Patten et al. (2014) — The Evolution of Genomic Imprinting

Patten et al. (2014) — The Evolution of Genomic Imprinting: Theories, Predictions and Empirical Tests

Bibliographic Reference (APA 7.0)

Patten, M. M., Ross, L., Curley, J. P., Queller, D. C., Bonduriansky, R., & Wolf, J. B. (2014). The evolution of genomic imprinting: Theories, predictions and empirical tests. Heredity, 113(2), 119–128. https://doi.org/10.1038/hdy.2014.29

Core Argument

Three evolutionary theories for genomic imprinting have withstood theoretical and empirical scrutiny: Haig and colleagues’ kinship theory, Day and Bonduriansky’s sexual antagonism theory, and Wolf and Hager’s maternal-offspring coadaptation theory. These theories share a critical common feature — some process creates a selective asymmetry between the maternally and paternally inherited allelic copies at a locus, causing selection to favor differential expression (typically silencing of one copy). However, they differ fundamentally in what generates this asymmetry and in their predictions about which allele should be silenced, in which tissues, and in which taxa. The paper clarifies the logic and assumptions of all three theories, derives discriminating predictions, and suggests tests (including experimental evolution and comparative approaches) to distinguish between them.

Methods

This is a review and conceptual synthesis. The authors:

• Reconstruct the logical structure of each of the three major theories, making assumptions explicit.
• Derive predictions from each theory in five domains: (1) direction of expression/silencing, (2) tissue specificity, (3) taxon specificity, (4) dosage sensitivity, and (5) loss of imprinting.
• Tabulate predictions across taxa and contexts (Table 1), enabling cross-theory comparison.
• Propose empirical tests organized into four categories: requisite genetic variation, experimental evolution, molecular evolution, and comparative phylogenetic data.

Key Findings

• The three theories differ in their core mechanism. The kinship theory views imprinting as a mechanism to change gene dosage driven by kin selection — matrigenic and patrigenic alleles experience different patterns of relatedness in the social environment, so their expression has different inclusive fitness consequences. The sexual antagonism and maternal-offspring coadaptation theories both view imprinting as a mechanism to modify resemblance to parents, increasing the probability of expressing the fitter of the two alleles at a locus.

• Under the kinship theory, growth-promoting genes are predicted to be paternally expressed and growth-inhibiting genes maternally expressed, because the patrigenic allele typically has lower relatedness to siblings competing for maternal resources. Reciprocally imprinted genes with opposing phenotypic effects (e.g., Igf2 and Igf2r) are a strong prediction unique to this theory.

• Under the sexual antagonism theory, imprinted expression arises from sex-specific selection pressure. Patrigenes are enriched for alleles of male benefit and matrigenes for alleles of female benefit. The theory predicts that imprinted genes influencing the same phenotype should be expressed from the same parental origin within a sex, and sex-specific imprinting (different expression patterns in males and females) can evolve. This theory has the least restrictive requirements and is expected to apply most broadly across taxa.

• Under the maternal-offspring coadaptation theory, imprinting evolves because the two alleles in an offspring code for non-equivalent fitness interactions with the mother. When positive genetic covariance between maternal and offspring traits produces higher fitness (phenotypic matching), selection favors expression of the matrigenic allele and silencing of the patrigenic allele in the offspring. The theory requires fitness interactions between mothers and offspring and is strongest in species with prolonged maternal care.

• The three theories make different predictions about loss of imprinting. Under the kinship theory, loss of imprinting is unlikely because biallelic expression would raise total expression above either parental origin’s optimum, disrupting dosage balance. Under the sexual antagonism and coadaptation theories, loss of imprinting carries no comparable cost, because neither views imprinting as a mechanism to modify total expression level from a locus, and imprinting status is predicted to be more evolutionarily labile.

Concepts Introduced or Used

• Genomic imprinting: Epigenetic phenomenon where expression of an allele depends on its parental origin, producing functional haploidy at imprinted loci.
• Matrigenic/patrigenic: Terminology (from Queller, 2003) describing alleles of a haploid genotype that derive from a diploid mother or father, respectively — distinct from “maternal genes” and “paternal genes.”
• Kinship theory (strong vs. weak versions): Strong version claims the origin of imprinted expression is a consequence of parental antagonism; weak version claims that once imprinted expression is established, an imprinted gene evolves subject only to the inclusive fitness consequences for whichever parental copy remains expressed.
• Parental antagonism (sensu Haig, 1997): A form of selection in which matrigenic and patrigenic alleles have opposing inclusive fitness effects due to asymmetric relatedness to interacting kin. Haig drew the analogy with sexual antagonism.
• Sexual antagonism theory: Intralocus sexual conflict creates different allele frequencies in eggs and sperm, making the two alleles at a diploid locus carry different adaptive information. Imprinting allows individuals to express the fitter allele for their sex.
• Maternal-offspring coadaptation theory: Fitness interactions between maternal and offspring phenotypes generate correlational selection, building linkage disequilibrium between loci affecting the interaction. The matrigenic allele is more likely to be coadapted with maternal genotypes.
• “Chip Off the Old Block” theory (Spencer & Clark, 2006): A related theory where selective asymmetry arises because the haploid genome from the parent of the less dispersive sex is enriched for locally adapted alleles. The paper discusses this as analogous to sexual antagonism but not as one of the three main theories.
• Evolutionarily stable strategy (ESS) for gene expression from the two parental origins under the kinship theory.
• Dosage sensitivity: Relevant specifically to the kinship theory, which predicts that imprinted genes must be dosage-sensitive at the origin of imprinted expression.
• Intragenomic conflict: In the kinship theory, imprinted genes and genes underlying the imprinting machinery may have conflicting interests, leading to antagonistic coevolution and elevated rates of molecular divergence.

Entities Referenced

• Igf2, Igf2r: Classic reciprocally imprinted genes — Igf2 is paternally expressed (growth promoter), Igf2r is maternally expressed (growth inhibitor). These are the canonical examples satisfying the kinship theory’s strong predictions (Haig & Graham, 1991).
• KLF14: Imprinted transcription factor identified as undergoing human-specific accelerated evolution.
• MEDEA: Imprinted locus in Arabidopsis showing evidence of positive Darwinian selection.
• Haig, D.: Originator of the kinship theory (Haig & Westoby, 1989; Haig, 1997; Haig, 2000a, 2004).
• Day, T. & Bonduriansky, R.: Originators of the sexual antagonism theory (Day & Bonduriansky, 2004).
• Wolf, J.B. & Hager, R.: Originators of the maternal-offspring coadaptation theory (Wolf & Hager, 2006).
• Spencer, H.G. & Clark, A.G.: Originators of the “Chip Off the Old Block” theory (Spencer & Clark, 2006) and authors of a companion review on non-conflict theories (Spencer & Clark, 2014, same journal issue).

Limitations (as stated by authors)

• The kinship theory’s parentally antagonistic selection pressure “exists only in hypothesis and has never been demonstrated empirically because of the difficulty of measuring inclusive fitness.”
• The complete knockout of imprinted genes does not provide the subtlety needed for examining fitness effects of expression level; more nuanced manipulations are required.
• For the maternal-offspring coadaptation theory, imprinted genes are expressed in the earliest tissues of both seed plants and mammals, before experimental cross-fostering is possible, posing a significant empirical challenge.
• All theories face uncertainty about how the acquisition of imprinted expression affects subsequent gene evolution, which influences predictions about the likelihood of reversion to biallelic expression.
• It is not known how dosage sensitivity evolves at a locus after it acquires imprinted expression — present dosage sensitivity does not necessarily indicate past dosage sensitivity.
• Deep knowledge of imprinting is available for only a few taxa (mouse, human, Arabidopsis, maize) and a few tissues (brain, placenta, endosperm, early-stage embryos), limiting comparative tests.
• Experimental evolution approaches are challenging because mammals (where most is known about imprinting) are unsuitable due to long generation times, and the strength of parental antagonism will tend to covary with the strength of sexual antagonism, making it necessary to derive fine-grained predictions.

Relevance to Clonal Evolution

This paper reviews why genomic imprinting evolved, which is directly relevant to understanding why imprinted genes are enriched for growth-regulatory functions and why loss of imprinting (LOI) at these loci is disproportionately associated with cancer. Three specific connections to clonal evolution research:

(a) Imprinted genes converge on growth regulation. The kinship theory predicts that growth-promoting genes will be paternally expressed and growth-inhibiting genes maternally expressed. The sexual antagonism theory similarly predicts that traits under male-benefit selection (including high growth rate) will show patrigenic expression. The coadaptation theory centers on resource transfer between mother and offspring. All three theories thus predict that imprinted genes will disproportionately affect growth, resource allocation, and cell proliferation — the very pathways whose dysregulation drives clonal expansion. This convergence of evolutionary theories on a growth-regulatory role provides the ultimate-level explanation for why LOI at imprinted loci (e.g., biallelic Igf2 expression via LOI) is oncogenic: it disrupts a gene dosage system that was shaped by organismal-level selection specifically to regulate growth.

(b) Functional haploidy alters mutation-selection dynamics. Imprinting creates effective haploidy at imprinted loci (only one allele is expressed). This means that a single mutational event can produce a full phenotypic effect, without buffering from the second, silenced allele. For biallelically expressed genes, a loss-of-function mutation in one copy is typically recessive and masked by the functional copy; at an imprinted locus, a mutation in the single expressed allele is immediately exposed to selection. This has implications for parameterization of branching-process-model frameworks: imprinted tumor-suppressor genes effectively require only one hit (rather than two) for complete inactivation if the expressed allele is the one mutated, and the selective advantage conferred by mutations at imprinted loci may differ from that at biallelic loci of similar function.

(c) Imprinting as a case study in organismal-level constraints on somatic evolution. The existence of imprinting illustrates how organismal-level selection can create a genomic architecture (monoallelic expression, parent-of-origin-specific silencing) that somatic evolution later exploits. The same molecular machinery that enforces imprinting (DNA methylation, histone modifications) is also co-opted or disrupted during tumorigenesis. The paper’s discussion of the evolutionary stability of imprinting (the kinship theory predicts strong selection against reversion to biallelic expression; the other theories predict lability) suggests that the cancer-relevant imprinted genes may be among the most deeply entrenched in the genome, making their dysregulation both potent and difficult to reverse therapeutically.

Revision history

• 2026-06-26 — Initial source-summary created from Patten et al. (2014) PDF.