A garland of guanine nucleotides, draped in just the right way, penetrates motor neurons and triggers the assembly of neuroprotective stress granules, according to a paper in the November 17 Proceedings of the National Academy of Sciences USA online. Such quadruplexes are normally made of fragmented transfer RNAs, cleaved in times of cellular distress. However, the authors, from Brigham and Women’s Hospital in Boston, found that they could substitute much more stable DNA versions to protect neurons in culture. These DNA quadruplexes might make treatments for neurodegeneration, suggested senior author Paul Anderson. They also seem to interact with another kind of G-quadruplex that inhibits the production of stress granules, those made by RNA repeats from the ALS and FTD gene C9ORF72.


Certain guanine-rich sequences can assemble into G-quartets. Stacks of these configurations are called G-quadruplexes. [Image courtesy of Xiang-Lei Yang, PNAS.]

Joint first authors Pavel Ivanov at Brigham's and Elizabeth O’Day of Boston Children’s Hospital found these G-quadruplexes when investigating the neuroprotective mechanism of angiogenin. Though angiogenin got its moniker for its role in promoting growth of blood vessels, it is also a ribonuclease that activates a stress response pathway. It targets transfer RNAs, cleaving them into fragments called transfer RNA-derived, stress-induced RNAs (tiRNAs). Two of these, the tiRNAs generated from the transfer RNAs for cysteine and alanine, work with the translational repressor Y-box binding protein 1 (YB-1) to halt protein synthesis. The shutdown of translation kicks off the formation of stress granules (Ivanov et al., 2011). These granules sequester mRNA and the ribosomal machinery during times of cellular stress, such as starvation. When conditions improve, the granules dissolve and the cell goes back to business as usual. 

Angiogenin treatment is neuroprotective (see Jul 2011 news story). Moreover, mutations that impede its ribonuclease activity have been reported to cause amyotrophic lateral sclerosis (Greenway et al., 2006). However, angiogenin would make a poor treatment, said Ivanov, because of side effects such as blood vessel growth. The researchers theorized that tiRNAs might do the job without so many side effects, but RNAs are notoriously unstable. Therefore, the authors made DNA analogs, tiDNAs, and compared them to the tiRNAs.

First, the scientists examined the structure of the tiRNAs and tiDNAs. The cysteine and alanine tiRNAs contain guanine-rich sections absent from other tiRNAs. This led the authors to hypothesize that they might form G-quadruplexes. They confirmed this with the dye N-methyl mesoporphyrin IX, which fluoresces upon binding to these structures. The tiDNAs also made G-quadruplexes, looking like a good stand-in for the tiRNAs.

Would the tiDNAs initiate stress granule formation? To find out, the researchers transfected a human bone cell line, U2OS, with the tiDNAs, tiRNA, or control nucleic acids they did not expect to affect the cell stress response. In cells treated with the control DNA, fewer than 5 percent of cells developed stress granules. Nearly 15 percent of the tiRNA-transfected cells made stress granules, and as did almost 20 percent of the tiDNA-treated cells.

The assembly of stress granules did not necessarily indicate that the tiDNAs would be neuroprotective. To check this, the authors treated human motor neuron cultures with three different stressors: the glutamate mimic AMPA, sodium arsenite, or hydrogen peroxide. They quantified cell survival by the presence of ATP. Each stressor hampered cell viability, but treatment with tiDNA G-quadruplexes protected the cells in each case.

Ivanov and colleagues concluded that tiDNAs can do the same job as tiRNAs, however, there was one more issue. In a therapeutic setting, they could not use transfection agents to get the tiDNAs into the cells that needed them. Similar G-quadruplexes enter tumor cells of their own accord, but it was not known if they could do the same in non-cancerous cells. Ivanov found that the tiDNAs could, in fact, spontaneously get into both the U2OS cells and human motor neurons.

“The [neurodegeneration] field has not really paid attention to these tiRNAs,” noted Ben Wolozin of Boston University, who was not involved in the work. “This paper brings up a potential therapeutic … for ALS, and perhaps for other [diseases] like stroke or Parkinson’s.”

G Does Not Always Stand for Good
Ivanov and colleagues initially focused on the G-quadruplexes made by tiRNAs, but lately ALS researchers have been enthralled by another RNA that makes G-quadruplexes: the C9ORF72 expansion. The first intron of this gene normally contains a small handful of repeats of the sequence GGGGCC, but in many people with inherited ALS, the number of repeats skyrockets into the hundreds or thousands. The repeated RNA forms a G-quadruplex in vitro, and researchers suspect this same shape may populate abnormal RNA foci in the neurons of people with C9ORF72 mutations (see Nov 2013 news story). Might these G-quadruplexes also influence assembly of stress granules?

Indeed they did. Ivanov co-expressed C9ORF72 repeat oligomers with alanine and cysteine tiDNAs. When cells contained tiDNA and four C9ORF72 repeats, about 15 percent of them made stress granules. That number dropped to half when there were 23 GGGGCC repeats. The extended repeat inhibited stress granule assembly, the authors concluded. They do not know how this might occur, but speculated the C9ORF72 repeats might sequester YB-1 or other RNA-binding proteins needed for stress granule formation and motor neuron survival.

What does that mean for the neurons of people with C9ORF72 expansions? It might mean they cannot properly assemble tiRNA-based stress granules—or it might mean nothing at all, Ivanov said. As yet, the authors have no evidence that the C9ORF72 repeats interact with the tiRNA pathway in human cases. Davide Trotti of Thomas Jefferson University in Philadelphia, who was not involved in the study, thought the data were solid but that it would be a “long stretch” to link the tiRNA stress response to C9ORF72-based ALS on the basis of this single experiment. He said he would like to see evidence of the interaction in a cell type more relevant to ALS, such as neurons, and with longer expansions. Ivanov said those experiments are underway.

If further studies prove that the GGGGCC repeats and tiRNAs or tiDNAs interact, it could have implications for the treatment of C9ORF72-based ALS, Xiang-Lei Yang of The Scripps Research Institute in La Jolla, California, suggested in an editorial accompanying the paper. “It would be interesting to test if the G-quadruplex-containing tiRNAs/tiDNAs could also be used to suppress the toxicity of the pathological GGGGCC repeats,” she wrote.—Amber Dance


  1. The work by Anderson and co-workers really is an exciting and highly interesting finding. We have previously shown that angiogenin is a protective factor for motor neurons (Kieran et al., 2008), so this paper provides new insights into potential mechanisms of action. We also showed that angiogenin is stress-induced and secreted by motor neurons during stress conditions, and then taken up by astrocytes (or endothelial cells). Angiogenin uptake then induces tRNA cleavage in astrocytes (Skorupa et al., 2012)—we proposed that this alters the translational output of astrocytes. Now, the work by Anderson and co-workers suggests another possibility—that there is a paracrine feed-back to motor neurons via G4 secretion and uptake.

    On a more cautionary note, some ALS-associated angiogenin mutations have no, or very little loss of, RNase activity, suggesting that the tRNA cleavage is not the only factor involved in angiogenin neuroprotection (or potentially that it is not involved at all).


    . Control of motoneuron survival by angiogenin. J Neurosci. 2008 Dec 24;28(52):14056-61. PubMed.

    . Motoneurons secrete angiogenin to induce RNA cleavage in astroglia. J Neurosci. 2012 Apr 11;32(15):5024-38. PubMed.

    View all comments by Jochen Prehn

Make a Comment

To make a comment you must login or register.


News Citations

  1. Trichostatin A, Angiogenin Keep Neurons Alive in ALS
  2. Researchers Revel in C9ORF72 Advances at RNA Symposium

Paper Citations

  1. . Angiogenin-induced tRNA fragments inhibit translation initiation. Mol Cell. 2011 Aug 19;43(4):613-23. PubMed.
  2. . ANG mutations segregate with familial and 'sporadic' amyotrophic lateral sclerosis. Nat Genet. 2006 Apr;38(4):411-3. PubMed.

Further Reading


  1. . Pathological stress granules in Alzheimer's disease. Brain Res. 2014 Oct 10;1584:52-8. Epub 2014 Aug 7 PubMed.
  2. . RNA metabolism in ALS: When normal processes become pathological. Amyotroph Lateral Scler Frontotemporal Degener. 2014 Sep;15(5-6):321-336. Epub 2014 Feb 20 PubMed.
  3. . Neurodegenerative diseases: G-quadruplex poses quadruple threat. Nature. 2014 Mar 13;507(7491):175-7. Epub 2014 Mar 5 PubMed.
  4. . C9orf72 nucleotide repeat structures initiate molecular cascades of disease. Nature. 2014 Mar 13;507(7491):195-200. Epub 2014 Mar 5 PubMed.
  5. . TMPyP4 porphyrin distorts RNA G-quadruplex structures of the disease-associated r(GGGGCC)n repeat of the C9orf72 gene and blocks interaction of RNA-binding proteins. J Biol Chem. 2014 Feb 21;289(8):4653-9. Epub 2013 Dec 26 PubMed.
  6. . The disease-associated r(GGGGCC)n repeat from the C9orf72 gene forms tract length-dependent uni- and multimolecular RNA G-quadruplex structures. J Biol Chem. 2013 Apr 5;288(14):9860-6. PubMed.

Primary Papers

  1. . G-quadruplex structures contribute to the neuroprotective effects of angiogenin-induced tRNA fragments. Proc Natl Acad Sci U S A. 2014 Dec 23;111(51):18201-6. Epub 2014 Nov 17 PubMed.
  2. . tRNA-derived G-quadruplex protects motor neurons. Proc Natl Acad Sci U S A. 2014 Dec 23;111(51):18108-9. Epub 2014 Dec 15 PubMed.