What would Gregor Mendel say? Modern genetics is finding exceptions to the rules of genetic inheritance that bear his name. Paramutation, an epigenetic change in an allele that is induced by another allele and passed to subsequent generations, occurs in maize and other plants. Rassoulzadegan et al. report a paramutation in the mouse and describe a role for RNA in its inheritance in a recent study in Nature.
Kit is a receptor tyrosine kinase important in developmental processes, including germ cell differentiation and melanogenesis. The Kittm1Alf allele contains a 3 kb lacZ-neo cassette inserted downstream of Kit's start codon, replacing all of Kit exon 1 and 200 bp of intron 1. Homozygous mutant mice (Kittm1Alf/tm1Alf) die shortly after birth, whereas heterozygous mutant mice (Kittm1Alf/+) survive and have white tail tips and feet. The authors bred heterozygous Kittm1Alf/+ mice and found an unusual percentage of progeny with white tail tips. Most of the wild-type (Kit+/+) progeny had the mutant phenotype.
The authors showed that this paramutation, designated Kit*, was unlikely to result from standard epigenetic changes, such as DNA or histone methylation. Methylated DNA sequences can be identified by sodium bisulfite mapping because this chemical converts unmodified cytosines to uracils. This technique showed similar levels of DNA methylation in heterozygous, paramutated and wild-type mice. Chromatin immunoprecipitation with antibodies to methylated lysines on histone 3 showed similar levels of histone methylation in the three groups.
The authors analyzed northern blots with a probe to the 5' region of Kit and found a 50% reduction in polyadenylated Kit mRNA in the brains of Kittm1Alf/+ and Kit* relative to Kit+/+ mice. Similar analysis of total RNA showed shorter Kit transcripts in the Kittm1Alf/+ brain and testis, suggesting that mature Kit transcripts may be degraded.
Intrigued by the shortened Kit transcripts in the testis, the authors did transcriptional run-on assays, which measure the incorporation of radioactive nucleotides into gene products in isolated nuclei, and found increased Kit transcription in Kittm1Alf/+ relative to Kit+/+ testes. Quantitative real-time PCR showed robust Kit mRNA expression in Kittm1Alf/+ sperm and nearly undetectable expression of Kit mRNA in Kit+/+ sperm. The authors note that the Kit gene is usually transcriptionally silent in sperm. Surprisingly, Kittm1Alf/+ sperm contained more RNA species than Kit+/+ sperm, as shown by both acridine-orange and EDTA regressive staining.
RNA therefore seemed to be the key to this paramutation. The authors microinjected RNA from Kittm1Alf/+ mice into mouse embryos. Nearly 50% of the microinjected embryos developed white tail tips. The authors also microinjected microRNAs reported to target Kit for degradation and found a high frequency of progeny with the white tail tip phenotype.
These data therefore suggest that in addition to DNA, RNA is a substance of heredity and challenge our definition of the term 'gene.' Although the prevalence of paramutations in the human genome is unknown, their discovery could revolutionize the search for disease-related genes.