 4th day of pregnancy and placed in culture on coverslips, first overnight in White’s medium (Knowland & Graham, 1972) and then into modified Eagle’s medium (Vogt & Dulbecco, 1963). Cole & Paul (1965) and Gwatkin & Meckley (1965) have already shown that giant cells appear in such cultures. Feulgen-stained nuclei in these blastocyst cultures were measured and the results are shown in Fig. 7. After one day in Eagle’s medium, nuclei as large as 16C were recorded; the ploidy reached 64C after 5 days in this medium. Higher ploidy values were not observed with further culture. This could well have been due to suboptimal culture conditions, as many of the nuclei appeared to lift off the coverslip after 5 or 6 days in culture. Some of the larger nuclei also appeared to be breaking up into smaller fragments.

Fig. 7.

DNA content of nuclei in cultured blastocysts. Blastocysts were obtained on the middle of the 4th day of pregnancy and cultured overnight in White’s medium (Knowland & Graham, 1972). They were then transferred in groups of about ten to drops of modified and supplemented Eagle’s medium (see Methods) maintained in place on glass coverslips with paraffin oil. After 24 h (Day 1) a coverslip was removed, rinsed in PBS and fixed in acetic acid/ethanol (1/3, v/v) for 10 min. After staining with Schiff’s reagent the nuclear DNA content was measured where possible in those blastocysts which had outgrown on the coverslip. This was repeated on coverslips removed from the dish after 48 h (Day 2), 72 h (Day 3), etc., in Eagle’s medium.

Fig. 7.

DNA content of nuclei in cultured blastocysts. Blastocysts were obtained on the middle of the 4th day of pregnancy and cultured overnight in White’s medium (Knowland & Graham, 1972). They were then transferred in groups of about ten to drops of modified and supplemented Eagle’s medium (see Methods) maintained in place on glass coverslips with paraffin oil. After 24 h (Day 1) a coverslip was removed, rinsed in PBS and fixed in acetic acid/ethanol (1/3, v/v) for 10 min. After staining with Schiff’s reagent the nuclear DNA content was measured where possible in those blastocysts which had outgrown on the coverslip. This was repeated on coverslips removed from the dish after 48 h (Day 2), 72 h (Day 3), etc., in Eagle’s medium.

Most of the smaller cells had already detached from the coverslips, as evidenced by the striking reduction in the proportion of diploid cells after the 4th day of culture in Eagle’s medium (Table 3, Exp. 1). In fact, the proportion of the diploid cell population up to the 5th day is under-represented in these histograms because many of the smaller (presumably inner cell mass) cells remain as a clump in the centre of the culture and are not easily measured. By the 6th day, however, these clumps have lifted off, leaving only the outgrowing polyploid cells. Consequently, while it would appear that embryonic precursor cells do not fare well under these culture conditions, giant trophoblast cells survive somewhat more successfully, and reach high levels of ploidy before they begin to degenerate.

Table 3.

Proportion of polyploid cells in blastocyst cultures

Proportion of polyploid cells in blastocyst cultures
Proportion of polyploid cells in blastocyst cultures

While these results demonstrate that contact with tissue is not required for the expression of polyploidy, it might still be argued that the onset of polyploidy inevitably resulted from a stimulus received by blastocysts while they were still in the uterus. To investigate this, eggs were removed from oviducts at the 2cell stage, maintained in culture (Whitten, 1971) until they were mature blastocysts (approximately 72 h), and then transferred to Eagle’s medium. Nuclei were measured for DNA content in these cultures, and as Fig. 8 and Table 3, Exp. 2 show, a pattern emerges very similar to that seen in Fig. 7. After 6 days in Eagle’s medium, nuclei with DNA contents in the 64–128C class were present. Once again, very few diploid cells remained after 6 days of culture (Table 3). We conclude from these experiments that trophoblast cells will undergo polyploidiza-tion independently of the maternal environment, at least from the two-cell stage.

Fig. 8.

DNA content in blastocyst cultures derived from two-cell embryos. Embryos at the 2-cell stage were removed from oviducts and cultured in Whitten’s medium (1971) until they reached mid to late blastocyst stage. They were then grown on coverslips as described in Fig. 7 for 2, 4, 5, 6, or 8 days. DNA contents were then determined as in Fig. 7.

Fig. 8.

DNA content in blastocyst cultures derived from two-cell embryos. Embryos at the 2-cell stage were removed from oviducts and cultured in Whitten’s medium (1971) until they reached mid to late blastocyst stage. They were then grown on coverslips as described in Fig. 7 for 2, 4, 5, 6, or 8 days. DNA contents were then determined as in Fig. 7.

Other factors involved in the onset of polyploidy

The above experiments have demonstrated that neither a uterine signal nor contact with foreign cells is responsible for the onset of polyploidy in trophoblast. We have considered the possibility that in blastocyst cultures, contact with the culture dish and the concomitant spreading of trophoblast along the surface would be sufficiently analogous to events during implantation in utero to provide the necessary stimulus for the onset of polyploidy. We have thus examined the DNA content of blastocyst nuclei under conditions that preclude attachment to the culture dish. Barlow et al. (1972) were unable to detect blastocyst cells containing more than 4C amounts of DNA up to the fifth day of development. On the fifth day, blastocysts which have not yet implanted or which are just in the process of implantation can be flushed out of the uterus. We have examined the DNA contents of cells from such blastocysts, and the results may be seen in Fig. 9. Of the 336 nuclei measured, a very small number, 6, had DNA contents significantly greater than 4C, between 5C and 7C. Furthermore, while some or most of the population of nuclei at 4C may be in the G2 phase of the diploid cell cycle, others may be held in readiness for polyploidiza-tion, i.e. in G1 of the first endoreduplication cycle. These results confirm similar studies by Graham (1971 b) ; the timing of the onset of polyploidization is described in Barlow et al. (1972).

Fig. 9.

DNA content of nuclei in 5th day blastocysts. Blastocysts were removed early on the 5th day of gestation and spread on slides as indicated in Methods. DNA determinations were then carried out on the Feulgen-stained nuclei.

Fig. 9.

DNA content of nuclei in 5th day blastocysts. Blastocysts were removed early on the 5th day of gestation and spread on slides as indicated in Methods. DNA determinations were then carried out on the Feulgen-stained nuclei.

When carefully prepared Eagle’s medium is used for blastocyst culture (see Methods), over 80 % of the blastocysts will shed their zona pellucidae, stick to, and. grow out along the surface of the dish or coverslip within 2 days of culture. The majority of the blastocysts which do not behave in this way will have failed to hatch from their zona pellucida. These do not, however, appear to be degenerate (Fig. 10). We do not yet know the reason behind the failure to hatch, but spreads of these unhatched blastocysts after 4 days in culture reveal that they contain from 64 to 250 cells (average of 12 samples = 130). All have giant cells, and over 80 % of them have cells in mitosis. They are able to grow out along the culture dish after enzymic removal of the zona. Finally, they appear to have the same life-span of normal blastocyst cultures, i.e. they degenerate after 8–10 days.

Fig. 10.

Unhatched blastocysts. Photomicrograph of blastocysts removed from the uterus on the 4th day of development and cultured in modified, supplemented Eagle’s medium for 4 days. Scale line = 25 μm.

Fig. 10.

Unhatched blastocysts. Photomicrograph of blastocysts removed from the uterus on the 4th day of development and cultured in modified, supplemented Eagle’s medium for 4 days. Scale line = 25 μm.

We have measured DNA contents in nuclei of unhatched blastocysts after 4 and 7 days in culture. As seen in Fig. 11, there are a substantial number of polyploid cells, even after 4 days in culture. Indeed, the patterns are strikingly similar to those seen in normal blastocyst cultures (cf. Figs. 7, 8). The exception, however, is that there are proportionately more diploid cells in unhatched blastocysts although the proportion of diploids falls significantly between 4 and 7 days in culture. These results confirm that implantation, contact with any surface, and indeed, even rupture of the zona pellucida are not essential for the onset of polyploidy.

Fig. 11.

DNA contents in nuclei from unhatched blastocysts. Unhatched blastocysts after (a) 4 or (b) 7 days in culture were treated as were uncultured blastocysts (Fig. 9).

Fig. 11.

DNA contents in nuclei from unhatched blastocysts. Unhatched blastocysts after (a) 4 or (b) 7 days in culture were treated as were uncultured blastocysts (Fig. 9).

The ploidy of the trophoblast stem cell

Zybina has indicated that in the rat, by the 15th day of pregnancy, all of the nuclei in the trophoblast are polyploid (Zybina, 1970; Zybina & Grischenko, 1970). It is generally assumed that the ectopiacental cone is the stem cell region of the trophoblast and that it derives from the presumptive trophoblast layer in the blastocyst. If this is so, and if all the cells of this layer become polyploid before implantation, then the stem cells for the trophoblast could not be diploid. But Fig. 5 and Table 2 show that diploid cells are present in excess of polyploids between the 7th and 11th days of gestation. Fig. 12 shows a study of the small cells in trophoblast preparations on the 10th and 11th days of gestation. Clearly many cells with DNA values of 4C or less are present. If these diploid cells are in fact stem cells, their disappearance in the rat near the end of pregnancy (Zybina, 1970; Zybina & Grischenko, 1970) would be consistent with the termination of proliferation of trophoblast.

Fig. 12.

DNA content of small nuclei in 10th and 11th day trophoblast. Trophoblast from (a) 10th or (b) 11th day embryos was dissected free of other tissues and smeared on slides. DNA determinations were carried out on a random sample of Feulgen-stained nuclei. The histogram includes only those nuclei with DNA contents equal to or less than 8C.

Fig. 12.

DNA content of small nuclei in 10th and 11th day trophoblast. Trophoblast from (a) 10th or (b) 11th day embryos was dissected free of other tissues and smeared on slides. DNA determinations were carried out on a random sample of Feulgen-stained nuclei. The histogram includes only those nuclei with DNA contents equal to or less than 8C.

Fig. 12(a) shows that the proportion of 4C to 2C nuclei on the 10th day of pregnancy is greater than on any of the other days studied (cf. Figs. 5 and 12b). Table 2 shows that the mitotic index falls to a minimum value on the 8th and 9th days and then rises markedly on the 10th and 11th days. It may be that the large number of 4C nuclei seen on the 10th day represents a population of nuclei that are just about to divide, so contributing to the increasing proportion of mitotic figures on this and the subsequent day.

In this paper we have demonstrated that in the mouse, as is the case in the rat (Zybina, 1960, 1963), giant trophoblast cells reach very high levels of ploidy. Indeed, the levels reached are comparable with those found in the rat, i.e. between 512 and 1024 times the haploid amount of DNA. As Table 1 illustrates, the ‘cell cycle’ time for the largest class of trophoblast nuclei is of the order of 24 h between the 6th and 11th days of pregnancy. The mechanism by which giant trophoblast nuclei accumulate DNA is still open to some question; although the presence of discrete nuclear size classes (Fig. 3a) would be consistent with Zybina’s (1960) proposal that endoreduplication is responsible, this histogram represents a population of selectively chosen cells (i.e. those whose nuclei rupture upon separating trophoblast from the decidua, possibly establishing these as cells in direct contact with the maternal part of the placenta). Cells outside of this population, as well as more randomly chosen samples, do not show this same type of distribution (e.g. Fig. 3b). Because the type of distribution illustrated in Fig. 3b might occur as a result of specific amplification of only some parts of the genome, we have been carrying out biochemical studies designed to distinguish between DNA accumulation by endoreduplication and specific gene amplification. Preliminary results (M. I. Sherman, A. McLaren and P. M. B. Walker, in preparation) are consistent with the endoreduplication hypothesis.

The possibility that polyploidy arises by engulfment of maternal cells following implantation (Galassi, 1967) is unlikely because trophoblast can proceed to relatively high ploidy values in the absence of any foreign cells. Our data do not exclude, however, some contribution of trophoblast nuclear fusion to the formation of giant nuclei (Avery & Hunt, 1969).

There is no evidence from our studies that the maternal environment is directly responsible or even required for the onset of polyploidy. Polyploidization is evidenced in trophoblast in ectopic pregnancies and under a number of culture conditions. Especially pertinent are the findings that many polyploid cells are present in embryos cultured from the two-cell stage (Fig. 8) and in unhatched blastocysts (Fig. 11). In the former case the embryos, removed from the oviducts, had not been exposed to any specific uterine stimuli such as oestrogen sensitization (see Shelesnyak & Marcus, 1971). In the latter case, neither release from the zona pellucida nor contact with any surface had occurred.

Although trophoblast nuclei in culture reach ploidy values as high as 128C, they do not achieve the levels of nuclei in vivo or in ectopic pregnancies. Neither are trophoblast esterase and alkaline phosphatase expressed to the full extent in blastocyst cultures (Sherman, 1972 a, b). These results suggest that although trophoblast can begin to differentiate in culture it does not reach its full developmental potential, possibly because blastocyst cultures have a restricted life-span or because of the diminution of some influence supplied by the inner cell mass, which, as we have shown, degenerates in culture prior to the giant cells (Figs. 7, 8).

In so far as the onset of polyploidy is independent of the implantation event and indeed appears to precede it (Fig. 9; Graham, 1971a; Barlow et al. 1972), we have considered other developmental events which might have provided the stimulus for polyploidization. We find the formation of the blastocoel cavity a particularly attractive candidate. The blastocoel may confer upon the presumptive trophoblast cells surrounding it special properties not shared by the other cells of the embryo, as these cells are the only ones not in contact with the inner cell mass, and they may also be uniquely exposed to specific physical stresses involved in the cavitation process. Although cells adjacent to the blastocoel could thus be specifically stimulated to undergo polyploidization, that part of the presumptive trophoblast layer in apposition to the inner cell mass may be protected from the stimulus. By remaining diploid, these cells could then serve as precursors of the ectoplacental cone. This is consistent with the observation by Dickson (1963) that presumptive trophoblast cells surrounding the blastocoel appear enlarged during implantation while those adjacent to the inner cell mass do not. Furthermore, in sections of recently implanted blastocysts, Amoroso (1952) has noted that giant cells with enlarged nuclei are first found at the abembryonic pole.

The data in Table 2 suggest that the trophoblast is again exposed to a stimulus later in development. The mitotic index of trophoblast before the 10th day of gestation is extremely low. On the 10th day there appears to be an increase in the percentage of cells with a 4C amount of DNA (Fig. 12 a), accompanied by a marked increase in the mitotic index (Table 2). This activation of the diploid population could be related to the formation of the labyrinth and spongy layers of the trophoblast which are first evidenced at about this time in development (Amoroso, 1952). Because the proportion of diploid cells in our cultures is always substantially lower than that found in vivo (cf. Tables 2 and 3), it is likely that the activation does not occur under our in vitro conditions.

This work was done, in part, in the Paediatric Research Unit, Guy’s Hospital, London, and later completed in the laboratory of Dr C. F. Graham, Zoology Department, Oxford, We wish to thank both Professor P. E. Polani and Dr Graham for their hospitality in the respective laboratories. We also thank Dr Graham for his suggestions and for his comments on the manuscript. We are grateful to the Medical Research Council and the Medical Research Council of Canada for support. We also wish to thank Dr P. Macdonald for help with computer programming, Professor Henry Harris for reading the manuscript, and Miss Z. White for excellent technical assistance.

Amoroso
,
E. C.
(
1952
).
Placentation
.
In Marshall’s Physiology of Reproduction
(ed.
A. S.
Parkes
), pp.
127
311
.
London
:
Longmans, Green and Co
.
Avery
,
G. B.
&
Hunt
,
C. V.
(
1969
).
The differentiation of trophoblast giant cells in the mouse studied in kidney capsule grafts
.
Transplantation Proc
.
1
,
61
66
.
Barlow
,
P.
,
Owen
,
D. A. J.
&
Graham
,
C.
(
1972
).
DNA synthesis in preimplantation mouse embryo
.
J. Embryol. exp. Morph
.
27
,
431
445
.
Beermann
,
W.
(
1962
).
Riesenchromosomen
.
Protoplasmatologia
6 D
,
1
160
.
Cole
,
R. J.
&
Paul
,
J.
(
1965
).
Properties of cultured and preimplantation mouse and rabbit embryos, and cell strains derived from them
.
In Preimplantation Stages of Pregnancy (eds
G. E. W.
Wolstenholme
and
M.
O’Connor
), pp.
82
112
.
Boston
:
Little, Brown and Co
.
Dickson
,
A. D.
(
1963
).
Trophoblastic giant cell transformation of mouse blastocysts
.
J. Reprod. Fert
.
6
,
465
466
.
Duval
,
M.
(
1891
).
Le placenta des rongeurs: le placenta de la souris et du rat
.
J. Anat. Physiol. Paris
27
,
24
106
.
Fawcett
,
D. W.
(
1950
).
The development of mouse ova under the capsule of the kidney
.
Anat. Rec
.
108
,
71
91
.
Galassi
,
L.
(
1967
).
Reutilization of maternal nuclear material by embryonic and trophoblastic cells in the rat for the synthesis of deoxyribonucleic acid
.
J. Histochem. Cytochem
.
15
,
573
579
.
Geitler
,
L.
(
1953
).
Endomitose und endomitotische Polyploidisierung
.
Protoplasmatologia
6C
,
1
89
.
Graham
,
C. F.
(
1971a
).
The design of the mouse blastocyst
.
In Control Mechanisms of Growth and Differentiation
(eds
D. D.
Davies
and
M. J.
Balls
), pp.
371
378
.
Cambridge
:
Cambridge University Press
.
Graham
,
C. F.
(
1971b
).
Virus assisted fusion of embryonic cells
.
Acta endocrin. (Kbit) Suppl
.
153
,
154
167
.
Gwatkin
,
R. B. L.
&
Meckley
,
P. E.
(
1965
).
In vitro culture of mouse ova beyond the blastocyst stage: effect of substrate and medium components
.
J. Cell Biol
.
27
,
136a
(Abstr.)
.
Hunt
,
C. V.
&
Avery
,
G. B.
(
1971
).
Increased levels of deoxyribonucleic acid during trophoblast giant-cell formation in mice
.
J. Reprod. Fert
.
25
,
85
94
.
Kirby
,
D. R. S.
(
1960
).
The development of mouse blastocysts beneath the kidney capsule
.
Nature, Lond
.
187
,
707
708
.
Knowland
,
J.
&
Graham
,
C. F.
(
1972
).
RNA synthesis at the 2-cell stage of mouse development
.
J. Embryol. exp. Morph
.
27
,
167
176
.
Nagl
,
W.
(
1963
).
Über Endopolyplodie, Restituiionskernbildung und Kernstrukturen in Suspensor von Angiospermen und einer Gymnosperme
.
Österr. bot. Z
.
109
,
431
494
.
Runner
,
M. N.
&
Palm
,
J.
(
1953
).
Transplantation and survival of unfertilized ova of the mouse in relation to postovulatory age
.
J. exp. Zool
.
124
,
303
316
.
Shelesnyak
,
M. C.
&
Marcus
,
G. J.
(
1971
).
Steroidal conditioning of endometrium for nidation
.
Advances in the Biosciences
6
,
303
313
.
Sherman
,
M. I.
(
1972a
).
The biochemistry of differentiation of mouse trophoblast: esterase
.
In preparation
.
Sherman
,
M. I.
(
1972b
).
The biochemistry of differentiation of mouse trophoblast: alkaline phosphatase
.
Devi Biol. (In the Press
.)
Tschermak-Woess
,
E.
(
1967
).
Der eigenartige Verlauf der I. meiotischen Prophase von Rhinanthus, die Riesenchromosomen und das besondere Verhalten der kurzen Chromosomen in Mitose, Meiose und hoch endopolyploiden Kernen
.
Caryologia
20
,
135
152
.
Vogt
,
M.
&
Dulbecco
,
R.
(
1963
).
Steps in the neoplastic transformation of embryonic cells by polyoma virus
.
Proc. natn. Acad. Sci. U.S.A
.
49
,
171
179
.
Whitten
,
W. K.
(
1971
).
Nutrient requirements for the culture of preimplantation embryos
.
Advances in the Biosciences
6
,
129
139
.
Zybina
,
E. V.
(
1960
).
Sex chromatin in the trophoblast of early white rat embryos
.
Dokl. Akad. Nauk SSSR
130
,
633
-
635
.
(In Russian.)
Zybina
,
E. V.
(
1961
).
Endomitosis and polyteny of trophoblast giant cells
.
Dokl. Akad. Nauk SSSR
140
,
1177
-
1180
.
(In Russian.)
Zybina
,
E. V.
(
1963
).
Cytophotometric determination of DNA content in nuclei of trophoblast giant cells
.
Dokl. Akad. Nauk SSSR
153
,
1428
-
1431
.
(In Russian.)
Zybina
,
E. V.
(
1970
).
Anomalies of polyploidization of the cells of the trophoblast
.
Tsitologiya
12
,
1081
-
1093
. (Tn Russian.)
Zybina
,
E. V.
&
Mos’Yan
,
I. A.
(
1967
).
Sex chromatin bodies during endomitotic polyploidization of trophoblast cells
.
Tsitologiya
9
,
265
-
272
.
(In Russian.)
Zybina
,
E. V.
&
Grischenko
,
T. A.
(
1970
).
Polyploid cells of the trophoblast in different regions of the white rat placenta
.
Tsitologiya
12
,
585
-
595
.
(In Russian.)