Todd A. Gray

The putatively functional Mkrn1-p1 pseudogene is neither expressed nor imprinted, nor does it regulate its source gene in trans

A recently promoted genome evolution model posits that mammalian pseudogenes can regulate their founding source genes, and it thereby ascribes an important function to “junk DNA.” This model arose from analysis of a serendipitous mouse mutant in which a transgene insertion/deletion caused severe polycystic kidney disease and osteogenesis imperfecta with ≈80% perinatal lethality, when inherited paternally [Hirotsune, S., et al. (2003) Nature 423, 91–96]. The authors concluded that the transgene reduced the expression of a nearby transcribed and imprinted pseudogene, Mkrn1-p1. This reduction in chromosome 5-imprinted Mkrn1-p1 transcripts was proposed to destabilize the cognate chromosome 6 Mkrn1 source gene mRNA, with a partial reduction in one Mkrn1 isoform leading to the imprinted phenotype. Here, we show that 5’ Mkrn1-p1 is fully methylated on both alleles, a pattern indicative of silenced chromatin, and that Mkrn1-p1 is not transcribed and therefore cannot stabilize Mkrn1 transcripts in trans. A small, truncated, rodent-specific Mkrn1 transcript explains the product erroneously attributed to Mkrn1-p1. Additionally, Mkrn1 expression is not imprinted, and 5’ Mkrn1 is fully unmethylated. Finally, mice in which Mkrn1 has been directly disrupted show none of the phenotypes attributed to a partial reduction of Mkrn1. These data contradict the previous suggestions that Mkrn1-p1 is imprinted, and that either it or its source Mkrn1 gene relates to the original imprinted transgene phenotype. This study invalidates the data upon which the pseudogene trans-regulation model is based and therefore strongly supports the view that mammalian pseudogenes are evolutionary relics.

Structure and function correlations at the imprinted mouse Snrpn locus.

Abstract. The human SNRPN gene maps within Chromosome (Chr) 15q11-q13, the region responsible for Prader-Willi syndrome (PWS) and Angelman syndrome (AS). As one of several 15q11-q13 transcripts expressed from the paternal allele-only, SNRPN is a candidate gene to explain at least some of the PWS phenotype in human and in genetic mouse models. The promoter and first exon of the SNRPN gene also correspond to an imprinting center element responsible for resetting of the maternal to paternal imprints within 15q11-q13 during spermatogenesis. Through characterization of the imprinted murine Snrpn locus in mouse Chr 7C, we have found that the gene structure is very similar to the human, with ten conserved exons spanning 22 kb, the last seven of which are tightly clustered. The promoter of Snrpn is differentially methylated in ES cells and adult tissues, supporting a role for DNA methylation at this site in somatic establishment and/or maintenance of Snrpn imprinting. The first intron of the mouse and human genes contains structurally conserved G-rich clustered repeats which may play a role in establishing DNA methylation patterns associated with imprinting of this gene. On the basis of the conserved structural and imprinted features of the human SNRPN and mouse Snrpn genes, we suggest that imprinting mechanisms are conserved between human and mouse.

Genetic abnormalities in Prader‐Willi syndrome and lessons from mouse models

Prader‐Willi syndrome is a multigenic disorder with developmental and neurobehavioural abnormalities. There are multiple genetic causes, although all ultimately involve the loss of paternally derived gene expression of chromosome region 15q11‐q13. Multiple imprinted genes expressed only from the paternal allele have been identified in the specific region of human chromosome 15q associated with Prader‐Willi syndrome and in the syntenic mouse chromosome 7C region, including a novel polycistronic gene (SNURF‐SNRPN) that encodes two independent proteins. The latter genetic locus may play a key role in Prader‐Willi syndrome and the evolution of imprinting in this domain, because it is uniquely involved with mutations in the imprinting process and balanced translocations in this syndrome. Indeed, based on the co‐localization of SNURF and SNRPN within the imprinting control region critical to Prader‐Willi syndrome, evolutionary arguments would suggest that this genetic locus is a prime candidate for mutations producing the failure‐to‐thrive phenotype of neonates with this syndrome and of corresponding mouse models. Hence the SNURF‐SNRPN gene may encode a paternally derived postnatal growth factor.