I am indebted to Mary Lyon as her X-inactivation hypothesis stimulated my mentor, Barton Childs, and in turn, myself, to think about the consequences of X-inactivation in heterozygous females. I often reread her original papers setting forth the single active X hypothesis, and still marvel at the concise and compelling exposition of the hypothesis and the logical predictions which seemed prophetic at my first reading, and have survived the test of time.

My contribution to this Festschrift reviews evidence derived from studies of DNA methylation, species variation and DNA replication that reveals an important role for methylated CpG islands and suggests a role for late DNA replication in propagating X inactivation from one cell to its progeny. These studies also show that X inactivation is a powerful research tool for identifying the factors which program and maintain developmental processes.

The phenomenon of X-chromosome inactivation in female mammals, whereby one of the two X chromosome present in each cell of the female embryo is inactivated early in development, was first described by Mary Lyon in 1961. Nearly 30 years later, the mechanism of X-chromosome inactivation remains unknown. Strong evidence has accumulated over the years, however, for the involvement of a major switch or inactivation centre on the mouse X chromosome. Identification of the inactivation centre at the molecular level would be an important step in understanding the mechanism of X-inactivation. In this paper we review the evidence for the existence and location of the X-inactivation centre on the mouse X-chromosome, present data on the molecular genetic mapping of this region, and describe ongoing strategies we are using to attempt to identify the inactivation centre at the molecular level.

Spontaneously cycling LT/Sv strain female mice were mated to hemizygous Rb(X.2)2Ad males in order to facilitate the distinction of the paternal X chromosome, and the pregnant females were autopsied at about midday on the tenth day of gestation. Out of a total of 222 analysable embryos recovered, 165 (74·3%) were diploid and 57 (25·7%) were triploid. Of the triploids, 26 had an XXY and 31 an XXX sex chromosome constitution. Both embryonic and extra-embryonic tissue samples from the triploids were analysed cytogenetically by G-banding and by the Kanda technique to investigate their X-inactivation pattern. The yolk sac samples were separated enzymatically into their endodermally-derived and mesodermally-derived components, and these were similarly analysed, as were similar samples from a selection of control XmXp diploid embryos. In the case of the XmXmY digynic triploid embryos, a single darkly-staining Xm chromosome was observed in 485 (82·9%) out of 585, 304 (73·3%) out of 415, and 165 (44·7%) out of 369 metaphases from the embryonic, yolk sac mesodermally-derived and yolk sac endodermally-derived tissues, respectively. The absence of a darkly staining X-chromosome in the other metaphase spreads could either indicate that both X-chromosomes present were active, or that the Kanda technique had failed to differentially stain the inactive X-chromosome(s) present. In the case of the XmXmXp digynic triploid embryos, virtually all of the tissues analysed comprised two distinct cell lineages, namely those with two darkly-staining X-chromosomes, and those with a single darkly staining X-chromosome. Four X-inactivation patterns were consequently observed in this group, namely, (XmXp)Xm, (XmXm)Xp, (Xm)XmXp and XmXm(Xp) in which the inactive X is enclosed in parentheses. The incidence of these various classes varied among the tissues analysed. There was, however, a clear pattern of non-random selective paternal X-inactivation in yolk sac endodermally-derived samples which possessed two inactive X-chromosomes. This finding contrasts with the situation observed in the yolk sac mesodermally-derived and embryonic samples which possessed two inactive X-chromosomes, where the ratio of (XmXm)Xp:Xm(XmXp) was 1:1·20 and 1:1·03, respectively, being clear evidence that random X-inactivation had occurred in these tissues.

Regardless of their origins, functions, and base compositions, all DNAs are scriptures written following the same grammatical rule. At the level of syllables, two, CG and TA are seldom used, while three, TG, CT and CA are utilized with abundance. Accordingly, at the level of three-letter words, two complementary base trimers, CTG and CAG, invariably enjoy frequent usage. Inasmuch as two of the three frequently used syllables, TG and CA are complementary to each other, while two seldom used syllables, CG and TA, are both palindromes, two complementary strands of DNA are inherently symmetrical with each other. Consequently, palindromic sequences as favourite targets of DNA-binding proteins occur at unsuspectedly high frequencies, if they contain TG and CA or CTG and CAG. Nevertheless, there are grammatical rules operating among these high frequency palindromes as well; e.g. the palindromic tetramer TGCA occurs nearly two times more often than its reciprocal; CATG. Thus, DNA-binding proteins are provided with a wealth of abundant targets whose densities are influenced by a regional difference in GC/AT ratios to variable degrees. One palindromic heptamer CAGNCTG is an ideal target of one DNA-binding protein engaged in chromosome packaging and in generation of banding patterns. This heptamer occurs once every 1000 bases in moderately GC-rich sequences, while its incidence is reduced to once every 3000 bases in extremely AT-rich sequences. The above must be the very reason that a solitary human X-chromosome DNA coated with mouse DNA-binding proteins in mouse-man somatic hybrids still maintains the original banding pattern and that the inactive X remains inactive, while the active X remains active.

X/Y male mice carrying the sex reversal factor, Sxr, on their Y chromosomes typically produce 4 classes of progeny (recombinant X/X Sxr ♂♂ and X/Y non-Sxr ♂♂, and non-recombinant X/X ♀♀ and X/Y Sxr ♂♂) in equal frequencies, these deriving from obligatory crossing over between the chromatids of the X and Y during meiosis. Here we show that X/Y males that, exceptionally, carry Sxr on their X chromosome, rather than their Y, produce fewer recombinants than expected. Cytological studies confirmed that X-Y univalence is frequent (58%) at diakinesis as in X/Y Sxr males, but among those cells with X-Y bivalents only 38% showed normal X-Y pseudo-autosomal pairing. The majority of such cells (62%) instead showed an illegitimate pairing between the short arms of the Y and the Sxr region located at the distal end of the X, and this can be understood in terms of the known homology between the testis-determining region of the Y short arm and that of the Sxr region. This pairing was sufficiently tenacious to suggest that crossing over took place between the 2 regions, and misalignment and unequal exchange were suggested by indications of bivalent asymmetry. Metaphase II cells deriving from meiosis I divisions in which the normal X-Y exchange had not occurred were also found. The cytological data are therefore consistent with the breeding results and suggest that normal pseudo-autosomal pairing and crossing over is not a prerequisite for functional germ cell formation. The data support the concept that Y short arm-Sxr pairing and crossing over may be the mechanism responsible for the occurrence of the Sxr variants reported in the literature.

Karyotypic investigation of a phenotypically normal but sterile male mouse showed the presence of an XYY sex chromosome constitution. The synaptic behaviour of the three sex chromosomes was examined in 65 pachytene cells. The sex chromosomes formed a variety of synaptic configurations: an XYY trivalent (40%); an XY bivalent and Y univalent (38·5%); an X univalent and YY bivalent (13·8%); or X, Y, Y univalence (7·7%). There was considerable variation in the extent of synapsis and some of the associations clearly involved nonhomologous pairing. These observations have been compared with previously published information on chromosome configurations at metaphase I from other XYY males.

The parental origin of 3 de novo X-autosome translocations in females with Duchenne Muscular Dystrophy (DMD) was studied by means of methylation analysis using the X-linked probe M27ß. In all three the translocation was found to be paternal in origin. The parental origin of X-autosome translocations in females with and without DMD is compared with other structural abnormalities of the X and with autosomal translocations.

Several genes expressed in kidney and other tissues determine phosphate homeostasis in extracellular fluid. The major form of inherited hypophosphatemia in humans involves an X-linked locus (HPDR, Xp22.31-p21.3). It has two murine homologues (Hyp and Gy) which map to closely-linked but separate loci (crossover value 0·4%–0·8%). Both murine mutations impair Na+-phosphate cotransport in renal brush border membrane; an associated renal disorder of 1, 25-dihydroxyvitamin D3 (1, 25(OH)2D) metabolism has been characterized in Hyp mice. Whereas experiments with cultured Hyp renal epithelium indicate that the gene is expressed in kidney, studies showing the development of the mutant renal phenotype in normal mice parabiosed to Hyp mice implicate a circulating factor; these findings can be reconciled if the humoral factor is of renal origin. The gene dose effect of HPDR, Hyp and Gy on serum phosphorus values is consistently deviant and heterozygotes resemble affected hemizygotes. The deviant effect is also seen on renal phosphate transport; all mutant females (Hyp/Hyp and Hyp/+) have similar phenotypes. On the other hand, there is a normal gene dose effect of HPDR in mineralized tissue; tooth PRATIO (pulp area/tooth area) values for heterozygotes are distributed between those for affected males and normals. The tooth data imply that the X chromosome locus is expressed in both renal and non-renal cells. The polypeptide product of the X chromosome gene(s) is still unknown.

The locus for properdin (properdin factor complement, Pfc), a plasma glycoprotein, has been mapped to band A3 of the mouse X chromosome by in situ hybridization to metaphase spreads containing an X;2 Robertsonian translocation. The X-linkage of the locus has also been confirmed by analysis of Mus musculus x Mus spretus interspecific crosses. The XA3 localization for Pfc places it in the chromosomal segment conserved between man and mouse which is known to contain at least six other homologous loci (Cybb, Otc, Syn-1 Maoa, Araf, Timp).

Linkage data relative to the markers tabby and glucose-6-phosphate dehydrogenase are presented to locate X-linked cataract (Xcat) in the distal portion of the mouse X-chromosome between jimpy and hypophosphatemia. The human X-linked cataract-dental syndrome, Nance–Horan Syndrome, also maps closely to human hypophosphatemia and would suggest homology between mouse Xcat and human Nance-Horan Syndrome genes. In hemizygous males and homozygous females penetrance is complete with only slight variation in the degree of expression. Phenotypic expression in Xcat heterozygous females ranges from totally clear to totally opaque lenses. The phenotypic expression between the two lenses of a heterozygous individual could also vary between totally clear and totally opaque lenses. However, a correlation in the degree of expression between the eyes of an individual was observed. A variegated pattern of lens opacity was evident in female heterozygotes. Based on these observations, the site of gene action for the Xcat locus is suggested to be endogenous to the lens cells and the precursor cell population of the lens is concluded to be small. The identification of an X-linked cataract locus is an important contribution to the estimate of the number of mutable loci resulting in cataract, an estimate required so that dominant cataract mutagenesis results may be expressed on a per locus basis. The Xcat mutation may be a useful marker for a distal region of the mouse X-chromosome which is relatively sparsely marked and the X-linked cataract mutation may be employed in gene expression and lens development studies.

A gene, Hst-3, responsible for sterility in F1 males from crosses between Mus spretus and laboratory strains of mice such as C57BL/6, has been localized on the distal part of the X chromosome, using both DNA probes and biochemical markers on a panel of Fl(C57BL/6 × SEG) × C57BL/6 backcross males. This gene may be a model for studying mammalian hybrid sterility.

The mouse mutation hairy ears (Eh) originated in a neutron irradiation experiment at Oak Ridge National Laboratory. Subsequent linkage studies with Eh and other loci on Chr 15 suggested that it is associated with a chromosomal rearrangement that inhibits recombination since it shows tight linkage with several loci occupying the region extending from congenital goiter (cog) distal to caracul (ca). We report here (1) linkage experiments confirming this effect on recombination and (2) meiotic and mitotic cytological studies that confirm the presence of a chromosomal rearrangement. The data are consistent with the hypothesis of a paracentric inversion in the distal half of Chr 15. The effect of the inversion extends over a minimum of 30 cM, taking into account the genetic data and the cytologically determined chromosomal involvement extending to the region of the telomere.

We present data on the effects of three chromosome 17 deletions on transmission ratio distortion (TRD) and sterility of several t-haplotypes. All three deletions have similar effects on male TRD: that is, Tdel/tcomplete genotypes all transmit their t–haplotype in very high proportion. However, each deletion has different effects on sterility of heterozygous males, with Tor/t being fertile, Thp/t less fertile, and Torl/t still less fertile. These data suggest that wild-type genes on chromosomes homologous to f-haplotypes can be important regulators of both TRD and fertility in males, and that the wild-type genes concerned with TRD and fertility are at least to some extent different. The data also provide a rough map of the positions of these genes.

We previously identified a primary sex-determining locus, Tas, on mouse Chr 17 that causes ovarian tissue development in C57BL/6J Thp/ + and Torl/ + individuals if the AKR/J Y chromosome is present. We hypothesized that Tas is located within the region of Chr 17 deleted by Thp and Torl and that C57BL/6J carries a diagnostic Tas allele, based on the observation that ovarian tissue develops in XY mice when Thp is on a C57BL/6J inbred strain background, whereas normal testicular development occurs when Thp is on a C3H/HeSnJ inbred strain background. To test this hypothesis, we mated (C57BL/6J × C3H/HeSnJ)F1 females to C57BL/6J Thp/ + hermaphrodites. As expected, half of the XY Thp /+ offspring developed ovarian and testicular tissue while half developed exclusively testicular tissue. Unexpectedly, the inheritance of selected Chr 17 molecular loci was independent of gonadal development, as half of the male and hermaphroditic offspring inherited C3H/HeSnJ-derived Chr 17 loci and half inherited C57BL/6J-derived Chr 17 loci. We conclude that for ovarian tissue to develop in an XY Thp/ + or XY Tort/ + individual (1) Tas must be present in a hemizygous state, which is accomplished by heterozygosity for the Thp or Tort deletions; (2) the AKR/J-derived Y chromosome must be present; and (3) an additional locus involved in primary sex determination must be present in a homozygous C57BL/6J state. This newly identified gene may be one of the previously defined loci, tda-1 or tda-2.

Transmission ratio distortion (TRD) is a property of the complete t-haplotype which results in the preferential transmission of the t-haplotype chromosome from heterozygous t/+ males to the majority of the progeny. Available data suggest that in t/+ males, a dysfunction of the wild-type sperm within the female reproductive tract is responsible for the observed deviation from Mendelian segregation ratios. Genetically, Lyon has shown that multiple loci within the t-complex are required for maximum levels of TRD. These loci include multiple t-complex distorters (Teds) which act upon a single t-complex responder (Ter). Testis-expressed genes have been cloned which map to the same subregions of the t-complex as the Teds and Ter and are thus considered candidate genes for the products of these loci. To begin to understand how the products of these loci biochemically control TRD, the expression of three TRD-candidate genes (Tcp-1, D17Leh117c3, and D17Leh66) has been determined in populations of spermatocytes and differentiated spermatids purified to near homogeneity by unit gravity sedimentation. Fractions covering the entire gradient were analysed resulting in a more accurate picture of the precise timing of expression than previously reported. Transcription of all three genes was up-regulated in pachytene primary spermatocytes and persisted at stable levels through the haploid spermatid stages. Significantly, only levels of mRNA encoded by D17Leh66, the candidate gene for Tcr, increased from early round to elongating-stage spermatids. If this pattern of expression does, in fact, represent Tcr, these data provide the first direct evidence that wild-type and t-haplotype Tcr elements could be differentially expressed in haploid spermatids.

Mammalian genetics now allows a molecular study of genomic regions previously analysed by genetic and embryological techniques. To simplify such an analysis, we have established a number of libraries of mouse DNA in Yeast Artificial Chromosome (YAC) vectors, constructed either by partial digestion with EcoRI, or by complete digestion with enzymes which cut rarely in the mammalian genome. In this paper we report the construction of complete digest libraries prepared from mouse genomic DNA using the rare cutter enzymes NoiI and BssHII, and the detection of gene loci from the H-2 complex, the t–complex, and other loci from the mouse genome. Due to their large insert size, YAC clones simplify the cloning of extended regions of the mouse genome surrounding known developmental mutations and should, after introduction into the germ line, offer a high probability of correct expression of the genes contained within the cloned region. We hope that this will allow the use of YAC clones to scan regions of interest such as the t–complex for specific genes by testing DNA introduced into transgenic mice for the ability to complement mutations localised to this region.

Diploid mouse conceptuses lacking a paternal genome can form morphologically normal but small fetuses of up to 25 somites, but they invariably fail to develop beyond mid-gestation. Such conceptuses differ from normal most notably in the poor development of extra-embryonic tissues which are largely of trophectodermal and primitive endodermal origin. However, it is not clear whether the demise of diploid parthenogenetic (P) or gynogenetic (G) conceptuses is attributable entirely to the defective development of these two tissues or whether differentiation of the primitive ectoderm, the precursor of the foetus, extra-embryonic mesoderm and amnion, is also impaired by the absence of a paternal genome. Therefore, a new blastocyst reconstitution technique was used which enabled primitive ectoderm from P blastocysts to be combined with primitive endoderm and trophectoderm from fertilization-derived (F) blastocysts. One third of the ‘triple tissue’ reconstituted blastocysts that implanted yielded foetuses. However, all foetuses recovered on the llth or 12th day of gestation were small and, with one exception, either obviously retarded or arrested in development. The exception was a living 44 somite specimen which is the most advanced P foetus yet recorded. Foetuses were invariably degenerating in conceptuses recovered on the 13th day. In contrast, at least 16% of control reconstituted blastocysts with primitive ectoderm as well as primitive endoderm and trophectoderm of F origin developed normally on the 13th day of gestation or to term. Hence, the presence of a paternal genome seems to be essential for normal differentiation of all 3 primary tissues of the mouse blastocyst.

The P foetuses that developed from reconstituted blastocysts were so closely invested by their membranes that they often showed abnormal flexure of the posterior region of the body. Several also showed a deficiency of allantoic tissue. Therefore, the possibility that the defect in development of P primitive ectoderms resided in their extra-embryonic tissues was investigated by analysing a series of chimaeras produced by injecting them into intact F blastocysts. The foregoing anomalies were not discernible even when P cells made a large contribution to the extra-embryonic mesoderm or amnion plus umbilical cord. Furthermore, selection against P cells was no greater in extra-embryonic derivatives of the primitive ectoderm than in the foetus itself.