1. On the 4th day of gestation, embryos were recovered from mice of the C57BL, RIII, JU, C3H and Q strains, and cell counts were carried out. Significant differences between strains were seen, both in the percentage of embryos which had reached the blastocyst stage, and in the mean number of cells per embryo. C57BL embryos had most cells, and C3H embryos fewest.

  2. Examination of earlier stages of C57BL and C3H development showed that the proportionate difference in cell number remained constant, so that the difference involved the time at which cleavage began, and not the rate of cleavage. Activation of the eggs and the formation of pronuclei also occurred earlier in C57BL than in C3H females.

  3. The difference in cell number between C57BL and C3H embryos did not depend on a difference in time of mating, nor on the genotype of the male, since reciprocal crosses were similar to the maternal strain. The difference was maintained in culture from the two-cell stage.

Implantation is a critical stage of gestation, which the embryo often fails to survive. In several species (mouse, rat, rabbit, sheep, cow) egg transfer experiments have established the necessity for a close degree of synchronization between the development of the embryo and of the uterus if implantation is to succeed. If the embryo is ahead of the uterus in development, it can remain at the blastocyst stage (at least in the mouse and rat) until the uterus catches up ; but if the embryo develops too slowly, it misses the short period of uterine receptivity during which implantation is possible.

The tempo of development during the pre-implantation period may therefore be an important factor in prenatal mortality. Little is known of its genetic control. Gates, Doyle & Noyes (1961) found that (BALB/C× 129) F1 mouse blastocysts contained more cells than did those of the maternal BALB/C strain, but with 129 females the F1 combination showed no superiority over the inbred. Whitten & Dagg (1962) also found a difference in stage of development between (BALB/C ×129) embryos and BALB/C embryos, as well as between BALB/C and 129 embryos, at both the blastocyst and the 8-cell stage, which they interpreted as a difference in cleavage rate. Bowman & McLaren (1970a) found no difference in embryonic cell number 3·5 days post coitum between strains of mice differing fourfold in adult size.

The present work is concerned with cell number in embryos of one randomly bred and four inbred strains of mice, at various times after ovulation.

The mice belonged to the randomly bred Q strain and to the inbred strains C57BL/McL, C3H/BiMcL, JU/Fa and RIII/Fa. Females were paired with males, and examined for vaginal plugs daily. For purposes of dating, coitus was assumed to occur at 1 a.m. on the morning that the plug was found; this day is termed the first day of gestation. In some experiments, females were paired with males at 9 a.m. and examined for plugs at 10 a.m.

Embryos were recovered from the oviducts or uterus in phosphate-buffered saline, and examined with a dissecting microscope. Up to the 8-cell stage, cell number could be determined by inspection; at later stages, embryos were classified into morulae or blastocysts according to whether cavitation had occurred, and were then placed in 0·25 % sodium citrate for 5 min at room temperature, before being transferred to aceto-carmine for 24 h. Squash preparations were made, stained with basic fuchsin, and examined by phase microscopy.

Embryos were cultured as described by Bowman & McLaren (1970b).

In the first series, embryos of the five strains were recovered 3-5 days post coitum (p.c.) and cell counts carried out. The results are given in Table 1 and Fig. 1. A preliminary account of the data on inbred embryos was given in McLaren (1968).

Table 1.

Cell number and stage of development 3·5 days post coitum in one random-bred (Q) and four inbred (C57BL, RIII, JU, C3H) strains

Cell number and stage of development 3·5 days post coitum in one random-bred (Q) and four inbred (C57BL, RIII, JU, C3H) strains
Cell number and stage of development 3·5 days post coitum in one random-bred (Q) and four inbred (C57BL, RIII, JU, C3H) strains
Fig. 1.

The distribution of cell numbers in embryos of five strains of mice 3·5 days post coitum. (Black areas represent morulae, hatched areas blastocysts; the arrows indicate mean cell number for each strain.)

Fig. 1.

The distribution of cell numbers in embryos of five strains of mice 3·5 days post coitum. (Black areas represent morulae, hatched areas blastocysts; the arrows indicate mean cell number for each strain.)

Although the distributions of cell number were somewhat skew, the departures from normality were not enough to affect an analysis of variance, and accordingly transformation of the data was not undertaken.

There was significant heterogeneity in cell number among strains. C57BL embryos had the highest mean cell number and the highest percentage of blastocysts ; they also showed the greatest variability with respect to cell number. The difference in variability was entirely accounted for by the greater variation between females in the C57BL strain: within-female variance was similar in all strains. Variation between females exceeded variation within females for all five strains, significantly so except for the JU strain. JU morulae and blastocysts had more cells than RIII morulae and blastocysts; but since the percentage of blastocysts was significantly higher in RIII than in JU females, the overall cell number averaged slightly less in JU than in RIII embryos. C3H embryos had the lowest mean cell number and the lowest percentage of blastocysts of any of the strains.

The two extreme strains, C57BL and C3H, were chosen for further study. The differences between them, both for cell number and blastocyst percentage, were significant at the 0·001 level. To determine whether the difference in cell number 3·5 days p.c. reflected a difference in cleavage rate or in time of commencement of cleavage, females were killed at intervals during the first 4 days of pregnancy. To determine whether the effect depended on the embryonic genome or on the maternal environment or cytoplasm, some females of each strain were mated to males of the other strain, and the F1 hybrid embryos examined. Unpublished data of McLaren & Michie, cited by McLaren (1968), had already established that the percentage of blastocysts and the location of the embryos (oviduct or uterus) at 3·5 days p.c. depended on the genotype of the mother. The results of the present series of cell counts are shown in Table 2.

Table 2.

Cell number of C57BL and C3H inbred and F1 hybrid embryos during the first four days of gestation

Cell number of C57BL and C3H inbred and F1 hybrid embryos during the first four days of gestation
Cell number of C57BL and C3H inbred and F1 hybrid embryos during the first four days of gestation

From the morning of the 2nd day onwards, cell number in all four groups showed a linear increase on a logarithmic scale (Fig. 2), as reported by Bowman & McLaren (1970c) for embryos of the Q strain. The data for this period were subjected to an analysis of variance, using the logarithms of litter means, rather than cell counts on individual embryos, in order to allow for the within-female ‘clumping’ effect mentioned above. The four groups did not differ from one another with respect to the slope of the regression line, showing that the differences in cell number by the 4th day of development were not due to any differences in the rate of cleavage. In position, the (C57BL ♀ ×C57BL ♂) line did not differ significantly from the (C57BL ♀ ×C3H ♂) line, nor the (C3H♀ ×C3H♂) from the (C3H♀ ×C57BL♂) line; however, the two inbred strains differed significantly from one another (P < 0·001), as also did the two reciprocal hybrids (P < 0·01). Embryos from C57BL females, whatever the paternal strain, were consistently about 4·5 h ahead of embryos from C3H females in their development.

Fig. 2.

Cell number in embryos of two inbred strains and their reciprocal crosses, during the first 4 days of pregnancy. The solid line represents the calculated regression from the 2nd to the 4th day for the two groups of embryos from C57BL females, which are homogeneous and have been combined. The broken line represents the equivalent regression for C3H females. The two lines differ significantly in position (P < 0·001) but not in slope.

Fig. 2.

Cell number in embryos of two inbred strains and their reciprocal crosses, during the first 4 days of pregnancy. The solid line represents the calculated regression from the 2nd to the 4th day for the two groups of embryos from C57BL females, which are homogeneous and have been combined. The broken line represents the equivalent regression for C3H females. The two lines differ significantly in position (P < 0·001) but not in slope.

To see whether the rate of cleavage of the two strains was also similar in vitro, C57BL and C3H embryos were removed on the afternoon of the 2nd day of pregnancy, at the 2-cell stage, and cultured for 68 h. A higher proportion of C57BL than C3H embryos survived in culture (41/46 v. 25/48; P < 0·001): this probably reflects the fact that the C57BL 2-cell embryos were a few hours further on in development when they were put into culture. Within a strain, the later in the afternoon of the 2nd day that 2-cell embryos are removed from the oviduct, the higher the proportion that develops in vitro (McLaren & Naysmith, unpublished). At the end of the culture period, 95 % of the surviving C57BL embryos were blastocysts, with a mean cell number of 49·6 ± 3·80 (S.E.), while 80 % of the C3H embryos had cavitated, with a mean cell number of 29·1 ± 1·91 (S.E.). Thus cell number attained in vitro was similar to that of the corresponding strain in vivo on the 4th day of pregnancy (Tables 1, 2) and differed significantly between strains (P < 0·001).

The difference in cell number thus reflects a difference in the time of commencement of cleavage, and is maternally determined. To test the possibility that C57BL and C3H females differed in the time at which they mated, mice of the two strains were paired with males for one hour only. Cell counts carried out on the morning of the 3rd day of pregnancy gave means of 7·0 (50 embryos from 8 mice) and 3·9 (26 embryos from 4 mice) cells for C57BL and C3H embryos respectively. As expected, these values are both slightly lower than those observed after overnight mating (see Table 2), but still differed highly significantly from one another (P < 0·001). The difference between the strains therefore lies not in the time of mating, but in the interval between mating and first cleavage.

In order to determine whether the retardation in C3H eggs begins at first cleavage, or affects all post-fertilization stages of development, groups of C57BL and C3H females were mated from 9 to 10 a.m. and killed between 12.30 and 5 p.m. The eggs were squashed, stained, and examined for the presence of maternal chromosomes in the second meiotic division, and for pronuclei. In order to exclude infertile matings, females in which all the eggs still contained chromosomes in the second meiotic metaphase were rejected. The rest of the data are summarized in Table 3. Using individual females as the unit of analysis, regression analysis following angular transformation was applied both to the proportion of eggs in which chromosomes were no longer visible, and to the proportion showing pronuclei. The rate of change with time (i.e. the slopes of the regression lines) did not differ significantly between the two strains, but C3H eggs were consistently retarded in their development, relative to C57BL eggs (P < 0·05 for each criterion). Indeed, Table 3 shows that the proportion of C3H eggs (12/25) still showing meiotic chromosomes between 3 and 5 p.m. was similar to the proportion of C57BL eggs in this condition (14/33) between 12.30 and 1.30.

Table 3.

The tempo of development of C3H and C57BL fertilized eggs on the first day of pregnancy {each group contained 5–7 females)

The tempo of development of C3H and C57BL fertilized eggs on the first day of pregnancy {each group contained 5–7 females)
The tempo of development of C3H and C57BL fertilized eggs on the first day of pregnancy {each group contained 5–7 females)

By the fourth day of gestation, embryos of the five strains included in the present survey contained significantly different numbers of cells. A similar difference has been reported for the comparison between BALB/C embryos and either strain 129 embryos or (BALB/C♀ ×129♂ ) F1 embryos (Gates et al. 1961 ; Whitten & Dagg, 1962). It therefore appears certain that genetic factors may affect early development in the mouse. On the other hand there is no evidence, either in published reports or in the present work, that genetic factors affect the rate of cleavage. On the contrary, our detailed analysis of the early development of one strain combination has shown that the 50 % superiority in cell number shown by C57BL over C3H embryos results not from any superiority in cleavage rate, but from an advantage of some four hours in the time at which cleavage begins.

We do not know whether the genetic effect is exerted through the cytoplasm of the egg, or through the environment in which pre-cleavage development takes place, the ampullary region of the oviduct. For later cleavage stages, Wintenberger-Torres (1964) reported that rate of development of sheep blastocysts was significantly increased either after superovulation, when there is an abnormally high number of corpora lutea, or during treatment with progesterone. For the pre-cleavage period, Krzanowska (1964) noted that fertilization was completed earlier after mating with F1 than with inbred males. In our C3H/C57BL combination, not only was there no difference in cleavage rate, but the difference in cell number was maintained in the standard environment of in vitro culture, and was unaffected by the genotype of the male, in that (C57BL♀ ×C3H♂) and (C3H ♀ ×C57BL ♂) embryos showed as great a difference from one another as did inbred C57BL and C3H embryos. Dickson (1967) claimed that some strain differences in tempo of development were due merely to variations in the time of copulation; in our C3H/C57BL combination this factor can be excluded, since the difference was maintained even when time of copulation was controlled to within an hour.

Since activation of the egg and the formation of pronuclei also appear to take place earlier in C57BL than in C3H eggs, the difference in timing must trace back to some event between copulation and activation. C57BL females must surpass C3H females in the rate at which spermatozoa reach the site of fertilization, or penetrate the eggs or their membranes, or induce activation.

We are grateful to the Ford Foundation and the Lalor Foundation for financial support. We thank Dr Mia Buehr for her assistance with some of the C3H and C57BL one-celled stages.

Bowman
,
P.
&
Mclaren
,
A.
(
1970a
).
Cell number in early embryos from strains of mice selected for large and small body size
.
Genet. Res
.
15
,
261
263
.
Bowman
,
P.
&
Mclaren
,
A.
(
1970b
).
Viability and growth of mouse embryos after in vitro culture and fusion
.
J. Embryol. exp. Morph
.
23
,
693
704
.
Bowman
,
P.
&
Mclaren
,
A.
(
1970c
).
Cleavage rate of mouse embryos in vivo and in vitro
.
J. Embryol. exp. Morph
.
24
,
203
207
.
Dickson
,
A. D.
(
1967
).
Variations in development of mouse blastocysts
.
J. Anat
.
101
,
263
267
.
Gates
,
A. H.
,
Doyle
,
L. L.
&
Noyes
,
R. W.
(
1961
).
A physiological basis for heterosis in hybrid mouse fetuses
.
Am. Zool
.
1
,
449
. (Abst.)
Krzanowska
,
H.
(
1964
).
Time interval between copulation and fertilization in inbred lines of mice and their crosses
.
Folia Biológica, Praha
12
,
231
244
.
Mclaren
,
A.
(
1968
).
Mechanisms affecting embryo development
.
In The Mammalian Oviduct
(ed.
E. S. E.
Hafez
&
R. J.
Blandau
), pp.
477
490
.
University of Chicago
.
Whitten
,
W. K.
&
Dagg
,
C. P.
(
1962
).
Influence of spermatozoa on the cleavage rate of mouse eggs
.
J. exp. Zool
.
148
,
173
183
.
Wintenberger-Torres
,
S.
(
1964
).
Influence de l’équilibre hormonal sur la vitesse de segmentation des œufs de Brebis
.
C. r. hebd. Séanc. Acad. Sci., Paris
259
,
1660
1662
.