In neurons, not all introns are left on the nuclear cutting-room floor. According to a paper in the March 10 Neuron, some rat introns that survive nuclear editing direct mRNAs toward neuronal dendrites, where the RNAs fully mature. These intronic tags appear to have evolved from retrotransposons (so-called “jumping genes”) sprinkled liberally throughout the genome. The researchers suspect that human cells have also co-opted retrotransposons for RNA-directing purposes, and that a similar mechanism could work to traffic immature transcripts to many cellular locations.

An RNA’s function depends not only on its sequence, but also on other factors: The transcript must be in the right place at the right time and at the right concentration, said senior author James Eberwine of the University of Pennsylvania in Philadelphia. Transporting an mRNA, rather than a ready-made protein, to a specific place means the cell gets more protein bang for its trafficking buck, he added: One mRNA can make hundreds of proteins on location. Eberwine led the work with co-senior author Junhyong Kim and co-first authors Peter Buckley and Miler Lee, who has since moved to Yale University in New Haven, Connecticut. Eberwine’s lab works on RNAs in dendrites, but he suspects the mechanism of intron-directed targeting will be widespread. Although the current work did not address neurodegenerative disease, the authors speculated that disruption of normal mRNA targeting could lead to neural problems. Both anterograde and retrograde transport problems are linked to various neurological disorders (see ARF related news story).

While RNA splicing occurs mainly in the nucleus, Eberwine’s group previously noticed that some RNAs undergo splicing in dendrites (Glanzer et al., 2005). More recently, the group found that the mRNA for a calcium channel subunit sometimes retains introns in the cytoplasm, and that these introns are important for regulating translation of that mRNA in the dendrites and promoting normal neural firing (Bell et al., 2008 and Bell et al., 2010). In the current study, Buckley and colleagues hunted for other dendritic mRNAs that might also be regulated by introns.

The researchers severed dendrites from the cell bodies of primary rat hippocampal neurons and amplified the RNA therein. They used both microarrays and high-throughput sequencing to identify introns present in the dendritic RNAs. They discovered several transcripts that still had introns, including those for fragile X mental retardation protein (FMRP) and the calcium/calmodulin-dependent protein kinase II β (CAMK2β).

The scientists analyzed the intron sequences, looking for similarities that might indicate a directional tag. They discovered several possibilities, including miRNA complementary sequences that might regulate the cytoplasmic splicing process. In this study, they chose to focus on ID elements, common intronic sequences already implicated in dendrite targeting (Muslimov et al., 1997). ID elements are a rodent class of retrotransposon, or mobile RNA element, all derived from the parent gene BC1. BC1 generates a non-coding RNA that targets dendrites (Kim et al., 1994). Over time, the retrotransposons were copied and reinserted into the genome thousands of times through the activity of reverse transcriptase. ID elements are approximately 74 base pairs long, and some form a hairpin structure that the researchers suspected might be involved in dendrite homing.

To test their hypothesis, the researchers engineered artificial green fluorescent protein (GFP) constructs with an ID element artificially tacked onto the 5’ end. They transfected these plasmids into primary rat hippocampal neurons and used in situ hybridization to locate the transcripts. While wild-type GFP RNA was mostly perinuclear, the version with the ID element extended out into the dendrites, confirming that the ID element was a dendrite-specific localization sequence. Versions with mutations in the RNA hairpin were less able to reach the dendrites.

Although the researchers have not yet identified the other elements of the ribonucleoprotein complex that presumably transports ID-containing transcripts to dendrites, they suspected that the cell must have a finite supply of the machinery. They showed that when they transfected their ID-GFP constructs into primary hippocampal neurons, it blocked dendrite targeting of the endogenous transcripts CAMK2β and FMR1—confirming that the transfected gene was using up all the shared dendrite-targeting tools. Further, transfection of the GFP construct containing the ID domain from FMR1 prevented normal dendritic localization of the protein FMRP—thus, altered mRNA trafficking had downstream consequences for the cell.

Eberwine and colleagues hypothesize that rodent neurons co-opted the dendrite-targeting ability of BC1 as it hopped into different genes. Being able to direct mRNA traffic could be an evolutionary advantage, Buckley suggested. The cell leaves the ID-containing introns unspliced in the nucleus, and those introns direct the transcript to the dendrites. The final steps to translation might even depend on electrical signals reaching the dendritic spines, suggested Gregor Sutcliffe at The Scripps Research Institute in La Jolla, California, who was not involved with the study. In that way, the incoming action potential might signal protein synthesis, thus strengthening synapses that are used frequently and contributing to memory. Indeed, ID elements were first discovered as common identifiers of brain-specific RNAs (Sutcliffe, 1982). Further, Sutcliffe noted that both BC1 and FMRP are components of RNA-toting complexes that cruise to the ends of dendrites.

Although ID elements are specific to rodents, humans have a similar set of retrotransposons. Called Alu elements, they are derived from the original gene BC200. Based on sequence, BC200 is recognizable as a BC1 cousin, and it also interacts with FMRP, Sutcliffe noted. The researchers suspect retrotransposons such as Alu elements could be involved in RNA localization in people. They also posit that different retrotransposons could drag mRNA to different addresses. “I think this is going to be a general cellular phenomenon,” Eberwine said.

Retrotransposons are enriched in the brain, where they may contribute to the diversity of neural types (see ARF related news story on Coufal et al., 2009). Mechanisms such as ID elements, splice variants, and microRNAs, which apparently serve to diversify gene expression, have been key in the evolution of the vertebrate brain, wrote Clive Bramham of the University of Bergen in Norway in an e-mail to ARF. “This makes our neurons sophisticated and adaptable, and there is good reason to believe this is important for higher cognitive functions,” he wrote. However, Bramham cautioned that the current study only addressed mRNA targeting in embryonic cells. “It remains to be seen whether this extends into adulthood,” he wrote.

In fact, some evidence suggests ID elements do not act as dendrite-targeting tags in adult rodents. Tasneem Khanam and colleagues at the University of Münster, Germany, found no evidence that ID sequences work that way in transgenic mice (Khanam et al., 2007). Eberwine and colleagues suggest the in vivo tagging did not work because Khanam put the ID elements in the 3’ untranslated regions (UTRs) of the genes. In an e-mail to ARF, Khanam questioned that explanation because most dendritic targeting elements, she wrote, appear in 3’ UTRs. However, it is difficult to compare mice and rats—rats have some 150,000 examples of ID elements, the study authors wrote, while mice have fewer than 1,000. Khanam also added, “Experimental evidence is lacking to show that deletion of ID elements would render the transcripts non-dendritic.”

RNA splicing factors have become a hot area for researchers studying neurodegeneration. Two genes associated with amyotrophic lateral sclerosis—TDP-43 and FUS—regulate splicing (see ARF related news story on Kwiatkowski et al., 2009 and Vance et al., 2009), and alternate splicing of tau is thought to play a role in frontotemporal dementia. Given the current results, Eberwine hypothesized that altered splicing could affect mRNA and protein localization, too. Scientists further found that mouse models of the neurodevelopmental disorder Rett syndrome are more likely to have a different retrotransposon, L1, jump around in their genome (Muotri et al., 2010).—Amber Dance


  1. This article in Neuron published by Buckley et al. describes findings in explanted and cultured neurons (ex vivo) that intron-retained, repetitive sequences, so-called identifier elements (ID elements), confer dendritic targeting competence to neuronal transcripts, partially in agreement with previous ex-vivo findings by Muslimov et al. (Muslimov et al., 1997). However, the in-vivo findings by our group using transgenic mouse models do not support the idea that ID elements could act as dendritic targeting elements for mRNAs (Khanam et al., 2007). The authors have claimed that the position of the ID elements in a transcript is crucial for such a phenomenon, and that the location of the ID elements in 3’ UTR is not ideal. This is quite intriguing, as most of the dendritc targeting elements identified so far are located in 3’ UTRs.

    It is noteworthy that ID elements are restricted to the mammalian order of rodentia (rodents). The question of whether a similar mechanism, if it exists at all, is in place in primates, including humans, is difficult to address.

    Transposed elements, including short interspersed elements (SINEs), have the potential to impart novel functions to existing genes including regulatory elements or novel exons (Brosius, 1991). Splice variants can have altered localization and/or expression and functional consequence. Splice variants are presumably one of the many potential causes in genetic disease, possibly including Alzheimer’s and Parkinson’s disease. On the other hand, it is not clear if the relevant mRNAs (APP and tau) are transported into dendrites, as the RNA localization data in the Allen Human Brain Atlas do not reveal dendritic staining.

    While the idea is interesting that the intron retained sequences impart dendritic targeting competence, the data presented do not fully support the hypothesis. For example, experimental evidence is lacking to show that deletion of ID elements would render the transcripts non-dendritic. If the hypothesis is correct, then it would raise the question as to how dendritic transport and its regulation is achieved in non-rodent species, where clearly it must also occur for a functional mammalian nervous system. There is no evidence so far that the 5’ Alu domain retrosposon found in the human BC200 gene has any role in the transport of BC200 RNA.


    . RNA transport in dendrites: a cis-acting targeting element is contained within neuronal BC1 RNA. J Neurosci. 1997 Jun 15;17(12):4722-33. PubMed.

    . Can ID repetitive elements serve as cis-acting dendritic targeting elements? An in vivo study. PLoS One. 2007;2(9):e961. PubMed.

    . Retroposons--seeds of evolution. Science. 1991 Feb 15;251(4995):753. PubMed.

    View all comments by Tasneem Khanam

Make a Comment

To make a comment you must login or register.


News Citations

  1. Chicago: Axonal Transport Not So Fast in Neurodegenerative Disease
  2. Bottom LINE—Retrotransposons Sway Human Neural Progenitors?
  3. New Gene for ALS: RNA Regulation May Be Common Culprit

Paper Citations

  1. . RNA splicing capability of live neuronal dendrites. Proc Natl Acad Sci U S A. 2005 Nov 15;102(46):16859-64. PubMed.
  2. . RNA transport in dendrites: a cis-acting targeting element is contained within neuronal BC1 RNA. J Neurosci. 1997 Jun 15;17(12):4722-33. PubMed.
  3. . Rodent BC1 RNA gene as a master gene for ID element amplification. Proc Natl Acad Sci U S A. 1994 Apr 26;91(9):3607-11. PubMed.
  4. . Common 82-nucleotide sequence unique to brain RNA. Proc Natl Acad Sci U S A. 1982 Aug;79(16):4942-6. PubMed.
  5. . L1 retrotransposition in human neural progenitor cells. Nature. 2009 Aug 27;460(7259):1127-31. Epub 2009 Aug 5 PubMed.
  6. . Can ID repetitive elements serve as cis-acting dendritic targeting elements? An in vivo study. PLoS One. 2007;2(9):e961. PubMed.
  7. . Mutations in the FUS/TLS gene on chromosome 16 cause familial amyotrophic lateral sclerosis. Science. 2009 Feb 27;323(5918):1205-8. PubMed.
  8. . Mutations in FUS, an RNA processing protein, cause familial amyotrophic lateral sclerosis type 6. Science. 2009 Feb 27;323(5918):1208-11. PubMed.
  9. . L1 retrotransposition in neurons is modulated by MeCP2. Nature. 2010 Nov 18;468(7322):443-6. PubMed.

Further Reading


  1. . Frontotemporal dementia and amyotrophic lateral sclerosis-associated disease protein TDP-43 promotes dendritic branching. Mol Brain. 2009;2:30. PubMed.
  2. . Depletion of TDP 43 overrides the need for exonic and intronic splicing enhancers in the human apoA-II gene. Nucleic Acids Res. 2005;33(18):6000-10. PubMed.
  3. . miR-375 inhibits differentiation of neurites by lowering HuD levels. Mol Cell Biol. 2010 Sep;30(17):4197-210. PubMed.
  4. . The FMR1 gene and fragile X-associated tremor/ataxia syndrome. Am J Med Genet B Neuropsychiatr Genet. 2009 Sep 5;150B(6):782-98. PubMed.

Primary Papers

  1. . Cytoplasmic intron sequence-retaining transcripts can be dendritically targeted via ID element retrotransposons. Neuron. 2011 Mar 10;69(5):877-84. PubMed.