A cycle of inactivation and reactivation of one X chromosome in the female (XX) germ line is shown by analysis of gene dosage effects on activity of an X-linked enzyme. The ratio of activities of the X-linked enzyme HPRT and an autosomal enzyme APRT are determined in XX and XY germ cells from embryonic gonads from the 12th to the 17th day of pregnancy. Mitotic stages of XX and XY germ cells on the 12th day have similar HPRT: APRT ratios, but on the 13th day the ratios are significantly higher in XX than XY germ cells. As the XX germ cells enter meiosis they show a marked increase in HPRT:APRT ratio which is primarily due to a rise in X-linked HPRT activity. Comparisons are made with XO germ cells on the 12th and 14th day. On the 12th day, XO do not differ from XX and XY germ cells, suggesting that only one X chromosome is active in XX germ cells at this stage. On the 14th day, on the other hand, the HPRT:APRT ratios in XO and XY germ cells are similar but in XX germ cells the ratio is significantly higher. The twofold difference between the ratio in XX and XO germ cells suggests that by this stage both X chromosomes are active in XX germ cells. The subsequent large increase of the ratio in XX relative to XY germ cells is thought to reflect their differing cell states.

The only cells in female mammals that are known to have both X chromosomes active are the pluripotential cells of the early embryo, up to the epiblast stage (Monk & Harper, 1979), and the oocytes (Epstein, 1969; 1972; Gartler, Liskay & Gant, 1973; Mangia, Abbo-Halbasch & Epstein, 1975; Kozak, McLean & Eicher, 1974; Monk & Kathuria, 1977). A continuous cell lineage might link these two developmental stages, with both X-chromosomes active in the primordial germ cells as well as in their precursors in the epiblast and their successors throughout oogenesis. If, on the other hand, one X chromosome is inactivated in all cells before the emergence of the primordial germ cells, or in the germ cells during their migratory phase or after entry to the genital ridges, then a mechanism for X-chromosome reactivation during oogenesis must be sought.

A little evidence exists in favour of the second alternative. Ohno (1964) claimed to see a heteropyknotic X in primordial germ cells of female mouse embryos during their period of migration; and Semyonova-Tian-Shanskaya & Patkin (1978) reported the presence of sex chromatin and heterochromatization of the X chromosome in female germ cells of human embryos after they had reached the genital ridge. Andina (1978) examined the level of activity of two enzymes coded by the X chromosome, glucose 6-phosphate dehydrogenase (G6PD, E.C. 1.1.1.49) and hypoxanthine phosphoribosyl transferase (HPRT, E.C. 2.4.2.8). In foetal germ cells of XX and XO mouse embryos he found that G6PD activity was similar in the two types of embryo before birth, but the ratio of XX to XO activity increased to nearly two-fold after birth. For HPRT, however, the ratio was between 1-5 and 2 for all prenatal stages examined. If inactivation of the HPRT locus had occurred, then reactivation must have taken place before days post coitum (p.c.), the earliest stage examined.

P. Johnston (personal communication), using an electrophoretic variant of PGK, found evidence for only one active X-chromosome in mitotic germ cells and reactivation of the second X-chromosome before entry into meiosis. A hybrid band for dimeric G6PD, bearing witness to the activity of both X chromosomes, has been reported by Gartler, Andina & Gant (1975) in ovarian extracts from heterozygous human foetuses at 14 – 21 weeks, but no such band was found at an earlier stage (12 weeks) although some of the germ cells were already in early prophase of meiosis. Migeon & Jelalian (1977), on the other hand, were able to detect a hybrid G6PD band as early as 8 weeks and are unconvinced that X inactivation occurs in the female germ line at any stage.

We decided to reinvestigate this question in foetal germ cells of the mouse with particular emphasis on the early period of germ-cell development within the genital ridges ( days p.c.), since the study of Andina (1978) which had yielded somewhat ambiguous results, did not begin until days p.c. We used the very sensitive assay utilized by Monk (1978) in which X-chromosome function is assessed by determining the relative activity of an X-chromosome-coded enzyme, HPRT, to an autosome-coded enzyme, adenine phosphoribosyl transferase APRT (E.C. 2.4.2.7.). Daily determinations were made on normal female (XX) and male (XY) germ cells, before, during and after the period when meiosis is initiated in the female. Some determinations were also made on XO germ cells in female embryos.

Mouse embryos were obtained from randomly bred Q-strain XX females or from XO females (Institute of Animal Genetics, Edinburgh) on successive days from the 12th day of pregnancy ( days post coitum) to term. The gonads were removed into PB1 (Whittingham & Wales, 1969) containing 0·4% polyvinylpyrrolidone instead of albumin (PBl-PVP)and sexed by their characteristic morphology (developing testes show cords) from the 13th day onwards, and by chromosome sexing or by presence or absence of sex chromatin bodies in amnion preparations for 12th-day embryos. XO embryos were distinguished from XX embryos by chromosome counts or by amnion preparations. These two methods were checked against one another on 14th-day embryos and invariably gave the same result in that female embryos lacking sex chromatin proved always to have 39 chromosomes. In embryos from XX females sexchromatin constitution (scored blind) was always concordant with phenotypic sex.

Cells were released from the gonad by mashing it with a needle, the residue of the gonad was then discarded. The large round germ cells were easy to recognize; their identity was confirmed initially by testing for alkaline phosphatase activity (Brinster & Harstad, 1977). The germ cells were washed in PB1-PVP, collected in batches of 50 or 100 in approximately 5 μl. of supporting medium in 10 μl microcaps, and stored at −70°C. Somatic cells released from the gonad in the same way included a high proportion of foetal blood cells. Some samples of these somatic cells were also collected in microcaps. Extracts were prepared by freeze-thawing in liquid nitrogen three times, followed by centrifugation. Approximately 4 μl of supernatant of each sample was transferred to 50 μl of reaction mix (see legend to Fig. 1) at 37°C for 3 h to assay the X-linked enzyme HPRT and the autosomally-linked enzyme APRT. The enzyme reactions were linear over this period (Fig. 1). Details of the reaction have been previously described (Monk & Kathuria, 1977; Monk & Harper, 1978). The results are expressed as a ratio of HPRT: APRT per 100 germ cells thus monitoring X-chromosome:autosome dosage and eliminating sample errors.

Fig. 1

HPRT and APRT activities with increasing time of reaction in a sample of XX germ cells collected 1512 days p.c. 100 germ cells were collected in 5μof PB1-PVP and an extract prepared by freeze thawing. After centrifugation the supernatant was added to 250 of reaction mix and incubated at 37°C. The reaction mix contained [3H]guanine (10μM specific activity 610 Ci/M), [14C]adenine (10μM, specific activity 276 CI/M), phosphoribosyl pyrophosphate (2 mM), magnesium chloride (5 mM), and phosphate buffer (50 mM pH 7·4). Aliquots of 50 μI were removed at times shown and added to ice cold lanthanum chloride (0·1 M) containing adenine and guanine (100 μM). The precipitates were collected on glassfibre filters and the reaction products, [3H]guanine monophosphate and [14C]-adenine monophosphate, determined in a Packard scintillation counter.

Fig. 1

HPRT and APRT activities with increasing time of reaction in a sample of XX germ cells collected 1512 days p.c. 100 germ cells were collected in 5μof PB1-PVP and an extract prepared by freeze thawing. After centrifugation the supernatant was added to 250 of reaction mix and incubated at 37°C. The reaction mix contained [3H]guanine (10μM specific activity 610 Ci/M), [14C]adenine (10μM, specific activity 276 CI/M), phosphoribosyl pyrophosphate (2 mM), magnesium chloride (5 mM), and phosphate buffer (50 mM pH 7·4). Aliquots of 50 μI were removed at times shown and added to ice cold lanthanum chloride (0·1 M) containing adenine and guanine (100 μM). The precipitates were collected on glassfibre filters and the reaction products, [3H]guanine monophosphate and [14C]-adenine monophosphate, determined in a Packard scintillation counter.

In the Q-strain of mice, primordial germ cells first enter the genital ridges at 10– 11 days post coitum (Figure 2); at days post coitum the gonads can for the first time be sexed by their appearance, since under a dissecting microscope cords can be seen in the developing testes; 24 h later some of the germ cells in the ovary have entered the leptotene stage of meiotic prophase. At days zygotene stages are the most common, and at and days p.c. pachytene stages prevail. In the male, by days p.c., most of the XY germ cells in the testis are T-prospermatogonia and are in a state of mitotic arrest.

Fig. 2

Stages of germ cell development in mouse embryos of the Q strain. Entry into meiosis and the progression of female germ cells through the first stages of meiotic prophase are shown with reference to number of days’ gestation. The times given are a guide to when the majority of the germ cells are in a particular stage. In fact cells enter into meiosis continuously over a period of at least 2 days. Individual germ cells are in leptotene or zygotene for periods of several hours and in pachytene for several days (see also Borum, 1961; Peters, 1970).

Fig. 2

Stages of germ cell development in mouse embryos of the Q strain. Entry into meiosis and the progression of female germ cells through the first stages of meiotic prophase are shown with reference to number of days’ gestation. The times given are a guide to when the majority of the germ cells are in a particular stage. In fact cells enter into meiosis continuously over a period of at least 2 days. Individual germ cells are in leptotene or zygotene for periods of several hours and in pachytene for several days (see also Borum, 1961; Peters, 1970).

Figure 3 shows the HPRT: APRT ratio for XX female and XY male germ cells as a function of gestational age of the embryo. XX germ cells have equivalent HPRT:APRT ratios to XY germ cells at days but show approximately seven times the XY value by days. The ratios of HPRT: APRT from gonadal somatic cells show little difference between male and female embryos (pooled in Figure 3) and no significant change with gestational age.

Fig. 3

HPRT:APRT ratios as a function of gestational age in germ cells from female XX and male XY embryos and in gonadal (somatic) cells. Duplicate samples of 100 germ cells were collected from between two to six male or female embryos from pregnant females on successive days p.c. Extracts were prepared, added to 50 μl of reaction mix and assayed as described in the legend to Fig. 1 for HPRT and APRT activities ([3H]guanine, specific activity 700 CI/M, [,4C]adenine> 287 CI/M). The reaction time was 3 h. Samples of somatic cells taken from male and female embryonic gonads showed no significant difference in HPRT: APRT ratios and are pooled in the figure. A repeat of this experiment using embryonic germ cells at 1112, 1212, 1312, 1512, and 1612 days p.c. gave similar results.

Fig. 3

HPRT:APRT ratios as a function of gestational age in germ cells from female XX and male XY embryos and in gonadal (somatic) cells. Duplicate samples of 100 germ cells were collected from between two to six male or female embryos from pregnant females on successive days p.c. Extracts were prepared, added to 50 μl of reaction mix and assayed as described in the legend to Fig. 1 for HPRT and APRT activities ([3H]guanine, specific activity 700 CI/M, [,4C]adenine> 287 CI/M). The reaction time was 3 h. Samples of somatic cells taken from male and female embryonic gonads showed no significant difference in HPRT: APRT ratios and are pooled in the figure. A repeat of this experiment using embryonic germ cells at 1112, 1212, 1312, 1512, and 1612 days p.c. gave similar results.

Figure 4 shows that the difference between XX and XY HPRT: APRT ratios in XX and XY germ cells in Figure 3 is a consequence of changes in activity of the X-linked enzyme HPRT rather than in the activity of the autosomally-linked APRT.

Fig. 4

HPRT and APRT activities in XX and XY germ cells as a function of gestational age of the embryos. Assays were performed for 3 h as described in the legend to Fig. 1. The results for the two enzymes looked at separately rather than as a ratio show considerable scatter due to sampling errors introduced at some stage of germ cell collection or preparation of extracts.

Fig. 4

HPRT and APRT activities in XX and XY germ cells as a function of gestational age of the embryos. Assays were performed for 3 h as described in the legend to Fig. 1. The results for the two enzymes looked at separately rather than as a ratio show considerable scatter due to sampling errors introduced at some stage of germ cell collection or preparation of extracts.

Four additional experiments were done to see whether the significant difference between XX and XY germ cells at days post coitum (Figure 3) could be confirmed. Table 1 shows that the divergence is indeed significant by days p.c.

Table 1

HPRT/APRT ratios in XX and XY foetal germ cells 1212 days post coitum

HPRT/APRT ratios in XX and XY foetal germ cells 1212 days post coitum
HPRT/APRT ratios in XX and XY foetal germ cells 1212 days post coitum

XX and XY germ cells have a similar HPRT:APRT ratio at days p.c.

If we assume that XX and XY germ cells at this stage are equivalent in all respects other than sex chromosome constitution, the results strongly suggest that only one X chromosome is active in the XX germ cells.

This interpretation is confirmed by comparison of XX, XO and XY germ cells in embryos from an XO mother at days p.c. (Table 2). A statistical analysis of the data in Table 2 shows no significant difference in HPRT: APRT ratios for XO and XX female and XY male germ cells.

Table 2

HPRT/APRT ratios in foetal germ cells from an XO mother 1112 days post coitum

HPRT/APRT ratios in foetal germ cells from an XO mother 1112 days post coitum
HPRT/APRT ratios in foetal germ cells from an XO mother 1112 days post coitum
Table 3

HPRT/APRT ratios in foetal germ cells from an XO mother 1312 days post coitum

HPRT/APRT ratios in foetal germ cells from an XO mother 1312 days post coitum
HPRT/APRT ratios in foetal germ cells from an XO mother 1312 days post coitum

A similar comparison of XX, XO and XY germ cells in embryos from an XO mother at days p.c. is given in Table 3. The XO still resemble the XY germ cells, but now differ significantly from the XX germ cells. Indeed, the difference between XX and XO germ cells in HPRT:APRT ratio approaches the twofold value that we would expect to see on reactivation of the silent X chromosome in the XX cells.

The ratio of HPRT to APRT activity gives a good relative index of X-chromosome function when essentially similar cell types are being compared, for example male and female embryos at the same stage of cleavage (Monk & Kathuria, 1977; Monk, 1978). The more dissimilar the cell types, the less valid the comparison becomes, since changing enzyme activities may reflect some functional requirement characteristic of a particular cell type. In the present work, the comparison of XX and XY germ cells may be considered to give a reasonable indication of X-chromosome function at and days p.c., when both are proliferating mitotically as gonia, but is unlikely to be valid later in gestation, when XX germ cells have entered the prophase of meiosis and XY germ cells are in mitotic arrest. A more reliable picture is given by the comparison of XX with XO germ cells, since the developmental pathway is the same.

We have shown that the HPRT: APRT ratio in XX germ cells is similar to that in XY germ cells at days p.c. The conclusion that only one X-chromosome is expressed in the XX oogonia is confirmed by the observation that XO oogonia at this stage also have a similar HPRT:APRT ratio. We cannot say whether X inactivation occurred in germ cell precursors at the same time as in other epiblast cells (i.e. by 6 days p.c., Monk & Harper, 1979) or at some later stage, between 6 and days. An alternative explanation that HPRT activity is regulated independently of functional gene dosage, is made less likely by the absence of such regulation in early embryogenesis and later stages of oogenesis.

By days p.c. the HPRT: APRT ratio in XX germ cells has increased to a level significantly higher than that in XY germ cells, suggesting that the second X chromosome is now being expressed. This conclusion is supported by the two-fold difference in ratio seen between XX and XO germ cells at days p.c. The significantly lower values shown by the XO germ cells cannot be accounted for by developmental retardation: with respect to the onset of meiosis XO germ cells are delayed at most a few hours (McLaren unpublished observations), and at days p.c. there has been little change in HPRT: APRT ratio in the previous 24 h.

The very rapid increase in HPRT:APRT ratio in XX relative to XY germ cells after days p.c. is due to an increase in X-linked HPRT activity. This probably corresponds to some functional requirement of meiotic prophase rather than reflecting X-chromosome dosage.

With respect to HPRT activity, reactivation of the silent X chromosome has occurred by days p.c. We cannot say exactly how much earlier it occurs, because we are ignorant of the kinetics of HPRT and APRT synthesis and degradation in germ cells. However, one may speculate, as did Gartler et al (1975), that it is linked to the onset of meiosis. There is some evidence (McLaren, 1981) that germ cells with two X chromosomes (XX) enter the leptotene stage of meiotic prophase in response to the influence of Meiosis-Inducing Substance (Byskov & Saxen, 1976; O & Baker, 1976) more readily than do those with only a single X (XO or XY). This would imply that the second X in XX germ cells is already expressed at the time that this response occurs. It may be that reactivation of the inactive X in female embryos is initiated by the same event that initiates the meiotic process, and indeed that reactivation is an integral part of this process, but precedes the entry into meiotic prophase. This reactivation during oogenesis is the only known example of the reversal of the non-functional state of the silent X chromosome in female mammals.

Andina
,
R. J.
(
1978
).
A study of X-chromosome regulation during oogenesis in the mouse
.
Expl Cell. Res
.
111
,
211
218
.
Borum
,
K.
(
1961
).
Oogenesis in the mouse. A study of the meiotic prophase
.
Expl Cell Res
.
24
,
495
507
.
Brinster
,
R. L.
&
Harstad
,
H.
(
1977
).
Energy metabolism in primordial germ cells of the mouse
.
Expl Cell Res
.
109
,
111
117
.
Byskov
,
A. G.
&
Saxen
,
L.
(
1976
).
Induction of meiosis in fetal mouse testis in vitro
.
Devl Biol
.
52
,
193
200
.
Epstein
,
C. J.
(
1969
).
Mammalian oocytes: X-chromosome activity
.
Science
163
,
1078
1079
.
Epstein
,
C. J.
(
1972
).
Expression of the mammalian X-chromosome before and after fertilisation
.
Science
175
,
1467
1468
.
Gartler
,
S. M.
,
Andina
,
R.
&
Gant
,
N.
(
1975
).
Ontogeny of X-chromosome inactivation in the female germ line
.
Expl Cell. Res
.
91
,
454
457
.
Gartler
,
S. M.
,
Liskay
,
R. M.
&
Gant
,
N.
(
1973
).
Two functional X-chromosomes in human fetal oocytes
.
Expl Cell Res
.
82
,
464
466
.
Johnston
,
P. G.
(
1979
).
Mouse Newsletter, no
.
61
,
July
1969.
Kozak
,
L. P.
,
McLean
,
G. K.
&
Eicher
,
E. M.
(
1974
).
X-linkage of phosphoglycerate kinase in the mouse
.
Biochem. Genet
.
11
,
41
47
.
Mangia
,
F.
,
Abbo-Halbasch
,
G.
&
Epstein
,
C. J.
(
1975
).
X-chromosome expression during oogenesis in the mouse
.
Devl Biol
.
45
,
366
368
.
McLaren
,
A.
(
1981
).
The fate of germ cells in the testis of fetal Sex-reversed mice
.
J. Reprod. Fert., in press
.
Migeon
,
B. R.
&
Jelalian
,
K.
(
1977
).
Evidence for two active X-chromosomes in germ cells of female before meiotic entry
.
Nature, Lond
.
269
,
242
243
.
Monk
,
M.
(
1978
).
Biochemical studies on mammalian X-chromosome activity
.
In Development in Mammals, vol. m
(ed.
M. H.
Johnson
).
Amsterdam
:
North Holland
.
Monk
,
M.
&
Harper
,
M.
(
1978
).
X-chromosome activity in preimplantation mouse embryos from XX and XO mothers
.
J. Embryol. exp. Morph
.
46
,
53
64
.
Monk
,
M.
&
Harper
,
M. I.
(
1979
).
Sequential X-chromosome inactivation coupled With cellular differentiation in early mouse embryos
.
Nature
281
,
311
313
.
Monk
,
M.
&
Kathuria
,
H.
(
1977
).
Dosage compensation for an X-linked gene in preimplanation mouse embryos
.
Nature, Lond
.
270
,
599
601
.
O
,
W.
&
Baker
,
T. G.
(
1976
).
Initiation and control of meiosis in hamster gonads in vitrol
.
J. Reprod. Fert
.
48
,
399
401
.
Ohno
,
S.
(
1964
).
Life history of female germ cells in mammals
.
2nd Intern. Conf. Congenita Malformations
, 1963. pp.
36
40
. Intern. Med. Congr., Ltd., New York.
Peters
,
H.
(
1970
).
Migration of gonocytes into the mammalian gonad and their differentiation
.
Phil. Trans. R. Soc. Lond. B
259
,
91
101
.
Semyonova-Tian-Shanskaya
,
A. G.
&
Patkin
,
E. L.
(
1978
).
Changes in gonocyte nuclei at different stages of their differentiation in early human female embryos
.
Arkh. Anat. Gistol. Embriol
.
74
,
91
97
. (In Russian.)
Whittingham
,
D. G.
&
Wales
,
R. G.
(
1969
).
Storage of two cell mouse embryos in vitro
.
Aust. J. Biol. Sci
.
22
,
1065
1068
.