A Subclass of mRNAs Use a Different Nuclear Export Machine

It's been a while since I've written about mRNA and mRNA export. There has been lots of CPEB papers (cytoplasmic polyadenylation element binding protein), but nothing fundamentally new at the molecular level. As for mRNA export, the Reed lab will have a big paper out soon, and when that comes out I'll write a blog about how each step of mRNA metabolism is thought to promote the next step in mRNA metabolism (so that mRNA transcription promotes splicing which in turn promotes nuclear export.)

But for now lets turn our attention to a new JCB paper that claims that a subset of mRNAs, which encode for proteins important for the cell cycle, have their own nuclear export pathway.

First you may ask, what is the current model for nuclear export?

As pre-mRNA is transcribed by Pol II (as in RNA polymerase II), the TREX complex (TREX = Translation Export) is loaded onto the growing mRNA. This and the act of splicing recruits the protein Aly to the mRNA. Aly then in turn recruits the TAP/p15 heterodimer (sometimes TAP is referred to as NXF1) which allows the mRNA to interact with the the Nuclear Pore Complex and thus catalyzes nuclear export. Now obviously mRNA export must be powered by an energy utilizing process. Enter Dbp5, an RNA helicase that is situated in the NPC. Helicases are enzymes that use ATP derived energy to walk along DNA or RNA chains (conceptually similar to a myosin motor using ATP to walk along an actin filament). Dbp5 is thought to pull the mRNA out of the nucleus and the whole thing may be regulated by IP6 (read this entry about IP6 and mRNA export).

But is this "canonical mRNA export pathway" true for all mRNAs? Well the Silver lab had published a paper which demonstrated that Yra1 and Mex67 (the yeast versions of Aly and TAP) only bind to a subset of all yeast mRNA transcripts, indicating that other mRNA transcripts may recruit alternative mRNA export factors.

Another little bit of background. mRNAs get modified with a cap. In the nucleus this small modification (Gppp-) recruits the cap binding complex (CBP20/CBP80) to the beginning of the mRNA. Eventually the CBP20/CBP80 heterodimer is replaced with eIF4e, (eukaryotic translation initiation factor 4e) which plays a part in recruiting the ribosome to the mRNA in the cytoplasm.

Now the new paper:

Culjkovic and colleagues found that eIF4e is recruited to a subset of mRNAs in the nucleus (presumably eIF4e binds to most mRNAs in the cytoplasm). The cytoplasmic/nuclear ratio of these mRNAs was dependent on eIF4e suggesting that eIF4e promoted their export. The subset of mRNAs all contain a stem loop pair that could enhance eIF4e binding and localized the mRNA to eIF4e loci within the nucleus. Interestingly all the mRNAs that are affected by eIF4e encode proteins that are important for the cell cycle (some examples are various cyclin transcripts, and the myc transcript). Importantly, these mRNAs could still be exported in cells that had reduced levels of TAP (by RNAi ... incidentally the authors never quantified whether TAP was knocked down, let alone by how much, I hate that!) but were affected by the exportin CRM1. So while most RNAs use the TAP pathway to get out, certain RNAs use the CRM1 export machinery and this may be catalyzed by eIF4e. Don't forget that CRM1 is part of the Ran nuclear export/import pathway that gets proteins, ribosomal RNA and other things out of the nucleus. CRM1 and Ran were once thought to mediate RNA export, however about 10 years ago the mRNA export did a 180 degree turn and found that the export of most mRNAs did not use this pathway. It's funny that a subset of mRNAs are indeed relying on CRM1 for export.

Conceptually this paper demonstrates that subclasses of mRNAs may be regulated separately by regulating their spatial distribution (i.e. their nuclear/cytoplasmic location). This differentiation within mRNA export gives the cell an extra layer of translational control. Thus cells can regulate and modulate the activity of all cyclin mRNAs, but not other mRNAs, by regulating nuclear eIF4e levels.

One note: the paper measures export of mRNA by measuring the cytoplasmic/nuclear ratio of mRNA. This should reflect export HOWEVER it may reflect other aspects of mRNA metabolism, such as mRNA stability or mRNA processing. I beleive that a lot of information about mRNA transport is yet to be revealed and that this will require RNA researchers to monitor mRNA dynamics.

eIF4E is a central node of an RNA regulon that governs cellular proliferation
Biljana Culjkovic, Ivan Topisirovic, Lucy Skrabanek, Melisa Ruiz-Gutierrez, and Katherine L.B. Borden
J. Cell Biol. (06) 175:415-426

Haley Hieronymus & Pamela A. Silver
Genome-wide analysis of RNA-protein interactions illustrates specificity of the mRNA export machinery
Nature Genetics (03) 33:155-161