Genomic imprinting: A revolution in conceptualisation of sex differences.

Read Also

 Sexual imprinting of genes or portions of chromosomes is part of a more general phenomenon called "genomic imprinting."  Broadly speaking, it refers to reversible, differential expression of a gene or set of genes in an offspring individual according to whether the genes came from the mother or from the father.  It's a very widespread and normal process, with a large literature. Apparently the phenomenon was first discovered in insects. Genetic imprinting is defined as a reversible, differential marking of genes or chromosomes that is determined by the sex of the parent from whom the genetic material is inherited. Imprinting was first observed in insects where, in some species, most notably among the coccoids (scale insects and allies), the differential marking of paternally and maternally transmitted chromosome sets leads to inactivation or elimination of paternal chromosomes. Imprinting is also widespread in plants and mammals, in which paternally and maternally inherited alleles may be differentially expressed. Despite imprinting having been discovered in insects, clear examples of parental imprinting are scarce in the model insect species Drosophila melanogaster. There is however a case of imprint-mediated control of gene expression in Drosophila. The imprinted gene - the white+ eye-color gene - is expressed at a low level when transmitted by males, and at a high level when transmitted by females. Thus, in common with coccoids, Drosophila is capable of generating an imprint, and can respond to that imprint by silencing the paternal allele. Golic (1998) has recently published a fascinating review of this litterature.

One of the basic mechanisms of mammalian genomic imprinting seems to be differences in the methylation of cytosine residues in the DNA in developing sperm and egg.  Note that this occurs in the parents.  The result is that although the sperm and egg carry the same genes -- that is, the same base sequences in their DNA -- these sequences have been handled differently during the development of the egg and of the sperm.  The result is stated blandly by Scott Gilbert: "It appears that during germ cell formation, previous information is erased, and then, during meiosis, new information is introduced into the genome.  The pattern of methylation on a given gene can differ between egg and sperm, and these gene-specific methylation differences are seen in the chromosomes of the embryonic cells.  Thus, methylation differences between sperm and egg genes may specify whether a gene came from the father or mother. This maternal and paternal imprinting adds additional information to the inherited genomes, information that may regulate spatial and temporal gene activity and chromosome behavior. Gilbert concludes even more blandly that gamete-specific methylation and the resulting genomic imprinting  "... also provides a reminder that the organism cannot be explained solely by its genes.  One needs knowledge of developmental parameters as well as genetic ones." 

The Prader-Willi and Angelman syndromes are textbook examples of genomically imprinted disorders in humans.  In humans, the loss of a particular gene in the long arm of chromosome 15 results in different phenotypes depending on whether the loss is in the male-derived or the female-derived chromosome.  If the defective or missing gene comes from the father, the child is born with Prader-Willi syndrome, a disease associated with mild mental retardation, obesity, small gonads, and short stature.  If the defective or missing gene comes from the mother, the child has Angelman syndrome, characterized by severe mental retardation, lack of speech, and inappropriate laughter (see Gilbert, 1994). Prader-Willi syndrome (PWS) and Angelman syndrome (AS) are distinct neurodevelopmental disorders with interrelated genetic mechanisms because genomic imprinting within the chromosome 15q11-13 region affects both the PWS and the AS locus. One study carried out by Buchholz and colleagues (1998) presents clinical and molecular data on a large series of 258 referred patients, evaluated with methylation analysis: 115 with suspected PWS and 143 with suspected AS. In these patients, the clinical phenotype was graded into three groups: classical (group 1); not classical but possible (group 2); not classical and unlikely (group 3).  For PWS, a fourth group consisted of hypotonic babies.  DNA methylation analysis confirmed the diagnosis of PWS in 30 patients (26%) and AS in 28 patients (20%). For 21 PWS patients the mechanism was established: 15 had deletions, 4 had uniparental disomy (UPD) and 2 a presumed imprinting defect. Clinically all those with an abnormal methylation pattern had the classical phenotype and none of those with a normal methylation pattern had classical PWS. For 23 AS patients in whom a mechanism was established, 17 had a deletion, 3 had UPD and 3 had a presumed imprinting defect. There was greater clinical overlap in AS, with 26 classical AS patients having a normal methylation pattern while an abnormal methylation pattern was seen in one patient from group 2. In addition, there were a further 40 patients with a normal methylation pattern in whom AS was still a possible diagnosis. Our conclusion is that methylation analysis provides an excellent screening test for both syndromes, providing approximately 99% diagnosis for PWS and for AS, a 75% diagnostic rate, supplemented for the remaining 25% with an essential basic starting point to further investigations.

Though Prader-Willi and Angelman syndromes were among the first brain disorders to be identified as involving genomic imprinting,  human genetic studies have since directed attention to genomic imprinting in a number of syndromes involving brain dysfunction, such as congenital adrenal hyperplasia, Turner's syndrome, bipolar depression, dyslexia,  autism, schizophrenia and even homosexuality. Molecular genetics is providing insight into the complexity of these imprinting mechanisms, while experimental studies are revealing the differential roles that maternal and paternal genomes may play in brain development and growth.

Congenital adrenal hyperplasia (CAH) is an inherited recessive disorder of adrenal steroidogenesis caused by mutations in the steroid 21-hydroxylase gene (CYP21) in more than 90% of affected patients. The CYP21 gene is located within the HLA complex locus on chromosome 6 (6p21.3). During a molecular characterisation study of a group of 47 Mexican families with 21-hydroxylase deficiency, Lopez-Gutierrez and colleagues (1998) identified nine in which the mutation or mutations found in the patient did not appear to originate from one of the parents. Through DNA fingerprinting, paternity was established in all nine families with a probability of non-paternity in the range of 10(-19) to 10(-23). Among these families, Lopez-Gutierrez and colleagues identified one patient with exclusive paternal inheritance of all eight markers tested on chromosome 6p, despite normal maternal and paternal contributions for eight additional markers on three different chromosomes. Lopez-Gutierrez and colleagues did not identify duplication of paternal information for markers in the 6q region, consistent with lack of expression of transient neonatal diabetes owing to genomic imprinting in this patient. These results substantiate evidence for the existence of different genetic mechanisms involved in the expression of this recessive condition in a substantial portion (approximately 19%) of affected Mexican families. In addition to the identification of a patient with paternal uniparental disomy, the occurrence of germline mutations may explain the unusual pattern of segregation in the majority of the remaining eight families.

Turner's syndrome is a sporadic disorder of human females in which all or part of one X chromosome is deleted. Intelligence is usually normal but social adjustment problems are common. Skuse and colleagues (1997) recently published a  study of 80 females with Turner's syndrome and a single X chromosome.  In 55 cases the X was maternally derived (45,X[m]) and in 25 it was of paternal origin (45,X[p]). Members of the 45,X[p] group were significantly better adjusted, with superior verbal and higher-order executive function skills, which mediate social interactions. These observations suggest that there is a genetic locus for social cognition, which is imprinted and is not expressed from the maternally derived X chromosome. Neuropsychological and molecular investigations of eight females with partial deletions of the short arm of the X chromosome indicate that the putative imprinted locus escapes X-inactivation, and probably lies on Xq or close to the centromere on Xp. If expressed only from the X chromosome of paternal origin, the existence of this locus could explain why 46,XY males (whose single X chromosome is maternal) are more vulnerable to developmental disorders of language and social cognition, such as autism, than are 46,XX females.

Skuse and colleagues have even proposed that basic normal sex differences in cognitive ability,  women’s verbal advantage and men’s spatial advantage for example,  are also tributary to genomic imprinting.   This puts into question classical doctrine according to which the sole vector of gender-specific brain differentiation,  aside from experience,  is hormonal.    This represents a revolutionary way of thinking about biological bases of brain sex differences.