X-Inactivation

X-inactivation (also termed X-chromosome inactivation [XCI] or lyonization) is the chromosome-wide epigenetic mechanism by which female therian mammals (XX) transcriptionally silence one of their two X chromosomes in every somatic cell, equalizing X-linked gene dosage with heterogametic males (XY). While the human X chromosome spans ~156 megabases (Mb) and encodes approximately 850 to 900 protein-coding genes involved in housekeeping, neurodevelopment, and coagulation, the male-determining Y chromosome spans ~57 Mb and retains fewer than 70 protein-coding genes. Without dosage compensation, XX females would express twice the transcript and protein abundance of nearly a thousand X-linked loci relative to XY males, disrupting multi-subunit stoichiometric complexes across somatic tissues.

In 1949, Canadian neuroanatomists Murray L. Barr and Ewart G. Bertram observed a dense, perinuclear heterochromatic structure—now designated the Barr body—exclusively within the interphase nuclei of female cat motor neurons. Twelve years later, in 1961, British geneticist Mary F. Lyon synthesized cytogenetic observations of X-autosome translocations and dappled coat-color phenotypes in heterozygous female mice to formulate the Lyon hypothesis: during early embryogenesis, each somatic cell independently and randomly inactivates either the maternally inherited (X^m) or paternally inherited (X^p) X chromosome, and once established, that choice is stably propagated through all subsequent mitotic divisions, rendering mammalian females somatic clonal mosaics for X-linked inheritance.

Chromosome Counting, Choice, and XIST Coating in cis

Random X-inactivation is initiated within the inner cell mass (epiblast) of the implanting blastocyst through a bipartite regulatory locus on the long arm of the X chromosome (Xq13.2 in humans) known as the X-inactivation center (XIC). Somatic cells follow the empirical "n − 1 rule": regardless of the total number of X chromosomes present in a diploid nucleus, one X chromosome remains transcriptionally active (Xa) per diploid autosome set, and all remaining X chromosomes (n − 1) are converted into inactive Barr bodies (Xi). Thus, a 46,XY male or 45,X female possesses zero Barr bodies, a normal 46,XX female or 47,XXY male possesses one Barr body, and a 47,XXX female possesses two Barr bodies.

The master molecular effector within the XIC is XIST (X-inactive specific transcript), a 17- to 19-kilobase (kb) capped, spliced, and polyadenylated long non-coding RNA (lncRNA). Prior to implantation, both X chromosomes express low basal levels of XIST; in rodents (and to a lesser degree through XACT competition in primates), Xist is antagonized in cis by its antisense lncRNA transcript Tsix. Upon differentiation of the epiblast, transient physical pairing of the two XIC loci and stochastic redistribution of pluripotency factors (OCT4, SOX2, NANOG, and REX1) breaking symmetry with the E3 ubiquitin ligase RNF12 (RLIM) lock in mutually exclusive fates:

• On the future active X (Xa), the XIST promoter is silenced and methylated.
• On the future inactive X (Xi), XIST transcription is strongly upregulated. Rather than diffusing into the cytoplasm for translation, XIST RNA molecules bind the nuclear matrix protein hnRNP U (SAF-A) and spread three-dimensionally across the territory of the chromosome from which they were transcribed, coating the future Xi in cis from telomere to telomere.

Sequential Chromatin Remodeling and DNA Methylation

Coating of the future Xi by XIST RNA triggers a tightly ordered, multistep chromatin remodeling cascade mediated by modular conserved repeat elements (Repeats A through F) within the XIST transcript:

1. Transcriptional Silencing and Histone Deacetylation (Repeat A / SPEN): Within hours of XIST coating, the 5′ A-repeat domain of XIST directly binds the RNA-binding protein SPEN (SHARP), which recruits the SMRT/NCoR corepressor complex and histone deacetylase 3 (HDAC3). HDAC3 strips activating acetyl groups from histone tails (specifically deacetylating H3K27ac and H4Kac) and evicts RNA Polymerase II from gene promoters.
2. Polycomb Repressive Complex Recruitment (Repeat B / PRC1 & PRC2): The B-repeat of XIST binds hnRNP K, which recruits non-canonical Polycomb Repressive Complex 1 (PRC1) to monoubiquitinate histone H2A at lysine 119 (H2AK119ub1). H2AK119ub1 in turn recruits Polycomb Repressive Complex 2 (PRC2)—containing the catalytic methyltransferase EZH2—which trimethylates histone H3 at lysine 27 (H3K27me3) across the inactive X chromosome, while demethylases remove activating H3K4me2/3 marks.
3. Nucleosome Compaction and Late Replication: During subsequent cell cycles, canonical H2A histones within nucleosomes on the Xi are replaced by the repressive variant macroH2A1/2, the structural maintenance protein SMCHD1 compacts the chromosome into two megadomains separated by the DXZ4 boundary element, and the Xi shifts its DNA replication timing to late S-phase.
4. Irreversible Lock-In via CpG Island DNA Methylation: Finally, de novo DNA methyltransferase DNMT3B introduces 5-methylcytosine (5mC) across promoter CpG islands on the Xi, which is subsequently copied across every mitotic division by maintenance methyltransferase DNMT1 (DNA methylation). Once locked by DNA methylation and macroH2A, transcriptional silencing of the Barr body persists throughout the somatic lifespan even if the XIST gene is experimentally deleted in adult cells.

Imprinted vs. Random X-Inactivation and Germline Reactivation

In eutherian (placental) mammals such as mice, X-inactivation occurs in two developmentally distinct waves. During the pre-implantation cleavage stages (4- to 8-cell embryo), imprinted X-inactivation selectively silences the paternally inherited X chromosome (X^p) across all blastomeres. This paternal imprint is retained within the extraembryonic trophectoderm and primitive endoderm lineages that form the placenta and yolk sac (just as marsupials exhibit exclusive paternal X-inactivation across all somatic tissues). However, inside the epiblast cells of the inner cell mass that give rise to the embryo proper, Xist is transiently downregulated and the paternal X is reactivated, erasing the parent-of-origin mark before random X-inactivation silences either X^m or X^p with equal 50:50 probability at implantation. In human embryos, extraembryonic XCI is predominantly random rather than strictly imprinted. A second major X-chromosome reactivation occurs exclusively in female primordial germ cells (PGCs) migrating to the genital ridge, ensuring that every mature oocyte transmits an active, unmethylated X chromosome to the next generation.

Somatic Clonal Mosaicism in Heterozygous Females

Because random X-inactivation occurs when the epiblast consists of roughly 20 to 100 progenitor cells and is clonally inherited thereafter, heterozygous females are patchwork mosaics of two functionally distinct cell populations:

• Calico and Tortoiseshell Cats (Felis catus): The Orange (O) coat-color locus is X-linked, with the X^O allele specifying phaeomelanin (orange fur) and the X^o allele permitting eumelanin (black fur). As melanocyte neural crest precursors migrate over the embryonic dermis, clones that inactivated X^o produce orange patches while clones that inactivated X^O produce black patches (with white spotting determined by the autosomal KIT piebald gene). Consequently, calico cats are almost always 46,XX females or sterile 47,XXY Klinefelter males.
• Human X-Linked Heterozygote Phenotypes: In X-linked anhidrotic ectodermal dysplasia (caused by mutations in the EDA gene), heterozygous carrier women display mosaic patches of skin lacking sweat glands alongside normal perspiring skin, corresponding to Blaschko's developmental lines. Similarly, females heterozygous for X-linked glucose-6-phosphate dehydrogenase (G6PD) deficiency or hemophilia A circulate two distinct populations of erythrocytes or hepatocytes, usually buffering them against severe disease unless skewed X-inactivation (by random bottleneck or selective cell survival) silences the wild-type allele in >80% of progenitor cells.

Pseudoautosomal Regions and Escape from X-Inactivation

Importantly, the Barr body is not completely silent. In humans, 15% to 23% of X-linked genes permanently escape X-inactivation and are expressed biallelically from both Xa and Xi (with an additional ~10% showing variable tissue-specific escape), compared to only ~3–7% escape in mice. Escape genes cluster primarily in two evolutionary strata:

1. The Pseudoautosomal Regions (PAR1 and PAR2): Spanning 2.7 Mb at the tip of Xp (PAR1, containing 24 genes including the stature gene SHOX) and 330 kb at the tip of Xq (PAR2), these regions retain full sequence homology with the Y chromosome and undergo obligatory meiotic crossover in males. Because XY males carry two functional copies of PAR genes (one on X, one on Y), XX females must express both X-linked copies to maintain equal dosage.
2. Non-PAR Escapees with Y-Homologs: Genes located on the recently evolved short arm (Xp) that still retain functional X-Y gametolog pairs (such as ZFX/ZFY, KDM6A/UTY, and USP9X/USP9Y), as well as XIST itself (which is expressed exclusively from the Xi).

The existence of these ~150+ escape genes explains why human sex-chromosome aneuploidies produce clinically recognizable syndromes despite the presence of the n − 1 Barr body mechanism: Turner syndrome (45,X) results from haploinsufficiency (1× dosage instead of 2×) of PAR1/escape genes like SHOX, causing short stature and ovarian dysgenesis, whereas Klinefelter syndrome (47,XXY) results from overexpression (3× dosage) of escape genes.

How many Barr bodies are present in individuals with sex chromosome aneuploidies?

Mammalian somatic cells follow the n − 1 rule, keeping exactly one X chromosome active per diploid set of autosomes and condensing all surplus X chromosomes into Barr bodies. Therefore, a Turner syndrome female (45,X) has 0 Barr bodies, a Klinefelter male (47,XXY) has 1 Barr body, a Triple X female (47,XXX) has 2 Barr bodies, and a 48,XXXX female or 48,XXXY male has 3 Barr bodies per interphase nucleus.

If extra X chromosomes are inactivated into Barr bodies, why do 45,X (Turner) and 47,XXY (Klinefelter) individuals exhibit phenotypic abnormalities?

Roughly 15% to 23% of human X-linked genes—including all genes in the pseudoautosomal regions (PAR1 and PAR2, such as the bone-growth transcription factor SHOX)—escape X-inactivation and are transcribed from the Barr body. A 45,X individual expresses only one copy of these escape genes (haploinsufficiency) instead of the normal two copies shared by 46,XX and 46,XY individuals, whereas a 47,XXY individual expresses three copies, altering skeletal, gonadal, and neurodevelopmental gene dosage.

Why can female carriers of X-linked recessive disorders occasionally show clinical symptoms?

Random X-inactivation occurs in a small pool of embryonic epiblast cells (~20–100 cells). By binomial sampling variance—or due to secondary post-inactivation cell selection—a heterozygous female can experience skewed X-inactivation in which 80% to 95% of cells in a target tissue (such as skeletal muscle in Duchenne muscular dystrophy or liver in hemophilia) happen to silence the X chromosome carrying the wild-type allele, unmasking the recessive mutation.

1. Lyon, M. F. (1961). "Gene action in the X-chromosome of the mouse (Mus musculus L.)." Nature, 190(4773), 372–373. https://doi.org/10.1038/190372a0
2. Heard, E., Clerc, P., & Avner, P. (1997). "X-chromosome inactivation in mammals." Annual Review of Genetics, 31, 571–610. https://doi.org/10.1146/annurev.genet.31.1.571
3. Galupa, R., & Heard, E. (2018). "X-chromosome inactivation: A crossroads between chromosome architecture and gene regulation." Annual Review of Genetics, 52, 535–566. https://doi.org/10.1146/annurev-genet-120116-024611
4. Carrel, L., & Willard, H. F. (2005). "X-inactivation profile reveals extensive variability in X-linked gene expression in females." Nature, 434(7031), 400–404. https://doi.org/10.1038/nature03479