Worm Breeder's Gazette 17(2): 24 (April 1, 2002)
These abstracts should not be cited in bibliographies. Material contained herein should be treated as personal communication and should be cited as such only with the consent of the author.
Max Planck Inst. for Medical Research, Jahnstrasse 29, Heidelberg 69120 Germany
Since the identification of smg-4, questions of how the domains of the protein may function to mediate mRNA surveillance have remained. In spite of the beautiful results with the human orthologues, hUpf3p and hUpf3-X, showing these nucleocytoplasmic shuttling proteins bind with Y14 to mRNAs 20nt upstream of exon-exon boundaries [Kim,V. N. et al. (2001)Science 293: 1832-1836, and Lykke-Andersen, J. et al. (2001)Science 293: 1836-1839], and in an RNP in association with CBP80 help mediate a 'pioneer' round of translation [Ishigaki Y et al. (2001)Cell 106:607-617], the domains responsible for these activities are not yet fully elucidated. Although overall sequence identity between the worm and human proteins is only 20%, there is 64% sequence similarity over one 336 a.a. stretch, and many potential functional domains are highly conserved. Interestingly, part of the most conserved domain (Aronoff, R. et al. (2001) Gene 268:153-164), similar to an RNP-1 -like RNA binding motif and essential for rescue of smg-4 mutant animals, is spliced out from one variant, hUpf3delta, of the two human genes. Complex regulation of the encoded protein products is probable.
The proline rich carboxy-terminal domain of SMG-4 is particularly intriguing, with its alternatively spliced extension and some similarity to a regulatory domain of MAP-4, containing hallmarks of a substrate for phosphorylation. The recent report of a proline rich domain responsible for nuclear export that is regulated by phosphorylation [Catez et al. (2002) MCB 22: 1126-1139], caused me to reexamine the SMG-4 sequence. It is clear that sequence identity (although only 15.5% overall) is shared between the two proteins (see Figure), and conserved elements include potential MAP Kinase sites. Perhaps when smg-4 is phosphorylated in this domain, it will be actively exported along with its associated RNP to the cytoplasm. Whether the virus coopts the mRNA surveillance machinery in its lifecycle is a new question raised by this apparent similarity.
Figure: Alignment of HSV-1 US11 protein and SMG-4 protein.
If the serine at position 129 of the US11 sequence is mutated to phenylalanine the protein is retained in nucleoli.