19.01.03 · eco-evo-bio / mendelian

Sex-linked inheritance, dosage compensation, and X-inactivation

stub3 tiersLean: nonepending prereqs

Anchor (Master): Lyon, M. F. — Nature 190 (1961) 372-373; X-inactivation

Intuition Beginner

Some traits are carried on the X chromosome. Because males have only one X (XY) and females have two (XX), X-linked recessive traits like color blindness and hemophilia appear more often in males. A single copy is enough to show the trait in males, while females need two copies. Females can be carriers. In mammals, one X chromosome in each female cell is randomly turned off (X-inactivation) to balance gene dosage between the sexes.

Visual Beginner

Worked example Beginner

Red-green color blindness is an X-linked recessive trait. A woman with normal vision but whose father was color-blind is a carrier ( $X^{+} X^{c}$ ). She has children with a man who has normal vision ( $X^{+} Y$ ).

The Punnett square:

	$X^{+}$	$Y$
$X^{+}$	$X^{+} X^{+}$ (normal daughter)	$X^{+} Y$ (normal son)
$X^{c}$	$X^{+} X^{c}$ (carrier daughter)	$X^{c} Y$ (color-blind son)

Each son has a 50% chance of being color-blind. Each daughter has normal vision, but 50% are carriers. No daughter is affected because she inherits a normal $X^{+}$ from her father. This pattern — affected males transmitting the trait through carrier daughters to affected grandsons — is called criss-cross inheritance.

Check your understanding Beginner

Formal definition Intermediate+

X-linked recessive inheritance

A trait is X-linked recessive if the causative allele is located on the X chromosome and is recessive to the wild-type allele. Males (XY) are hemizygous for X-linked loci: a single copy of the allele determines the phenotype. Females (XX) must be homozygous for the recessive allele to express the trait.

Key features of X-linked recessive inheritance:

Affected females are rare; they must inherit the recessive allele from both parents.
Affected males transmit the allele to all daughters (who become carriers) and to no sons.
Carrier females transmit the allele to 50% of sons (who are affected) and 50% of daughters (who become carriers).
There is no male-to-male transmission.

Common X-linked recessive disorders: red-green color blindness (protanopia/deuteranopia), hemophilia A (factor VIII deficiency), hemophilia B (factor IX deficiency), Duchenne muscular dystrophy (dystrophin), Becker muscular dystrophy, G6PD deficiency, Lesch-Nyhan syndrome.

X-linked dominant inheritance

A trait is X-linked dominant if the causative allele on the X chromosome is dominant. Both males and females can be affected, but the inheritance pattern differs from autosomal dominant traits:

Affected males transmit the trait to all daughters and no sons.
Affected females (heterozygous) transmit the trait to 50% of sons and 50% of daughters.
Affected females are approximately twice as common as affected males in the population (because females have two chances to inherit the dominant allele).
In males, X-linked dominant disorders are often more severe and may be lethal (because there is no second X to moderate the effect).

X-linked dominant disorders: vitamin D-resistant rickets (X-linked hypophosphatemia, PHEX gene), Rett syndrome (MECP2, almost exclusively in females because it is lethal in hemizygous males), incontinentia pigmenti (IKBKAP, lethal in males), fragile X syndrome (expansion in FMR1, technically X-linked dominant with incomplete penetrance).

Y-linked (holandric) inheritance

Y-linked traits are carried on the Y chromosome and are transmitted exclusively from father to son. All Y-linked genes show a patrilineal inheritance pattern. The Y chromosome carries relatively few genes (approximately 70 protein-coding genes), and most are involved in male sex determination and spermatogenesis.

The SRY (sex-determining region Y) gene triggers testis development. Other Y-linked genes include $D A Z$ (deleted in azoospermia, involved in spermatogenesis) and $U S P 9 Y$ . Hairy ear rims (hypertrichosis pinnae auris) were historically cited as a Y-linked trait, though the evidence is inconsistent.

Because the Y chromosome does not recombine along most of its length (the non-recombining region, NRY, or male-specific region, MSY), Y-linked variation accumulates only by mutation, making the Y chromosome a useful tool for tracing patrilineal ancestry.

Inheritance pattern recognition: criss-cross inheritance

Criss-cross inheritance is the hallmark of X-linked recessive traits: an affected male transmits the trait through his daughters (who are phenotypically normal carriers) to affected grandsons. The trait appears to "cross" the sex line in alternate generations. This pattern distinguishes X-linked recessive inheritance from autosomal recessive inheritance, where both sexes are equally affected and carrier status is not sex-specific.

Punnett squares for X-linked crosses

For an X-linked cross, the genotypes must include the sex chromosomes:

Carrier female $\times$ normal male: $X^{+} X^{c} \times X^{+} Y$

	$X^{+}$	$Y$
$X^{+}$	$X^{+} X^{+}$	$X^{+} Y$
$X^{c}$	$X^{+} X^{c}$	$X^{c} Y$

Offspring: 1/4 normal female, 1/4 carrier female, 1/4 normal male, 1/4 affected male.

Affected male $\times$ homozygous normal female: $X^{c} Y \times X^{+} X^{+}$

	$X^{+}$	$Y$
$X^{+}$	$X^{+} X^{c}$	$X^{+} Y$

All daughters are carriers ( $X^{+} X^{c}$ ); all sons are normal ( $X^{+} Y$ ).

X-inactivation mechanism

In placental mammals, one of the two X chromosomes in each female somatic cell is transcriptionally silenced early in development (around the 8-cell stage in mice, the blastocyst stage in humans). This process, called X-inactivation (or lyonization after Mary Lyon), equalises the dosage of X-linked gene products between males (one active X) and females (one active X, one inactive X).

The steps of X-inactivation:

Counting: The cell senses the X ratio. The X-inactivation center (Xic, located at Xq13) on each X chromosome produces the long non-coding RNA XIST (X-inactive specific transcript).
Choice: In random X-inactivation, one X chromosome is chosen to remain active (Xa) and the other is chosen for inactivation (Xi). In imprinted X-inactivation (found in marsupials and in the extraembryonic tissues of mice), the paternal X is preferentially inactivated.
XIST coating: The XIST RNA coats the future inactive X in cis, spreading along the chromosome from the Xic.
Chromatin remodeling: The coated X undergoes heterochromatin formation: histone H3 trimethylation at lysine 27 (H3K27me3) by Polycomb repressive complex 2 (PRC2), histone H2A ubiquitination, DNA methylation at CpG islands of X-linked promoters, and incorporation of the histone variant macroH2A.
Condensation: The inactive X condenses into a densely packed Barr body, visible at the nuclear periphery in interphase cells.

Escape from X-inactivation

Not all genes on the inactive X are fully silenced. Approximately 15% of X-linked genes in humans escape X-inactivation to some degree, and another 10% show variable escape across individuals or tissues. Escape genes are more common in the pseudoautosomal regions (PAR1 at Xp22.33 and PAR2 at Xq28), which have homologous sequences on the Y chromosome and undergo pairing and recombination during male meiosis.

Escape from inactivation has clinical significance: in Klinefelter syndrome (47,XXY), the extra X carries escape genes whose overexpression contributes to the phenotype. In Turner syndrome (45,X), haploinsufficiency of escape genes (e.g., $S H O X$ in the PAR1 region) contributes to short stature and other features.

Dosage compensation across species

Different organisms solve the X-dosage problem in different ways:

Mammals (X-inactivation): One X chromosome is transcriptionally silenced in each female cell. The active X produces the same amount of transcript as the single X in males.

Drosophila (hypertranscription): The single X chromosome in males (XY) is transcribed at approximately twice the rate of each X in females (XX), mediated by the MSL (male-specific lethal) complex. The MSL complex binds along the male X chromosome and acetylates histone H4 at lysine 16 (H4K16ac), decompacting chromatin and increasing transcription. Females lack the MSL complex because the $m s l - 2$ gene is repressed by the RNA-binding protein Sxl (Sex-lethal).

C. elegans (downregulation): Both X chromosomes in hermaphrodites (XX) are transcribed at approximately half the rate of the single X in males (XO), achieving a 1:1 dosage ratio. The condensin-like complex DC (dosage compensation complex) binds both X chromosomes and reduces transcription, partially by altering chromosome architecture.

Ohno's hypothesis: Susumu Ohno (1967) proposed that the ancestral X chromosome carried genes whose products were optimised for expression from two copies. After the Y chromosome degenerated, losing most of its genes, a dosage compensation mechanism evolved to restore the 2:1 expression ratio. The Ohno hypothesis predicts that the expression level of X-linked genes in females (after compensation) should equal the expression level of each autosomal gene in both sexes (i.e., X-linked expression should be comparable to autosomal expression on a per-gene basis). Empirical data from RNA-seq studies broadly support this, though the match is imperfect.

Key result Intermediate+

Result (Hemizygous exposure of X-linked recessive alleles). For an X-linked recessive allele at frequency $q$ in a population at Hardy-Weinberg equilibrium, the proportion of affected males is $q$ and the proportion of affected females is $q^{2}$ . The ratio of affected males to affected females is $1/ q$ for $q ≪ 1$ , so rare X-linked recessive disorders appear predominantly in males.

Proof. Males are hemizygous ( $X^{A} Y$ or $X^{a} Y$ ), so the probability a male is affected equals the allele frequency: $P (affected male) = q$ . Females must be homozygous: $P (affected female) = q^{2}$ . The ratio is $q / q^{2} = 1/ q$ . For hemophilia A, where $q \approx 1/5000$ , the male ratio is approximately 5000:1. $□$

Result (The $n - 1$ rule for X-inactivation). In mammals, the number of active X chromosomes per cell is always one, regardless of the total number of X chromosomes. Cells with $n$ X chromosomes inactivate $n - 1$ of them. In 45,X (Turner syndrome), no X is inactivated and the single X remains active. In 47,XXY (Klinefelter syndrome), one X is inactivated. In 48,XXXX, three X chromosomes are inactivated.

Evidence. Barr body counts in interphase nuclei confirm the $n - 1$ rule: normal male cells show 0 Barr bodies, normal female cells show 1, 47,XXY cells show 1, 47,XXX cells show 2, and 48,XXXX cells show 3. The X-inactivation center counts the X ratio and inactivates all X chromosomes beyond the first. $□$

Result (Morgan's rules for sex-linked inheritance). T.H. Morgan (1910) established the following rules from the white-eye mutation in Drosophila: (1) an X-linked mutation in a male is transmitted to all daughters and no sons; (2) heterozygous carrier females transmit the mutation to half their sons and half their daughters; (3) the pattern of inheritance correlates with the sex chromosomes. These rules distinguished sex-linked inheritance from autosomal inheritance and provided the first experimental evidence that genes reside on chromosomes.

Exercises Intermediate+

Exercise 3 (medium, short answer).

A woman with normal vision has a color-blind father and a color-blind maternal grandfather. Her husband has normal vision. What is the probability their first son is color-blind?

Hint

Determine the mother's genotype from her pedigree. Her father is color-blind ( $X^{c} Y$ ), so she inherited $X^{c}$ from him. Her maternal grandfather is color-blind, so her mother could be $X^{+} X^{c}$ or $X^{c} X^{c}$ .

Answer

The woman inherited $X^{c}$ from her color-blind father, so she is either $X^{+} X^{c}$ or $X^{c} X^{c}$ .

Her mother's father was color-blind, so her mother inherited $X^{c}$ from him. Her mother's genotype is either $X^{+} X^{c}$ or $X^{c} X^{c}$ depending on the maternal grandmother's allele.

If the mother is $X^{+} X^{c}$ : the woman has probability 1/2 of being $X^{+} X^{c}$ and 1/2 of being $X^{c} X^{c}$ .

If the woman is $X^{+} X^{c}$ : probability of color-blind son = 1/2 (inherits $X^{c}$ ). If the woman is $X^{c} X^{c}$ : probability of color-blind son = 1.

If the mother is $X^{c} X^{c}$ : the woman is $X^{c} X^{c}$ and the probability of a color-blind son = 1.

Without further information about the maternal grandmother, the woman's carrier status depends on the mother's second allele. If the maternal grandmother is assumed normal ( $X^{+} X^{+}$ ), then the mother is $X^{+} X^{c}$ . The woman is then $X^{+} X^{c}$ with probability 1/2 and $X^{c} X^{c}$ with probability 1/2. The probability of a color-blind son is $(1/2) (1/2) + (1/2) (1) = 3/4$ .

Exercise 4 (medium, short answer).

Distinguish between X-linked dominant and autosomal dominant inheritance based on pedigree analysis. What key feature would you look for?

Hint

Consider what happens when an affected male has children. In autosomal dominant inheritance, affected males can transmit the trait to sons. In X-linked dominant inheritance, affected males transmit to all daughters and no sons.

Answer

The critical feature is male-to-male transmission. In autosomal dominant inheritance, affected males transmit the trait to approximately 50% of sons and 50% of daughters. In X-linked dominant inheritance, affected males transmit the trait to all daughters and no sons (because they pass their Y chromosome to sons, not their X). A single instance of an affected father with an affected son rules out X-linked dominant inheritance.

Additionally, in X-linked dominant inheritance, affected females are roughly twice as common as affected males (because females have two X chromosomes), and males are often more severely affected (no second X to compensate).

Exercise 5 (medium, short answer).

In a female somatic cell, the gene for G6PD (glucose-6-phosphate dehydrogenase, X-linked) is heterozygous: one allele produces a functional enzyme (G6PD+) and the other produces a non-functional enzyme (G6PD-). After X-inactivation, what is the expected ratio of cells expressing G6PD+ versus G6PD- in a tissue?

Hint

X-inactivation is random in each cell. Once an X is inactivated in a cell, all daughter cells maintain the same inactive X.

Answer

Approximately 50:50. In each cell, one X is randomly inactivated. Cells that inactivate the X carrying G6PD+ express only G6PD-, and cells that inactivate the X carrying G6PD- express only G6PD+. The result is a mosaic of two cell populations. This mosaicism was one of the key pieces of evidence for Mary Lyon's X-inactivation hypothesis: heterozygous females show patchy expression of X-linked markers, consistent with random inactivation early in development followed by clonal expansion.

Exercise 6 (hard, short answer).

A woman is heterozygous for an X-linked recessive disorder. Due to skewed X-inactivation, 90% of her cells inactivate the X carrying the normal allele. What phenotype would you expect, and why?

Hint

Skewed X-inactivation means the random choice of which X to inactivate is biased. If 90% of cells inactivate the normal X, then 90% of cells express only the mutant allele.

Answer

The woman would show symptoms of the disorder (or at least attenuated symptoms), despite being heterozygous. With 90% of cells expressing only the mutant allele, the effective gene dosage of the normal product is reduced to approximately 10% of normal. The severity depends on the gene and the threshold of functional protein needed. For some X-linked disorders (e.g., Duchenne muscular dystrophy), even mild skewing can produce symptoms in carrier females. Skewed X-inactivation is one reason some female carriers of X-linked recessive disorders show variable or unexpected phenotypes.

Advanced topics in sex-linked inheritance Master

The X-inactivation center and XIST regulation

The X-inactivation center (Xic) is a region on the X chromosome (Xq13 in humans) that orchestrates X-inactivation. The key gene is XIST (X-inactive specific transcript), which produces a 17 kb long non-coding RNA that coats the future inactive X in cis. XIST is the master regulator: deletion of XIST from an X chromosome prevents that chromosome from being inactivated, while ectopic insertion of XIST into an autosome can trigger silencing of that autosome.

XIST regulation involves several anti-sense transcripts and regulatory elements:

TSIX (XIST antisense) is transcribed from the opposite strand of XIST and represses XIST expression on the future active X. TSIX and XIST are mutually exclusive: the X that continues expressing TSIX remains active, while the X that silences TSIX upregulates XIST and becomes inactive.
JPX is an activator of XIST that acts as a molecular switch: JPX RNA accumulates and promotes XIST upregulation on the future inactive X.
FTX (five prime to Xist) is another lncRNA that positively regulates XIST.
RLIM (Rnf12) encodes an E3 ubiquitin ligase that promotes XIST expression. RLIM is an X-linked gene that is subject to auto-regulatory degradation, creating a feedback loop that limits X-inactivation to a single X.

The counting mechanism senses the ratio of X chromosomes to autosome sets. In diploid cells, one X is kept active per two autosome sets. In triploid cells (69,XXY), the counting mechanism still results in a single active X.

The $n - 1$ rule and X chromosome number

The $n - 1$ rule states that in a cell with $n$ X chromosomes, exactly $n - 1$ are inactivated. This has been verified across sex chromosome aneuploidies:

Karyotype	Active X	Inactive X (Barr bodies)	Phenotype
46,XY	1	0	Normal male
46,XX	1	1	Normal female
45,X	1	0	Turner syndrome
47,XXY	1	1	Klinefelter syndrome
47,XXX	1	2	Triple X (often mild or asymptomatic)
48,XXXX	1	3	Tetrasomy X
49,XXXXY	1	4	Pentasomy X (severe)

The increasing severity with additional X chromosomes suggests that escape genes on the inactive X(s) are overexpressed, and the process of inactivating multiple X chromosomes itself disrupts nuclear architecture.

Reactivation in germ cells

X-inactivation is reversed in the female germline. During oogenesis, the inactive X is reactivated so that the oocyte carries two active X chromosomes. Reactivation occurs at the onset of meiosis and involves the loss of XIST RNA coating, removal of repressive histone marks, and DNA demethylation. This ensures that the oocyte can properly regulate X-linked gene expression and that the resulting gamete carries an X chromosome in a transcriptionally competent state.

In the male germline, the single X chromosome is transiently silenced during spermatogenesis (meiotic sex chromosome inactivation, MSCI), along with the Y chromosome, because the X and Y lack a homologous pairing partner for most of their length. MSCI prevents aberrant recombination and maintains genomic stability.

Skewed X-inactivation

Random X-inactivation produces an approximately 50:50 ratio of cells expressing the maternal versus paternal X. Skewed X-inactivation occurs when the ratio deviates significantly from 50:50 (commonly defined as >75:25 or >90:10). Causes include:

Stochastic skewing: The smaller the number of cells at the time of inactivation (the "bottleneck"), the greater the variance in the final ratio. If inactivation occurs when only 10-20 cells are present, random drift alone can produce substantial skewing.
Positive selection: If one X carries alleles that confer a proliferative advantage, cells with that X active will outgrow cells with the other X active. Mutations in X-linked genes involved in cell growth can drive selection.
Negative selection: If one X carries a deleterious allele, cells expressing that allele may be selectively eliminated, skewing the ratio toward the normal X. This is the basis of "non-random" X-inactivation observed in some carrier females who show no symptoms despite carrying a severe X-linked mutation.
Primary skewing: A mutation in the X-inactivation center itself (e.g., in XIST or TSIX) can bias the choice of which X to inactivate.

Skewed X-inactivation has clinical implications: a female carrier of an X-linked recessive disorder with skewed inactivation favoring the normal X may be asymptomatic, while skewing toward the mutant X can produce symptoms.

X-autosome translocations

When a segment of the X chromosome is translocated onto an autosome (or vice versa), X-inactivation can spread from the inactive X into the autosomal segment, silencing autosomal genes on the translocated chromosome. Conversely, the translocated autosomal segment may be insulated from inactivation.

X-autosome translocations are a diagnostic consideration in females with X-linked disorders: if the translocation breakpoint disrupts an X-linked gene, and the normal X is preferentially inactivated (to maintain dosage balance for the autosomal genes), the female will express only the disrupted allele. This "balanced X-inactivation" pattern (where the normal X is consistently inactivated) is a hallmark of X-autosome translocations involving a disease gene and is used diagnostically in Duchenne muscular dystrophy.

Dosage-sensitive genes and the X chromosome

The X chromosome is enriched for genes involved in brain function and cognitive development. Over 100 X-linked genes are associated with intellectual disability (X-linked mental retardation, XLMR), including:

FMR1 (fragile X syndrome): CGG trinucleotide expansion in the 5' UTR leads to methylation and silencing. Most common inherited cause of intellectual disability.
MECP2 (Rett syndrome): Methyl-CpG-binding protein involved in transcriptional repression. Duplications of MECP2 cause X-linked intellectual disability in males.
ARX (X-linked lissencephaly, infantile spasms): Aristaless-related homeobox gene.
OPHN1 (X-linked intellectual deficit with cerebellar hypoplasia): Oligophrenin-1, a Rho-GAP protein.
L1CAM (X-linked hydrocephalus, MASA syndrome): Neural cell adhesion molecule.

The concentration of brain-expressed genes on the X chromosome may reflect evolutionary selection: genes that are advantageous in heterozygous state (due to dosage sensitivity) may be preferentially retained on the X because hemizygous exposure in males allows rapid selection. An alternative hypothesis is that the X chromosome evolved as a "safe haven" for genes that benefit from hemizygous expression during brain development.

Escape from inactivation and cancer

Genes that escape X-inactivation are overexpressed in females relative to males, creating a baseline sex difference in gene expression that can influence cancer susceptibility. The KDM6A (UTX) gene on the X chromosome escapes inactivation and functions as a tumor suppressor. Because females have two active copies (one on each X, due to escape) while males have one (on the X; the Y-linked paralogue UTY has reduced activity), females have a dosage advantage for KDM6A tumor suppression.

Conversely, reactivation of the inactive X is observed in some cancers, particularly breast and ovarian cancer. Loss of XIST expression and reactivation of the inactive X can lead to overexpression of X-linked oncogenes. In some contexts, the inactive X is partially reactivated as part of the epigenomic dysregulation characteristic of cancer cells.

Fragility of the X chromosome

The X chromosome has several fragile sites, most notably FRAXA (fragile X syndrome) at Xq27.3, caused by a CGG trinucleotide expansion in the 5' UTR of the FMR1 gene. Normal alleles have 5-44 repeats; premutation alleles have 55-200 repeats; full mutation alleles have >200 repeats, which trigger methylation and silencing of FMR1. Fragile X syndrome is the most common inherited cause of intellectual disability and the most common single-gene cause of autism spectrum disorder.

The FRAXE fragile site (AFRAX, Xq28) involves a similar CCG expansion in the FMR2 gene. Other fragile sites on the X chromosome are less well characterised.

The Ohno hypothesis and X chromosome evolution

Susumu Ohno (1967) proposed that the X chromosome originated as a pair of autosomes in the common ancestor of mammals. One member of the pair acquired a sex-determining gene (SRY) and became the Y chromosome. Over evolutionary time, the Y chromosome lost most of its genes through deletions and Muller's ratchet (the irreversible accumulation of deleterious mutations in non-recombining regions), retaining only a small subset of genes.

As the Y degenerated, the X chromosome (which still recombines in females) retained its original gene complement. Dosage compensation evolved to restore the 2:1 expression ratio between the X and autosomes. The Ohno hypothesis predicts that X-linked gene expression in females (after one X is inactivated) should match autosomal gene expression levels, and this is largely confirmed by transcriptomic data.

Comparative genomics shows that the X chromosome is conserved across eutherian mammals (the same genes are X-linked in humans, mice, dogs, and cows), indicating that the X chromosome was established before the divergence of placental mammals approximately 100 million years ago. Marsupials share part of the X chromosome with eutherians (the "conserved" region) but have an additional region that is autosomal in eutherians and X-linked in marsupials (the "added" region), reflecting an X-autosome translocation after the divergence of metatherians and eutherians.

Result (X-linked allele frequency dynamics). For an X-linked recessive allele at frequency $q_{m}$ in males and $q_{f}$ in females, the allele frequency in males at generation $t + 1$ equals the female frequency at generation $t$ (because males inherit their X from their mother): $q_{m}^{(t + 1)} = q_{f}^{(t)}$ . The female frequency is the average of the male and female frequencies in the previous generation: $q_{f}^{(t + 1)} = (q_{m}^{(t)} + q_{f}^{(t)}) /2$ . This generates oscillatory convergence to equilibrium with a period of approximately 3 generations (two steps forward, one step back), slower than the convergence of autosomal allele frequencies.

Connections Master

Mendelian genetics 19.01.01. Sex-linked inheritance extends Mendelian principles to genes located on the sex chromosomes. The key departure from autosomal Mendelian ratios is the asymmetry between sexes: males are hemizygous for X-linked genes, producing different phenotypic ratios in male and female offspring.
Linkage and genetic maps 19.01.02 pending. The first genetic map, constructed by Sturtevant (1913), mapped six X-linked genes in Drosophila. Sex-linked genes are mapping-rich because the hemizygous state in males makes all alleles phenotypically visible, simplifying recombination analysis. The sex chromosomes also introduced the concept of linkage to a specific chromosome.
Population genetics 19.02.01. X-linked alleles have different allele frequency dynamics than autosomal alleles because males inherit their X only from their mother. The effective population size for X-linked loci is 3/4 that of autosomal loci (assuming equal numbers of breeding males and females), which affects the rate of genetic drift and the efficacy of selection at X-linked loci.
Molecular genetics 17.01.01. X-inactivation is one of the best-studied examples of epigenetic gene regulation. The mechanisms involved (long non-coding RNA-mediated silencing, histone modification, DNA methylation, chromatin remodeling) are central to the broader field of epigenetics.
Evolution of sex chromosomes 19.03.01. The evolution of dosage compensation mechanisms is tightly linked to the degeneration of the Y chromosome. Understanding why different lineages evolved different compensation strategies (inactivation vs. hypertranscription vs. downregulation) provides insight into the constraints and contingencies of chromosome evolution.
Human genetics and medicine 19.06.01. X-linked disorders are a major category of Mendelian disease. Carrier testing, genetic counselling for X-linked disorders, and prenatal diagnosis all rely on understanding sex-linked inheritance patterns. X-inactivation mosaicism also complicates gene therapy approaches for X-linked diseases.

Historical notes Master

Thomas Hunt Morgan discovered sex linkage in 1910 when he observed a single white-eyed male in a culture of red-eyed Drosophila melanogaster. When this male was crossed to red-eyed females, all F1 offspring had red eyes, but the F2 generation showed a peculiar pattern: all females were red-eyed, while half the males were red-eyed and half were white-eyed. Morgan recognised that the white-eye gene ( $w$ ) was located on the X chromosome, and that the pattern of inheritance reflected the transmission of X-linked genes ^{[Morgan 1910]}.

This discovery was the first experimental evidence that a specific gene resides on a specific chromosome, providing crucial support for the chromosome theory of inheritance (proposed independently by Sutton and Boveri in 1902-1903). Morgan's student Alfred Sturtevant used X-linked genes to construct the first genetic map in 1913 ^{[Sturtevant 1913]}, and Calvin Bridges demonstrated non-disjunction of the X chromosome in 1916, providing cytological proof of the chromosome theory.

Mary Lyon proposed the X-inactivation hypothesis in 1961 based on studies of coat colour mosaicism in mice ^{[Lyon 1961]}. She observed that female mice heterozygous for X-linked coat colour genes showed a mottled (dappled) coat pattern, while males were uniformly coloured. Lyon proposed that one X chromosome is inactivated in each female somatic cell early in development, and that the inactivation is random with respect to parental origin. The inactivated X is maintained through all subsequent cell divisions, producing clonal patches of tissue expressing one or the other X chromosome.

Lyon's hypothesis explained several previously puzzling observations: the Barr body (a condensed chromatin body in female interphase nuclei, first described by Murray Barr and E.G. Bertram in 1949), the dosage compensation of X-linked enzymes (showing similar activity in males and females despite different X chromosome numbers), and the variable expressivity of X-linked disorders in heterozygous females.

The molecular mechanism of X-inactivation was elucidated over the subsequent decades. The XIST gene was identified in 1991 by Brown et al. as a gene expressed exclusively from the inactive X chromosome. XIST was the first long non-coding RNA shown to have a regulatory function in chromosome-wide silencing. The discovery that XIST RNA coats the inactive X in cis, recruiting chromatin-modifying complexes to establish and maintain the silenced state, opened the field of lncRNA biology.

Susumu Ohno's 1967 book "Sex Chromosomes and Sex-Linked Genes" established the theoretical framework for understanding X chromosome evolution and the origin of dosage compensation. Ohno argued that the X chromosome originated from an autosome pair and that dosage compensation evolved to compensate for Y chromosome degeneration. His prediction that X-linked gene expression should equal autosomal gene expression (after compensation) has been largely confirmed by modern transcriptomic studies, though with notable exceptions for escape genes and tissue-specific effects.

The discovery that different organisms use fundamentally different dosage compensation mechanisms (X-inactivation in mammals, hypertranscription in Drosophila, downregulation in C. elegans) demonstrated that the same evolutionary problem (balancing X-linked dosage between the sexes) can be solved by different molecular strategies. This convergence on function from divergent mechanisms is a textbook example of convergent evolution at the molecular level.

Bibliography Master

@article{Morgan1910,
  author = {Morgan, Thomas Hunt},
  title = {Sex-linked inheritance in {Drosophila}},
  journal = {Science},
  volume = {32},
  pages = {120--122},
  year = {1910},
}

@article{Lyon1961,
  author = {Lyon, Mary F.},
  title = {Gene action in the {X}-chromosome of the mouse ({Mus musculus L.})},
  journal = {Nature},
  volume = {190},
  pages = {372--373},
  year = {1961},
}

@article{Brown1991,
  author = {Brown, Carolyn J. and Ballabio, Andrea and Rupert, J. L. and Lafreniere, R. G. and Grompe, M. and Tonlorenzi, R. and Willard, Huntington F.},
  title = {A gene from the region of the human {X} inactivation centre is expressed exclusively from the inactive {X} chromosome},
  journal = {Nature},
  volume = {349},
  pages = {38--44},
  year = {1991},
}

@article{Sturtevant1913,
  author = {Sturtevant, Alfred H.},
  title = {The linear arrangement of six sex-linked factors in {Drosophila}, as shown by their mode of association},
  journal = {J. Exp. Zool.},
  volume = {14},
  pages = {43--59},
  year = {1913},
}

@book{Ohno1967,
  author = {Ohno, Susumu},
  title = {Sex Chromosomes and Sex-Linked Genes},
  publisher = {Springer-Verlag},
  year = {1967},
}

@article{BarrBertram1949,
  author = {Barr, Murray L. and Bertram, E. G.},
  title = {A morphological distinction between neurones of the male and female, and the behaviour of the nucleolar satellite during accelerated nucleoprotein synthesis},
  journal = {Nature},
  volume = {163},
  pages = {676--677},
  year = {1949},
}

@book{FutuymaKirkpatrick2017,
  author = {Futuyma, Douglas J. and Kirkpatrick, Mark},
  title = {Evolution},
  edition = {4th},
  publisher = {Sinauer Associates},
  year = {2017},
}

@book{HartlClark2007,
  author = {Hartl, Daniel L. and Clark, Andrew G.},
  title = {Principles of Population Genetics},
  edition = {4th},
  publisher = {Sinauer Associates},
  year = {2007},
}

@article{CarrelWillard2005,
  author = {Carrel, Laura and Willard, Huntington F.},
  title = {X-inactivation profile reveals extensive variability in {X}-linked gene expression in females},
  journal = {Nature},
  volume = {434},
  pages = {400--404},
  year = {2005},
}

@article{Lucchesi1978,
  author = {Lucchesi, John C.},
  title = {Gene dosage compensation and the evolution of sex chromosomes},
  journal = {Science},
  volume = {202},
  pages = {711--716},
  year = {1978},
}

Intuition Beginner

Visual Beginner

Worked example Beginner

Check your understanding Beginner

Formal definition Intermediate+

X-linked recessive inheritance

X-linked dominant inheritance

Y-linked (holandric) inheritance

Inheritance pattern recognition: criss-cross inheritance

Punnett squares for X-linked crosses

X-inactivation mechanism

Escape from X-inactivation

Dosage compensation across species

Key result Intermediate+

Exercises Intermediate+

Advanced topics in sex-linked inheritance Master

The X-inactivation center and XIST regulation

The n−1 rule and X chromosome number

Reactivation in germ cells

Skewed X-inactivation

X-autosome translocations

Dosage-sensitive genes and the X chromosome

Escape from inactivation and cancer

Fragility of the X chromosome

The Ohno hypothesis and X chromosome evolution

Connections Master

Historical notes Master

Bibliography Master

The $n - 1$ rule and X chromosome number